For Math Guru 

Sports
Economics
APPLE USER VERSION BETA (PREVIOUSLY PUBLISHED BY PEARSON/PRENTICE HALL)
RODNEY FORT

SPORTS ECONOMICS
Preface

For the Student
WELCOME TO THE REVOLUTION!
This is the first sports economics textbook to completely
embrace electronic presentation. I hereby dub it Sports
Economics 1.0. The decision was carefully considered given
the traditional print alternatives. In my humble opinion
(teaching and writing now for over 25 years in the area), this
is the way to go.
The disadvantage first. No printed pages and some students
are geared that way in their study approach. The advantage?
No printed pages! So the cost is decidedly lower and my
flexibility as a student of sports to bring you the very best and
most recent advances is now a reality. No more waiting years
for new print editions.
Given that standardization is still the battleground, there is
one version for iPad users and one version for the Kindle
community. But I have managed to make the two versions
nearly identical and all page number references work with
either version. The Kindle users will find the tables and
figures for each chapter at the end on unnumbered pages.
Later, under “Special Features”, I list more added abilities.
INTENDED AUDIENCE
This book is intended primarily for upper-division
undergraduate classes aimed at a general audience (usually
composed of business and sports management majors) or for
a specialized seminar for economics majors. It also provides
the basics required for a graduate-level treatment of the
subject, and the chapter references provide a good starting
point for graduate students on any team sport topic. Students
using this text should have an understanding of the principles
of microeconomics.
These days, the business side of the playing field dominates
the sports pages at times. Players, owners, local elected
officials, and fans all seem at odds. This is a confusing state of
affairs because Americans appear to have a nearly insatiable
demand for sports, and there appears to be more than enough
money to go around. This book aims to clear up this messy
topic by applying a dose of economic thinking to the business
of sports. Sports truly are business, and off-field economic
decisions determine on-field outcomes.
NEW TO THIS EDITION
My first aim was to make sure that the book transported
textually intact. So the highlights from the last print edition
remain for this first electronic edition. But more to come on
this in “Special Features” below.
Highlights of the last print edition included the following:
ii

• The extensive data from the second edition are all updated.
(I’m working on the next update of all of the data for the
next version.)
• All dollar comparisons now show both the original nominal
amounts and 2009 dollars, adjusted by the consumer price
index.
• Extended coverage of the relevant professional research for
each chapter.
• New and updated Learning Highlights add significantly to
the discussions. And some previous highlights are updated
to reflect evolving circumstances. Examples include
Performance Enhancement (Chapter 2); Malls, MLB, and
Disney (Chapter 10); and The Arms Race in College Sports
Spending (Chapter 13).
Additional updates for the last print edition included the
following:
• Chapter 1: Rottenberg’s two main ideas—the uncertainty of
outcome hypothesis and the invariance principle—are
consolidated into a presentation in Chapter 1. This sets the
stage from the outset for repeated references to his
fundamental contributions throughout the text.
• Chapter 2: A new section on performance enhancement and
a new Learning Highlight on the subject.
• Chapter 3: The role of electronic media is introduced for the
first time.
• Chapter 4: The more modern case of the Seattle Supersonics
replaces the old case of the San Antonio Spurs.
• Chapter 5: For the current economic situation, there is
coverage of team and league responses to trying economic
times.
• Chapter 6: Extended coverage of (1) local revenue sharing to
cover the NBA and NHL and (2) the analysis of cap
effectiveness to show how distributional consequences affect
league choices.
• Chapter 7: Revised the section on International Issues to
include the hot loonie and international talent migration to
MLB and other leagues.
• Chapter 9: The logic of the NHL lockout, 2004–2005, is
added.
• Chapter 10: The entire subsidy issue is cast in the broader,
more well-known context of all benefits and costs—external
benefits and external costs receive treatment under those
headings explicitly.
SPECIAL FEATURES
Now for the fun part!
As with the last print version, these features will facilitate
student learning and understanding:
• Learning Objectives at the start of each chapter preview key
content.
iii

• Did You Know? margin facts help pique and retain student
interest. [I’ve eliminated the “Pull Quotes”; never liked them
in the first place.]
• Learning Highlights at the end of some sections present
interesting and detailed examples and information relevant
to the text material. These are intended to be informative
and entertaining, often concerning sports business
personalities and economic explanations of their actions.
Three levels of end-of-chapter questions:
• Review Questions are memory joggers that remind students
of the important elements of the readings.
• Thought Questions require students to work through the
chapter content that requires higher-level thinking.
• Advanced Questions push students to actually apply sports
economics rather than just think about it.
But here is how the electronic version now outshines anything
that could be done in print:
• Navigation hints and other notes are just a tap away.
• Learning objectives are now linked via bookmarks to
relevant sections throughout the text. UNDER
CONSTRUCTION.
• “Did You Know” (DYK?) are now are popovers and no longer
distract from the text readings (just tap).
• Hints are now available as pop-overs as well, including
reference back to the relevant areas in the text that you
might wish to review. UNDER CONSTRUCTION.
• Tables and Figures take advantage of popovers as well; no
more flipping back and forth from reference to the content
of Tables and Figures; just tap.
And there is more! You will surely notice that there is no
longer an index. But why should there be? You may now just
use the search device in your reader to find any string of text
or topic of your choosing. Cooler and faster than thumbing
through an index and then thumbing through to the relevant
page, yes? (The glossary is coming; again, I needed to make
sure you had the book first, then the trimmings.)
FEEDBACK
If you have any questions related to this edition don’t hesitate
to contact me (this is the format for links; just tap and off you
go!).
iv

mailto:rodfort@umich.edu?subject=Sports%20Economics%20Feedback

mailto:rodfort@umich.edu?subject=Sports%20Economics%20Feedback

For Instructors
WELCOME TO THE REVOLUTION!
This is the first sports economics textbook to completely
embrace electronic presentation. I hereby dub it Sports
Economics BETA Version. The decision was carefully
considered given the traditional print alternatives. In my
humble opinion (teaching and writing now for over 25 years
in the area), this is the way to go.
I hope you will read “For the Student” and I won’t repeat what
is there already.
The disadvantage for you first. The pagination of course is all
busted from the print version. But that is the only
disadvantage I can think of from my perspective as a teacher.
Given that standardization is still the battleground, there is
one version for iPad users and one version for the Kindle
community. But I have managed to make the two versions
nearly identical and all page number references are the same
for either version. The Kindle version is just a PDF version of
the iPad version, and putting the tables and figures at the end
on unnumbered pages is the only difference between the two.
You will notice the cost is decidedly lower. When I broke with
Prentice Hall, they intimated that a 4th print edition would
have run in the neighborhood of $400. Incredible and
avoidable.
The most important advantage to the electronic approach is
that I can now bring you and your students the very best and
most recent advances that I discover in sports economics. No
more waiting years for new print editions.
I know there may be some cringing since there is the
temptation to change things so often that your notes and
presentation approaches would be rendered obsolete. I will
only do the update on an annual basis and I will make every
effort to make changes as seamlessly as possible vis a vis
presentation. I am a teacher as well and rest assured that any
and all changes will be updated in my presentations. And
those have always been available to adopting instructors.
That said, please note that I have reorganized Chapter 10.
The section names and content are identical, it is just that the
sections have been reorganized more consistently along the
lines of costs and benefits. I have also renamed Chapter 11;
but that is it, just a rename. There were also just two fixits for
a couple of Figures (7.10 correct and 13.7 deleted).
As always, instructors can also visit Professor Fort’s Web Site
for additional material and teaching tools related to the text.
In the Instructor section, you’ll find links to Professor Fort’s
Sports Economics Data Directory, a set of Sports Economics
v

https://sites.google.com/site/rodswebpages/research

https://sites.google.com/site/rodswebpages/research

Projects, and chapter-by-chapter additional Web Links,
Answers to Thought and Advanced Questions from the text,
Answers to the Sports Economics Projects, PowerPoint
lectures, and figures and tables from the text.
Everything else that you would expect to find to recommend
to students is already in the book! That’s the beauty of the
electronic text idea.
That said, it is a BETA Version. Doubtless errors in
transporting Tables and Figures to the new popovers. There
also will be glitches in iBooks; I hope you will let me know
what you find.
FEEDBACK
If you have any questions related to this edition don’t hesitate
to contact me.
vi

mailto:rodfort@umich.edu?subject=Feedback

mailto:rodfort@umich.edu?subject=Feedback

Acknowledgements
All shortcomings in this book are my fault, but many of the
good things about the book came about with the help of
others. I am fortunate and grateful to have learned from the
best, Roger Noll and James Quirk. While Simon Rottenberg is
the “father” of sports economics, all sports economists should
admit that Gerald Scully is the true pioneer. Whether it is
demand, team and league behavior, the market for players, or
the analysis of sports over time, Scully was nearly always first
on the scene. His death was a great personal loss, and I will
miss him. The book benefited greatly from a number of
reviewers. Their thorough comments helped shape this book
and (hopefully) future editions. My heartfelt thanks to:
Jeff Ankrom, Wittenberg University
Katie Baird, University of Washington, Tacoma
Ross Booth, Monash University
Roger Blair, University of Florida
Edward Coulson, Pennsylvania State University
Karl Einolf, Mount St. Mary’s University
Stanley Engerman, University of Rochester
John Fizel, Pennsylvania State University, Erie
Jahn Hakes, Clemson University
Bruce Johnson, Centre College
Phillip Lane, Fairfield University
Daniel Marburger, Arkansas State University
Kevin Quinn, St. Norbert College
Tom Regan, University of South Carolina
Allen Sanderson, University of Chicago
John Siegfried, Vanderbilt University
Paul Staudohar, California State University, Hayward
Lee Van Scyoc, University of Wisconsin-Oshkosh
Jason Winfree, University of Michigan
If you decide to engage the text at a level where you offer me
substantial feedback, I’ll be glad to add you to the list.
Even though they don’t know it, all of my sports economics
students over the years contributed fundamentally to this
book. As I honed my lectures and approaches, the
organization of the book came into being.
vii

CHAPTER 1
Warm-Up:
The Business of Sports
If we do everything right this year and win
again, we probably will be able to break even.
And I’ll tell you this much: I didn’t buy this
team to break even.
–Carl Pohlad, Minnesota Twins Owner
Sporting News, August 24, 1992, p. 10.
CHAPTER OBJECTIVES
After reading this chapter, you should be able to:
• Explain why there is more to sports than the
revenues they generate.
• Understand that because sports are actually
businesses, standard economic tools can enhance
your understanding of sports off the field.
• Know why the business side of sports often
befuddles fans.
• Understand the skepticism that players voice
toward owners’ claims of poverty.
• Explain why owners typically are exasperated at
their treatment by players and fans.

SECTION 1
Introduction
In 1915, the New York Yankees were purchased for $460,000
($9.9 million; throughout the text, a dollar value in
parentheses represents inflation adjustment to dollars in
2009). Eighty-three years later, in 1998, a bona fide offer of
$631.1 million ($845.7 million) was recorded for the team.
Adjusting for inflation, that’s a 5.5 percent return per year
[9.9(1 + r)83 = 845.7 implies that r, the constant rate of
growth, is 5.5 percent]. The inflation-adjusted growth rate in
the U.S. economy at large is typically about 3 percent
annually. Thus, owning the Yankees was 1.8 times as valuable
as a diversified investment portfolio.
The Yankees’ value serves to introduce the following warm-up
on the business of sports. These days, the sports pages look
more like the business pages, and judging from media reports,
nobody seems happy about it. Players and owners fight over
money; owners fight with each other over money; and the fans
seem more hardened and cynical every day. To top it all off, a
dizzying array of terminology drowns everybody—free agency,
luxury taxes, lockouts, salary caps, arbitration, the National
Labor Relations Board (NLRB), franchise free agency, local
host blackmail, and public–private partnerships, to name just
a few. Any sports fan needs more than the analysis offered in
the newspaper to make sense of it all. In this chapter, I’ll
discuss these issues and a few basic ideas to set up the
economic analysis of American sports that appears in the rest
of the book.
2

SECTION 2
Putting the Business of
Sports Into Perspective
Analysts differ in their opinions about the place of sports in
society. Some argue that America’s priorities are out of whack
and that our preoccupation with sports is unhealthy. For
example, in 1994, the U.S. House of Representatives held
hearings on the labor–management tensions that had led to
the Major League Baseball (MLB) strike that year. Brookings
Institution economist Henry Aaron (believe it or not, there
really is an economist with that storied sports name) almost
scolded the House of Representatives subcommittee for
spending its time on sports. He pointed out a number of facts
about the relative economic importance of sports: MLB was
about a $2 billion ($2.9 billion) industry in terms of total
revenue (at that time), and any given sports team typically
represented only a fraction of 1 percent of the economic
activity in its county. Professor Aaron went on to note that the
envelope industry generated about $2.6 billion ($3.8 billion)
and the production of cardboard boxes generated $7.6 billion
($11.2 billion) in annual revenues. Surely, he urged the
committee, there had to be bigger things to worry about than
baseball!
It cannot be denied that the importance of sports exceeds its
total revenues. For example, there is no cardboard-box page
in the daily paper or on the news each night. Further, the last
time I checked, there was no cardboard-box equivalent of a
cable network like ESPN devoted to reporting the minutiae of
the industry. These news reports make it clear that the level of
interest in sports goes beyond just their monetary
contribution to team owners, employees, or local
governments.
Two of the many keys to understanding the way the sports
world works are a little bit of economics and a modicum of
common sense. Of course, having applied this approach to
sports for over 20 years, I speak from experience. My aim is to
show you how the business side of sports affects the game you
see at the stadium or in the arena. So let’s go ahead and start
applying economics and common sense to the world of sports.
3

SECTION 3
Sports Really Are Big
Business
Sports are big business. Legends of owners who don’t care
about profits or players who play just for the love of the game
are overblown. The romanticized Shoeless Joe Jackson
character in the film Field of Dreams would have us believe he
would have played for free. Well, sport was fun (and still is),
but the money has never been bad either. And I don’t mean
the money that Jackson and his teammates allegedly took for
throwing the 1919 World Series. The Sporting News (October
17, 1994, pp. 38–40) reported that Jackson made $6,000 that
year ($75,720). One estimate put the average earnings of
farmworkers at $675 ($8,518) and mine workers at $1,283
($16,191), and $1,148 ($14,488) as the average earnings in
manufacturing that year (King, Knauth, and Macauley, 1922).
Jackson made over 4.5 times the highest of these alternatives.
Of course, star athletes now make hundreds of times the
average pay in the United States but the point here is that
earning money from a contest has always been the way in
sports. As we’ll see in the last chapter, this goes for so-called
amateur sports as well.
Interestingly, many sports fans have a hard time
understanding that sports really are big business. Some
evidence and a little common sense help to drive the point
home. Historically, teams that spend the most tend to win the
most. So why don’t team owners, some of the most successful
and wealthy of all Americans, just buy a winning team or
invest in winning players? We don’t see wealthy owners toss
their money around in this way very often. Even when we do,
the strategy doesn’t always work. Take Wayne Huizenga, who
in 1997 invested big money in the Florida Marlins. By all
accounts, he bought enough talent that year to win the World
Series. In fact, the team did go on to win the World Series, but
Huizenga sold off most of the important members of the team
the following year. The reason? The city of Miami made it
clear that part of the revenue Mr. Huizenga hoped to capture
with his world champions, in the form of a new publicly
financed stadium, was not going to be delivered. In the end,
Huizenga’s investment didn’t pay off.
The stadium issue brings up more evidence that sports really
are business. Why don’t owners just build their own stadiums
rather than pressuring state and local governments to foot the
bill? Surely, these extremely wealthy people can afford their
own stadiums. For example, Paul Allen, who owns both the
National Basketball Association’s (NBA) Portland Trailblazers
and the National Football League’s (NFL) Seattle Seahawks,
built and owned (indirectly through another organization) the
Trailblazers arena. However, he only exercised his option to
buy the Seahawks after public money was put toward a new
football stadium in Seattle. Perhaps it had something to do
with the fact that an arena can be built at a little less than half
4

the cost of a stadium. That is clearly a bottom-line
consideration important to a business decision.
For that matter, why do these eminently rich people even
charge admission or sell broadcast and sponsorship rights?
Why would they ever lock out players, as happened with the
loss of the entire National Hockey League (NHL) season
(including the Stanley Cup Finals!) in 2004–2005? In fact,
why would the owners negotiate player contracts at all? If
sports are not businesses, wealthy owners could just give the
players what they wanted, let fans into stadiums for free, and
everybody would be happy. Players would be richer and fans
wouldn’t suffer any interruption in play. Speaking of players,
if sports are not businesses, then why don’t they just play for
expenses?
THE POWER 100
The importance of the business aspect of sports can also be
seen in the Sporting News “Power 100” list. Each year, the
magazine lists the 100 most powerful people in sports.
Typically, fewer than 10 of these dominant sports
personalities actually participate in their sport. The Sporting
News list is dominated by sports officials (commissioners and
league officers in pro sports, conference presidents and bowl
officials for college sports, and presidents and organizers for
the Olympics). Media moguls and other network officers are a
close second. CEO/owners, agents, sponsors, and participants
(players and coaches) round out the field. Fewer than half of
these power brokers are directly responsible for the actual
operation and playing of sports (officials, CEO/owners, and
participants).
In fact, in 2006, NBA commissioner David Stern topped the
list. The rest of the top ten contained the three other major
league sports commissioners (third, fourth, and sixth), the
chairman and CEO of NASCAR (fifth), and media moguls.
Incidentally, in 1995, Stanford University economist—and my
former professor—Roger Noll made the list, ranked number
93. Although my old professor does have a pretty decent
baseline jump shot, he didn’t make the list because of his
athletic prowess. Professor Noll’s services are in demand by
nearly every players’ union in sports, as well as by the U.S.
Congress for expert testimony. In fact, his analysis was
instrumental in court considerations that eventually resulted
in NFL free agency beginning in the 1994 season and during
the MLB strike of 1994.
So sports are business, and big business at that. Sometimes
owners make statements that appear to contradict this. Even a
billionaire owner like Mark Cuban, owner of the NBA Dallas
Mavericks, doesn’t let his fans in for free, nor is he going to
mismanage his personnel (the team did not make the play-
offs in the 10 years prior to purchase by Cuban in 2000; they
have made the play-offs every year since and made the finals
in 2005–2006). He is going to treat his team as an
investment. Part of the return may be in terms of his love of
sports, but there is an undeniable business aspect to the
investment as well. Cuban intends to make money on his
5

team. In the next section, we’ll look at the problems that the
sports business poses for fans, owners, and players.
6

SECTION 4
Befuddled Fans
Media accounts seem to indicate that the greed and hypocrisy
of those involved in the sports business generate a great deal
of fan confusion and skepticism. These common feelings
about the sports business are worth mentioning. A review of
the causes of these views provides a nice point of departure
for the economic explanations that we will discuss later in the
book.
We’ve all heard sports fans lament that sports are not about
the game anymore but all about money. Fans also believe that
players and owners are overpaid—a belief that has been
remarkably consistent over time. Table 1.1 shows the results
of two polls conducted about 10 years apart on fans’ opinions
of player and owner earnings. As you can see from Table 1.1,
the vast majority of fans agrees that owners and players are
overpaid, and players more so than owners. But note that
these feelings have eased a little over the years in Table 1.1.
However, these feelings do still run strong. An ESPN/Seton
Hall University Poll in 2009 found that 40.3 percent of
respondents still felt that overpaid ballplayers were “The
Biggest Problem with Major League Baseball” (“Steroids”
came in second with only 22.5 percent). A Rasmussen Reports
Survey in 2009 found that 30 percent of respondents thought
sports stars should not be allowed to make more than $1
million. It can’t be known whether fan survey responses
reflect concern that payment issues are spoiling the games or
whether such responses just reveal fan envy.
Although overpayment may be in the eye of the beholder,
there is no doubt that athletes make a lot of money. As you
can see in Table 1.2, according to the Forbes Top 100
Celebrities list for 2006, top-earning athletes can make tens
of millions of dollars annually. Even Yankees third baseman
Alex Rodriguez and New England quarterback Tom Brady,
tied for 45th on the list by pay, made $29 million in 2006.
Compare this with the income of top family practice
physicians at the time, about $240,000. At that rate, the
average doctor wouldn’t earn A-Rod’s $29 million for 121
years. A teacher with an average pay at that time of around
$49,000 would take 592 years to match A-Rod’s pay for just
one year.
The wealth of owners also befuddles fans. Forbes
(www.forbes.com/lists/2005) reported in 2005 that Paul
Allen, owner of the Portland Trailblazers and Seattle
Seahawks, had a net worth near $22.5 billion ($25.2 billion).
Indeed, according to Forbes, there were more than a dozen
billionaire owners in 2003, and many of them owned more
than one pro sports team. Few of us can even imagine this
type of buying power. On top of it all, these owners make
choices that irritate fans. Fans perceive that owners let money
get in the way of fielding winning teams. Then the owners
7

http://www.forbes.com/lists/2005

http://www.forbes.com/lists/2005

have the gall (in the eyes of fans) to hold fans and local
governments hostage in negotiations over public subsidies for
stadiums and arenas. Some owners have even blamed their
fans for lack of support when their subsidy demands are not
met. Meanwhile, fans surmise that all of the money comes out
of their pockets. The typical view is that greedy players hit up
rich owners who bow to their demands and, in turn, raise
prices for the fans. A little economic thinking shows that this
commonly held path of causality is flat backwards: It just
doesn’t fit reality (Chapter 7).
Imagine walking into your boss’s office and demanding a pay
increase that isn’t tied to some sort of extra revenue that you
have generated for the firm. Good luck! Unless the firm is
earning more money and unless some of the increase can be
directly tied to your performance, your demand won’t be met.
The same is true of sports stars. Team revenues continue to
grow, providing the basis for such demands in the first place.
Nonetheless, fans are befuddled and, at times, angry, and not
all of them are taking it lying down. Fan consumer
movements are detailed in this section’s Learning Highlight:
Consumer Movements in Sports.
8

LEARNING HIGHLIGHT: CONSUMER MOVEMENTS IN
SPORTS
In 1994, MLB fans lost the end of the regular season, both
League Championship Series, and the World Series to a
players’ strike. NBA fans lost the first half of the 1998–1999
season to a lockout by the owners. NHL fans lost the entire
first season, including the championship, in the modern
history of professional sports in 2004–2005. Understandably,
fans were upset. But not all fans are willing to take strikes and
lockouts sitting down. A few have begun to organize via the
Internet. Here are descriptions of a few of these groups:
• We the Fans. They’re mad as hell and they aren’t going to
take it anymore. Their aim is getting fans united, organized,
and “into the game” of sports business. Their view is that
fans are the most important part of sports and the sports
business (wethefans.com).
• League of Fans. Founded by Ralph Nader, this sports reform
project encourages social and civic responsibility in the
sports industry and culture (www.leagueoffans.org).
• Unfortunately, as we’ll explore in later chapters, forming a
successful political organization can be a daunting task. For
example, The Baseball Fans Union was listed as part of this
consumer movement in the last edition of this textbook. It
has since disappeared.
[You used to have a picture of a crazed fan.]
9

http://www.leagueoffans.org

http://www.leagueoffans.org

SECTION 5
Player Skepticism and
Owner Exasperation
As anyone who reads the sports pages knows, players are
often portrayed as greedy in their pay negotiations and
skeptical of owners’ claims of poverty. Every year, players
with multimillion-dollar contracts hold their teams and fans
hostage during prolonged preseason holdouts. This lack of
player loyalty is a common fan lament. But think about this
from the players’ perspective. There is plenty of justification
for player skepticism. The value of teams continues to
increase over time. In our first book, Pay Dirt (1992), James
Quirk and I demonstrated that franchise sale values typically
outperformed a portfolio of common stocks from the 1940s
through the 1980s (for more information on this, see the
Learning Highlight: Buying the New York Yankees at the end
of the section). In our second book, Hardball (1999), Quirk
and I showed that, while the increase in sale prices slowed
during the 1990s, teams still offered a pretty attractive
investment. Team sale prices in the 1990s increased 11.3
percent in MLB, 12.7 percent in the NFL, 17.7 percent in the
NBA, and 10.7 percent in the NHL. Just recently, record
payments for teams were set in every league—the NHL’s
Edmonton Oilers in 2008 sold for $200 million ($202
million), the NBA’s Cleveland Cavaliers in 2005 for $375
million ($420 million), MLB’s Washington Nationals in 2007
for $450 million ($472.5 million), and the NFL’s Minnesota
Vikings in 2005 for $600 million ($672 million).
Further, as Table 1.3 shows, expansion franchise prices have
grown astronomically. The amounts in Table 1.3 show what a
new expansion team would have cost over the decades in the
four major professional team sports. But let’s be careful
making comparisons over a 40-year history. In Table 1.3,
expansion fees that involved league mergers in hockey and
basketball in 1979 are excluded. On an annual basis,
expansion franchise prices in the major leagues have
increased between 2.7 percent annually in MLB and 10.0
percent annually in the NFL over the past 4 decades.
From the players’ perspectives, it is easy to see why owners’
claims of poverty are met with skepticism. These increases in
sale values are public knowledge, and it is only natural for
players to focus on their contribution to the rising values.
After all, nobody pays to watch an owner own. So players are
just like other workers when they try to collect their share of
the value they help to create, and those values have been
rising handsomely for quite a long while.
Owners receive the most mixed treatment from fans and the
media. It is easy to paint a clear picture of owner exasperation
with the business of sports. On the one hand, they pay an arm
and a leg to get the team in the first place. Further, payments
to players are increasing over time. It is small wonder owners
claim that they aren’t getting rich running these teams. From
10

their perspective, being investment savvy, taking risks, and
having administrative skill all contribute to success and
increasing team values. So they, too, want their fair share.
Many owners seem genuinely exasperated with the
disapproval their actions garner from fans and players.
11

LEARNING HIGHLIGHT: BUYING THE NEW YORK
YANKEES
The New York Yankees is one team that has been, for the most
part, kind to its investors. This Table 1.4, adjusted to 2009
dollars from figures in Sports Illustrated (November 2, 1998,
p. 40), shows the sale prices of the Yankees over 83 years.
Remember that the real annual growth rate in the economy at
large is about 3 percent per year.
The growth in sales value, even with the disastrous ownership
tenure of the Columbia Broadcasting System CBS, is truly
amazing. A simple formula shows the annual rate of return:

where P0 is the original sale price, r is a constant rate of
return, t is the length of time to next sale, and Pt is the latest
sale price. From this, it is clear that

Ruppert and Huston enjoyed about a 4.2 percent real annual
return to sale in 1945, while the Triumvir enjoyed about a 5.1
percent real annual return to sale for their roughly 20 years of
ownership. Both of these early annual rates are substantially
more than the usual 3 percent annual growth rate in the
economy. CBS lost about 7.5 percent annually (covered in
more detail in Chapter 4), but then look at George
Steinbrenner’s reign—the franchise increased in value
conservatively at a whopping 12.0 percent per year adjusted
for inflation based on an actual offer by one media provider.
This is four times the typical 3 percent growth rate in the
economy at large. Overall, then, the 83-year average of about
5.5 percent is the real annual increase in the value of the New
York Yankees. Nice work, if you can get it when the general
growth rate of the economy is 3 percent.
[You used to have a picture of the 1915 Yankees: In 1915, the
New York Yankees were sold for $460,000 ($9.9 million). In
1998 that figure had risen to about $631.1 million ($845.7
million).]
12

SECTION 6
Sports Economics: Making
Sense of It All
In 1956, economist Simon Rottenberg published the first and
most cited sports economics article, “The Baseball Players’
Labor Market,” in the Journal of Political Economy. In it, he
offered two hypotheses that have been fundamental to
research in the area. First—important in the next chapter—he
suggested that competitive balance should matter to pro
sports leagues because it matters to fans. Competitive balance
is the idea that fans prefer their team to win but they also
prefer close game outcomes and season finishes to blowouts.
Second—for Chapters 6, 7, and 8—Rottenberg suggested that
player drafts and the “player reserve system” then in place did
nothing to competitive balance. Instead, these mechanisms
simply redistributed the value of talent away from players and
toward owners.
Many of us with an interest in sports economics have been
developing and testing Rottenberg’s ideas ever since. And
many more have been applying microeconomics to even more
sports topics. The rest of this book, you might notice, follows
the outline of a typical “introduction to microeconomics”
textbook. In as many ways possible, given my experience with
the topic, we will examine together the ways that this powerful
way of thinking has helped explain the often confusing and
sometimes maddening business side of sports.
13

SECTION 7
Chapter Recap
Professional team sports generated revenues of less than $30
billion in 2008. This isn’t much compared to other industries,
but interest in sports goes way beyond revenues. Perhaps
Americans are obsessed with sports to the detriment of other
important areas of endeavor. The business of sports is the
object of our analysis.
The bulk of the evidence shows that sports are big business.
Even billionaire owners treat their endeavor as a business.
The bottom line matters, revenues are of paramount
importance, and costs are fought with grim determination.
Millionaire players certainly behave the same as any labor
supplier seeking to maximize the return on their efforts. At
times sports may be about more than just money, but by and
large, business aspects dominate off the field.
Fans believe it’s all about money. They’re jealous and amazed
at players’ salaries and believe that both owners and players
make too much from their sport. Finally, amid all this
moneymaking, the price of enjoying sports continues to
increase. Part of this fan befuddlement has turned to anger
and action in the form of fan advocacy groups.
Players and owners are also skeptical and exasperated.
Despite the arguments of owners that they aren’t getting rich
at their sport, team values and expansion franchise prices rise
at handsome rates, and players know that they contribute to
these increases. Owners counter that high team prices are the
albatross they bear through their entire term of ownership.
They take all the business risks and provide no small amount
of administrative skills, and they expect profits in return.
14

SECTION 8
Key Terms and Concepts
You should have run into each of these in pop-ups in the text
of this chapter:
• Sports are big business
• Fan confusion
• Fan envy
• Fan consumer movements
• Player loyalty
• Player skepticism
• Franchise sale values
• Expansion franchise prices
• Owner exasperation
15

SECTION 9
Review Questions
1. Why do some argue that America’s preoccupation with
sports is unhealthy? Think in terms of economics.
2. Why are sports important beyond the basic revenues that
they generate? Why are they important to you?
3. What are some of the indicators of the importance of
sports beyond the money?
4. List reasons why sports really are businesses.
5. Why do rich owners insist that the public share some of
the burden of stadium remodeling or construction?
6. Do you believe that Mark Cuban, owner of the Dallas
Mavericks, really doesn’t care if he blows $100 million in
management mistakes? What else is behind his
statement?
7. List three reasons why fans are befuddled by the current
sports landscape.
8. Why are players skeptical of owners’ claims of poverty?
9. Why do owners have chips on their shoulders about team
ownership?
10. With whom do you most identify—fans, players, or
owners? Why?
16

SECTION 10
Thought Problems
1. Boston Bruins’ owner Jeremy Jacobs once suggested that
contemporary business practices had been an essential
contributor to the popularity and vitality of pro sports in
America (Sporting News, 1992). How do you assess
Jacobs’s claim? What are the good things that
“contemporary business practices” have brought to pro
and college sports? What are the bad things?
2. Why is it that fans perceive that the game is all about
money? How would you respond if one of your friends
started to spout off about how sports are all about the
money? Be sure to include both pro and college sports in
your answer.
3. In Table 1.1, what happened to the “no opinion”
percentage between 1984 and 1994? Who bears the
greater brunt of fan dissatisfaction with earnings over
time, fans or owners? Why do you think this has
happened?
4. How do you explain the rise of fan consumer movements
on the Internet? Be sure to think both in terms of fan
attitudes and the cost involved in organizing a group
using the Internet. Check out some fan consumer
movements, and decide which one you would join. What
influenced your selection?
5. If you were a professional athlete (pick your sport), would
you behave any differently than today’s players in your
pursuit of pay? If so, why? How are you different from
players, economically speaking?
17

SECTION 11
Advanced Problems
1. How would you go about assessing the overall value of
sports? For example, one component might be the
difference between the size of circulation and price of
newspapers in cities that have pro sports and those that
don’t. Can you think of other elements to get a handle on
this value?
2. Can you think of actions by owners and players that
indicate there is more to sports than just money? Are
these actions outside the bounds of economic analysis?
3. How is player pay just like pay for, say, management
information systems jobs?
4. Do owners have any incentive to misrepresent the
economic welfare generated for them by team ownership?
Think of the big picture in terms of all of the relationships
that owners have with players, fans, and the state and
local government officials with whom they must deal.
5. Why are the NHL expansion fees in Table 1.3 so much
lower than those for the rest of the major pro sports
leagues? The data in Table 1.3 for all leagues besides MLB
share something in common. Identify the commonality
and offer an explanation.
18

SECTION 12
References
Hanel, Marnie. “Ted Turner on Murdoch, CNN, and Jacques
Cousteau.” Vanity Fair (on line), April 2, 2009. URL last
checked April 27, 2009: http://www.vanityfair.com/online/
politics/2009/04/listen-up-rupert-ted-turner-would-like-a-
word.html.
King, Willford I., Knauth, Oswald. W., and Macauley,
Frederick R. Income in the United States: Its Amount and
Distribution, 1909–1919. Vol. II, Detailed Report. New York:
National Bureau of Economic Research, 1922.
Quirk, James, and Rodney Fort. Pay Dirt: The Business of
Professional Team Sports. Princeton, NJ: Princeton
University Press, 1992.
Quirk, James, and Rodney Fort. Hardball: The Abuse of
Power in Pro Team Sports. Princeton, NJ: Princeton
University Press, 1999.
Rottenberg, Simon. “The Baseball Players’ Labor Market.”
Journal of Political Economy 64 (1956): 242–258.
19

http://www.vanityfair.com/online/politics/2009/04/listen-up-rupert-ted-turner-would-like-a-word.html

http://www.vanityfair.com/online/politics/2009/04/listen-up-rupert-ted-turner-would-like-a-word.html

http://www.vanityfair.com/online/politics/2009/04/listen-up-rupert-ted-turner-would-like-a-word.html

http://www.vanityfair.com/online/politics/2009/04/listen-up-rupert-ted-turner-would-like-a-word.html

http://www.vanityfair.com/online/politics/2009/04/listen-up-rupert-ted-turner-would-like-a-word.html

http://www.vanityfair.com/online/politics/2009/04/listen-up-rupert-ted-turner-would-like-a-word.html

SECTION 13
Suggestions for Further
Reading
Costas, Bob. Fair Ball: A Fan’s Case for Baseball. New York:
Broadway Books, 2000.
Helyar, John. Lords of the Realm: The Real History of
Baseball. New York: Ballantine Books, 1994.
Klein, Gene. First Down and a Billion: The Funny Business of
Pro Football. New York: St. Martin’s Press, 1987.
Miller, Marvin. A Whole Different Ball Game: The Inside
Story of Baseball’s New Deal. New York: Simon and Schuster,
1991.
Noll, Roger G., and Andrew Zimbalist, eds. Sports, Jobs, and
Taxes: The Economic Impact of Sports Teams and Stadiums.
Washington, D.C.: Brookings Institution, 1997.
20

CHAPTER 2
Demand and Sports
Revenue
I remember Arthur Wirtz (former owner of the
Chicago Blackhawks, deceased, 1983)
delivering his most famous line. Somebody
said, “We need some paint for the seats,” and
Arthur said, “We don’t need paint, we need
asses.
–Stan Mikita, Former Blackhawk Great
Sports Illustrated, June 1, 1992, p. 68.
CHAPTER OBJECTIVES
After reading this chapter, you should be able to:
• Explain the concept of demand for sports, why it
varies for different types of sports, and why
teams have market power.
• Explain how demand functions yield estimates of
sports fan welfare.
• Discuss the relationship between demand,
elasticity, total revenue, and marginal revenue in
sports.
• Explain why fans with more inelastic demand can
be charged higher prices than those with more
elastic demand.
• Explain why revenue variation is the source of
unbalanced competition on the field and tension
between teams in a given league.

SECTION 1
Introduction
Fan decisions to attend games are one of the most important
elements of the sports business, but fan decisions are a
complex issue. Stephen Jay Gould, the eminent natural
scientist, emphasized this as follows:
Old fans like me often argue that the hoopla and
commercialism—with rock music between innings at the
park, organized cheering and chanting, hyped selling of
all conceivable paraphernalia—have greatly
compromised the gentler game we loved (which can still
be so beautiful in its execution between the between-
innings hype). But who is buying the T-shirts, fawning
(and pawing) over every utility infielder for autographs,
acquiescing in each successive move to eliminate the free
broadcasts we once enjoyed and substitute some new
scheme of selective payment? Us, of course; nobody but
us, or at least most of us. How can we moan that
baseball has fallen to mammon when we willingly
rushed into each new purchase? (The New York Review,
November 5, 1992, p. 44).
We analyze these interesting fan choices in this chapter. We
will examine scarcity and rationing in sports, as well as the
demand theory as it applies to sports. Because there are few
close substitutes for sports team output, we will look closely at
market power and its effect on sports and discuss willingness
to pay and consumers’ surpluses. Price elasticity, total
revenue, and marginal revenue are also covered in this
chapter. We will also explore price discrimination, one of the
most misunderstood elements of sports pricing decisions.
Finally, we will close our discussion by analyzing revenue
data.
22

SECTION 2
Scarcity in Sports
Scarcity makes the economic world go round. Indeed, it is the
foundation of the economic chain of events—the trinity, if you
will—summarized in Figure 2.1. Scarcity leads to rationing,
and rationing leads to competition. Although the press usually
portrays owners in pursuit of fan dollars, it actually is a two-
way street: Sports fans demand team output and are willing to
pay for it.
But just what is it that is so scarce when we talk about sports?
The answer, of course, is that all of the characteristics of the
sports product—the beauty of athletic prowess, absolute and
relative team quality, the commonality sports provide, and the
thrill of victory—are scarce.
ATHLETIC PROWESS
We have all played the games and understand that we are
observing far more than mere proficiency when we watch
professional athletes in action. All of the other scarce
elements aside, fans appreciate the beauty and seeming
impossibility of athletic prowess. One of my favorite
statements on this was by Frederick Exley, describing a
Sunday afternoon football game at the Polo Grounds in 1954
as something like art. Nearly every fan has felt what Exley has
verbalized, and we are willing to pay in order to see it.
QUALITY OF THE TEAMS AND COMPETITION
The quality of teams, both absolute quality and relative
quality, is a scarce commodity that generates the competition
that fans enjoy. The absolute level of quality simply refers to
the level of quality; for example, the difference between minor
and major leagues. The relative level of quality describes the
competitiveness of a team once the absolute level is
determined; for example, the difference between a cellar
dweller and a division champion.
Although fans enjoy the quality of competition at any level,
they are willing to pay more for higher quality levels. Thus,
the absolute level of quality matters. The level of competition
that particular fans will eventually enjoy will depend on their
willingness to pay for team quality. Those willing to pay more
get to enjoy a higher level of quality than those willing to pay
less. We will wait until Chapter 4 to analyze this in detail, but
just think of the reason why one city enjoys only minor league
competition whereas another enjoys a World Series
champion. All of the elements behind that outcome concern
the ability and willingness of fans to pay for different levels of
competition in a given league.
Once the absolute level of quality is determined, relative
competition becomes the object of fan attention. For the most
part, an operational hypothesis in sports economics is that
23

fans want their team to win in close games. And when it
comes to the postseason, fans expect a reasonable chance that
their team will appear, not every year, but often. This is
Rottenberg’s uncertainty of outcome hypothesis from the
previous chapter. Even if the best that fans can support at a
particular location is minor league sports, they still will care
deeply about the competition at that level of play.
The uncertainty of outcome hypothesis is just that, a
hypothesis. It is a prediction about preferences that cannot be
examined directly. So those interested in the hypothesis must
approach it indirectly by examining data related to fan choice.
For example, in academic studies of the demand for
attendance, investigators attempt to determine the impact of
the usual demand variables on attendance, such as price,
income, and winning, but also include various measures of
how close games might be expected to be, or how close the
regular season ended up, or turnover in the teams that make
the play-offs. Professor Stefan Szymanski (2003) shows that
the results of these studies are mixed. Some analysts find that
these types of uncertainty of outcome variables matter, others
don’t. In addition, outcome uncertainty, itself, matters for
some types of sports leagues but not for others.
Table 2.1 provides a bit of simple evidence concerning the
uncertainty of outcome hypothesis and serves to emphasize
that it is, after all, a hypothesis. Table 2.1 relates fans’
willingness to pay (in terms of the average price of
attendance), winning percent (wins divided by total games),
and close games (plus or minus a touchdown). As you can see,
there is some support for the uncertainty of outcome
hypothesis. Teams with winning percents above 0.600
charged higher ticket prices and had a much higher
percentage of close games compared to those teams below
0.400. If you think this is due only to the outliers (New
England among the best teams and Detroit among the worst),
then recalculate the comparison by throwing them out. You’ll
see that the results for close games remain essentially
unchanged.
This brings up an important operating principle for sports
leagues that we will return to repeatedly in this book. If fans
care about relative outcomes, there must be some balance in
competition between teams. If the same teams always
dominate the play-offs and the competitors in the
championship are a foregone conclusion, then fan interest will
wane. At the end of this section, the Learning Highlight:
Michael Jordan’s Retirement provides an introduction to why
competitive balance is a big deal. In 1992, my coauthor and I
put it this way:
For every fan who is a purist who simply enjoys
watching athletes with outstanding ability perform
regardless of the outcome, there are many more who go
to watch their team win, and particularly to watch their
team win a close game over a challenging opponent. In
order to maintain fan interest, a sports league has to
ensure that teams do not get too strong or too weak
24

relative to one another so that uncertainty of outcome is
preserved. (Quirk and Fort, 1992, p. 243)
COMMONALITY
Another scarce sports characteristic is the commonality it
provides. Yesterday’s game and its coverage by the media
provide a common bond among people. How often have you
heard, “Hey, how about those [favorite team here]?” The
common bond of sports is also apparent in poetic epics such
as Ernest Lawrence Thayer’s “Casey at the Bat” (1888) and
shared personal reflections such as Donald Hall’s Fathers
Playing Catch with Sons (1998). Finally, motion pictures
ranging from the emotional (Hoosiers or Field of Dreams) to
the tragic (Brian’s Song) to the lighthearted (Tin Cup) bind us
together. Once you’ve heard it, you’ll never forget Terrence
Mann (played by James Earl Jones) explaining the binding
power of baseball to farmer Ray Kinsella (played by Kevin
Costner) in Field of Dreams:
The one constant through all the years, Ray, has been
baseball. America has rolled by like an army of
steamrollers. It’s been erased like a blackboard, rebuilt
and erased again. Baseball marked the time. This field,
this game, is a part of our past, Ray. It reminds of all
that once was good … that could be again. Ohhhh, people
will come, Ray. People will most definitely come.
WINNING
Let’s not forget about winning. Depending on how you may
have been coached, you probably remember those immortal
words attributed to Vince Lombardi, NFL coaching legend
and namesake of the Super Bowl Trophy: “Winning isn’t
everything. It’s the only thing.” Fans love the home team, but
they love it even more when it is a winner. The more bitter the
rivalry, it seems, the sweeter the win. Thus, winning is very
closely related to competition in the sense that victory over a
bitter rival is the sweetest of all. Winning is such a pervasive
motivation for fans that later in the book we will use it as a
catchall for the characteristics of sports demanded by fans.
RATIONING IN SPORTS
From an economic perspective, the factors we just described
are all scarce commodities; that is, there are not enough of
them freely available to satisfy all desires. Thus, rationing
must occur; some mechanism must be chosen to help
discriminate between those who want sports. There are many
rationing devices, but the most prominent is price. If fans are
willing to pay ticket prices, they get to see the sports event.
The more fans in a given location are willing to pay for quality,
the higher the quality of the team they will get to enjoy.
Further, as long as fans pay a price through product
purchases, advertisers will give media providers (networks,
cable, and satellite stations) the financial incentive to put
sports on television.
Those who are unhappy with price as a rationing device often
seek to alter the terms of trade. In many lease agreements
25

between professional teams and their city hosts, clauses either
restrict ticket prices or require that teams keep some seating
sections affordable. In cases where the dollar price is kept
artificially low due to agreements between the teams and the
government, some fans are happier than they would be if
ticket prices were higher. But the inescapable observation is
that a price will be paid, either in dollars or as a combination
of dollars and waiting. Many fans who are relatively slow to
get in line lose out on attendance altogether.
Waiting in line is just one option; the black market is another.
Black markets arise whenever the price mechanism is
exchanged for another rationing method. Scalping reflects a
fully functioning black market. Scalpers buy the cheap tickets
and sell them to people who are willing to pay high prices in
order to avoid waiting in lines. Such arbitrage from lower- to
higher-price buyers usually is viewed as an important function
of markets. However, scalping is often viewed so scornfully
that it is illegal in many places.
Let’s now turn to a more rigorous specification of scarcity and
rationing, namely, demand theory. The basic idea of relating
quantity to price generates the entire foundation of the
revenue side of sports. If you seek answers to the many
puzzles surrounding sports, you’ve got to follow the advice
that “Deep Throat” gave to Bernstein and Woodward in All
the President’s Men (1974): “Follow the money.” Following
the money requires a thorough understanding of the demand
for sports.
26

LEARNING HIGHLIGHT: MICHAEL JORDAN’S
RETIREMENT
When Michael Jordan retired in 1998, reactions were mixed.
On the one hand, the NBA lost arguably its greatest player
ever, long before his talents were in decline. The beauty of the
game suffered. Many argued that the absolute level of talent
declined with the loss of just one player. In addition, one of
the NBA’s biggest draws was gone, leading to much hand-
wringing among observers of the financial welfare of the
game. All of this was on the heels of the loss of half a season to
a labor dispute that led to the first lost games in the league’s
history.
But fans in New York, Phoenix, San Antonio, and Utah had an
entirely different take on MJ’s retirement. With the demise of
the Bulls, no longer led by Jordan, fans in these cities could
look forward to an increased probability of successful
postseason play for their own teams. Without Jordan, the
absolute level of competition in the NBA may have decreased,
but relatively speaking, more teams were in the hunt for the
championship. The revenue impact across so many cities
could potentially outweigh any other decline in fan interest
due to the loss of Jordan to the league. If more fans are more
interested in the enhanced chances for a wider variety of
teams, the NBA could actually be better off without Jordan.
When Michal Jordan retired in 1998, many believed the
absolute level of talent in the NBA declined. [A picture of
Michael Jordan wearing number 45.]
27

SECTION 3
Demand Theory in Sports
Fans want to share in the excellence, competition,
identification, and the vicarious “thrill of victory and agony of
defeat” associated with sports teams. They are willing to pay
in order to do so. The economic concept of demand quantifies
willingness to pay and the scarcity implicit in that willingness.
Demand is simply the relationship between prices and
quantities demanded. But let’s not forget that willingness to
pay—and all of the elements that determine willingness to pay
—is also captured in this deceptively simple concept. For the
rest of this chapter, we will use attendance to measure the
consumption of the scarce items in sports events. In later
chapters, we’ll expand to other measures, but for now will
stick to just attendance.
Suppose we want to discover the relationship between price,
or willingness to pay, and the number of season tickets to your
school’s men’s and women’s basketball teams that would be
purchased by the people in your class at different prices. The
data points that we would collect, relating prices and
quantities of attendance demanded, are called a demand
schedule. A typical pair of demand schedules for men’s and
women’s basketball looks like the pair shown in Table 2.2.
Graphing the data shown in Table 2.2 (remember, in
economics the convention is to plot price against quantities
demanded), one finds the demand functions for men’s hoops
and women’s hoops shown in Figure 2.2.
DEMAND DETERMINANTS
The two-dimensional graph in Figure 2.2 can show only the
relationship between two things. The demand for season
tickets to men’s basketball and women’s basketball shows only
how many tickets will be purchased at a variety of prices.
Thus, if ticket price changes, movement along the given
demand function shows changes in quantity demanded. Prices
and quantities are inversely related; as the price of tickets
increases, the quantity demanded decreases. However, what
happens if something else changes?
Changes in other factors will cause a change in demand itself.
What other factors determine the demand for attending the
two types of basketball for this particular group of buyers?
There are five main categories of demand shifters:
• Preferences, or sports fan tastes (especially for team quality
and outcome uncertainty)
• Fan income
• The price of other goods fans enjoy (especially
entertainment substitutes)
• Fan expectations about the future
28

• Population in a team’s city
FAN PREFERENCES
Identifying the source of fan preferences for sports raises
questions of both the “nature” and “nurture” variety. The
natural joy of coordinated physical activity is apparent in the
youngest of children. Are people competitive by nature, or is it
a learned behavior? Certainly, experience contributes to our
preferences for sports. It appears that fans love the games that
they learned to play while growing up (baseball, basketball,
soccer, football, and hockey) and continue to play as they age
(tennis, golf, and bowling). Participation among high school
students in various sports is shown in Table 2.3. Clearly,
once grown, American men and women are fans of the games
they played when they were younger.
The relative number of people who have experience in
different sports helps explain why it is only very recently that
soccer (what the world calls football) has become an
important spectator sport in the United States. Experience
may also help explain why more sports fans seek the level of
excellence and competition offered in men’s sports rather
than in the same sport played by women. Not that long ago,
interest in women’s basketball was so low that athletic
departments did not even charge fans for admission! This
would be one part of the explanation for the shape and
location of the demand for women’s hoops and men’s hoops
in Figure 2.2.
It is essential to remember that preferences can change.
Judging by the increasing fan interest in women’s basketball
at both the college and professional level, as experience with
women’s sports has increased, so has the attention of fans.
Thus, one would expect that the demand for women’s
basketball has been increasing over time, shifting to the right
and outward toward the demand for men’s basketball. The
recent staying power of women’s professional basketball, the
Women’s National Basketball Association (WNBA), stands as
testimony to the importance of changing preferences.
Quality is one of the more important preference elements in
the determination of demand. Although all of the other
elements matter, they can be overcome by the intensity of
preferences for quality. All fans enjoy higher quality relative to
lower quality, but the intensity of that desire is reflected in
willingness to pay. Fans in two different locales may be
identical in all other regards, but if one set of fans puts a
higher premium on winning than another, then they will have
a greater willingness to pay for higher quality. This greater
willingness to pay will be reflected in a greater demand than
enjoyed at the other location.
It is a simple extension of this idea to see that preferences for
quality can actually overcome other demand factors when
comparing two locations. Even though the population may be
larger in one city, fans in another city may have such intense
preferences for quality, and willingness to pay for it, that their
team is competitive even though it has a smaller population.
29

For example, the NFL’s Green Bay Packers come to mind.
We’ll see the details of how this happens later in the chapter.
To emphasize an earlier point, a second important preference
element is outcome uncertainty. Again, while the data show
mixed results on whether and how outcome uncertainty
impacts demand, clearly it is a preference for just how
competition actually occurs. And since it is a preference
element, theory does not predict when it will matter or not,
either by sport or among fans in a given sport.
One final note about preferences: Whether they are a matter
of nature or nurture doesn’t matter to economists, who simply
take preferences as given. This bothers people who are more
interested in the question of preference formation than in the
question of how preferences affect choices. From the
economic perspective, if preferences for a particular sport
change over time, then the demand for that sport will shift. A
change in preferences probably explains the increase in
demand for NBA basketball over time relative to the demand
for MLB baseball. NBA basketball and MLB baseball are
entertainment substitutes. This partially explains why fan
interest in basketball increased at the same time as a decline
in the interest in baseball. And, at least to hear MLB
marketers tell it, this was especially true of young African
American sports fans.
FAN INCOME
Income changes also cause sports demand functions to shift.
An increase in income can cause demand to either increase for
so-called normal goods or decrease for so-called inferior
goods. If incomes rose and some sports fans shifted their
purchases from college basketball to professional basketball
available in the same area, then the pro version would be
income normal and the college version would be income
inferior for this group of consumers. It is extremely important
to point out that any given sport can be a normal good for
some fans and an inferior good for other fans. After all,
income simply allows fans to pursue their preferences, and
preferences certainly vary between sports for a given fan and
among different sports for different fans. We will further
discuss the influence of fan income later in this chapter.
PRICE OF OTHER GOODS
Changes in the price of other entertainment alternatives also
shift sports demand functions. All entertainment options can
be considered as alternative consumption possibilities, from
opera to Little League. Even professional sports in different
leagues with different seasons can be substitutes. It all
depends on how you view the planning process for sports
fans. If they think over an extended period about how they
will spend their money, then the NBA and MLB can be
substitutes, even though they are played at different times
during the year. Sports fans confront trade-offs as they plan
their annual spending.
Whether demand decreases or increases with the price of
another good depends on whether a sport and its
entertainment alternative are substitutes or complements. If
30

the price of season opera passes (a substitute) fell
dramatically, some sports fans might well choose more opera
and reduce sports attendance. On the other hand, if the price
of a complement such as parking were to increase, fewer
people would attend sports.
FAN EXPECTATIONS
Expectations about future prices also shift demand itself. If
fans expect the price of a particular type of sports event to
increase in the future, their demand for that event might
increase today. Fans might consume more of it while it is
cheaper. One example of the impact of expectations is the
purchase of “lifetime” reservation rights. Some fans of pro
teams will pay for a personal seat license, or PSL, that
guarantees access to a particular location for a specified
period of time. PSLs vary in price by seat location and the
amenities offered for that particular type of seat. Fans would
pay this price if they expected that the price of obtaining the
same seat was at least as high as the PSL price. If they
expected that the price would increase even more, then the
license would be a bargain. Because fans can just wait and buy
seats on an annual basis, we are led to conclude that they
must be getting a price break on their seat, given their
expectation that prices will increase over time.
POPULATION
The impact of population on demand is simple arithmetic.
Suppose the class size in the example portrayed in Table 2.2
and Figure 2.2 was to double. With 200 students, there simply
would be more people willing to pay at every price (as long as
some of the added students were basketball fans!). Thus, as
population increases, so does demand. Recall that an increase
in demand is a shift to the right in the entire demand
function.
Table 2.4 lists areas with major league teams and the
number of teams, population, and per capita income for each
area. The areas are listed in order of population. Because all of
the factors that determine the shape and location of demand
functions are highly interactive, it isn’t always just the greatest
population or the highest income that leads to the highest
demand. However, you should be able to trace some of the
ideas we have just discussed. For example, the New York–
New Jersey area has the greatest population, the largest
number of teams, and the ninth highest per capita income of
the 41 listed areas. Green Bay, at the other end, has the
smallest population, only one team (the NFL’s Packers), and
one of the smallest median family income levels, well below
average.
Although there is quite a bit of variation in Table 2.4, a look at
the top 10 and bottom 10 cities proves insightful. The top 10
population areas average just over 4.0 million in population,
4.6 teams, and median family income of $70,833. The bottom
10 population areas (all below the average of median family
income, except for Raleigh) average 595,164 in population, 1.5
teams, and median family income of $61,486. Based on these
averages, the top 10 population areas have 6.8 times the
31

population, 1.2 times the income, and 3.1 times as many
teams as the bottom 10 areas.
However, population is not static. Some areas grow in
population while others decline. Because demand represents
willingness to pay, owners are quite interested in the demand
prospects in different locations as population changes. Given
the limited number of teams, areas that are growing faster
represent greener pastures for current sports team owners.
Therefore, population also helps determine the migration of
teams as owners seek areas with greater willingness to pay.
32

SECTION 4
Lessons from Demand
Let’s pause and reflect on what you have learned thus far. The
first important concept, to which we will often return, is that
nearly all of the money in sports comes from fans. The single
qualification to this statement concerns the money
transferred through the political process. Typically, this
money comes from nonsports fans and goes to fans, owners,
players, and others who earn income from the presence of
sports teams in a given location. We will spend a significant
amount of time on this exception in Chapters 10 and 11. As for
the rest of the money, even though it is collected from a
variety of sources, such as ticket revenue, broadcast revenue,
shares of stadium revenues, sponsorships, and the sale of
licensed merchandise, the source is fans willing to pay for
sports. In the next chapter, we’ll talk extensively about
advertising revenue, and even that comes from fans.
MARKET POWER
Perhaps the first thing you noticed about the demand
functions in Figure 2.2 is that they are negatively sloped; that
is, the functions slope downward. As price increases, fewer
season tickets are demanded. That these demands slope
downward implies that there are substitutes, although not
perfect ones, for attendance at these college games. In the
case of perfect substitutes, the market is perfectly competitive.
Therefore, it follows that the demand curves in Figure 2.2
come from a less than perfectly competitive situation. To see
this distinction based on economic competition, let’s think
about the substitutes for attendance.
Consumers view all manner of other entertainment as
substitutes for sports. Opera, theater, movies, television
shows, and concerts—all of these activities entertain us and
are substitutes for sports. In fact, some sports are even
substitutes for each other. For some fans, men’s basketball
and women’s basketball may well be substitute goods. If there
were perfect substitutes for sports attendance, one would not
find the demand functions sloping downward, as in Figure
2.2. At some given price for season tickets, if prices were
increased, fans would simply turn to the perfect substitutes
for men’s basketball or women’s basketball. In this case, the
athletic department would have no power over the price they
could charge. This would be a perfectly competitive situation.
But why aren’t these other nonsports entertainment
alternatives good substitutes for sports? In his book Sports
Marketing (1999), Matthew Shank offers one explanation:
Sports are the most spontaneous of all entertainment:
A play has a script, and a concert has a program, but the
action that entertains us in sport is spontaneous and
uncontrolled by those who participate in the event. When
we go to a comedic movie, we expect to laugh, and when
33

we go to a horror movie we expect to be scared even
before we pay our money. But the emotions we may feel
when watching a sporting event are hard to determine. If
it is a close contest and our team wins, we may feel
excitement and joy. But if it is a boring event and our
team loses, the entertainment benefit we receive is quite
different. (p. 3)
Put another way, the character Mimi will always die in the
opera La Boheme (unless it is shamelessly altered as it was in
the stage play Rent). But, as the old sage goes, “On any given
day, any team can beat any other.” That, after all, is why the
games must be played!
Economists seldom delve into such explanations, but they are
important in determining the shape of demand. Figure 2.3
shows the variety of shapes for demand functions based on
the closeness and availability of substitutes. Suppose the
demand function were like D1. If a team raised its price by
even a penny, it would lose all of its fans. This is the perfectly
competitive demand function. If there were absolutely no
substitutes, the attendance demand function would be
vertical, like D2. Even if a team raised its price substantially, it
would lose no fans. But for the teams depicted in Figure 2.3,
the case is in between, more like D3. There are substitutes for
sports attendance but not perfect substitutes. It is this last
type of demand that characterizes sports attendance.
In sports, this market power is derived from the fact that
teams occupy exclusive geographic territory. At all levels of
sports, even down to youth sports, there is a decision-making
body that decides the number and location of teams in the
organization. In baseball, for example, organizations such as
Little League, Protect Our Nation’s Youth (PONY) League,
and Babe Ruth Baseball all grant specific locations to a certain
number of teams. Because there are many competing youth
sports organizations, which may sometimes include local park
departments, there is little reason to suspect that any of these
organizations are able to exercise much power over the price
that they charge local sports organizations to belong.
Can the same thing be said as we move up the sports ladder?
The Professional Baseball Agreement (PBA) governs the
relationship between the minor leagues, organized as the
National Association of Professional Baseball Leagues
(NAPBL), and MLB. The PBA also governs the relationship
between all of the various levels of minor league baseball,
from short-season “rookie” leagues to AAA leagues, the level
just below MLB. Although the PBA is complicated, the final
result is very clear. Each league can assign exclusive
geographical rights to a particular team. In the event of
territory conflict between leagues, the PBA gives a clear
prescription on how to settle the issue.
Outcomes under the PBA are well-defined, exclusive rights to
a particular territory for a particular team. No other team may
enter the territory and compete, economically speaking, with
the exclusive franchise holder. In the case of minor league
baseball, the lack of substitutes is engineered under the PBA
to make sure that the franchise holder is the only game in
34

town. Although there are some other entertainment
substitutes for the local team, there are no other professional
baseball substitutes. Here is an example with which I am
familiar from some consulting work in the past.
In 1993, the AAA Salt Lake City Gulls folded and were
replaced by the Salt Lake City Trappers. The Trappers were
members of the Pioneer League, a short-season rookie league.
They were phenomenally successful, hiring pretty good talent
in a weak league, developing a large following, and even
making the national news for the longest winning streak in
professional baseball (a record that stands to this day).
The Portland Mavericks, playing at the AAA level just below
Major League Baseball, weren’t faring nearly as well. The
Mavericks played in an oversized stadium in front of few fans.
Under the rules of the PBA, all the owner of the Mavericks
needed to do was inform the Trappers that the Mavericks
would be taking over their territory. Under those same PBA
rules, however, the Mavericks were required to compensate
the Trappers, but the two teams could not reach an agreement
as to the amount. In this case, the PBA rules dictate a formal
arbitration hearing. Parties represent their cases and agree to
live with whatever the arbitrator decides. Eventually, nearly
$2 million in total compensation went to the Pioneer League,
and the Mavericks moved to Salt Lake City. Under the
arbitration settlement, the Pioneer League was allowed to
place a team (the Raptors) in nearby Ogden, Utah, but the
Trappers could no longer operate in Salt Lake City. This was
because a Pioneer League team too close to the Mavericks in
Salt Lake City would compete for fans.
The highest-level professional leagues have complete control
over the number and location of teams. The league decides
when to expand, and rules for team movement require more
than a simple majority vote among league members. Leagues
(and the NCAA in the college case, as we’ll see in Chapter 13)
offer a number of explanations for their exclusionary
practices. Essentially, the leagues’ arguments all boil down to
quality control—major leagues determine who gets to count
themselves as major. Regardless of the justification, the result
is that the number of close substitutes is reduced and
individual teams are endowed with some market power.
Through careful management by the league, the location and
number of teams are controlled over time (one of the subjects
covered in Chapter 5).
Consider the recent expansion of the NFL. The NFL’s decision
came down to two candidate host cities, Los Angeles and
Houston. The league chose the latter, guaranteeing the
Houston area to a single NFL team. In the meantime, fans in
the Los Angeles area are without an NFL team. Surely, if the
world were a bit more economically competitive, another
league would have put a team in Los Angeles, the second
largest market in the United States (Table 2.4). As a result, the
only real chance for economic competition facing teams in a
given league comes from outside the sport. We’ll see in-depth
how this control affects league outcomes in Chapters 5 and 6
35

and just why it is that they continue to have this control in
Chapter 12.
Given this careful management of team location to ensure
market power, how can we explain multiple teams from the
same league operating in the same location? There is a two-
part explanation. First, there is the simple power of numbers.
Some cities have more than enough people for the economic
survival of more than one team. For example, in Table 2.4, the
smallest population area supporting an MLB team is
Milwaukee (808,367). Thus, the greater Southern California
area clearly can support more than one team, as can the San
Francisco Bay area in Northern California, the New York/New
Jersey area, and Chicago.
Second, just because artificial boundaries define a particular
city or metropolitan area, that doesn’t mean that these areas
are just one big happy single entity. Chicago’s North and
South Sides, for example, are economically and ethnically
distinct. So are the boroughs of New York City. San Francisco
and Oakland couldn’t be more distinct in terms of their
economic situation. As the old joke among San Franciscans
goes, there’s a good reason why they collect the Bay Bridge toll
as you leave Oakland. Nobody would pay to get in. And
Oakland residents probably feel the same way about San
Franciscans.
The upshot of all of this is that attendance demand for any
particular sports team, at any level, slopes downward. This is
the definition of market power. A firm has market power
when it has few close substitutes for its product. A sports team
does not have to be a technical monopoly in the strictest
sense, literally the only game in town, in order to have market
power.
PREFERENCES AND DEMAND FOR WOMEN’S SPORTS
Let’s return to Figure 2.2. It is clear that the demand for men’s
basketball tickets is greater than the demand for women’s
basketball tickets—at every price, a greater quantity of men’s
than women’s tickets is demanded. What factors explain this
difference in demand? The number of students is constant,
and the rest of the factors influencing demand (income, the
price of other goods, and expectations) do not change when
turning from the demand for men’s hoops to the demand for
women’s hoops. The answer lies with preferences. We will
spend quite some time on sex and race issues in sports in
Chapters 7 and 13. The important lesson here is that
preferences are at the heart of this difference in demand.
PREFERENCES, PLAYER INCENTIVES, AND
PERFORMANCE ENHANCEMENT
For a given absolute level of play, fans demand the highest
relative competition they can afford. As we will see in depth in
Chapter 7, it is willingness to pay by fans that partly
determines the values that players hope to earn. If fans value
relative competition, players will invest in ways to make
themselves relatively more competitive. Enter performance
36

enhancement and its most recent troubling variation,
performance-enhancing drugs.
Make no mistake about it. Ever since there was a reward to be
had, athletes have sought performance enhancement. And the
devotion to those efforts has been pushed by the size of the
rewards. Believe it or not, just a few decades ago training was
a part-time endeavor for professional athletes primarily
during the season. All but the highest-paid players also had
off-season jobs. That was the original reason for required
preseason workouts (e.g., baseball’s spring training)—physical
preparation for the upcoming season was needed after a long
off-season of inactivity. Of course, in the modern context with
tens of millions of dollars per year riding on physical
preparedness and relative physical advantage, athletes now
devote full time to their sport. But note that is because the
rewards have risen so dramatically! In the 1960s, many pro
athletes made tens of thousands of dollars, by comparison.
Some approaches to enhancement are viewed almost
universally as acceptable—full-time general physical training,
specialized weight training, and adherence to performance
diets. Indeed, this type of approach is often viewed as
exemplary behavior—the most dedicated players earn the
fruits of strenuous preparation. Even working out in
hyperbaric chambers that cheat nature by simulating actual
environmental conditions that will occur under competition
are viewed as acceptable.
Interestingly, other performance enhancement approaches
are not so universally acceptable. For example, equipment in
sports is typically regulated, presumably because performance
should be about the athletes, not the equipment they use.
Recent high-performance swimsuits fit this case.
Performance-enhancing drugs also fit this case. Now, just why
it should be the case that hyperbaric chambers that cheat
nature are acceptable but performance enhancing drugs are
not, from the economic perspective, only goes to show that
this is strictly a preference issue. Indeed, some fans do not
care at all whether athletes use performance-enhancing drugs.
More on this controversial issue will be discussed later in
Chapters 4, 6, and 7.
Another point of economic interest is completely ignored in
the debate over performance-enhancing drugs (in my humble
opinion). Most of the focus is on the impact of performance-
enhancing drugs on relative competition; the unfair advantage
gained by some players over others. But if enough players take
the drug route to performance enhancement, eventually
absolute competition is raised as well. The crucial insight
from the economic perspective is that some fans will be
willing to pay more for this higher level of absolute quality.
Thus, the use of performance-enhancing drugs creates value
to some fans via their enjoyment of higher levels of absolute
competition. Whatever the values of intervention against the
use of performance-enhancing drugs, a complete assessment
of the benefits and costs of such intervention must include the
reduced value of absolute competition to some fans.
37

CHANGE IN DEMAND VERSUS UPWARD SLOPING
DEMAND
Note the difference between a change in the quantity
demanded and a change in demand itself. It sometimes seems
that the more expensive an event becomes, the greater the
number of people who attend. Before jumping to the
conclusion that demand slopes upward, remember that a
given demand curve can change over time. The more likely
outcome is that demand simply has increased over time and
one has observed two different prices on two different
demand functions.
The law of demand states that consumers buy fewer units
when prices rise, but the typical observation is that both
attendance and price increase over time. Indeed, the data
show that this has been true over some time periods. Covering
the period 2000-2007, in Table 2.5 the correlation is
between attendance and prices (adjusted for inflation). Note
that the majority of values are positive and just as strong as
the negative correlations. These data appear to be consistent
with the idea that higher prices go along with greater
quantities.
However, our analysis of demand suggests a different answer.
Be careful to distinguish between changes in the quantity
demanded and changes in demand itself. To assume that the
relationship between attendance and price over time
represents a single demand function is to assume that no
other demand shifters have changed over time. That seems
pretty unreasonable. Instead, it’s highly likely that income,
the price of other goods, preferences, expectations, or
population might have changed over the eight years in the
sample. Among these, it is most likely that an increase in
income or population would lead to higher prices and
attendance over time. The confusion is shown in Figure 2.4.
At one time, an observed attendance–price pair might be (A1,
P1) on demand function D1. At another time, after any number
of demand shifters may have changed, an observed
attendance–price pair is (A2, P2) on demand function D2. It
would be careless to just connect the two points, as the dotted
line in Figure 2.4 does, when the two different points might lie
on different demand curves.
DEMAND PERSPECTIVES ON “LARGE” AND “SMALL”
MARKETS
Population is part of the explanation for what is known as the
large market–small market distinction in sports. We’ll return
to this distinction throughout the remainder of the text, but
here is a good place to introduce it. Historically, large-market
teams win more often than small-market teams. Part of the
reason has to do with the fact that teams with larger
populations to draw from have the opportunity to earn higher
revenues for any given level of quality. This means that teams
in larger markets will buy better players because they can earn
higher revenues from those players than teams in smaller
markets could earn. Although it is revenue that ultimately
38

drives this wedge between small- and large-market teams,
population is one of the determinants of revenue differences.
But remember this: Population is not the only factor that
determines the demand for sports in a given location. A team
owner might consider moving to another location, even if its
population is smaller, as long as other factors such as income
and preferences result in greater demand. This is especially
true if this greater demand manifests itself through a city’s
offer to build a new stadium or arena to house the team.
To see this point plainly, return to the case of Green Bay in
Table 2.4. It is clearly the smallest population area in all of
sports (the next lowest is the Raleigh-Cary area, 3.3 times
larger than Green Bay). In addition, Green Bay’s median
family income of $61,644 is well-below average. Often, the
NFL’s Green Bay Packers are referred to as a “small-market
miracle” because of the team’s historical NFL success in the
face of these limitations. But a closer look, along the lines of
demand-driven spending by fans, reveals this isn’t really the
case. Forbes annual reports on sports team finances (all
available and cited at the web page) put the Packers’ 2007
revenues at $218 million, 16th in the league of 32 teams, just
below the league average of $222 million, but well in excess of
the smallest revenue of the Minnesota Vikings at $195 million.
Since income and population are both quite low in Green Bay,
this fan spending result must be due to fan preferences, lack
of substitutes, and management and coaching. But clearly
Green Bay is not a “small” market from the perspective of
demand!
39

SECTION 5
Willingness to Pay and
Consumers’ Surpluses
The market demand function for sports reveals consumers’
willingness to pay. For example, in our derivation of the
attendance demand functions for men’s basketball and
women’s basketball in Figure 2.2, willingness to pay is higher
for the men’s version for every single level of season tickets.
WILLINGNESS TO PAY
In Figure 2.2, the demand curve for women’s hoops shows
that the most any person is willing to pay for the fifth
women’s ticket is $40. The demand curve for men’s hoops
shows that somebody would be willing to pay about $65 for
the fifth men’s ticket. Some people would be willing to pay
even more than that for the first through fourth tickets for
each type of basketball output. Moving farther down along the
demand functions, some fan is willing to pay only $10 for the
70th women’s season ticket, whereas the greatest willingness
to pay for the 70th men’s ticket would be $20.
These marginal changes in willingness to pay show us one
measure of value of season tickets to fans. At the margin, the
amount given by the two demand functions allows us to
compare the dollar value that fans put on the two different
versions of basketball. There is considerable variation in even
this simple example. At the end of this chapter, we’ll examine
revenues within a sport and across the four major pro sports
leagues. Considerable variation in willingness to pay,
measured by actual revenues collected by teams, exists in all
sports.
CONSUMERS’ SURPLUSES
A closely related idea concerns just how happy consumers are
with their consumption. A dollar measure of this welfare is
called consumers’ surpluses. Consumers’ surpluses are the
sum of the difference between willingness to pay and the
actual market price on all units of consumption. Figure 2.5
demonstrates consumers’ surpluses for the demand for
women’s season tickets. In the figure, the shaded area shows
the consumers’ surpluses for women’s tickets purchased at
$20. Clearly, all 40 buyers are willing to pay more than $20
for each season ticket. The consumers’ surpluses for women’s
tickets totals $450 (just add up the differences between
willingness to pay on the demand function and the actual
price of $20). Remember, these values are dollar measures of
welfare. In the most real sense of the word, they represent
precisely how much buyers would be willing to give up to
enjoy season tickets over and above the price, in dollar terms.
It is also possible to put a dollar measure on the change in
welfare associated with a change in the price of tickets. This
idea is demonstrated in Figure 2.6 for women’s season
tickets. At a price of $20, we’ve already seen that the
40

consumers’ surpluses are equal to the shaded area below the
demand function and above the price of $20, totaling $450.
Now, suppose the price falls to $10 per ticket. Intuitively,
buyers will be happier at a lower price. We can use consumers’
surpluses to attach a dollar value to that increased welfare.
The first part of the increased welfare concerns the savings
that occur for those purchasing the original 40 tickets. Each
ticket now costs $10 less. That’s a dollar measure of increased
welfare equal to $400, or the area in Figure 2.6 at the bottom
left filled in with vertical stripes. The second part of the
increased welfare occurs because the number of tickets
demanded increases when the price falls to $10. In fact, 30
more tickets would be purchased at the lower price. The
dotted area shows the consumers’ surpluses on these tickets.
A careful calculation of this area using the data in Table 2.2
yields an increase of $150. All told, the increased welfare
enjoyed by buyers of women’s tickets when the price falls to
$10 is equal to $550. Added to the $450 already enjoyed
before the price fall, that’s a total of $1,000.
Knowing the level of surpluses is interesting enough. But
throughout this book, we will use surpluses to analyze a
number of important sports issues. For example, what could
be more important to team owners than the fans’ willingness
to pay? Suppose owners decide to raise ticket prices.
Consumers’ surpluses give a measure of how this makes ticket
buyers worse off. From the policy perspective, if a team is
leaving a location, consumers’ surpluses give one estimate of
the losses to some fans. Another example of the usefulness of
surpluses is in the analysis of the impact of market power on
consumers, which is covered in Chapters 5 and 12. Also,
owners of sports teams often demand subsidies from local
communities. Elected officials might want to know if their
constituents find a particular subsidy level to be worth it.
Surpluses are one way to estimate willingness to pay for
subsidies. We’ll examine this issue in Chapter 10.
41

SECTION 6
Price Elasticity, Total
Revenue, and Marginal
Revenue
The price elasticity of demand is one of the most important
bits of information captured by the demand function. In this
section, we develop the idea generally for use in subsequent
sections to gain insight into particular sports economics
issues, including attendance pricing by teams.
PRICE ELASTICITY
Let’s stick with look at gate attendance, denoted A. From basic
economics, the price elasticity of demand relates the
percentage change in consumption to a percentage change in
price. It is free of units of measurement. The elasticity of
attendance for a sports event can be expressed as follows:

Because ∆A/∆P is always less than zero for a negatively sloped
attendance demand function, the negative sign in the
elasticity expression guarantees that we are only talking about
the magnitude of elasticity. Notice that the elasticity of
attendance demand contains the slope of the demand
function, ∆A/∆P, and depends upon where the starting price
and quantity for the elasticity calculation, P0/A0.
The elasticity of demand is not constant over the entire range
of attendance. At the attendance level associated with the
linear midpoint of demand, ? = 1. The percentage change in
attendance is just equal to the percentage change in price;
demand is called unit elastic at this attendance level. At
attendance levels below the unit elastic level, the percentage
change in attendance exceeds the percentage change in price,
? > 1. Here, demand is called elastic. At attendance levels
beyond the unit elastic level, the percentage change in
attendance is less than the percentage change in price, ? < 1. In this range, demand is called inelastic. Keep these ideas about elasticity in mind when we discuss revenues. TOTAL REVENUE Total revenue is one of the two most important pieces of information that owners of a sports team must have. If we remember that demand slopes downward for sports teams with market power, then total revenue is defined as follows: Where R is total revenue, P is the price of attendance, and A is gate attendance. That price is a function of attendance incorporates the fact that sports team owners have market power. If the market were competitive, price would not depend on attendance. As you can see, total revenue, R(A), is inextricably linked to demand, P(A). If sports team owners 42 know their demand function, then they know their total revenues for any level of attendance that might occur. Even though demand is usually a curve, it is easy to work with the linear version. It supports everything we will do in this book. Such a demand function is defined as follows: This demand function is in y-intercept form, with price on the y-axis. The parameter a is the y-intercept and parameter –b (with its negative sign) is the slope of the linear attendance demand function. Figure 2.7 shows a generic attendance demand function, along with the elasticity regions described in the previous section. Let’s examine total revenue for this linear demand function. As just noted, total revenue is defined as price multiplied by attendance. Multiplying the linear price function by attendance, we get the following: This quadratic function is a parabola, concave to the x-axis, starting out at zero with a unique, positive maximum point. Thus, total revenues increase at first as price falls, but eventually, past some level of attendance (A > a/2b), total
revenues fall back to zero.
MARGINAL REVENUE
Important economic intuition comes from the slope of the
total revenue function, defined as ∆R/∆A. You probably
recognize this as the definition of marginal revenue, or MR(A)
for short. If we go ahead and calculate this from the revenue
function, we find the following (taking the first derivate):

Notice that the only difference between marginal revenue and
the demand function is that marginal revenue is twice as
steep. Figure 2.8 shows the relationship between demand,
marginal revenue, and total revenue. In addition, based on
our findings in the previous section, regions of the elasticity of
demand also are identified in Figure 2.8.
Another important thing to notice is that the following three
things all occur at attendance level A0 in Figure 2.8:
• R(A0) is at its maximum level
• MR(A0) = 0
• ? = 1 (i.e., demand is unit elastic)
Given Figure 2.8, it is easy to tie some economic insight to the
mathematics. In the elastic range of attendance demand (? >
1), MR(A) is positive but falling. Even though price falls,
attendance increases enough in percentage terms to increase
revenue. But because marginal revenue is falling MR(A) is
positive but getting smaller in this region of demand, total
revenue will continue to increase but at a decreasing rate.
Once attendance reaches the unit elastic point (? = 1) at A0,
43

marginal revenue is zero. No addition to total revenue is
made, and we must be at the maximum point of total revenue.
Once into the inelastic region of demand, added attendance
no longer is large enough to overcome the decrease in price, in
percentage terms. Marginal revenue is negative, and total
revenue must start to fall away from its maximum. As price
falls even more, marginal revenue becomes increasingly
negative. Total revenue must fall at an increasing rate to zero.
MARGINAL REVENUE, PRICE, AND ELASTICITY
There is one last concept to develop that will be useful later in
the chapter. Generally speaking, price is just a function of
attendance, say, P = P(A). Total revenue is just price
multiplied by quantity, that is, R(A) = P(A) × A. Marginal
revenue is the change in total revenue when output increases
by a unit. Using the basic chain rule for a derivative,

If we pull a P out of the right-hand side, then we get the
following:

The first term inside the parentheses is the reciprocal of the
negative of the elasticity of demand. Remember? The
elasticity of demand is as follows:

In summary, we can write the following:

This is just another outcome that demonstrates how prices,
revenues, and the elasticity of demand are inextricably linked.
Remember this result for later use when we discuss price
discrimination.
44

SECTION 7
Lessons from Total and
Marginal Revenue
Let’s review some important insights from what you have
learned so far. The relationship between elasticity and total
revenues offers important insights for team owners. Owners
may only know a few price and quantity pairs for a very
limited portion of their demand function, but elasticity can
always be calculated. The result of that calculation reveals two
important things: where the owner’s pricing decision lies on
the demand function and the impact of changing price.
For example, suppose you own a sports team and wonder
what the effect on your revenues would be if you raised your
ticket price. Your research staff investigates ticket pricing for
the team. There isn’t much to go on because prices really
haven’t changed much over time. But the small amount of
information they do have allows them to calculate the
elasticity of demand. It ends up at about 1.6. What will
happen to revenues if you raise the ticket price? Refer to
Figure 2.9 and you will remember that revenues will
decrease if you increase your price in the elastic portion of
demand. You don’t have much to go on, but at least
calculating elasticity gives you a clue as to what will happen if
you increase the ticket price.
PRICING IN THE INELASTIC REGION OF DEMAND
Here’s another insight from the relationship between
elasticity and revenue. I’ll pose it as a puzzle. The economics
literature that analyzes the demand for attendance
consistently finds inelastic ticket pricing. But in the inelastic
portion of the attendance demand function, marginal revenue
is negative. Reducing output will increase total revenue. In
Figure 2.9, attendance level A1 is in the inelastic portion of
demand. Notice that its associated marginal revenue, MR (A1),
is less than zero. As output is reduced toward attendance level
A0, total revenue increases. Even though we haven’t developed
the cost side of sports, it is intuitive that costs fall as
attendance is reduced (if for no other reason than fewer
service personnel are needed to serve fewer fans). So if
revenues are rising and costs are falling, profits must be
increasing. The puzzle is just why owners would price
attendance in such a fashion if by increasing price they can
increase profits. But pricing in the inelastic portion of demand
is just what demand analysts have discovered.
Of course, one explanation for this could be that owners don’t
care about profits. If this were true, then pricing where
marginal revenue is negative, that is, MR(A) < 0 can happen based on owner preferences alone. But if owners care about profit, like all good businesspeople do, then such a pricing choice can only make sense if there is some other, nongate revenue gained as an offset. There certainly are other types of revenue tied to attendance. If an owner lowers ticket prices to increase attendance, then attendance revenue falls in the 45 inelastic range of demand. However, parking, concessions, and licensed merchandise sales may more than make up for the reduction in gate revenue. The owner will continue to reduce ticket prices as long as the added revenue from parking, concessions, and merchandise sales makes up the difference. With these other revenue sources, ticket pricing in the inelastic portion of attendance demand is consistent with profit pursuits. One simply must count all of the sources of revenue. CHANGES IN DEMAND AND CHANGES IN TOTAL REVENUE There is one more essential observation to be had here. What happens when the demand function shifts? Let’s suppose that income or population increases or the price of parking falls or prices for sports are expected to increase in the future. The demand function would shift to the right, say, from D0 to D1 in Figure 2.10, so that greater attendance would be demanded at every ticket price. What does this do to total revenue? Clearly, total revenue must increase, as illustrated in Figure 2.10. As the revenue function expands from R0 to R1, both the range along the x-axis and the new maximum of total revenue increase. Why is A1 > A0? Remember, the maximum of total
revenue occurs at the unit elastic level of output, and that
point on the demand curve has increased from A0 to A1.
Putting all of these pieces together, you can probably fill in
the last insight. A common lament is that a sports team would
be much better off if it could only increase attendance. As
developed here, increasing attendance can mean one of two
things, either increasing the quantity of attendance demanded
or increasing demand for attendance itself. Let’s look at each
of these.
Does it make sense to increase attendance along a given
demand curve? Let’s refer back to Figure 2.9 and assume that
stadium capacity is at fixed size, A1. Now, given capacity A1,
does it make sense to fill the stadium? Suppose that ticket
revenue increases at a greater rate than that at which other
revenues are lost. In that case, the answer would be no. If a
team inherits such a capacity, it will never admit more than
the revenue maximizing level of attendance, A0. In fact, as
long as costs increase with output, the team will operate to the
left of the maximum revenue point. Teams like this will
simply have excess capacity as long as profits drive their
decisions.
CAN TEAM OWNERS SHIFT THEIR DEMAND
FUNCTION?
Another way to increase attendance would be to try to shift
the demand function to the right, rather than lowering price
along a given demand. If demand shifts right, both the range
and the maximum of total revenue shift to the right. If the
increase is large enough, stadium capacity may be below the
revenue maximizing level of attendance. Even though profits,
not revenues, drive decisions, there is at least a chance that
filling the stadium can make sense. One surefire way to
increase demand is to increase the quality of the team—lay out
46

enough money to bring a winner to the fans. But as we’ll see in
subsequent chapters, this can be risky if demand doesn’t
increase enough. Other ways of increasing demand are well
known. Many of them were created by one of sports’ most
interesting entrepreneurs, Bill Veeck (see the Learning
Highlight: Bill and Mike Veeck at the end of this section).
STADIUM RECONFIGURATION
Another approach to excess stadium capacity is simply to
reconfigure the stadium—make it smaller. This idea doubtless
is behind the downsizing of stadiums in MLB as the cookie-
cutter facilities of the early 1970s have been replaced. The
Ballpark at Arlington for the Texas Rangers, Jacobs Field for
the Cleveland Indians, and Safeco Park for the Seattle
Mariners all followed the important example set by Camden
Yards for the Baltimore Orioles. All seat in the 40,000 to
50,000 range, down dramatically from the monsters they
replace. More recently, the same is true of the new Yankees
and Mets stadiums in New York. In addition to getting out
from under other types of costs, these capacities are more in
keeping with the market characteristics of these areas.
SELLOUTS AND OTHER ODDITIES
There are other interesting puzzles, such as the phenomenon
of sellouts. First, from our discussion thus far, it’s unlikely
that sellouts ever should happen because stadium capacity is a
short-run outcome. For sellouts to occur, profit maximization
would have to coincide with a very particular level of
admissions. There is the additional question of why it would
make sense to price tickets so that waiting in line is the norm,
either for any given game or for years in the case of some
season tickets. Filling the stadium is one thing, but filling it at
a lower price than you could otherwise get begs some
extensions of the analysis. We’ll get to that analysis in
subsequent chapters.
47

LEARNING HIGHLIGHT: BILL AND MIKE VEECK
As owner of the Cleveland Indians, the St. Louis Browns, and
the Chicago White Sox at various times from 1946 to 1980,
Bill Veeck (rhymes with wreck) was a marketing genius who
loved the game of baseball. Veeck introduced to baseball such
attractions as ladies night, straight-A night (kids just had to
bring their report cards), and scoreboard fireworks. On
grandstand management night, fans “helped” the manager
make important calls like pitching changes by holding up
cardboard signs supplied under their seats. Perhaps his most
memorable gambit was signing the adult, 3′ 7″, 65 lb., Eddie
Gaedel to a major league contract and sending him to the
plate as a pinch hitter in a game between his St. Louis Browns
and the Detroit Tigers in 1951. Gaedel drew a walk and was
pulled from the game. His career still stands as the shortest
one in MLB, according to his listing in The Baseball
Encyclopedia.
Following in his father’s footsteps, Mike Veeck built the St.
Paul Saints into one of the most profitable independent minor
league teams in history, and right in the backyard of the MLB
Minnesota Twins! He once hired five mimes at $300 each to
act out instant replays (reported by Tim Kurkjian on
ESPN.com, February 4, 1999, “Baseball’s Back, from A to Z”).
“The crowd hated them, but we sold 42 hot dogs per paid
customer, and they threw everything they had at the mimes,”
Veeck recalls. “They left in the fifth inning crying.” Veeck also
served as marketing man for MLB’s Tampa Bay Devil Rays for
a spell. There, Veeck touted his team’s newest promotion,
lawyer appreciation day (Sports Illustrated, February 22,
1999, p. 34): “We’re going to charge them double, bill them by
the third of an inning, and generally berate them.” His father
would have been so proud.
Bill Veeck, the self-described baseball hustler, added a great
deal of business creativity to the sport of baseball. [Picture of
Veeck on the phone.]
Sources: Veeck (1962) and Sports Illustrated (February 22,
1999, p. 34).
48

SECTION 8
Elasticity and Price
Discrimination
Elasticity is the primary determinant of attendance-pricing
decisions by team owners. The most notable characteristic of
price discrimination is price variation between fans. In its
most basic form, price discrimination means charging
different people different amounts for the same consumption.
However, we must exercise caution in determining whether
price variation really is price discrimination in any particular
case. For example, a difference in price to different people
may reflect that it costs different amounts to serve different
customers. Or different fans may not actually be buying the
same quality experience even though they enjoy the event at
the same time. Thus, if quality or costs of service vary by
consumer, charging them different prices need not be price
discrimination.
In sports, the variation in price at an event based on seat
location is not price discrimination because different seats
have different qualities (view, closeness to the action, and
access to other amenities). It is also the case that, on average,
walk-up customers are more expensive to serve than fans who
buy their tickets in advance. In order to save seats for walk-up
buyers, owners must estimate the right number of tickets to
reserve for walk-up fans. If they guess wrong, seats may sit
empty. Thus, walk-up fans must be charged the implicit cost
of saving them a seat as opposed to selling it in advance,
including the risk of seats sitting empty.
One of the trickiest pricing decision from the price
discrimination perspective is season tickets. If the difference
compared with the price of single-game tickets were price
discrimination, we would expect to see two-for-one discounts
and discounts for multiyear advance purchases. But we don’t
see these types of offers. Further, teams actually facilitate the
resale of season tickets through centralized “official resellers”
like StubHub. But resale is typically thought to be anathema
to successful price discrimination. So, it is most likely that
season tickets are not price discrimination but, rather, simply
selling the additional right to occupy the same seat for an
entire season (sometimes rolled over to future seasons as
well). Further, season tickets often come with access to other
amenities, so they represent a different quality fan experience
and are not price discrimination.
However, the cost of providing sports does not vary by the
time of day, day of the week, or ages of fans. Neither does a
person who attends a baseball game between the same two
teams, sitting in a particular seat, in a particular section, and
on a particular day consume a different quality good when the
same consumption happens on another day. Variation in price
according to these factors is price discrimination.
TRUE PRICE DISCRIMINATION
49

Let’s look in detail at a case of true price discrimination,
namely, lower attendance prices for elderly fans. We’ll see that
the elasticity tool proves valuable once again. What is
different about elderly fans that would lead an owner to
charge them differently to sit side by side with younger fans at
the same game? Recall that the determinants of elasticity are
pretty much the same ones that determine willingness to pay
along a demand function—income, closeness and availability
of substitutes, expectations, and adjustment time. Because the
elderly have lower disposable income, they are much more
responsive to price than younger fans.
Using some of our previous results, we can see that owners
will charge more elastic demanders, such as the elderly, lower
prices. In an earlier section, we showed that:

Suppose for simplicity that an owner confronts two types of
demanders, the elderly, denoted E, and the rest of the younger
fans, denoted Y. It is clear that the owner will make as much
money as possible by selling seats so that:

So revenues are as large as possible when seats are allocated
between the two types of buyers so that the marginal revenue
from the last elderly fan through the gate equals the marginal
revenue from the last younger fan. (Suppose instead that
MRE(AE) > MRY(AY). In that case, the owner rearranges seat
sales away from younger fans toward elderly fans for a net
increase in revenues.)
Let’s substitute in for the MR earned from our two types of
buyers and set the two MR relations equal to each other. In
order to make as much money as possible, the owner will find
that

Clearly, if ?E = ?Y, then PE = PY. Unless there is some
difference in the elasticity of demand between the two types of
buyers, the owner will charge them the same price. But
suppose that ?E > ?Y, so that PE < PY. Elderly fans have more elastic demand than younger fans, and owners charge the elderly the lower price. Remember, this differential price outcome is based on willingness to pay for an identical product, not seat quality or differential cost of service. Does price discrimination describe any actual outcomes in sports? Elderly (and very young) fans pay less than the rest of the fans. They have lower disposable income and so have more elastic demand; an increase in ticket price will drive more of them away than the rest of the demanders. Price discrimination can also be used to sell weekday tickets. The value of foregone alternatives is higher during the week than on the weekend. Prices are lower for weekday buyers in the form of special deals off the regular 50 ticket price. This is especially true for daytime games on weekdays. LOWERING PRICES Another type of price discrimination is ticket deals that appear to lower the average price of a ticket. Roll-your-own multiple ticket packages during the season are of this variety. In order to get the seemingly lower unit price, a fan must buy a specified number of tickets. However, because demand slopes downward, subsequent tickets purchased at the same point in time are worth less than the first ones bought, and there really is no deal for the fan. The appearance is a lower average ticket price, but fans don’t get to see what price would be charged in the absence of the owner’s knowledge about willingness to pay. As always, be careful. Advance season tickets are not a form of price discrimination because they have a reservation of a given seat included in the price (and other amenities as well). This is a distinct quality difference. PERSONAL SEAT LICENSES While we’re on the topic of reserved season tickets, let’s examine a recent seat pricing phenomenon in pro sports, the personal seat license (PSL) (in Chapter 13, we’ll see that PSLs are quite similar to a very old college pricing approach, “donation” seating). Let’s suppose that an owner has already taken advantage of price discrimination and is charging accordingly. For some given fan, the result might be as shown in Figure 2.11. At price P*, from our previous development, the consumers’ surpluses for this fan are the triangle XP*Z. Suppose the owner discovers which fans, in general, enjoy this type of surplus. Relative to just selling tickets for P*, the owner might charge P* for the tickets and then charge an annual fixed fee equal to XP*Z. If you think this sounds just like a PSL on season tickets, it is. The owner has been able to extract just a bit more revenue from each fan. The additional fixed charge, as close as possible to XP*Z for fans like the one depicted in Figure 2.11 in addition to the ticket price, P*, will exhaust the fan’s willingness to pay. This two-part pricing mechanism is just an extension of price discrimination to collect even more of the consumers’ surpluses across all fans. Different fans pay different ticket prices and different PSLs based on willingness to pay. But as always, exercise care in this determination. Some of the difference might be due to seat quality or the value of reserving a particular seat forever where it might only have been a one-year right previous to the PSL, but the remainder is price discrimination. 51 SECTION 9 Revenue Data The pro team sports revenue data that are generally available appear in Forbes (earlier versions of the data were published in the now defunct Financial World) (We’ll treat college sports separately, in Chapter 13.) Revenues come from ticket sales at the gate, TV (from national contracts for all sports and from local contracts for a few sports), shares of league property sales (licensed merchandise and memorabilia), sponsorships, and from the nature of the team’s stadium arrangement. Some teams own their own stadium so that concession and parking revenues and revenues from any other use of the stadium belong to the owner. However, even if a team doesn’t own its own stadium, the same types of revenues can flow to a team, depending on how profitable an arrangement the team has made with its host city. We’ll get into the gruesome details of leases in Chapters 10 and 11. The data in Forbes come from a survey of team owners, who often are bargaining with their hosts for subsidies. Thus, it is in the owners’ best interests to understate revenues and overstate costs in order to convince their hosts of the need for such assistance. This means that we should view the survey responses with some skepticism. We could try to guess a team’s revenue by using other available data. Gate revenues can be “guesstimated” pretty easily from attendance and posted ticket prices. TV revenues are also commonly known and published in a variety of places. Shares of league merchandise revenues are less well publicized, and the only way to know the details of stadium revenues is to dig into actual stadium lease documents. However, leases are difficult to obtain and, perhaps, even more difficult to understand. So the survey data are the best numbers that we have. Table 2.6 and Figure 2.12 show a summary of the revenue data available from Forbes for 2007. Clearly, there are haves and have-nots. In each league, that is, there is significant within-league revenue variation. Those at the low end of the scale can have as little as 65 percent of the mean, such as in the NBA, whereas those at the top end typically have revenues of well over 1.5 times the mean. The difference between the haves and have-nots is approximately similar in all sports except the NFL, the league with the most equal sharing of revenues (a topic for Chapter 6). While the same could be determined from the data in Table 2.6, Figure 2.12 easily shows that there is substantial between- league revenue variation. Some sports are simply worth more to their fans than others. For example, at the respective league averages, football fans spend over twice as much on their sport as hockey fans spend on theirs (including the value of TV advertising). At the top end, football and baseball fans spend just over twice as much as hockey fans. With an 52 understanding of the basics of demand, this is no mystery. Fan preferences, population, and income situations drive these types of demand differences. It is perfectly in keeping with demand theory, for example, that New York can be a great baseball town (Yankee spending is tops for MLB) but not really a great hockey town (for the Islanders, spending is at the bottom for the NHL). Also in keeping with demand theory, it is important to remember that things change over time! In 1990, the Braves and Mariners were at the bottom of MLB’s revenue heap. By 2005, the Atlanta Braves ranked 12th in reveue among the 30 MLB teams, above the median. The Seattle Mariners had risen from literally last to 8th of 30! This revenue disparity is the source of a lot of tension in sports. Although the world of sports is an uncertain one, revenues translate into strong on-field performance when owners use revenue to purchase talent. Owners in large revenue market locations can earn higher revenues from a given level of talent than can owners in smaller revenue markets. As long as revenues are unbalanced, it is quite likely that there will be unbalanced talent choices between teams and unbalanced outcomes on the field. If these types of imbalances become too great, some teams will be unable to survive. But these are issues for Chapters 4 and 6. 53 SECTION 10 Chapter Recap The beauty, competition (both absolute and relative), commonality, and vicarious enjoyment of winning are scarce characteristics of sports. Fans pay the most for the highest level of absolute competition, but once that is decided, relative competition is extremely important. Scarcity requires some means of rationing. In sports, both price and waiting time are common rationing devices. The demand function is a representation of fan willingness to pay for different quantities of sports. The demand for any given team’s sports output slopes downward, meaning that there are only imperfect substitutes for that team’s output. As price increases, fans demand less of the sports output. Changes in fan preferences, income, prices of other goods, expectations about the future, and population all will shift the demand for sports. Examples that appear to contradict the law of demand in sports can be explained by changes in demand over time. Market power dominates sports demand functions. The source of the market power lies mostly in the fact that individual teams are granted exclusive franchise areas. Demand functions contain an astonishing amount of information. The economist’s tool of consumers’ surpluses produces one type of dollar estimate of the value of sports consumption. The demand function also demonstrates responsiveness, or elasticity, of fan demand to changes in price. An important piece of information that comes from a knowledge of demand is total revenue. Price decreases through the elastic portion of demand will increase total revenue. At the unit elastic point on the demand function, total revenues are at a maximum. Even lower prices increase quantity, but total revenues fall through the inelastic portion of demand. Marginal revenues are defined as the change in total revenue as output increases. Marginal revenues are positive but falling through the elastic portion of demand. They are zero at the unit elastic point on the demand function (i.e., the maximum of total revenue). Finally, marginal revenues are negative through the inelastic portion of demand. Even though owners may not know their entire attendance demand function, they will know the impact on revenue of changing their ticket price by a small amount if they know the elasticity of demand. Sports teams would only price in the inelastic portion of attendance demand if other revenues offset reduced gate revenues. Finally, price discrimination involves charging different fans different prices for the same attendance. When real price discrimination occurs, those fans with the highest elasticity of demand receive the lower price. 54 The data on sports team revenues reveal substantial variation between teams in the same pro sports league. The data on revenues also reveal that there is substantial variation in revenues between different sports. It is important to remember that distinguishing “small” and “large” markets is economically most interesting based on fans’ willingness to pay. 55 SECTION 11 Key Terms and Concepts You should have run into each of these in pop-ups in the text of this chapter: • Scarcity • Rationing • Absolute level of quality • Relative level of quality • Uncertainty of outcome • Demand • Movement along the given demand function • Demand shifters • Market power • Exclusive geographic territory • Performance enhancement • Large market–small market distinction • Consumers’ surpluses • Price elasticity of demand • Total revenue • Marginal revenue • Inelastic ticket pricing • Price discrimination • Within-league revenue variation • Between-league revenue variation 56 SECTION 12 Review Questions 1. What are the scarce elements of sports demanded by fans? 2. What are the typical rationing devices used in sports markets? If price is eliminated as a rationing device, is competition for tickets eliminated? Why or why not? 3. Explain Professor Rottenberg’s uncertainty of outcome hypothesis. According to this hypothesis, what will happen to a league that becomes too competitively unbalanced on the field? 4. Explain the effect that each of the following would have on the demand for NFL football games in Chicago: a. All else constant, some Bears fans find the violence in football less to their liking. b. All else constant, incomes of Chicago residents increase. c. All else constant, the price of a season ticket to the Chicago Opera falls to equal the price of a Bears season ticket. d. All else constant, Bears fans expect season ticket prices to increase next year. e. All else constant, due to hard times, the Chicago population is expected to decline. 5. Why does the demand for attendance at sports events slope downward? 6. Define consumers’ surpluses for those attending sports events. Discuss the two components of changes in consumers’ surpluses when the price of attendance falls. 7. Define the price elasticity of attendance demand. What does it measure? 8. Define total revenue. In a graph, show and explain the relationship between a downward sloping demand curve and a team’s total revenues. Identify the elastic, unit elastic, and inelastic regions of attendance on the total revenue function you drew. 9. On a graph of total revenue, show what happens to total revenue if demand increases. 10. On a graph of total revenue, show a team that has stadium capacity below the maximum of total revenue. Should this team operate to capacity? Why? If demand falls drastically, so that capacity is now above the maximum of total revenue, should this team still operate to capacity? Use the graph to demonstrate your answer. 57 11. Define marginal revenue. Beginning at a price where quantity demanded equals zero and lowering the price to zero, describe how marginal revenues change along the demand function. 12. Where is marginal revenue equal to zero? Are total revenues positive or negative in the inelastic portion of demand? 13. Will sports teams ever price in the inelastic portion of demand? Explain. 14. What is price discrimination in sports attendance? Why do senior citizens pay a lower ticket price than other fans for a given game and seat? 15. Why do members of a sports league keep a close eye on within-league revenue variation? Relate this to your answer in Review Question 3. 58 SECTION 13 Thought Problems 1. Think about your favorite sports team using the scarcity, rationing, and competition outline (Figure 2.1). What are the scarce commodities supplied by the team? Is attendance rationed by price only? If the price of attending a game were reduced to $1 for all seats, would any of your team’s fans be unhappy? Explain. 2. Consider the following statement, paraphrasing one lament by fans: “Fans can never catch up with the vicious cycle of sports event pricing. As fan income rises, they can finally afford the high price of sports events. But this increase in demand for sports just raises prices again, and fans are right back where they started. It’s a vicious circle that drives the everyday fan out of the arena.” What is the error in this statement? Explain. 3. Discuss why competitive balance is an important characteristic of team sports leagues. What role does Rottenberg’s uncertainty of outcome hypothesis play? In your discussion, cover what would happen if three teams came to dominate a league. In closing your discussion, address whether the end of the NBA Bulls dynasty of the 1990s was a good or bad thing for the NBA. 4. Explain why it must be preferences that explain the difference between the demand functions in Figure 2.2. 5. If you look at the movement of teams from one city to another, you will find that some teams move to cities with smaller populations. For example, the owner of the NFL’s Los Angeles Rams moved her team to St. Louis for the 1997 season (examine Table 2.4 for population and income characteristics of areas). Is this a bad business decision? Explain. 6. A student intern discovers the following about the demand by local businesses for attendance at a pro sports team’s games: where PB is the ticket price paid by businesses, measured in dollars, and AB is their attendance, measured in thousands of fans. Draw this demand function. Does this team have market power over business buyers? Why or why not? 7. In the most recent PBA between the majors and the minors, MLB imposes minimum stadium quality specifications for all affiliated minor league teams. What does this restriction accomplish from the perspective of market power? 59 8. The consumers’ surpluses for the women’s basketball demand functions in Figure 2.5 when the price was $20 calculated earlier in the chapter. At the same price of $20, calculate the consumers’ surpluses for the men’s basketball attendance demand function in the same figure. Is the amount larger or smaller than for women’s games at this price? Why? 9. Calculate the price elasticity of attendance demand for women’s hoops for a price decrease of $30 to $20 in Figure 2.2. Will revenues for the team rise or fall for this price change? Explain. 10. Teams earn revenue from a variety of sources: ticket sales, parking fees, concession sales, sales of licensed products, and broadcast rights sales. Explain how all of these revenues actually come from fans. 11. Explain the role of the inflation rate in comparisons of team revenues over time. 12. Return to the business attendance demand function in Thought Problem 6. Find the total revenue function. What is the shape of the total revenue function? What is the highest possible total revenue that the team can hope to collect? At what level of attendance? At what price? a. Find the marginal revenue function. At what level of attendance is marginal revenue equal to zero? At what price? b. What is the elasticity of demand at the revenue maximizing attendance level? c. If capacity at the team’s stadium is 25,000 seats, should the team owner fill the stands with business buyers? Why or why not? 13. PSLs reserve the rights to a particular seat for a lump sum payment in addition to the price of tickets. Is it price discrimination to charge for this extra reservation value through a PSL charge? Why? Is it price discrimination if two different people in approximately the same seating area pay different PSL charges? Why? 14. In addition to the business demand function in Thought Problem 6, our clever student also discovers that the demand by families for attendance at the same games and the same seat types is as follows: where PF is the price of a ticket for individual family members, measured in dollars, and AF is the attendance by individual family members, measured in thousands. Show why individual family members will be charged a lower price than business buyers for the same games and seat types. What is the ratio of business buyer to family member prices if the team owner cares about the bottom line? Is this price discrimination? 60 15. Using the data from Figure 2.12 and Table 2.6: a. What is the range of total revenues in the NHL? What is the ratio of highest to lowest total revenue? b. Identify the team with the smallest total revenues in the NHL. This team is not very successful, but the MLB teams for this city are very successful. How can this be? c. Given your answer in part a, do you expect play on the ice to be balanced in the NHL? Why or why not? d. Based on revenue dispersion, which of the leagues covered in Figure 2.12 and Table 2.6 do you think would be the most competitively balanced? Why? 61 SECTION 14 Advanced Problems 1. In his book Getting It Right (1997), economist Robert Barro argues that the only thing that matters to fans is the relative level of competition. Given this, many behaviors in sports, such as the escalation in training routines and the growing use of performance-enhancing drugs, are socially wasteful because they are costly but do not alter relative competition. At the heart of the issue is that fans would pay the same with or without these behaviors because the outcome on the field is unaltered. With these behaviors, resources are spent without any value being added from the fan perspective. a. Barro’s view is at odds with the arguments about absolute and relative quality in this chapter. What is the strongest argument in sports favoring the view that both absolute and relative quality matter? b. Using Barro’s logic, is the existence of both major league and minor league levels of sports socially wasteful? After all, there are really close minor league games, so why invest in attaining major league status? Do you reach the same conclusion using the argument given in this text? Why or why not? c. Barro uses his logic to argue that strict drug testing and punishment of offenders would be beneficial from society’s perspective, not because it punishes bad behavior, but because it reduces wasteful use of resources. Would Barro’s logic also support strict limits on the amount of time that school kids can spend on sports? How would the time limit be determined and by whom? Would parents of gifted student athletes agree with this idea? Why or why not? 2. How might you measure the uncertainty of outcome between your team and its next scheduled opponent? (Hint: Uncertainty probably increases with expected closeness of the outcome.) How would you measure it for teams in a league after a given season? How about over the longer run? 3. Is the price of attendance just the dollar value of the ticket and PSL (if any)? How would you measure the full price of attendance? (Hint: Be careful. Why are prices lower during the week than on the weekends? Is the price of a Los Angeles Lakers game higher for someone living in San Diego than it is for someone living in adjacent Orange County?) 4. The NFL’s Houston Oilers left Texas to become the Tennessee Oilers for the 1997 season. They changed their name to the Tennessee Titans and moved into their new 62 Adelphia Stadium for the 1999 season. Interestingly, their lowest-priced individual tickets were priced at $12, while their lowest-priced season ticket cost $250. Given that there are only eight home games, why didn’t fans just buy individual tickets rather than season tickets? Is this price discrimination or not? Explain. 5. In addition to the lowest-priced arrangements in the last question, the Tennessee Titans’ highest-priced individual tickets in 1999 were $52, while the highest-priced season ticket was $2,500. Relative to the lowest-priced arrangements, is the highest-priced arrangement price discrimination? Why? Suppose that the $2,500 included a $1,000 personal seat license. Is this price discrimination compared to charging no personal seat license? Explain. 6. If fans in one location are willing to pay more for higher quality than fans in another location, why will it never make sense for the point spread between their teams to be zero? 7. Thinking about Figure 2.1 and your basic economics training concerning price ceilings, evaluate the impact of a lease requirement that the price of some seats be kept at “a family-affordable” level. In particular, think about the following: a. The quality of family-affordable seats. b. The amenities that will be offered at family-affordable seats. Who will capture the difference between the family- affordable price and the price at which these seats would otherwise have sold? 8. Recall the business attendance demand function in Thought Problem 6. Answer the following questions. a. Calculate consumers’ surpluses at PB = $80 and PB = $60. b. What is the dollar value of the increased happiness enjoyed by business fans at the lower price of $60? c. Even though fans would be happier, why would you not expect the price to fall from $80 to $60? 9. Why would it be inadvisable for sports teams to sell all of their seats as season tickets even if they could? That is, why don’t as many teams as possible aim for season-ticket only sellouts for the season? 10. The full revenue data are at the Web site for this text (https://sites.google.com/site/rodswebpages/codes). Describe the relative contributions of gate, TV, and venue in pro sports. Which one has been growing relative to the others over time? Why? 63 figure:%23 figure:%23 SECTION 15 References Barro, Robert. Getting It Right: Markets and Choices in a Free Society. Cambridge, MA: MIT Press, 1997. Bernstein, Carl, and Bob Woodward. All the President’s Men. New York: Simon and Schuster, 1974. Exley, Frederick. A Fan’s Notes: A Fictional Memoir. New York: Harper & Row, 1968. Hall, Donald. Fathers Playing Catch with Sons. New York: North Point Press, 1998. Quirk, James, and Rodney Fort. Pay Dirt: The Business of Professional Team Sports. Princeton, NJ: Princeton University Press, 1992. Shank, Matthew D. Sports Marketing: A Strategic Perspective. Upper Saddle River, NJ: Prentice Hall, 1999. Sherman, Lauren. “Toughest NFL Waiting Lists,” Forbes.com, September 7, 2007. Last accessed May 11, 2009. URL: http:// www.forbes.com/2007/09/07/nfl-football-tickets-forbeslife- cx_ls_0907tickets.html Szymanski, Stefan. “The Economic Design of Sporting Contests,” Journal of Economic Literature XLI (2003): 1137– 1187. Thayer, Ernest Lawrence. “Casey at the Bat,” San Francisco Examiner, June 3, 1888. Veeck, Bill. Veeck as in Wreck. New York: Simon and Schuster, 1962. 64 http://www.forbes.com/2007/09/07/nfl-football-tickets-forbeslife-cx_ls_0907tickets.html http://www.forbes.com/2007/09/07/nfl-football-tickets-forbeslife-cx_ls_0907tickets.html http://www.forbes.com/2007/09/07/nfl-football-tickets-forbeslife-cx_ls_0907tickets.html http://www.forbes.com/2007/09/07/nfl-football-tickets-forbeslife-cx_ls_0907tickets.html http://www.forbes.com/2007/09/07/nfl-football-tickets-forbeslife-cx_ls_0907tickets.html http://www.forbes.com/2007/09/07/nfl-football-tickets-forbeslife-cx_ls_0907tickets.html SECTION 16 Suggestions for Further Reading Surely, I can come up with some. 65 CHAPTER 3 The Market for Sports Broadcast Rights It would not be good public relations for baseball to have the Series sponsored by the producer of an alcoholic beverage. –Happy Chandler, MLB Commissioner Refusing beer ads on the first televised World Series Helyar, 1994, p. 393. CHAPTER OBJECTIVES After reading this chapter, you should be able to: • Describe the functioning of advertising and the market for rights to sports programming. • Understand that the market power lies with leagues; broadcast, cable, and satellite media providers merely channel money competitively from advertisers to leagues. • Explain how buying advertising and sports rights may become very complicated. • Explain why advertising revenues can be large enough to cover a team’s entire payroll but also are highly variable across different leagues and across teams in a given league. • Understand that the commercialization of sports increases compensation to athletes, raising their incentive to excel, but that it can have an adverse effect on fan perceptions. • Recognize that outright team ownership by media providers may increase the distance between revenue haves and have-nots, leading to increasing competitive balance problems. SECTION 1 Introduction In its original bid to break into network TV, FOX spent so much obtaining NFL rights that its sports division took a $350 million ($504 million) loss on its 1994 contract. However, the network as a whole showed an increase in profits (after the $350 million loss solely by its sports division) from $511 million ($754 million) for the last six months of 1993 to $610 million ($878 million) for the last six months of 1994. Despite losing hundreds of millions of dollars on the NFL, FOX profits rose 16 percent (in 2009 dollars). As you saw in Chapter 2, teams generate revenue from a number of sources. In this chapter, we will focus on advertising because it offers interesting economic lessons and because media revenues are growing in importance for all sports leagues. We will take a look at the basics of the sports broadcast rights market and sports leagues’ rights. Our discussion will also cover media providers and advertising, as well as media ownership. We will examine firms and how they make their advertising choices, with a focus on the “beer wars” and sponsorship. Finally, we will look at the impact of all this money on sports. 67 SECTION 2 Setting the Stage: The Media in Sports Media are the second most important source of revenue for all but a few teams in all major sports. The willingness of advertisers to pay for sports programming long ago altered the revenue side of pro and college sports forever. It all started, of course, with radio broadcasts in the 1920s, but video has been the megaforce behind the explosion in media revenues in pro sports. Over-the-air (OTA) broadcasts on the three major TV networks (ABC, CBS, and NBC) was the first video revenue source, followed in the early 1980s by the advent of cable and, a bit later, satellite TV. Most recently, direct audio and video streaming over the Internet are beginning to take their place as revenue generators for all leagues. Individual team owners are richer by tens of millions of dollars due to the virtual explosion in media rights over time. Leagues now sell programming in four ways: • National packages to networks that go both OTA and on cable/satellite (e.g., ABC, CBS, ESPN, FOX, NBC, TNT, VERSUS). • Local/regional cable packages (except for the NFL, e.g., Fox Sports Network, MSG, TBS, YES, WGN). • Special premium packages that go directly to cable and satellite (MLB Network, NBA Network, NFL Network, and NHL Network on, e.g., Comcast and DirecTV). • Directly to subscribers over the Internet (MLB.TV at MLB.com, NBA League Pass—Internet at NBA.com, and NFL.com/Live). Not that long ago, the only source of league broadcast revenue was the first source listed earlier—the value of ad slots to advertisers on sports programs. In turn, this entire value rested on fans who also purchased advertised products. In the modern context, this is still the most important source of broadcast revenues to leagues, but gaining in leaps and bounds are revenues to leagues through direct sales to cable and satellite, the so-called premium package sales. Finally, while still in its infancy, the leagues, themselves, are streaming audio and video directly to subscribers over the Internet. This last one is in its infancy and causing growing pains for all team owners and their leagues. MLB, NBA, and NFL owners have all centralized their Internet offerings at their respective league web page. However, individual hockey owners have sued their own league over whether the rights to Internet streaming should properly reside with them, at their own individual web pages, or with their league. The increase in the number of cable and satellite TV providers, with league-operated Internet streaming services 68 joining the mix, all reflect underlying fan demand for sports viewing. There are two results of economic interest—sports rights values have been driven up, but media markets have also become extremely competitive. Given this dramatic competition, no single media provider, whether OTA, cable, satellite, or a league over the Internet, should have the power to extract more than a normal return to its endeavors. However, the story is much different for the programming sold by leagues through premium packages directly to cable and satellite and, of course, for Internet streaming direct to subscribers. For these revenue sources, leagues definitely have the upper hand. The sums that change hands for this type of programming include monopoly profits that can be charged by leagues. They are in the market power position in this case. A recent “battle” between the NFL and cable giant Comcast, addressed in a subsequent section, serves to emphasize this point. Sports leagues structure the bidding process to extract the highest-possible amount from media providers. The results can be staggering for sports leagues, but they are a mixed blessing. The variation in media revenue across teams in a league contributes to the current competitive balance woes of some pro leagues. Ultimately, revenue imbalances can lead to competitive imbalance in the eyes of fans. There’s also the issue of sorting out just how to collect the revenues in the modern context of media streaming on the Internet. This has led to hegemony between individual owners and their own leagues! Thus, the value of sports programming to media providers (OTA, on cable, on satellite, and over the Internet) is the ultimate value to fans directly or through the products they purchase. Given the massive sums that media providers send to leagues, some have argued that media providers are cursed to lose money if they win broadcast rights. We will see in this chapter that such an analysis fails to include the entire array of benefits that go to media providers when they win sports broadcast rights. Similar puzzles exist for firms that purchase ad slots from media providers. Turning to firms and their advertising choices, as with all inputs, economics suggests that ads are purchased until the extra benefit equals the extra cost. However, determining the marginal benefits of advertisements is complex. There may be direct results in terms of added sales. But in some situations, there is an added strategic aspect to advertising choices—a given firm’s advertising choice may well depend on the choices made by other rival firms. Finally, for firms, while hardly a new phenomenon, the sponsorship of events and arenas and the outright ownership of teams by media providers further complicate the relationship between sports and advertising. Although many lament the commercialization of sports through advertising revenue, it is valuable. The payoff to athletes engendered by this money generates much more intense competition on the field, producing athletes like none seen before. However, in addition to any possible over- commercialization, a shift to outright ownership by media 69 providers brings other concerns. Not the least of these concerns is that as the market for team ownership moves away from individual ownership, some teams are left behind in the short run. This could further exacerbate revenue disparity and the competitive balance problems that follow. 70 SECTION 3 Ads and Preference Formation To understand the power of advertising in sports, it is worth pausing for a moment to ponder just how ads work. When firms use ads, the direct impacts to consider are how the ads impact new consumers and consumers already purchasing their product (we’ll discuss strategic considerations about rival advertising later in the chapter). Of course, one objective is to reach consumers and increase sales. Evidence is mixed on whether ads generate new consumers who have never tried a product before. That is, do ads lead to start-up consumption? For example, noted sports management expert John Crompton (1993) observes that alcohol ad bans don’t appear to have any impact on consumption, because countries with ad bans show little difference in alcohol behavior than those without bans. Generally speaking, Crompton points out that the impact of ads on alcohol consumption is weak, at best. It appears that price, availability, and family/lifestyle are more likely to affect consumption. My reading of the more recent literature on start-up consumption and advertising changes nothing about Crompton’s observation. But be warned! Advertising is its own entire field of business analysis. If your level of interest is large, discuss the work on these topics with your marketing professors. Conversely, if individuals never have been exposed to a particular type of consumption, advertisements might make them aware, heighten their awareness, or shape their preferences. This certainly is the belief behind regulations aimed at stopping particular types of ads that may influence children. The ban on cigarette television advertising in 1971, the “Kid Vid” crusade of the 1970s and 1980s, and the more recent movement against advertising icons (such as Joe Camel) all are aimed at advertisements purported to lead young people (new consumers) to unhealthy consumption. If advertisements generate new consumers, then they clearly are valuable to the producers of these products. ADS AND BRAND CHOICE Advertisements play another important role. Once people have decided which goods to consume, advertisements clearly present the array of possible choices for a particular good. So, ads may help consumers make brand choices. How many brands of cigarettes, beer, or cars are there? Consumers appear to identify very heavily with being a particular kind of consumer; we might consider ourselves Pepsi drinkers as opposed to Coke drinkers. Thus, ads also generate brand loyalty. We will return to this second important function of advertising later in the chapter when we address advertising choices by firms. AIMING ADVERTISEMENTS AT THE RIGHT TARGET 71 Ads must be aimed at the group of consumers most likely to care about them. This group is referred to as a target demographic group. For sports programming, the typical viewer is a male between the ages of 18 and 34, with the upper endpoint sometimes stretched to 49 (women typically are less than one-third of a sports audience). Advertisers interested in reaching this male target demographic purchase sports program ad slots. Ads that are actually seen by large numbers of the intended demographic group are said to have good advertising reach. As you can see in Table 3.1, automobile and beer ads dominate sports broadcasts, with strong showing by fast- service food, because males aged 18–34 tend to buy cars, drink beer, eat fast-service food—and watch sports. More expensive consumption items, such as luxury vehicles, dominate golf advertising because golfing males are typically more affluent. Table 3.2 demonstrates the different types of sports viewers that advertisers try to reach when they allocate their advertising dollars. If you are catching on to the value of advertising, you should be able to explain why spending on football dwarfs spending on all other sports in Table 3.2. Of course, it’s not just who watches, but how many watch, as well. Figure 3.1 shows the results of a recent poll of fans’ favorite spectator sports. Football tops the poll by a wide margin. Baseball and football are next, but with some variation between men and women viewers. Indeed, both basketball and baseball trail “none” for women! The only overlap at the bottom of the poll numbers is auto racing for men and women. Interestingly, historically, from the 1930s to the 1960s, baseball topped all polls I have seen. Football overtook baseball in a 1972 poll and never lost top status from then on. In addition, basketball overtook baseball for the first time in 2000; they have traded second spot in polls since then. Table 3.3, showing Super Bowl ratings and ad rates, makes it clear just how popular football really is. A rating point equals one percent of the total homes with a television. By this measure, seven of the top 10 rated games were in the 1980s. Super Bowl XVI in 1982 was the highest-rated Super Bowl ever, drawing 49.1 percent of all households with a television. The “share” measure refines things a bit further. A share point is one percent of the televisions actually in use that were tuned to a given broadcast. By this measure, seven of the top 10 shares were in the 1970s. Super Bowl X in 1976 had the highest share of any Super Bowl; of the households with their sets turned on that day, fully 78 percent of them were tuned to the game. The highest-rated Super Bowl XVI in 1982 was no slouch in terms of share at 73 percent. Historically, Super Bowls dominate the top-rated broadcasts of all time, behind such monumental TV moments as the season-ending episode of M*A*S*H, the “Who Shot JR” episode of Dallas, and the finale of Roots (www.sportsbusinessnews.com, February 5, 2006). The data show that sports programs are very popular with the target demographic that advertisers wish to reach. The ad rates in Table 3.3 indicate how much networks think 72 http://www.sportsbusinessnews.com http://www.sportsbusinessnews.com advertisers are willing to pay for this type of reach. As you can see, measured in 2009 dollars, a 30-second Super Bowl ad was never worth more than in 2009 at $3 million, and those 30 seconds have been worth more than a million dollars since 1985. But note that it is not just time that tells the tale—if we rank the data by the size of the ad slot price, Super Bowl XLI in 2007 comes in 10th place while Super Bowl XXXV in 2001 comes in 4th. Overall, the real annual growth rate in the price of these slots, from 1967 to 2009, is an impressive 6.2 percent (when the typical growth rate in the economy at large is around 3 percent). Figure 3.2 shows advertisers by product category for Super Bowl XLIII in 2009, compared to Super Bowl XLII in 2008. Unsurprisingly, motion picture releases, beverages (alcoholic and otherwise), and car categories dominate ad space in these Super Bowls. But the full range of companies, in Table 3.4, also includes snack foods, “dot-coms,” fast-service food, tax preparers, cell phone service, fresh flowers, and amusement parks. The big spenders were clearly alcoholic beverages and soft drinks. 73 SECTION 4 Basics of the Sports Broadcast Rights Market Let’s begin our analysis of the market for sports broadcasting rights by identifying the participants and their roles, which are displayed in Figure 3.3. The figure appears complex, but it is easy to follow, participant by participant. At the top of Figure 3.3 sit professional sports leagues and their member teams. As you can see on the left side of the figure, leagues and teams sell a portion of their programming to media providers like networks, cable, and satellite companies. Lately, leagues and teams have also adopted a “media provider” role, sending programming directly to individual subscribers via their cell phones or the Internet. Now let’s track the rest of the participants. Networks (ABC, CBS, ESPN, NBC, TNT, and VERSUS) pay a rights fee for an exclusive portion of the league’s games in what is referred to as the national broadcasting contract. National contract games, of course, will have more than regional appeal for national network broadcast. These games are broadcast OTA by a few networks and on cable and satellite companies by all networks. All leagues sell a portion of their games this way, but the proportion sold varies by league. MLB teams sell only a few games nationally, NBA and NHL teams sell a larger share, and NFL owners sell all of their games through the league’s national contract. Following down the end of the network leg in Figure 3.3, we see that networks provide advertising slots sold to product advertisers. Advertisers, of course, know that their ads will be seen by potential customers both OTA and over all cable and satellite providers. In addition, networks sell the same programming to cable (e.g., Comcast and Time-Warner Cable) and satellite companies (DirecTV and Dish Network). The next leg to examine in the left portion of Figure 3.3 shows leagues and teams selling their local broadcasting contracts and premium packages to cable and satellite companies. The local contract is between individual team owners and (usually) regional cable companies. Remember, NFL owners do not sell any local broadcasting contract at all. The most well-known “local” cable offerings are MLB games by the divisions of the Fox Sports Network (e.g., Fox Sports Northwest carries the Mariners), MSG (Red Sox), and YES (Yankees). Indeed, Rupert Murdoch’s Fox Sports Network carries almost all of the individual local games offered by MLB owners. But there are always variations on a theme; MLB’s Braves and White Sox sell their “local” games over cable networks carried nationally (TBS and WGN, respectively). It is worth noting something truly interesting about the difference between a “national” and a “local” contract. Given the level of broadcast technology that exists, both leagues actually end up broadcasting a very similar product. In MLB, 74 where local contracts dominate, all locations get their local team nearly nightly, along with a general interest game a couple of times each week. In the NFL, where there is literally no such thing as a local television broadcast, fans get approximately the same thing. They can be reasonably sure that they will see their home team every week, along with a couple of games of broader interest. We’ll return to this outcome and discuss why teams sell some of their games through leagues when we cover the behavior of leagues in Chapter 5. Portrayed in this same leg of Figure 3.3, leagues offer parent cable (Comcast, or Time-Warner Cable, for example) and satellite (DirecTV and Dish Network) companies special premium packages. These special packages offer almost complete choice of any game broadcast on any day. These packages are usually offered at an additional premium to cable subscribers (MLB Network, NBA Network, and NHL Network). Lately, however, leagues have been pushing cable and satellite companies to add these special packages as a part of their broader “basic” offering to subscribers. An example is NFL Network now offered one tier above Comcast’s basic tier. Just how the NFL made that happen is detailed shortly. The last leg to cover in the left side of Figure 3.3 is the direct sale of programming to individuals by leagues, themselves. The Internet/broadband revolution has made it possible for leagues to compete directly with media providers by providing their own subscription programming. Only the NHL appears (at this writing) to be lacking in this regard. All of the other leagues offer streaming audio and video directly to individuals via their cell phones or over the Internet (MLB.TV at MLB.com, NBA League Pass—Internet at NBA.com, and NFL.com/Live). That wraps up the flow of programming, so let’s turn to the right side of Figure 3.3 and track the flow of money generated by this seeming myriad of programming offerings. Networks collect slot fees from advertisers and subscription fees from cable and satellite companies. Advertisers pay based on the value of advertising slots (detailed in a subsequent section). Table 3.5 shows the values that advertising creates for networks for a recent year. Cable and satellite companies pay based on value of programming to their subscribers, usually a flat rate per cable or satellite subscriber. Both of these revenue sources for networks determine the size of the rights fees that they, in turn, pay to pro sports leagues. We’ll return to the size of these rights fees at a more opportune point in this chapter. The remaining flow of money back to leagues and teams is through subscription fees. These are collected both from the premium offerings on cable and satellite and from direct sales to individual subscribers. Note that there is no box for fans in Figure 3.3. So how does the figure square with our view that all the money comes from fans? From the relationship shown in Figure 3.3 we can make the following observation: The only reason that advertisers buy ad slots is to increase and/or preserve the sales of their products to consumers. In the case of advertising on sports programs, the consumers also happen to be sports fans. 75 Therefore, it is fan willingness to pay that is the point of interest, albeit indirectly in this case, through their purchase of advertised products. Subscriptions make the link between payments to leagues and the sports broadcast market much clearer—this money, as promised, comes directly from fans. WHERE DOES THE MONEY GO? We’ll use Figure 3.3 to analyze sports advertising from the perspective of each participant in the figure later in the chapter, but let’s jump straight to the $64,000 question: “Who gets the money?” The answer lies in identifying the economically competitive sectors of the broadcasting rights market. Because audiences can be enormous, many people mistakenly assume that this means that individual media providers have market power. It is easy to confuse the power of the media to influence people with market power. However, my consulting experience with media providers, briefly working with lawyers for ABC, plus some particularly insightful real-world examples, indicates a high level of media provider competition. Remember, market power is driven by a lack of close, readily available substitutes. Back in the days when ABC, CBS, and NBC were the only major networks, there weren’t very many close substitutes for advertisers wishing to purchase ad slots. However, times have changed. Since the early 1980s, national and regional cable and satellite networks have proliferated. In addition, radio, the corner newsstand, and even the newspaper rack at the local library are all substitutes for each other (not necessarily perfect substitutes, but substitutes nonetheless). Can you believe that there were once only two dominant sports publications, Sporting News and Sports Illustrated? And these were usually obtained at specialty news stores or by subscription. Now you can find entire shelves of magazines devoted to the topic in the supermarket. TV media providers maintain their audiences by airing programs that audiences want to watch, and they do so in a very competitive environment. This means that, ultimately, the competitive media market has only one power, namely, the power to give viewers what they want as measured by ratings/shares. If a media provider loses in the ratings, advertisers move to higher-rated slots from other providers. The media provider that loses advertisers suffers lower profits. Consumers have the ultimate power of switching channels or turning off the set. This means that leagues, ultimately, wield the market power here. It is then reasonable to suspect that the values that are created for media providers ultimately are pulled through to leagues, and their member owners, through the competitive process. Here is the “insightful real-world example” I mentioned earlier (the example is also the “battle” mentioned in the introductory section of this chapter). The conflict arose between the NFL and Comcast over the NFL Network. To see that the leagues hold all the cards, consider the bargaining that went on. Comcast took the stand that it would provide the NFL Network only to subscribers willing to pay an additional premium over and above their basic charge. The 76 NFL immediately took the stance that its network must be provided in the basic subscription. Of course, that would mean that Comcast would have to raise basic rates to all, rather than surcharging just those willing to pay an added fee. In the settlement reached in May of 2009, Comcast made the NFL Network available on its additional charge tier (Digital Classic) one step above basic cable but not at an additional premium payment. At settlement time, that tier was already chosen by about 2/3 of Comcast customers. Also in the end, the NFL took about 50 cents per subscriber, down from its initial asking price of 70 cents. Chairman and CEO of Comcast, Brian Roberts, said of the agreement, “I think both sides may have had to give a little more than they intended” (www.foxnews.com, May 19, 2009). But a bit of basic economic intuition suggests that it was only Comcast that gave up much of anything in the final settlement. First and foremost, remember the NFL knows it can confront Comcast with an all-or-nothing proposition and rest assured of winning. The substitutes for the NFL Network are … absent; the NFL, after all, has a monopoly over the highest level of league football. Further, given the rabidity of fans, satellite TV offered a perfect substitute—both DirecTV and Dish Network already offered the NFL Network. Comcast might wish they could force the NFL to accept the fact that it would charge the extra premium price, but their bargaining stance was untenable from the outset. Second, at a particular bargained price, the NFL knows that its payment from Comcast under basic provision is greater than its payment if Comcast charges an additional premium for the NFL Network. Let S* be the profit-maximizing number of subscribers under basic provision, P* the basic subscription price charged to the S* subscribers, and AC* the average cost of service to the S* subscribers. Further, let the ultimate fee per subscriber that ultimately is paid to the NFL be F* (this would be the eventual 50 cents per subscriber). Profit per unit for Comcast is just P* – AC*. So Comcast would prefer F* = 0, and the NFL would prefer F* = P* – AC*. Ultimately, the actual bargained fee of 45 cents should be somewhere in between, that is, 0 < 45 < P* – AC*. From this perspective, Mr. Roberts’ statement rings a bit hollow. The only thing that probably did go Comcast’s way is that F* < P* – AC*; the NFL probably did not get all of the profit per subscriber earned by Comcast. We will see a similar episode in college sports (Chapter 13) between Comcast and the Big Ten Network with a similar outcome for Comcast. 77 http://www.foxnews.com http://www.foxnews.com SECTION 5 A Closer Look At Media Providers and Advertising With multiple media providers under each of the paths shown in Figure 3.3, all are involved in what can be characterized as an auction for broadcast rights. The most interesting thing about sports rights auctions is how much providers eventually end up paying for the rights. Many observers think that the rights go for more than they are worth based on their contribution strictly in terms of ad slot revenue. These same observers condemn such purchases as folly. Are such purchases really silly? If not, how have observers missed the point? We’ll discuss these questions in this section. From an economic perspective, media providers would consider the extra costs and benefits of rights fee purchases. As long as a sports broadcast right generates at least as much revenue as it costs, then it is worth buying. Otherwise, it’s not worth the price of the broadcast right. Table 3.5 reveals that these revenues are sizable. For example, three out of four of the general major networks (CBS, ABC, and FOX), along with the exclusively sports ESPN, generate revenues in excess of $1 billion ($1.2 billion) from their sports programming. On the other hand, besides the exclusively sports-programmed ESPN, only FOX collects more than 26 percent of its revenues from sports programming. So sports programming is important, but the rest of the programming purchased by media providers generates the vast majority of network revenues. Often, however, the sports division of a media provider actually loses money! Sports rights fees can be greater than the sports division revenues generated by selling ads on sports programming. Despite these losses, the rights are still worth having. The reason is that the value of broadcast rights to media providers is only partly in the advertising slot revenue that the programming generates. Thus, the marginal revenue product of broadcast rights extends beyond just the value of ad slots sold to advertisers. Let’s examine these other values. First, marginal revenue will be earned both by selling ad slots to advertisers and by advertising the provider’s own shows during the broadcast. In addition, placing the sports program close to other programs may provide some “sequencing value”; people who tune in to the broadcast may leave their sets tuned to the same channel and watch a subsequent show and vice versa. For example, Grey’s Anatomy followed Super Bowl XL in 2006. Its 36.0 rating was the series best ever and, in the 18–49 age group, it was the most-watched show since the Friends season finale in 2004 (www.tv.com, February 6, 2006). In addition, the media provider’s local affiliates receive revenues from local ads that appear during the sports program. The returns to the major media providers are earned here through franchise contracts with affiliates. The marginal revenue gained from all of these 78 http://www.tv.com http://www.tv.com sources must be included in the calculation of the value of a sports broadcast right. In a recent NFL broadcast rights go-round, NBC lost out to CBS. Dick Ebersol of NBC claimed that they bailed out of the bidding because, at the final level of bidding, the eventual winner would suffer annual “catastrophic losses” of $150 million to $200 million ($196 million to $262 million). Sean McManus, president of eventual bid winner, CBS Sports, argued, “We are not going to lose money on this deal because of the value it brings to our stations, the savings in promotional time, the extra value to our affiliates” (Quirk and Fort, 1999, p. 40). All of these factors generate revenue that appears on the overall CBS balance sheet but may not show up on the CBS Sports balance sheet. This is an essential point that bears repeating: Media providers place their bids based on the contribution that programming makes to their overall revenue structure, not just to the sports division from ad revenue. Another example, which we mentioned at the start of the chapter, is FOX. In its original bid to break into network TV, FOX spent so much obtaining NFL rights that it took a $350 million ($504 million) loss on its 1994 contract. But don’t shed any tears for Rupert Murdoch. The network as a whole showed an increase in profits (after the $350 million loss solely by its sports division) from $511 million ($754 million) for the last six months of 1993 to $610 million ($878 million) for the last six months of 1994 (Quirk and Fort, 1999, p. 41). Despite losing hundreds of millions of dollars on the NFL, FOX profits rose 16 percent (in 2009 dollars). Surely, some of that increase was from NFL spillover and affiliate values. THE WINNER’S CURSE Generally, apparent losses on sports rights actually turn into positive contributions to the overall revenues of the media provider once spillover values are included in the calculation. However, in some auction settings, rights prices might be forced to a level that exceeds their overall contribution to media provider revenues. Typically, auctions among well- informed buyers move broadcast rights to the highest bidder, with the media provider passing the vast majority of the value through to the league. A typical auction has many bidders who are both experienced and well informed about the product up for auction. The bids are essentially the same and close to the true underlying value of the rights. However, in some sports broadcast rights auctions, these usual circumstances don’t hold. Suppose that there is not much information on which to build any subjective evaluation of the probable value of the ads and other contributions to media provider revenues. This could happen for a number of reasons. Media providers may be facing very different situations than in the past, or there may be providers that are new to the process. In such cases, providers still must develop some sort of estimate of the broadcast rights value, but it is very likely that the bids won’t be bunched around the true expected value. Such a setting is ripe for what is called a winner’s curse. In such a setting, a competitive bidding 79 system would elicit the most optimistic—and therefore wrong —highest bid. The winner in such a situation would be cursed by having won because the true expected value is most likely much less than the winning bid. A recent possible example of the winner’s curse in action is covered in this section’s Learning Highlight: The Curse of Monday Night Football? 80 LEARNING HIGHLIGHT: THE CURSE OF MONDAY NIGHT FOOTBALL? At its inception in 1970, only (then) president of ABC Sports Roone Arledge saw what Monday Night Football (MNF) would become. Indeed, Arledge built part of his legend on the show. But the rights belong, of course, to the NFL, and the league put them up to bid for the 1998 season. ABC eventually won the MNF rights for eight years for $4.4 billion ($5.8 billion) in the face of a hard charge by NBC. Because NBC should have pretty good knowledge of the value of MNF rights, did ABC necessarily win? Or was ABC a cursed winner? One analysis would have it that ABC was cursed because of its winning bid. NBC had very experienced sports people in the bidding process and had carried NFL games for many years. Surely, those people must have known the value of these rights as well as ABC? Because NBC dropped out, ABC must have ended up paying too much for the rights. Reinforcing this view were public statements by NBC at the time that it expected ABC to lose more than $100 million per year with the winning bid. This is the usual notion of a winner’s curse and reminiscent of Dick Ebersol’s quote concerning the horrible prospects facing the winner in the battle over NFL rights between NBC and CBS. The argument against a winner’s curse for ABC rests solely on an evaluation of the setting because only ABC can know if they took a loss or not. First, that NBC dropped out doesn’t mean that it had a better guess than ABC. Think about it like this. If NBC had won the bid, they would have been the cursed winner by the same argument. Further, the setting here just doesn’t fit the requirements for a winner’s curse. How likely is it that either bidder would have only a weak estimate of the true expected value of the rights? In fact, ABC had the rights to sports programming in the past and MNF broadcasts in particular. So the bidders are neither uninformed nor neophytes, and the setting for a winner’s curse is questionable given that NBC and ABC were informed bidders. An alternative assessment is that, based on its expenditures, ABC expected to make about $500 million ($654 million) each year for the 16 MNF games that it would televise. That’s just a bit over $31 million ($41 million) per game. This seems a pretty tall order if one looks only at the revenues generated by MNF broadcasts. However, some of the MNF ad slots will be filled with ads for ABC shows. Further, one program leads into another, and the sequence of programming can affect viewing levels. For example, CBS had rights to NFC games before the most recent contract. CBS moved their Sunday Game of the Week to a lead-in position just before 60 Minutes, and ratings for the latter jumped. When FOX won the rights to the NFC games away from CBS in the recent contract, the ratings for 60 Minutes fell right back down again. Presumably, ABC knew that MNF could have the same effect on its Monday night programming. By this explanation, ABC must have expected that the actual MNF ad revenues, plus the value of advertising its own programs, plus the sequencing value of MNF, would exceed 81 the $31 million ($41 million) per game price tag. However, the winner’s curse cannot be dismissed without a close look at the specific setting and the outcome. ABC’s Steve Bornstein managed to hold on to Monday Night Football in the face of NBC’s hard charge in 1998, but was he the victim of the winner’s curse? [Photo of Bornstein.] Source: Inspired in part by “Thrown for a Loss,” Time, January 16, 1998, p. 52. 82 SECTION 6 Media Provider Ownership of Sports Teams The phenomenon of media provider ownership of teams provides one of the most interesting unsolved puzzles in the business of sports. The phenomenon itself is clear. Except in the NFL, corporations can own teams outright (we’ll talk about why it makes sense to restrict corporate ownership in the next chapter), and some corporate owners are media giants themselves. In MLB, Liberty Media owns the Braves, the Tribune Company owned the Cubs for years, and Rogers Communications owns the Blue Jays (FOX Sports owns about 14 percent of the Rockies and Nintendo American owns a share of the Mariners). In the NBA, Bell Globemedia owns the Toronto Raptors, Comcast owns the Philadelphia 76ers, and FOX and Cablevision have shares of the New York Knicks. In the NHL, Bell Globemedia also owns the Toronto Maple Leafs, and Cablevision owns the New York Rangers. Until recently, Rupert Murdoch’s News Corp. owned the Los Angeles Dodgers, and Disney, which owns ABC and ESPN, owned the NHL Anaheim Mighty Ducks and the MLB Anaheim Angels. Why buy the entire team instead of just its broadcast rights? What is gained? The decision to buy a sports team is just one of the many choices a media provider must make concerning the structure of its firm. All media providers obtain programming in one of two ways. They can purchase it from another programming producer (firms like Chuck Lorre Productions of Two and a Half Men and The Big Bang Theory or Brad Grey Television of The Sopranos), or they can produce it in-house (nearly all network news shows are in-house productions). A similar organizational choice confronts media providers when it comes to sports programming. Media providers can purchase the broadcast rights for a given team or purchase the team itself. In a sense, the latter choice can be thought of as bringing the production of actual sports programming in-house. Whether to buy a team rather than just its broadcast rights is quite close to the question of what to produce in-house versus what to outsource to other production companies. CBS AND THE YANKEES: A HISTORICAL EXAMPLE AGAINST MEDIA PROVIDER OWNERSHIP The puzzle is best presented with the single historical example of an economically tragic media provider ownership—CBS and the New York Yankees (the Yankee sale prices are in Quirk and Fort [1992], but the calculation of losses is original here). In 1964, then-owners Dan Topping and Del Webb sold 80 percent of the New York Yankees to CBS for about $11 million ($76 million). By 1967, CBS had bought them out completely for about another $3 million ($19 million), for a total of about $14 million ($95 million). CBS did so poorly with the team 83 that the new majority owner, George Steinbrenner, was able to put together a syndicate in 1973 and buy the team for $10 million ($48 million). This was exactly the price of an expansion franchise in 1968. Two existing teams, the Seattle Pilots and Washington Senators (arguably the worst teams ever in the history of MLB) each sold for about $10 million ($55 million) in 1970. In addition, under CBS ownership, the Yankees were a shadow of their former greatness. The Yankees led their division in 1963, with a 0.611 winning percent. However, they never led their division during the entire tenure of CBS (the best they did was second in 1970). In nominal terms, the $4 million loss represents about 29 percent of CBS’s original purchase price of $14 million. But this simple loss assessment ignores inflation and the opportunity cost of the funds. Putting it all into 2009 dollars, the 1964 amount is $76 million, the 1967 amount is $19 million, and the total is $95 million. The sale price in 1973 becomes $48 million. In addition, the original $76 million spent in 1964 would have accrued interest over CBS’s nine years of ownership. According to the usual formula, and using a 3 percent real interest rate, $76 million invested at the beginning of the ownership period would have been worth ($76 million) × (1.03)9 = $99 million. The additional $19 million invested at the end of 1967 would have been worth ($19 million) × (1.03)6 = $23 million. Over the entire ownership period, the opportunity cost of funds was $99 million + $23 million = $122 million. After selling the team to Steinbrenner’s group for $48 million, the actual loss compared to the opportunity cost of the funds was a whopping $122 million – $48 million = $74 million, or 78 percent of the opportunity cost of the investment. For a little perspective, on average, owners who bought teams in the 1960s essentially broke even at the next sale episode in real terms (Fort, 2006), even when the Yankees are included in the calculation, but CBS managed to suffer a 78 percent loss. Presumably, there is no reason to believe that CBS would be any better or worse at hiring front office talent, a field manager, coaches, and players than anybody else, especially for 10 years running. Just why the network failed so miserably as an owner of one of the most storied franchises in MLB history is something of a mystery. This is especially interesting because Steinbrenner turned the team around on the field and in the marketplace. In Chapter 1, we saw that an actual offer of $632 million ($834.2 million) was made for the team in 1998. Over that period of Steinbrenner’s leadership, the price of the Yankees grew at a real annual rate of 12 percent. His predecessors at CBS managed a 7 percent real annual loss. POSSIBLE ADVANTAGES TO MEDIA PROVIDER OWNERSHIP Buying a team secures local broadcast rights for the owner. These rights are less expensive than the only alternative to a network, namely, the national contract. Because many areas have no local team, a network-level media provider can attempt to fill that gap with a team that may have nationwide 84 appeal. All that those fans would get, otherwise, would be the scattered few games on national TV, plus cable superstation offerings. This is just what Ted Turner did when he bought the Atlanta Braves in 1976 through his Turner Broadcasting System (TBS). The Braves provided programming for Turner’s fledgling regional cable company that eventually grew into TBS. TBS reached millions of viewers nationwide who didn’t have any other “every day” baseball alternative. In addition, the Braves were cheap, making them especially appealing to Turner, whose staples were syndicated reruns. News Corp./FOX and Disney/ABC/ESPN were a completely different situation. In their media provider role, both News Corp. and Disney were already well established when they entered into team ownership. Their programming was of the more expensive, original variety rather than the syndicated reruns aired by TBS. Further, Turner and other superstations that followed already were beaming their teams nationwide. So News Corp./FOX and Disney/ABC/ESPN couldn’t fill the same niche with their baseball offerings. In addition, News Corp./FOX and Disney/ABC/ESPN did not purchase bargain- basement sports teams. Rupert Murdoch’s News Corp. paid $311 million ($407 million) for the Los Angeles Dodgers in 1998. This leads to the puzzle mentioned at the beginning of this section. News Corp./FOX or Disney/ABC/ESPN could pursue exactly the same policy just by buying a team’s broadcast rights rather than by buying the entire team outright. If these media providers just bought the rights to the games left over after the national contract, they could accomplish the same end without paying hundreds of millions of dollars for the team itself. So why buy a team? The obvious economic answer would be that News Corp. and Disney are of the opinion that it’s more profitable to own the team than it is to purchase residual broadcast rights. However, none of the reasons for increased profitability that come to mind are very convincing. We’ve already gone through the CBS/Yankee episode, so we know that corporate ownership doesn’t guarantee profits. Some of the possible reasons for corporate ownership include the following: • Avoidance of the high price of broadcast rights • Production efficiency gains through control of the process • Tax advantages • Alteration of the competitive environment confronting the media provider • Enjoyment gained from owning a team and making money AVOIDING THE HIGH PRICE OF BROADCAST RIGHTS Contrary to the opinion of some very good reporters (who turn out to be poor economists), ownership has nothing to do with whether a media provider can avoid the high price of broadcast rights. If the media provider buys the broadcast 85 rights on the open market, then it pays the market price. If, on the other hand, the media provider owns the team, it must implicitly account for this forgone price if it chooses to broadcast the programming itself. This is no different from the logic of opportunity cost underlying every decision that has to do with using something of value rather than selling it. GAINING PRODUCTION EFFICIENCY Does some gain in production efficiency reduce the cost of putting games over the air if a network owns the team? If so, then the network would get to keep the balance of value over the reduced cost. But how can owning the team lead to lower broadcast costs? One way would be if owning the team reduces contracting costs (e.g., legal costs between the media provider and the team as well as the costs of making sure that all elements are in place in order to produce a broadcast). However, this does not decrease the stages of production or reduce the number of people who must be hired. As we saw in the Yankees case, there is no reason to suspect that media providers have some sort of specialized talent that would lower costs. Further, it would seem that the same sorts of economies are available under long-term contracts. SECURING POSSIBLE TAX ADVANTAGES The tax benefits of owning a team are an interesting possibility. In this situation, there would have to be a connection between the media provider’s tax payments and the tax liability of the owned team. Suppose that the media provider itself (the parent company) makes high profits subject to the corporate income tax. If the team it owns shows a loss, then transferring some of the taxable profit to the team would reduce the tax obligation of the overall operation. This is a possibility, but there are other things to consider. First, for this tax avoidance scheme to be of value to the parent company, as opposed to the team itself, the savings would have to come back to it somehow. However, the revenue would have been shown on the team’s books, and that would go into the team’s net value. Indirectly, because the media provider owns the team and its net value would rise, the media provider now has a more valuable team in its overall portfolio. But because the revenue of the media provider fell, how can the overall value rise? Second, if there was a tax advantage to one type of firm–team relationship over another, then all team owners could simply incorporate, like Disney, and gain the same advantage. We simply don’t see this happening. Finally, the idea of transferring value would argue that media providers buy teams that are likely to show losses (or, at most, very low revenues). This hardly seems to characterize the Yankees, Braves, Dodgers, Angels, or Mighty Ducks. CHANGING THE COMPETITIVE ENVIRONMENT Sometimes owning the team may change the competitive environment of the media provider. Cornering the market on local rights can be of very high value. However, a few mitigating factors apply to this approach. First, it simply is very hard to do. Competition among media providers is fierce. 86 Second, such a move invites regulatory scrutiny. An international example of how media provider ownership can draw the scrutiny of antitrust oversight is featured in the Learning Highlight: Murdoch’s Bid for Manchester United at the end of this section. The Federal Trade Commission (FTC) and the Federal Communications Commission (FCC) closely watch the behavior of media providers in the United States. Finally, if team owners see what the media provider is up to, they will simply raise the price of the team to try to capture the expected future value of the reduced competition. The value of such an action just gets bid into the price of the team and the current owner collects it, rather than the media provider. Obtaining an environment of decreased competition is simply a difficult thing to pull off. HAVING FUN AND MAKING MONEY? There is one other explanation that works for the actual people behind the media provider firms. Perhaps they just wanted their firms to own baseball teams for the same reason that other owners buy their teams—team ownership has significant fun and profit aspects. The associated broadcast rights just follow along at the same cost as buying them, but the other aspects of ownership might have driven their choice. But don’t forget that a corporation cares only about profits. Fun doesn’t enter into the picture. And the people running these corporations could pay a high price for fun—if profits fall, they could lose control of their firms. Just why is it that owning the teams is more profitable than buying rights to their local broadcasts? Given that media provider ownership is taking on international dimensions, figuring out the answer takes on added significance. THE POSSIBLE PROBLEMS WITH MEDIA PROVIDER OWNERSHIP Whatever the reasoning by media providers, their ownership of sports teams may signal the beginning of a dilemma. If teams are worth more to media providers than they are to individual owners (and this is a big if because it is difficult to find any advantages to owning versus buying broadcast rights), then media providers will start bidding up the price of sports teams. If this really is the beginning of a trend, media provider ownership could lead to two problems. First, if it is true that teams are most valuable as a division of media providers, then revenues will be higher for those teams owned by media providers. In the short run, at least, the impact of media provider ownership may be to further exacerbate revenue differences among teams. Unless all teams end up as divisions of media providers, the revenue gap between teams would widen. Wide-enough gaps would lead to less competitive balance, risking lost fan interest, except for the fans of teams owned by media providers. The second problem concerns changes forecast by some observers. If the move toward media provider ownership is the front end of a trend, then current owners of teams will have to live with an interesting, mixed result. The price of 87 franchises will increase, and current owners will be wealthier, but they may well no longer be in the sports business. This is because they would have to sell their teams to collect this higher price. Although resources move to a more highly valued use, not all will be happy with the outcome. What will happen to the mix of OTA and cable or satellite sports, especially premium-priced special tier sports? Surely, those willing to pay the most will get the sports. But won’t some current fans be displaced? Finally, local fans may find it difficult to identify with absentee media provider owners who care only about ratings. However, let’s remember that a few instances (currently two in MLB, three in the NBA, and two in the NHL) do not, in and of themselves, signal a trend. Indeed, Disney sold the Angels to billboard magnate Arturo Moreno, and Frank McCourt recently completed the purchase of the Dodgers from News Corp. Bob Daly, chairman, general manager, and 5 percent owner of the Dodgers put it bluntly, “The Dodgers are not a core asset for News Corp. They are not making money, and you don’t have to control the team to control the sports rights” (sportsbusinessnews.com, accessed January 23, 2003). Only the passage of time will sort out these issues. 88 LEARNING HIGHLIGHT: MURDOCH’S BID FOR MANCHESTER UNITED None of the usual explanations for media provider team ownership seems to fit the case of Rupert Murdoch’s buying the Los Angeles Dodgers through his company News Corp. But additional insight into media provider ownership of teams lies in another of Murdoch’s moves. In 1998, his international ambitions led to his tendering an offer for Manchester United in the English Premiere League (European football), for $1 billion ($1.3 billion) through his international cable system, BSkyB. In this particular instance, the answer to why buy the team rather than just the rights is supplied by the British Monopolies and Merger Commission (MMC). The MMC concluded that ownership of the best team in the Premiere League might have reduced competition for the broadcasting rights to all English Premier League matches. After all, it is precisely the Manchester United home games that everybody wishes to see. The upshot might have been fewer choices for the Premier League in the broadcasting of soccer games. The MMC also concluded that the move would improve BSkyB’s ability to secure rights to Premier League matches, would reduce competition in the market for sports premium television channels, and would lead to reduced competition in the wider pay-TV market. These competition concerns were the main reasons for the MMC’s conclusion that the proposed merger was not in the public interest. The British secretary of state for trade and industry, Stephen Byers, nixed the sale, citing the MMC conclusions. Such an acquisition might have altered the competitive situation for broadcast rights, as BSkyB is the dominant player in that market in England, with a contract for £670 million with the Premiere League already. Eventually, through BSkyB, Murdoch did gain control of 10 percent of Manchester United, which he sold in October 2003, for around $113 million ($131 million). We can only contemplate how much more valuable the team would have become if competition had been reduced dramatically, as the MMC feared. Rupert Murdoch tendered an offer of around $1 billion for the Manchester United English football team. [Photo of Murdoch.] Sources: James Walker, “Give Us Back Our Man United,” SportsJones Magazine, www.sportsjones.com, April 23, 1999; sportsbusinessnews.com, October 9, 2003. 89 http://www.sportsjones.com http://www.sportsjones.com SECTION 7 Firms and Their Advertising Choice Let’s move on to the last (or first, depending on your perspective) rung on the ladder in Figure 3.1, firms that buy ad slots from networks (OTA and cable). At the most basic level of economic analysis, advertisements are just another input in any firm’s production process. How do firms determine the amount of advertising to buy and how much they will pay for it? Because advertisements are “just another input,” the usual logic of input hiring holds. If the revenue earned from the additional sales generated by ads exceeds the extra cost of the ads (production costs plus the purchase of ad slots), then the advertisements are worth it. The contribution to revenue made by an additional unit of input is called its marginal revenue product (MRP). In the case of advertising, the logic is precisely as stated, but figuring out the marginal benefits is complex. Here is an example that shows why. Firms often have a devil of a time figuring out the value of ads. Advertising results of the 1991 and 1992 Super Bowls are shown in Table 3.6. Clearly, the results are highly variable. Gillette and Advil advertised during both games and saw significant sales increases in 1992. Budweiser and Pepsi did the same but enjoyed a much smaller increase in 1992 compared to 1991. The interesting outcome concerns the remaining firms. Kellogg’s, Nuprin, and 7-Up did not advertise in 1991 but did in 1992. One would expect an increase in sales given the high price of Super Bowl ads. However, only Kellogg’s enjoyed any increase. The other two firms actually had much smaller sales despite their very expensive Super Bowl ads. How can a firm spend millions on Super Bowl advertising and not show an increase in sales? How did a firm decide whether a 30-second ad during the 1992 Super Bowl was worth $850,000 ($1.3 million)? While the data driving this question are older, the issue is not. As every Super Bowl draws near, questions are raised in the press about whether the ads are worth it or not. Marketing officers are fond of responding about increased “branding” and “impacting” for new offerings, especially from Internet firms. But the answer must eventually lie in a change in sales. Maybe the winner’s curse is in operation. There would have to be a very broad distribution of estimates of Super Bowl ad values among inexperienced bidders for ad slots. If the bidding for sports programming ad slots is very competitive, the most optimistic, highest—and therefore wrong—bid would be the winner. The winner would then be cursed by having won because the true expected value must be less. In this Super Bowl ad context, some firms guessed high and took the ads. Others, like Coke, took no ads. The outcome revealed that 90 the true expected value was lower than those who “bid”—or paid—the most. The sales just weren’t there. This is an appealing piece of logic but, as with our analysis of media providers, does it make sense? After all, who knows more about the value of advertising than firms like Kellogg’s and 7-Up? One might as well throw in Pepsi, because its increase in revenue was only about 3 percent. Surely these firms are neither naïve nor inexperienced. On the face of it, a winner’s curse explanation goes wanting. 91 SECTION 8 Extending the Definition of the MRP of Ads from the Advertiser’s Perspective An alternative explanation expands the idea of the MRP of ads in an important way. Implicitly, in assessing the value of Super Bowl ads, we have assumed that the MRP of ads is just in terms of additional sales per dollar spent. This misses an essential point about advertising that we learned in the first section of this chapter. Sales can come from both new consumers and established consumers that choose to switch brands. What if most of a firm’s advertising budget actually impacts the latter? In this situation, the net result of advertising by one firm depends on the advertising choice of other, competitive firms. Thus, the net result of advertising, given advertising by rivals, is the point of analysis. Let’s see where this leads. If one firm advertises its brand, some consumers will choose it. But if another firm advertises its competing brand, some consumers will choose it instead. If advertisements were at equal levels for the two firms, and each firm had equal access to the same quality of ads, equal numbers of consumers would choose each brand. Thus, the net result of one firm’s ads depends on how much advertising is done by its rivals. In this case, MRP can be portrayed as follows: A1 and A2 stand for the amount of advertising by the two firms. The other element in MRP is X1, the level of sales of the actual product being advertised. This is the general form of MRP because marginal revenue depends on the level of sales of the good generated by advertising. In this specification of advertising MRP for Firm 1, the firm’s additional sales revenue depends on its own advertising and advertising by the other firm. For any given level of advertising for Firm 1, greater advertising by Firm 2 would reduce sales. We could state the same type of MRP for Firm 2 that would depend on the advertising choices of Firm 1. GAME THEORY This insight leads us to cast the choice of advertising in strategic terms. This is the realm of game theory. In the portrayal of the game used here, two firms must choose simultaneously whether to increase their advertising. Simultaneous choices are easily portrayed in a game matrix, like the one shown in Figure 3.4. In this model of strategic advertising, our assumptions are (1) simultaneous choice by the two firms, and (2) the firms are trying to decide whether to increase their advertising spending by $5 million. Each of the cells in the matrix shows the net payoffs after paying the cost of advertising that will result from each firm’s choices. If neither firm advertises (bottom-right corner), then nothing happens, and each enjoys its current level of, say, $15 million in net revenue. 92 The rest of the payoffs make use of our insight about the interdependence of advertising. If Firm 1 were to increase its advertising and Firm 2 didn’t follow suit, then Firm 1 would increase its sales at Firm 2’s expense. In such a case, let’s put Firm 2’s hypothetical losses at $7 million. This would mean that the net position of Firm 2 in this case would be $15 million (its starting point) minus $7 million for a net of $8 million. But Firm 2’s loss is Firm 1’s gain. So Firm 1’s position is calculated as follows: $15 million (its starting point) plus $7 million gained at Firm 2’s expense minus the cost of its increased ads, $5 million. On net, Firm 1’s result is $17 million. This outcome is depicted in the upper-right cell of the diagram in Figure 3.4. Because we assume that the situations are exactly symmetric for the two firms, just the opposite outcome would happen if Firm 2 increased its advertising and Firm 1 didn’t. This is shown in the lower-left cell of Figure 3.4. That leaves the upper-left cell. If Firm 1 increased its advertising at the same time that Firm 2 increased its advertising, the following would occur. Starting from its initial $15 million position, Firm 1 would pay the $5 million ad price. It would gain the $7 million return from customers previously buying from Firm 2. One would think that the outcome would be, again, $17 million. But remember that ads are interdependent. Firm 2 would have also paid its $5 million ad cost and earned the $7 million return from customers previously buying from Firm 1. The gains of $7 million for each firm are canceled out by the identical amount lost to their rival. So for Firm 1, starting from $15 million, $7 million would be gained from previous customers of Firm 2, but an equal amount would be lost to Firm 2. There would be no net change in sales revenue. However, Firm 1 would have paid the $5 million in ad costs, so its net position would be $10 million. The same goes for Firm 2. This is depicted in the upper-left cell of Figure 3.4. THE BEST STRATEGY IN THE ADVERTISING GAME What are the strategies in this case, and is there an equilibrium result? From the matrix, Firm 1 knows that if Firm 2 were to increase its advertising, then Firm 1 would earn $10 million if it increased advertising and $8 million if it didn’t. Firm 1 would, therefore, increase its ad spending. Conversely, if Firm 2 stood pat, Firm 1 would earn $17 million if it increased its advertising and $15 million if it didn’t. Again, Firm 1 would increase its ad spending. Thus, Firm 1 has a dominant strategy. Regardless of what Firm 2 decides to do, Firm 1 will increase its advertising. Turning to the question of equilibrium, we need only note that the payoffs are exactly symmetric for Firm 2. Thus, increasing advertising is also the dominant strategy for Firm 2. Because increasing ads is the dominant strategy for both players, each firm increases its advertising, and the equilibrium result puts the players in the top left-hand corner of the matrix in Figure 3.4. Both increase their advertising and earn $10 million. For the purposes of discussion later in this section, note that both firms would rather be in the lower-right cell than in the 93 upper-left cell. However, their individual actions keep them in the inferior $10 million position. The game theory depiction of advertising choice reveals that the extra benefits of advertising can be difficult to figure out. Although, technically, the equilibrium follows from a consideration of extra revenue, the complicating factor is that the result for one firm depends on the choices of the other. Thus, one consideration can be that failing to advertise (or to increase advertising) means lost revenue as customers move to a rival. Simply put, increasing advertising yielded no increase in sales in our example in Figure 3.4. However, failing to advertise would have been even worse for each firm. Going back to our Super Bowl advertising puzzle, one could come away with the impression that the ads are useless, but this misses the important point that failing to advertise would lead to an even lower payoff. Perhaps this was the consideration by Pepsi and 7-Up, highly competitive soft- drink firms. Thus, an explanation of the outcome in our Super Bowl ad example doesn’t require a winner’s curse. All it requires is a careful consideration of the marginal value of ads, including the value of preserving sales in the face of rivals. THE ADVERTISING DILEMMA The game theory portrayal of advertising offers another insight. If they could, the two firms would rather not advertise, and each would enjoy a $5 million net revenue increase and no change in market share. Both firms would rather be in the lower-right cell of the payoff matrix than in the upper-left cell. If they could, both would stop and escape this advertising dilemma. However, each expects the other to simply continue the escalation, and neither can risk being left behind. In such a case, the two firms are trapped in an advertising dilemma that will be difficult to escape through their own actions. ESCAPING THE DILEMMA WITHOUT OUTSIDE INTERVENTION Escaping the advertising dilemma without outside intervention may be difficult, but it can occur. One of our assumptions is that the game is played one time. One way out of this dilemma involves learning over time. Each firm knows the game matrix in Figure 3.4. Thus, each can look forward to such a game and know the eventual, noncooperative outcome. There is some evidence that firms in this type of situation can tacitly reach an understanding that avoids escalation. A typical type of tacit cooperation would be to advertise alternately and split the market. Perhaps this tacit type of result also helps explain why Coke and Pepsi typically do not advertise in major events at the same time. Note that this is a tacit type of agreement; firms do not get together and collude, they simply realize independently that splitting the market is in their mutual interest. ESCAPING THE ADVERTISING DILEMMA THROUGH OUTSIDE INTERVENTION 94 There is another, more direct way out of this advertising escalation trap. The firms could hope for or obtain some sort of external constraint on advertising. For example, if the firms are beer companies, the government might be convinced to impose an ad ban on beer ads on TV. Government has already outlawed cigarette advertising on TV, as detailed in this section’s Learning Highlight: Is Beer Next? Lessons from the Cigarette TV Ad Ban. The result of an ad ban would be favorable to existing firms for at least two reasons: 1. They do not have to spend as much to maintain sales. They would continue to advertise, but in lower impact and, hence, lower cost media (print, radio, and billboards). 2. Because TV ads are necessary to introduce competing brands, the elimination of TV ads would raise a barrier to entry for new firms. For example, when cigarette ads were banned from TV, profits of existing firms rose and entry into the market essentially ended for many years. 95 LEARNING HIGHLIGHT: IS BEER NEXT? LESSONS FROM THE CIGARETTE TV AD BAN The cigarette industry was under attack from antismoking forces such as the American Lung Association and the March of Dimes. It also was stuck in the advertising game described in Figure 3.2. Then an amazing thing happened. The FTC stepped in to impose the surgeon general’s warning that “Smoking Cigarettes May Be Hazardous to Your Health.” Shortly thereafter, the FCC enforced the fairness doctrine in the case of cigarette ads. This doctrine stated that if local stations allowed any advocacy advertising, they must also allow equal access to opposing viewpoints. Equal access extended to helping the opposing group produce its response. Local stations were suddenly in the antismoking commercial business, and public service announcements took a novel step forward. Thus, in addition to its own advertising escalation dilemma, the cigarette industry now suffered lost sales due to antismoking ads produced at nearly no cost to advocacy groups by local stations. What happened next is that the FCC stepped in to ban cigarette ads on television. Among antismoking forces, there was much rejoicing. But analysts have since pointed out that a number of good things happened to cigarette companies as a result of the TV ad ban. Cigarette company profits soared; there was no decline in consumption; and there hasn’t been a challenge to the existing structure of the industry since. One wonders if the antismoking advocates ever saw what hit them. Antismoking ads may have taught the beer industry some important lessons. [Cigarette billboard ad.] Sources: Gideon Doron, The Smoking Paradox (New York: University Press of America, 1984), and A. Lee Fritschler, The Politics of Smoking (Upper Saddle River, NJ: Prentice Hall, 1995). 96 SECTION 9 The Beer Wars John Helyar, in his book Lords of the Realm: The Real History of Baseball (1994), details the strategic interplay between advertisers and a pro sports league, with media providers right in the middle. The example concerns the “beer wars” of the late 1970s and their impact on the value of MLB broadcast rights. MLB’s first commissioner, Kenesaw Mountain Landis, once said, “Not in my lifetime or yours will you ever hear a beer advertisement during a World Series broadcast.” Later, Commissioner Happy Chandler refused beer ads on the first televised World Series, saying, “It would not be good public relations for baseball to have the Series sponsored by the producer of an alcoholic beverage.” Helyar (p. 393) emphasized how the times had changed by the 1980s. By 1986, Anheuser-Busch sponsored all 26 teams in MLB, up from 12 just a decade before. But they were not alone. The war with Miller was responsible for driving the value of MLB broadcast rights to record highs. Miller and Anheuser-Busch competed fiercely for ad space, and the results were predictable. MLB knew that media providers were making more from this war and charged appropriately for its broadcast rights. Beer advertising was huge money indeed. In 1983, the new contract for MLB’s national broadcasting rights (to last over the period 1984–1989) went for $1.125 billion ($2.4 billion), approximately four times the previous contract. As you might guess, MLB played both ends against the middle and encouraged the escalation between Anheuser-Busch and Miller. First, NBC was offered the entire contract for $1 billion ($2.1 billion). They refused. Said Ken Schanzer, “number two” at NBC Sports at the time, on their beer wars strategy, “Either we succeeded in getting the whole package, or we so crippled our competitor, we made his life hard” (Helyar, 1994, p. 395). Eventually, the $1.125 billion ($2.4 billion) was paid in two pieces, $575 million ($1.2 billion) from ABC and $550 million ($1.2 billion) from NBC. By 1992, the beer wars ended. Their end signaled the end of the dramatic rise in MLB broadcasting rights values. Indeed, the MLB contract has never come close to the dramatic level reached during the beer wars. But the logic that fueled that escalation appears to be gaining steam in another sport. About Disney’s recent win in the competition over NHL broadcast rights, a senior vice president at Turner Sports said, “The money is so big, given the current ratings, that the (proposed) increase is amazing. But there are always two decisions in something like this. One is purely economical. The other is strategic: What does the other guy lose if we take something away from them?” (USA Today Online, www.usatoday.com, August 6, 1998.) 97 http://www.usatoday.com http://www.usatoday.com There couldn’t be a clearer statement of the strategic underpinnings of advertising escalation. The “purely economical” part is just the ad slot revenue. The strategic part is winning the rights away from rivals and the spillover values to the overall revenues of a network. The advertising war between Bud and Miller bid up the value of ad slots to levels never seen before or since. Media providers are perfectly happy to take more for the slots they have to offer. However, competition at the media provider level actually just sent a substantial portion of the overblown ad slot revenue to MLB through an extremely lucrative rights fee contract. The beer wars simply made MLB teams much richer than they already would have been. 98 SECTION 10 Sponsorship One other interesting item deserves treatment while we’re on the subject of advertisers, namely, sponsorship. Sponsorship is just another form of advertising by firms, where firms buy the right to have their name affiliated with a particular event or facility. Many fans find this form of commercialization especially irritating. Perhaps this is because we long for the nostalgic; people would much rather conjure up images of “The Stick” (Candlestick Park) and its San Francisco Giants history than the newer, growing legacy at AT&T Park. Despite the common perception that sponsorship is just the latest in commercialization, it actually has been around for a long time. The longest-standing examples are sponsors of individual sports (sponsorship and college sports are covered in Chapter 13). In men’s golf, just take a look at the PGA Tour Schedule (www.pga.com). The list includes important car companies (e.g., Buick), energy companies (e.g., the Shell Houston Open), communications companies (e.g., the AT&T Pebble Beach National Pro-Am), and a host of others. The list of major sponsors is just as impressive for women’s golf (LPGA Tour Schedule: www.lpga.com), and Nationwide Insurance sponsors an entire tour of men’s PGA hopefuls. Turning to pro team sports, sponsorship of a sort actually has existed for quite some time. The tie is in name recognition with a particular product. Busch Stadium, named after the previous owners of the MLB St. Louis Cardinals, has obvious name recognition with Anheuser-Busch beers. The same type of name recognition worked for Wrigley Field, named after the original owners of the MLB Cubs. The relationship to their line of chewing gum is inescapable. Sometimes name recognition fails. Labatt Breweries owned the MLB Blue Jays from 1991 to 2000. One of the owner’s major regrets is that the nickname Jays caught on instead of the Blues, which would have obvious ties to the brewery’s top- selling Labatt’s Blue Label. More recently, sponsorship has grown into an explicit, big business. In addition to sponsoring events or individual teams, many very large firms now purchase the right to have their name attached to stadiums and arenas to enhance their advertising. Table 3.7 shows recent stadium and arena sponsorship arrangements. It’s possible for such rights to go for as much as $11.3 million per year (adjusted to 2009 dollars). Although lease arrangements vary, sponsorship rights fees commonly go to the primary tenant of the facility. For baseball stadiums and arenas, this is always the single-tenant team owner. For arenas that may have double or triple occupancy, the owner with the highest revenue often receives the value of sponsorship rights. Even at $2.5 million 2009 99 http://www.pga.com http://www.pga.com http://www.lpga.com http://www.lpga.com dollars, the lower end of the values in Table 3.7, this type of revenue is more than enough to obtain an especially able coach, field manager, or front office general manager. 100 SECTION 11 Big Rights Money and Its Impacts In this closing section, we will look at the data on broadcast right fees and the impact of big money on sports. We will examine two important questions: 1. How much money is involved in broadcasting sports? 2. How does the level of this type of revenue shape both our perceptions and the actual outcomes in sports markets? Here’s how my coauthor and I characterized the contribution that media providers have made to the sports business: It’s a no-brainer: the networks, the superstations, the cable sports stations, and local TV stations have inundated pro team sports with a veritable monsoon of dollars, which has affected everything and everyone involved in pro team sports. Over the past 20 years, from 1980 on, the most important single factor responsible for the wild explosion in franchise prices and in player and coaching salaries is the huge increase in pro sports’ television income. (Quirk and Fort, 1999, p. 29) Table 3.8 shows all media revenues in professional sports (national and local, if any) from 1990 to 1996, plus additional years for MLB and the NFL (I know these data are a bit old, but they are the only data I know that cover this issue). For the NBA and the NFL, the increase in these values over time is phenomenal. On an annual average basis, adjusted to 2009 dollars, NBA media revenues grew 18 percent per year in real terms. Remember that, on average, this is a staggering six times the average rate of annual real growth in the U.S. economy. Interestingly, MLB and the NHL about equaled the economy-wide average growth rate. The NHL and MLB data offer an interesting comparison concerning the impacts of fan disillusionment over work stoppages. A strike occurred in MLB impacting both the 1994 and 1995 seasons (more in Chapter 9). Media revenues fell dramatically during both of those seasons and really hadn’t rebounded by 1996 (although, ultimately, media revenues are healthy recently). The lockout in the 1991–1992 NHL season had much less impact on hockey fans. Media revenues rebounded immediately and grew at a 4.3 percent real rate in the following four years. Perhaps it was the fact of a strike by players in MLB versus a lockout by owners in the NHL, but whatever the reason hockey viewers were more forgiving than baseball viewers. Table 3.9, which provides recent national contract amounts, shows that things have kept right on going into the present for the NBA and MLB, while the NFL and NHL have suffered a bit of late. The NFL remains the heavy hitter at a $2.1 billion 101 ($2.3 billion) annual average, followed by the NBA at $925 million ($950 million) annually and MLB at about $786 million ($814 million). The NHL has only a $24 million ($24 million; the value of the dollar fell in 2009 relative to 2008) annual promise from the new cable presence, VERSUS, and a revenue-sharing arrangement with NBC carried on from 2005. But when we turn to a detailed comparison to their previous contracts, both the NFL and the NHL actually suffered decreases in the value of their media contracts. The previous NFL contract (1998–2005) totaled $17.6 billion ($23.1 billion) so that the new contract diminished in value in real terms by 21 percent. While the revenue-sharing portion of the NHL’s contract with NBC is impossible to determine, the guaranteed portion with VERSUS represents a 12 percent decrease (the contract value is the same as before [with ESPN/ABC, 2004– 2006], but inflation has taken its toll). MLB and the NBA, on the other hand, both enjoyed large increase in percentage terms. The previous MLB contract (2001–2006) was $3.3 billion ($4.0 billion). The league’s new contract represents a 42 percent increase. And the contract before that (1996– 2000) generated about $1.69 billion ($2.3 billion), so there can be no doubt that MLB is doing just fine with their national contract (Quirk and Fort, 1999, p. 43). For the NBA, the previous (2002–2003 to 2006–2007) was $4.6 billion ($5.5 billion), so the new contract represents a 38 percent increase. Remember, the source of these truly impressive national contracts is the networks whose ad revenues, shown for 2003 in Table 3.5, drive broadcast rights payments to leagues. It is easy to see just how this market sorts itself out taking the NFL as an example. Table 3.9 shows that the NFL’s current national contract goes six years with ESPN, FOX, and CBS, and another two years after that with ESPN (have no doubt, the NFL will surely end up with new contracts with other networks for these last two years). In 2009 dollars, ESPN is obligated to $1.1 billion annually, FOX $717 million, and CBS to $617 million. From Table 3.5, clearly all of these networks made much more than that from advertising revenues. The only one remotely close is ESPN, but remember that it should be expected that their annual ad revenues increase over the duration of their contracts in Table 3.9. This outcome makes it clear that the bidding will remain competitive for these monopoly broadcast rights over time. First, Table 3.9 makes it clear that each pro sports league spreads the wealth among media providers, but there are always excluded providers. For example, in the most recent set of national contracts covered in Table 3.9, NBC is excluded almost entirely, managing only a revenue-sharing arrangement with the NHL. There always are multiple winners in the bidding game, but not all media providers win by obtaining a contract. They will be keen competitors the next time these contracts come up for bid. MEDIA REVENUES AND PLAYER COST COMPARISONS An even more important aspect of media contracts can be seen when we include all media revenues. Table 3.10 shows 102 all media revenues for MLB. (This really is a behind-the- scenes look at the totals listed in Table 3.8.) Tabled values are left in nominal terms since the point is a comparison between teams in that given year. The sum of media revenues would include individual team negotiations with regional and local media providers, as well as the national contracts. The table also shows total player costs and the ratio of media revenue to player costs in order to provide a comparative benchmark of the value of media revenues and spending on talent. A ratio equal to or greater than one means that media revenues, alone, are enough to cover player costs. All other revenues (gate, concessions, and venue) would simply go to pay the rest of the team costs, including remaining player costs and owner profits. As you can see from Table 3.10, about one-quarter of MLB teams had a ratio below 0.5. The top quarter of teams covered more than 75 percent of their player costs solely out of media revenues. Almost 80 percent of teams covered more than half of their player costs from this revenue source alone. Similar data exist for the NFL for 1999 due to a leak to the Los Angeles Times during one of Al Davis’ lawsuits against the NFL. The results are even more startling than for MLB. Every NFL team covered all player costs in 1999 just with their media revenues except for Green Bay and Kansas City (and they covered 90 percent)! All other revenues went to other costs and profits for owners. The NBA and NHL data for this purpose ended in 1996. In that year, all 29 NBA teams could cover better than 60 percent, and almost half could cover more than 80 percent of their player costs from media revenues. The story was dramatically the opposite of the other leagues for the NHL. Only two of the 26 teams (New York Islanders and Boston Bruins) could cover even half of their player costs with media revenues. Fewer than half the NHL teams could even cover 30 percent of player costs with media revenues. In summary, broadcast rights money is big. Further, certain sports are more important to advertisers and, consequently, to media providers and sports leagues. The NFL typically leads the way, followed by the NBA and MLB. The NHL always is a distant last. Why would a sport like the NFL, with so many fewer broadcast slots, make so much more money than the other sports in total? Of course, the answer is that the NFL provides better reach to the important target demographic of advertisers. Media providers collect this value and pass it through to the NFL. VARIATION IN MEDIA REVENUE FOR TEAMS AND LEAGUES In any given sport, the same type of variation witnessed earlier for total revenue also holds for media revenue. There is substantial media revenue disparity around the mean in any sport. In every sport except the NFL (with its nearly complete revenue sharing), the vast majority of teams are below the average in media revenue. Ultimately, this imbalance contributes to one of the major problems for sports. Recall the implications of the uncertainty of outcome hypothesis: More 103 balanced competition is more interesting to fans than less balanced competition. But when fans are willing to pay more money to watch high-quality teams, if owners give the fans higher quality, then the owners will have the money to spend on talent. As with the conclusion for revenue in general, variation in media revenues contributes to less balanced competition. If these differences continue to grow over time, as they have in recent years, then leagues run a real danger of losing one of the things that fans demand, namely, an acceptable level of competitive balance on the field. The common lament is that we all may end up watching nothing but larger-revenue teams in the championship games of every sport. This is a very real possibility if no other teams can compete economically. Leagues are quite aware of this danger and have invented a variety of mechanisms to take care of it. We’ll discuss this in more detail in Chapter 6. For now, you only need to understand the source of the potential problem. OTHER PROBLEMS WITH BIG MEDIA MONEY Competitive balance problems are just one issue concerning the commercialization of sports, but there are other big money problems in sports. Some argue that “It’s not about the game anymore,” it’s all about TV. However, sports have always been big business. The only difference from the past is in terms of magnitude. For example, TV does intrude on the actual play on the field, directly. There weren’t any TV time- outs through most of the history of sports broadcasts, and the length of games clearly diminishes fan satisfaction. Another downside of commercialization concerns the behavior of players. Players are freer to behave as they please because their value is so high. When problem players misbehaved in “the good old days,” getting rid of them wasn’t nearly so expensive as it is today. Leagues tend to put up with problem players who happen to be stars much more than in the past. The revenues they generate make it so. In this regard, they are no different than other entertainment megastars who misbehave. Another important talent issue is that the absolute level of competition would be lower if it weren’t for the commercialization of sports. According to Michael Roarty, executive vice president of Corporate Marketing and Communications for Anheuser-Busch, advertising support by American business has allowed sports to improve and expand so that there is greater participation as well as more events on TV (Sporting News, May 11, 1992, p. 9). Clearly, without this involvement and the subsequent elevation in player value to teams, we would not see the very high absolute level of quality that we currently enjoy. This is evident in the investment that athletes now make in their own careers, especially after they make it to the pros. Pro athletes used to consider their sport to be a part-time endeavor, and all had other jobs in the off-season. Given the immense returns driven in part by commercialization, training and 104 preparation are full-time occupations. Roarty goes on (remember, this is in 1992, and the $100 million ESPN package, $152 million in 2009 dollars, was a tremendous event): In 1979, Anheuser-Busch became the charter advertiser for a fledgling organization called ESPN, with a $1.4 million contract. Last year, our company and ESPN signed the largest sponsorship commitment in cable advertising history, a five-year, $100 million package. Measured in 2009 dollars, that’s an increase from $4.1 million to $152 million. Along with players, fans clearly are beneficiaries of this type of corporate commitment to sports. The pursuit of money by athletes has its downside. First, in order to win the big prize, nearly obsessive levels of training and practice are required of athletes from their earliest age. However, only a very few will ever actually win the prize, so much of this effort may be wasteful. Especially, because young people are assessing the possible outcomes, the losses are in terms of the fundamentals of education that lead to job skills. This lost lifetime earnings stream is extremely costly to society if young people are overly optimistic about their chances. We’ll revisit this in Chapter 7. Another problem occurs when the intensity of competition over a potentially lucrative prize leads to risky health choices by athletes. Performance-enhancing drugs and playing hurt can lead to long-term disability. It is always easy to say that it is the athlete’s choice in these matters. However, many of the athletes making these choices have not reached the age of legal consent. There is an economic logic to these choices. Performance- enhancing drugs, if universally adopted, would raise the absolute level of performance. Fans value higher levels of absolute performance over lower levels. However, if all athletes were using the drugs, relative outcomes wouldn’t change. The final question becomes, “How much is society willing to give up to obtain higher absolute quality levels?” An offset is the decreased value that fans might place on the behavior that leads to performance enhancement in the first place. If enough fans feel this is bad behavior, then any gains fans enjoy by increased absolute performance are offset. In any event, some fans may lament the huge amounts paid to athletes and even begrudge them that pay. Society may come down in its final analysis against the lamentable loss of human potential that goes into the making of a very few star athletes. We are forced to note our recurrent theme: The money comes from fans in the first place. Apparently, while it might be viewed as a mixed blessing, the vast majority of fans enjoy the fruits of commercialization in the form of higher- quality competition than they would enjoy otherwise. 105 SECTION 12 Chapter Recap The willingness of advertisers and, lately, of fans directly to pay for sports programming fundamentally shapes the revenue side of the sports business. Firms use advertising to sell their products. Some advertisers wish to reach particular target demographic groups most likely to watch sports. Media providers purchase broadcast rights and supply advertisers with ad space for that demographic. Sports leagues provide the programming to the media providers. All of the money that changes hands during sports advertising comes from fans, who purchase the goods and services that are advertised. Fans also now purchase programming directly through subscription services from media providers, including the leagues themselves, by cell phone and over the Internet. Sports teams possess the broadcast rights to their home games. Some games are sold through their leagues in the form of a national contract, while others are sold by the team through local contracts. The proportion of games sold in each type of contract varies by league. Media markets are extremely competitive. No single media provider can extract more than the normal return to its endeavors. The massive sums that change hands reflect the market power of teams and leagues. Media providers pay rights fees to leagues and teams for programming. Through an auction system, media providers compete for these rights. Some observers criticize this process because rights fees may end up being greater than the value of ad slots. However, programming provides other types of values to the media provider. Media providers advertise their own programming, as well as the products of paying advertisers, and sequence other programs around the sports broadcast. Finally, local affiliates also enjoy the same types of values in their own local advertising markets. The overall value of sports programming should be compared to rights fees if one wants to judge their net economic contribution to media providers. In some cases, the net value of sports broadcasting rights is negative, even from the overall perspective of media provider revenues. If bidders are inexperienced and do not have enough information to guess broadcast rights values, then a winner’s curse may arise. A broad range of estimates of the expected value of the rights and competitive bidding will elicit the highest bid and the wrong value. The winner is cursed to have paid too much relative to the true underlying value of the rights. One must evaluate each case in order to see if the special circumstances that would yield a cursed winner hold before assuming this explanation. 106 Much has been made of media provider ownership of teams. Why does it make sense for a media provider to actually own a team rather than just buy its broadcast rights? It is a difficult case to make on profit grounds. CBS’s ownership of the Yankees shows that profits are not guaranteed to a media provider owner. Subsequent sale of the Dodgers and Angels by media producers serves to reinforce this claim. The media provider pays the same cost for broadcast rights, and it is difficult to see any cost reduction in the actual production process. Any tax advantages would usually be offset. In addition, it would be both difficult and expensive to try to reduce competition with rival media providers by buying up teams. A much simpler explanation is that teams are just a good, solid financial investment when all of the values of team ownership are considered. Advertising MRP determines how much advertising a firm will purchase. At first look, some ads don’t appear to be worth it because they do not increase sales. In such cases, rival firms see the value of advertising as saving the current level of sales as opposed to increasing sales. If a firm doesn’t advertise, it may lose sales to its rivals. Bidding is structured by leagues and conferences to elicit the most that such advertising wars can generate for the ultimate holders of market power, the leagues themselves. Sometimes, firms can tacitly recognize ways to avoid such a disastrous situation, but at other times, only outside intervention can save them. Sponsorship is another form of advertising. In team sports, the newest manifestation of sponsorship is in the naming rights to venues. These rights can be upward of $11 million annually. These amounts certainly make a difference in terms of the financial health of team owners. Media revenues are often large enough to cover team player costs. In the NFL, it is common for team owners to cover all of their team payroll solely with TV revenues. On the downside, substantial variation in media revenue contributes to the current lack of competitive balance in many leagues. The commercialization of sports causes many a great deal of anguish. 107 SECTION 13 Key Terms and Concepts You should have run into each of these in pop-ups in the text of this chapter: • Start-up consumption • Brand choice • Target demographic group • Advertising reach • Ratings • Media providers • Rights fee • National broadcasting contract • Advertising slots • Local broadcasting contract • Slot fees • Media provider competition • Marginal revenue product of broadcast rights • Winner’s curse • Media provider ownership of teams • Game theory • Strategic advertising • Advertising dilemma • Tacit cooperation • Ad ban • Beer wars • Sponsorship • Media revenue disparity 108 SECTION 14 Review Questions 1. Explain the role of advertisements in both (a) start-up consumption and (b) brand choice after start-up consumption. What other factors matter in each case? 2. What determines advertising reach? 3. What are ratings? How are they measured? Why do they matter to advertisers? 4. Which of the participants in Figure 3.3 ends up receiving most of the money from advertising (over and above their costs)? Why? 5. Explain the difference between a national broadcasting contract and a local broadcasting contract. 6. Describe the different approaches to selling broadcasting rights used by the NFL and MLB. 7. Define the marginal revenue product of broadcast rights to media providers. Why does this marginal revenue product exceed the value of ad slots that media providers sell to advertisers? 8. What factors must exist in an auction to generate a winner’s curse? Do these factors characterize sports broadcast rights auctions? How? 9. What are the possible advantages to media providers of owning a team rather than just buying that team’s broadcast rights? What problems might be caused if this type of ownership takes over in sports? 10. Define the marginal revenue product of ad slots to advertisers. Why does this marginal revenue product exceed just the additional sales generated by running an advertisement in an ad slot? 11. In the advertising game matrix in Figure 3.4, describe each payoff pair in each cell, starting clockwise from the top-left cell. Why does one player do better than the other in the off-diagonal cells? 12. Explain Firm 1’s best strategy in the advertising game in Figure 3.4. Is it different from Firm 2’s best strategy? Why? 13. In what way is the equilibrium of the game in Figure 3.4 a dilemma? 14. Which team in MLB has the highest, middle, and lowest media revenue in Table 3.10? What types of competitive balance issues arise from this difference in media revenue? 109 15. Beside competitive balance, what other problems arise with big media money? 110 SECTION 15 Thought Problems 1. Try the following small experiment on your own. Of your friends and classmates, how many think that advertisements can induce start-up consumption? How many would fall into the brand-influence camp? 2. Males ages 18–34 drink beer and buy cars. But they also buy ice cream and nonathletic shoes. Why aren’t these types of companies among the top sports advertisers in Table 3.1? 3. Here are some companies that have measured the importance of their brand name and rate it highly (Sports Business Journal, November 6, 2000, p. 31): Intel, GE, Disney, Cisco Systems, Citibank, Kodak, Heinz, Xerox, The Gap, and Kellogg’s. Even though they seem just like the firms in Table 3.1, only Disney even advertises on the Super Bowl (Table 3.4). Why? 4. In Table 3.3, what was the largest percentage increase in Super Bowl advertising rates between years? What explains this large increase? 5. Why would a media provider advertise its own other shows during one of its sports broadcasts rather than sell the ad slots to paying advertisers? Why allow local affiliates to advertise on national broadcasts? 6. What characteristics of the ABC/NBC competition for Monday Night Football would lead to a winner’s curse? (Refer to the Learning Highlight: The Curse of Monday Night Football?) Do you think the result was a winner’s curse outcome? Why? 7. If you were a media provider, would you buy a team or just buy its broadcasting rights? Why? Are all your reasons economic? 8. Use the idea of the marginal revenue product of advertising slots to explain why advertisers put so much more into professional football than into other sports (see Table 3.2). Does this explanation also explain the amount advertisers put into the sport with the lowest amount? Explain. 9. How much more valuable in percentage terms is sports advertising to GM than to Anheuser-Busch (use the data in Table 3.1)? 10. Why isn’t a Super Bowl ad slot even more expensive than the amount listed in Table 3.3? (Be careful to think about both sides of the market! For example, Table 3.4 shows Toyota Motor Corp. spending on a recent Super Bowl, but, for example, sports advertising is often only 25 111 percent of Toyota’s total spending on advertising. What would happen if the price of an alternative to Super Bowl advertising fell?) 11. Does the outcome in Figure 3.4 depend upon the price of advertising? Suppose the price of ads rises to $8 million. What is the new outcome? Instead of an ad price increase, suppose an ad costs $5 million and that the ad brings $6 million in new start-up consumption for either firm that advertises. What is the new outcome? Given these results, do you always expect to find the kind of sports advertising dilemma described in the text? 12. Who gained most from the beer wars? Why? 13. What roles do attendance and TV play in the value of a sponsorship agreement? Are the data in Table 3.7 consistent with your answer? 14. Why can nearly all NFL team owners cover all of their player costs with media revenues, while nearly none of the owners in MLB can do so? 15. Does the ability to cover at least half of the total player cost also correlate to the ranking of revenues by league? Why or why not? 112 SECTION 16 Advanced Problems 1. Is there really any difference between the NFL and MLB broadcast rights approach? After all, in either case, fans are pretty much guaranteed to see nearly all of their home team’s games. Also in either case, fans will get other games of broad, national interest as well. What is the difference in their approach? 2. Using the data in Table 3.1, devise an index of the relative importance of sports advertising for Anheuser-Busch versus Proctor and Gamble (hint: compare their relative adjustments from 2008 to 2009). What determines this relative importance of sports advertising between different advertisers? 3. What determines the ranking of favorite sports to watch in Figure 3.1? Why isn’t soccer in the top five for men or women? If you took this same poll 10 years from now, what might be different? Why? 4. A number of very prominent firms are not even listed in Table 3.4 (e.g., other beer and soft drink firms), but both Universal Pictures and Sony Pictures (Columbia/Tri-Star) are there. Why? (Hint: Think strategically.) 5. How would one discover whether there really was a winner’s curse result in the ABC/NBC bidding over Monday Night Football? (See the Learning Highlight: The Curse of Monday Night Football?) 6. What was the key to Ted Turner’s success using the Atlanta Braves as a programming anchor for TBS? Why couldn’t Rupert Murdoch replicate this success by using Fox to purchase the Dodgers? 7. Refer to the Learning Highlight: Murdoch’s Bid for Manchester United. What lesson does the European example rejecting Rupert Murdoch’s bid to buy Manchester United hold for his purchase of the Dodgers through FOX? (Think carefully about the relationship between his regional FOX networks and local MLB broadcast contracts.) 8. Discuss the major impacts of TV on sports in terms of club revenues and profits, wages, and competitive balance. In your answer, distinguish between outcomes in the NFL and MLB. 9. Calculate the following ratios from Table 3.10: the highest to lowest media revenues and the highest to average media revenues. Show how these simple ratios help to enlighten a discussion of the relationship between competitive balance and media revenues across leagues. What data would you need in order to use these ratios to discuss the relationship between competitive balance and media revenues within a given league? 113 10. Economist Robert Barro, in his book Getting It Right (1997), has argued that it is efficient to regulate/ban the use of performance-enhancing drugs because they have no effect on relative performance by players. If all used drugs, the result would be that all players would be stronger and there would be no difference in outcome on the field. Critique this view. 114 SECTION 17 References Barro, Robert. Getting It Right: Markets and Choices in a Free Society. Cambridge, MA: MIT Press, 1997. Crompton, John L. “Sponsorship of Sports by Tobacco and Alcohol Companies: A Review of the Issues,” Journal of Sport and Social Issues 17 (1993): 148–167. Doron, Gideon. The Smoking Paradox. New York: University Press of America, 1984. Fort, Rodney. “The Value of Major League Baseball Ownership,” International Journal of Sport Finance 1 (2006): forthcoming. Fritschler, A. Lee. The Politics of Smoking. Upper Saddle River, NJ: Prentice Hall, 1995. Helyar, John. Lords of the Realm: The Real History of Baseball. New York: Villard Books, 1994. Quirk, James, and Rodney Fort. Pay Dirt: The Business of Professional Team Sports. Princeton, NJ: Princeton University Press, 1992. Quirk, James, and Rodney Fort. Hardball: The Abuse of Power in Pro Sports. Princeton, NJ: Princeton University Press, 1999. Walker, James. “Give Us Back Our Man United,” SportsJones Magazine, www.sportsjones.com, accessed April 23, 1999. 115 http://www.sportsjones.com http://www.sportsjones.com SECTION 18 Suggestions for Further Reading Surely, I can come up with some. 116 CHAPTER 4 Team Cost, Profit, and Winning Owners are never the most popular sports figures in a city. When a team loses consistently, the owner is the person the fans blame; when the team wins, the owner is the person the fans want to get out of the way so that the announcer can interview the coach and star players. There is a reason that no owner has ever been pictured on a football card. —Gene Klein, former owner of the San Diego Chargers Klein and Fisher, 1987, p. 12. CHAPTER OBJECTIVES After reading this chapter, you should be able to: • Explain the difference between the short- and long-run decisions that confront sports team owners. • Understand that, subject to the usual real-world allowances for uncertainty, the short- and long- run choices made by sports team owners are consistent with profit maximization. • Explain how profit maximization creates tension between those who want champions (fans, players, and on-field managers) and owners, who also want to win but have to pay attention to the revenues and costs associated with doing so. • Express how variation in profits across teams in a given league contributes to competitive balance problems within a league. • Identify how acceptable sports accounting techniques can make a profitable sports team appear unprofitable, and why sports team owners may prefer it that way SECTION 1 Introduction Wayne Huizenga, owner of the MLB Florida Marlins, a team in a major media market, claimed losses in 1997 of $34 million ($45.2 million) just after winning the World Series. When Broward and Miami-Dade counties refused to build him a new, retractable-roof ballpark, Huizenga made it clear that if the revenues he expected in order to field a champion weren’t forthcoming, he wouldn’t try to field a champion. He sold off the players responsible for the series victory, ostensibly to reduce losses. The Marlins’ payroll fell from $53 million ($70.4 million) in 1997 to $13 million ($17 million) the next year. The team fell from winning the World Series to last place and, with a winning percent of 0.333, was the worst team in baseball in 1998. Considering the introductory quote by Gene Klein, is it any wonder that the fans often despise owners for their business choices? In this chapter, we’ll explore the reasons behind team owners’ business choices. Our guiding assumption will be that team owners maximize profits. In exploring these choices, we’ll develop the cost side of producing sports outputs. Bringing in the revenue side developed in the previous chapter, our profit maximization assumption will allow us to identify some of the important business choices of owners. In particular, we’ll analyze the quality of the teams that they put on the field. One of our most important discoveries will be that there is a clear conflict between winning, costs, and profits. We will see that in some cities, where revenues are insufficient to cover the costs of fielding a winning team, there will always be unhappy fans, players, and coaches. We will also look at some real- world accounting data on sports teams and see how certain acceptable accounting practices can make a team worth millions of dollars look like it is losing money. 118 SECTION 2 Profit Maximizing Owners? Ultimately, we are going to need to know what motivates owners to understand the level of winning that they choose to put in front of their fans. Some view owners as gentle sportspeople. The Yawkee family, whose trust once owned the Boston Red Sox, and the Haas family, previous owners of the Oakland Athletics, are perfect examples. This view would have it that the fun of being involved with the sport, the coaches, and the players is the main payoff. In other words, owners would just be sports fans with bigger wallets—rather than just buying tickets they buy the whole team. Or perhaps owners want to be famous—many owners are better known as owners than they are as real estate or industry magnates. A variation on the gentle sportsperson view is that owners simply wish to maximize the number of wins their team achieves. One needn’t go far to find owners complaining about the lack of money involved, painting themselves as benevolent sportspeople. Gene Klein, the irascible former owner of the NFL’s San Diego Chargers, describes a love–hate relationship regarding his ownership of the San Diego Chargers. He makes it plain that the money wasn’t that great, claiming that one of the two happiest days of his life was when he sold the team (the other was when he bought it!). That might have been true in his case, but it doesn’t matter that profits weren’t large. What matters for our purposes is that profits are pursued. THE PROFIT MAXIMIZATION HYPOTHESIS AND ITS IMPLICATIONS It should be clear from the first three chapters that the most common economic view holds that owners pursue profits in the usual way, total revenues minus total costs. Under the profit maximization hypothesis, owners make decisions about their venues, players, and media contracts in order to maximize the difference between total revenues and total costs. Even if actual observed levels of profits are not always great, owners pursue them, and their choices are made in order to make profits as large as possible. Their choices involve the quality of the team to be fielded, the quality of the fan experience in the venue, and pricing. Owners would love to field winning teams if there is money in it. Otherwise, owners are quite content to keep costs low, field second-rate teams, and make as much profit as possible. The “as possible” part has to do with the willingness of fans to pay for some level of quality. Later, we will turn to some data to determine how much money is involved. Sports business analysts have gained important insights using the idea that owners are motivated by profit. This is true no matter how wealthy an owner might be. Profit maximization is a common and powerful abstract in economics, even if it 119 only approximates behavior or only covers the behavior of some. Its usefulness is in the insights it generates. Profit maximization can seem too strict an imposition on real- world decision makers. However, there is nothing in an economic view of team owners that precludes them from making mistakes. It may well be that an owner intends to win the pennant some year, makes the deals to put the level of talent on the field that is required, and is simply wrong about the talent that was hired. For example, as teams near the play- offs, they trade for particular players that are expected to put their receiving teams over the top and into postseason play. However, some of these players may be nursing injuries unknown to their receiving teams. As a result, owner expectations go unmet, and the trades and the final result disappoint fans. There is also room for plain old bad luck. Over time, mistakes and bad luck should even out so that talent choices meet expectations. If not, another owner would buy the team and do better. 120 SECTION 3 The Short Run versus the Long Run Every market has two sides. We discussed the demand side in Chapters 2 and 3. The other side, the supply side, requires producer knowledge of the costs of production. We will start by examining the basics of production. Then we will see how knowledge of players’ physical contributions to team success, along with the price of players, results in the costs of winning that owners face. Teams produce what fans want. As we saw in Chapter 2, fans want, among other things, grace and beauty, a sense of commonality, and relative and absolute competition. All of these items are also highly correlated with winning. Winning teams tend to have more grace and beauty, and winning tends to be a stronger common bond than losing (although fans of lovable losers also share commonality, albeit of the miserable type). We can all probably agree that competitiveness is the essence of a winning team and that winning stands as a reasonable proxy for team output. Because of this, we will use the terms winning percent and team quality interchangeably. Winning percent is the number of games won by a team divided by the total number of games played. However, this winning percent misses the championship incentive inherent in the structure of all team sports. Some think that fans only care about winning the championship. However, winning matters—a team cannot make it to the championship without a high winning percent. Owners actually can be in two different situations when they think about the level of winning they will put in front of fans, what economists refer to as the short run and the long run. In the short run, elements of the team production function are fixed. This includes anything covered by contractual obligation in force during production. For sports teams, the primary fixed inputs are talent, on and off the field, and the facility in which the team plays. One difficulty with this economic distinction concerns the relationship between the short run and actual physical time. In terms of actual physical time, the short run for team quality choices tends to correspond to a given season, from preseason on through the regular season, play-offs, and championships. In contrast, the short run for the stadium or arena corresponds to the economic life span of the facility. This means that the short run for a stadium might be 30 years or more. To clarify, if any of the resources used to produce winning are fixed, the owner is in a short-run situation. So, in the short run, the owner makes decisions about the number of season tickets to sell and the price to charge (because demand slopes downward) or the number of games to put on local television, given that the quality of the team is fixed; that is, quality has been more or less determined prior to league play. 121 However, teams often alter their level of on- and off-field talent during the season. Negotiating player trades, changing managers or front office personnel, and even moving into a new stadium will alter the quality of a team. When an owner changes the quality of his or her team, the team is in a new situation of different fixed inputs. Note that the short run is not associated with any specific period of time, but is a period where team quality is held constant. In the long run, everything about the team’s production process is variable—a new stadium may be built, the lineup may be completely altered, or the head coaches and management may be replaced. Perhaps the easiest way to think about the long run is in terms of an owner’s planning process. Despite the short-run situation confronting owners, they can always step back and think about how they would alter both team quality and the rest of the quality of the fan experience in the stadium. This long-run planning process captures the essence of exactly what the long run means economically. Whenever all elements of the process are variable, then the long run is under consideration. 122 SECTION 4 Short-Run Production and Costs The essence of the short run is that some decisions about putting winning in front of fans have already been made. In this section, we discuss in detail the elements of the winning production process that are fixed and the elements that remain variable in the short run. When combined with knowledge about the price of these inputs, short-run costs can be determined. FIXED INPUTS IN SPORTS PRODUCTION In the sports team production process, short-run fixed inputs include the player roster, the stadium, other contracted personnel such as the front office general managers and on- field managers and coaches, insurance coverage, and loans to cover investments. A team also has certain obligations to its league that are fixed, such as participation payments. All of these elements have costs attached to them. Because the factors are fixed, the costs associated with them are called short-run fixed costs. They must be paid whether or not the team ever plays a game. VARIABLE INPUTS IN THE SHORT RUN Other elements in the team production process are short-run variable inputs. These include team operations expenses. In the movie Major League, the owner decided to punish the team to get them to lose by busing them and, when she couldn’t bus them, flying them on a rickety airplane charter. She also cut off the hot water to the locker room. However, she didn’t cut off team spending on her own enjoyment and comfort. These are extremes, but you get the idea. These are all examples of short-run variable team-operating expenses. Teams often control stadium operations, at least during their tenancy and often for all nonsports events as well. These stadium expenses also are variable. A team can pour resources into quality stadium experiences for fans regardless of the quality of play it has chosen to put on the field. For example, entertaining mascots and between-innings music and video presentations enhance the stadium experience. Recently, some teams have begun to sponsor live musical entertainment after the game, even including portable dance floors. Because any of these inputs can be cancelled at any time, they are variable. Scouting and player development expenses also are variable. They have a great deal to do with the future quality of the team, but are not fixed for any given quality choice. For example, in the extreme, a team could stop all scouting and player development. Finally, advertising and marketing are variable inputs. The teams can market like crazy or not. 123 Just as with short-run fixed inputs, short-run variable inputs have costs attached. Because the factors are variable, the costs associated with them are called short-run variable costs. The level of these costs will vary, depending on the level of short- run outputs offered to fans. For example, the greater the level of attendance, the more labor that must be hired to manage ticket sales, turnstiles, clean-up, and parking and to sell concessions. SHORT-RUN TOTAL COSTS Short-run total costs are the sum of short-run fixed costs and short-run variable costs: Short-Run Total Costs = Total Fixed Costs + Total Variable Costs, or: Table 4.1 shows total operating expenses for the Seattle Mariners from 1993 to 1998. These data may seem old, but believe it or not this is one of only a couple of data sets generally available that provide the detail we need to investigate short-run costs. Because a strike shortened both the 1994 and 1995 seasons and revenue-sharing practices in MLB changed after the 1995 season, we’ll restrict ourselves to the period 1996 to 1998 in order to make observations about the levels of short-run costs. The time period is also short enough, so we’ll ignore inflation. Can you determine which spending categories are included in TFC and which are included in TVC? Table 4.2 shows which operating costs fall under the fixed, variable, and mixed headings. As you can calculate from Table 4.2, TFC averaged about $58.6 million a year for the Mariners from 1996 to 1998. General and administrative expenses averaged $7.2 million a year. These general and administrative expenses pose a problem in identifying fixed and variable costs. Although it is difficult to know which portions of general and administrative expenses are fixed or variable, it seems reasonable that the majority are fixed (e.g., legal expenses) and don’t change much over time. That puts TFC in the neighborhood of $65.8 million annually, whereas TVC averaged $20.8 million a year. Therefore, SRTC averaged around $86.6 million a year from 1996 to 1998. Tables 4.1 and 4.2 show that player compensation dwarfed all other cost elements, fixed or variable, at about $45.2 million on average per year. Player compensation thus represented about 52 percent of SRTC for the Mariners over the years 1996 to 1998, whereas TFC averaged about 76 percent of the SRTC, and TVC made up the remaining 24 percent. Although outside analysts are not privy to many sports teams’ expense ledgers, the little data that are known, combined with reports from consultants, indicate that this relationship holds for nearly all sports teams, with TFC dominating the expense side of the ledger. GENERAL GRAPHICAL ANALYSIS OF SHORT-RUN COSTS Table 4.2 shows just one example of short-run cost outcomes for a team, the Seattle Mariners. Note that it only presents 124 three years’ data. From Table 4.2, for the Mariners, on average, TVC = $20.8 million, TFC = $65.8 million, and SRTC = $86.6 million. In addition, let’s have attendance proxy for short-run output. Average yearly attendance for the Mariners from 1996 to 1998 was 2.9 million. If we use the typical economic reasoning about short-run costs, along with these averages from Table 4.2, we can make a rough approximation of how short-run costs might look, in general. Let’s think about how TFC, TVC, and SRTC vary with attendance. In general, depicting TFC is easy. These costs are fixed, as represented by the horizontal line in Figure 4.1. Using the Mariners’ data as our guide, fixed costs are about $65.8 million. However, this number is particular to our data. The Mariners were a pretty good team through the late 1990s, finishing second in the American League West in 1996, first in 1997, and third in 1998. Better teams will have a higher TFC because they will choose to put together a more expensive set of players during a given season. What about TVC? Diminishing returns to variable inputs characterize all short run production. Diminishing returns means that, given the level of fixed inputs, increasing the use of variable inputs will eventually result in decreased marginal product. For example, given that the dimensions of the stadium are fixed in the short run, adding more and more labor to the fan experience will mean that an additional vendor eventually will not enhance the stadium experience as much as the last one hired. If so, then TVC might rise at a decreasing rate over some low range of output, but once diminishing returns set in, TVC must begin to increase at an increasing rate. TVC are shown in Figure 4.1 for a reasonable hypothetical range of these types of costs relative to attendance. In Figure 4.1, diminishing returns set in when attendance is about 3.5 million. Because we don’t have much data, this is simply a rough guess. In actuality, diminishing returns could set in at higher or lower levels of attendance. However, it is true that attendance rarely exceeds 3.5 million for any team in the league. Finally, in Figure 4.1, SRTC is the vertical sum of total fixed and total variable costs. We used our data on the Mariners to anchor the location of these costs functions in Figure 4.1. At an average attendance of 2.9 million, TVC = $20.8 million, TFC = $65.8 million, and SRTC = $86.6 million. 125 SECTION 5 Long-Run Production and Costs In the long run, all inputs are variable, including the team’s level of talent. What must the team consider when it investigates altering its quality? There are two ways to look at this long-run concept. First, we could compare a series of short-run outcomes for a given team. In fact, we could use the Mariners’ data in Table 4.2 in just this way. By using winning percent as a proxy for quality, we would just look at teams that got better over time and see how much more they spent in terms of fixed costs as they improved. This series of short- run outcomes would be the result of long-run planning at the beginning of every season. This long-run choice pretty much boils down to purchasing more talent. The problem with this approach is that we would see only a restricted set of relationships between costs and winning percent. Why restrict our look at costs to just one team? Each season, we get to observe all of the teams and the variation in costs across teams of differing quality. There is an abundance of data to investigate the relationship between owners’ spending and team quality. Of course, one season will have its exceptions; some teams will be better than their payroll/talent choice would indicate, and some teams will be worse. To see how these factors even out over time, we’ll have to look at the same types of choices over a longer time period. TEAM QUALITY ANALYSIS IN A GIVEN SEASON The 2003 MLB season is as good as any for our purposes. As Table 4.3 shows, there were really bad teams, such as the Tigers (a team that won just over a quarter of its games), and very good teams like the Yankees, Braves, and Giants (who won just over 60 percent of their games). As the averages in Table 4.4 show, teams at the bottom of their divisions won about 38 percent of their games in 2003, that is, an average winning percent of 0.377. To move up to the fourth-place position in their division, the losing teams would, on average, need to increase their winning percent by about 0.078 points, to 0.455. An additional 0.061 points moves the team from fourth place to third place in its division, whereas adding 0.037 points moves the team from third place to second place. Adding 0.041 points facilitates the jump to first place at a winning percent of about 0.593, on average, across the divisions in 2003. Here is an essential point: The teams at the top of the winning percent heap are better than other teams because they hire better players. Sometimes, a historically dominant team ends up with interrupted success for a short period. Sometimes, a team that never had success has a brief, shining moment. Winning is uncertain, even for teams loaded with talent. 126 However, on average, over time, higher-quality teams win more games. TEAM QUALITY ANALYSIS OVER TIME Although analyzing statistics from one season provides interesting information, it doesn’t demonstrate how hiring talent averages out over time. So let’s look at the 1999 to 2003 seasons. Table 4.5 shows the same information regarding average team-winning percents by place in the final standings. It also shows the average increases teams need to advance through the standings for those five seasons. You can see that the pattern in Table 4.5 is similar to the one shown in Table 4.4. It would seem that a low-quality team has an average winning percent of about 0.408. Successively, such a team must add about 0.039 points in order to move up the ladder one rung, 0.054 points to move into third place, 0.050 points to move up to second place, and 0.046 points to make it to first place. Let’s use this historical view of winning to characterize how teams typically think about increasing their output of winning, that is, their long-run quality choice. Essentially, this involves adding increasingly better players who move teams up in the standings. Even teams that end up choosing a low winning percent can at least consider how they might have a higher winning percent. For our purposes, we’ll characterize these increasingly better players as stars. The idea of stars is clear—they are a cut above the average player. We will now examine the positive relationship between stars and winning percent. At the bottom rung on the winning percent ladder, a team either has no stars (but may produce a few in the future) or only older, faded stars at the end of their careers. The data in Table 4.6 are my best guess about how adding more stars to a team will increase that team’s winning percent. Looking at the data in the table, it seems reasonable that a team would have to add a star to get close to a winning percent of 0.429. Just as reasonably, to move from a weak 0.420 winning percent to contention, a 0.528 winning percent, a team probably would need to add five star players. Let’s be very clear: Table 4.6 is just an intuitive snapshot of the relationship between adding stars and winning; it is not the result of any more extensive analysis than you have just read. The data in Table 4.6 portray a long-run relationship between adding talent and increasing winning percent. Adding stars increases the output of winning that fans pay to see. The long- run winning percent production function in Figure 4.2 is a graphical presentation of this relationship. There must be some sort of limit to management ability in the long run once a large enough number of stars is on one team. After five stars are added to any team, winning percent increases at a decreasing rate. This is the same type of argument that is made of all production functions in the long run. Once a process reaches a particular scale, it becomes unwieldy. LOOKING INSIDE THE PRODUCTION OF WINNING 127 In our discussion thus far, culminating in Figure 4.2, we glossed over an underlying production process: In a nutshell, player talent and managerial or coaching skill are mixed to produce winning on the field. It takes a keen awareness of ability and the way to combine the abilities of the roster in order to produce winning. Is the bundle of talent in a given roster a power team (e.g., hitting or pitching in baseball and running in football)? Is the talent better suited to finesse (e.g., hitting and speed in baseball and deceptive passing offense in football)? An entire process underlies the observed winning percent produced by some level of talent. Economists have extensively analyzed this production activity. On the one hand is the analysis of individual player contributions in a team environment. Scully (1974) was the pioneer, as he often was as covered in this section’s Learning Highlight: Gerald Scully (1941–2009): Pioneering Sports Economist. His original ideas about how to estimate player marginal product form the basis of a large literature summarized and extended by Berri, Schmidt, and Brook (2006). We’ll also look at individual player contributions in Chapter 7. Equally important is the impact of managerial skill on the field. Again, Scully pioneered (Porter and Scully, 1982; Scully, 1989) the research on managerial efficiency in sports, followed rather quickly by Kahn (1993) and Singell (1993). Essentially, this work finds that baseball managers contribute substantially to the process of winning, that these managers improve with tenure on a given team, and that players appear to respond more to better managers. In the real world, stars are not added to a lineup in the nice orderly fashion portrayed in Figure 4.2. Some are added early on, others are added only if teams need them at particular times, for example, in the play-offs. Still others are added subject to market availability. In an attempt to gain understanding of a complex issue, economic analysis reduces the problem to a much simpler setting. At least we have captured just how it is that adding more stars will alter winning percent, even if we’ve abstracted a bit on when they are added. One way to keep this clear is to remember that we are thinking of production in the long run as a planning process. Player and coaching talent is used to produce winning on the field, on the ice, and on the court. Adding more stars raises winning percent. THE LONG-RUN COSTS OF TALENT AND WINNING A production function like the one shown in Figure 4.2 allows us to derive the cost of winning to an owner. Because talent is one of the most important long-run choices for owners, the cost of talent will represent the largest element in the team’s long-run total cost function. Cost functions relate costs to output. Earlier, we chose winning percent as the long-run team output. Thus, long-run total cost is the relationship between winning percent and the cost of obtaining it. Before we move on, a few comments about the costs of winning are in order. In economics, costs relate to output. However, many people mistakenly refer to costs in terms of the price of inputs. For example, you can often find 128 newspaper stories on the “costs” per at bat, a player’s salary divided by that player’s number of plate appearances. Similarly, columnists sometimes list “costs” per game pitched or the “cost” per minute played by basketball or hockey players. These are not costs but are, instead, just interesting ways to portray the price of the talent input. Teams do not produce at bats, pitching appearances, or minutes played any more than a fast-food restaurant owner produces clean tables. Player tasks are performed not as individual products but as part of an effort to win games. It is important not to confuse input prices with output costs. For example, the easiest way to make the “cost” per at bat smaller would simply be to have inexpensive players appear at the plate most often. However, such a strategy would prove very expensive in terms of long-run output, namely, winning. When one puts weak players at the plate, the team loses more often. While input prices are not the costs of winning, they do matter in terms of determining those costs. It’s easy to find the costs of winning once we introduce the price of star players. Continuing with our baseball example, at the top end, Alex Rodriguez of the New York Yankees earned about $28 million in 2008, whereas lesser stars earn less. The data for 2008 show that the top 10 percent of salaries run from $10 million per year to A-Rod’s lofty height. The data on 2008 team payrolls suggest that bottom-dwellers spend $45 to $55 million in total, but even these teams have a star or two. So let’s assume a lineup without any stars would cost $25 million. Finally, let’s measure winning percent in whole units, that is, multiply by 1,000 (e.g., a winning percent of 0.700 in whole units would be 700). Table 4.7 combines the long-run relationship between stars and winning with our assumptions about the price of obtaining stars. The cost with no stars is $25 million, the first few stars come at a price of $10 million each, and so on, up to the very last superstar at $20 million. The result in Table 4.7 is the team’s long-run total cost schedule (this hypothetical falls a bit short of the actual Yankees at $201 million in 2008). (Remember: All costs are variable in the long run, and the talent level of a team, or its quality, is a long-run consideration.) Long-run total costs are simply the sum of the player costs (salaries) as more stars are added and winning percent increases. It is the cost of winning percent that the team owner seeks to find. For example, in order to play at a winning percent of 528, the team would have to add five stars to a “no- star” lineup. The total cost of playing 528 ball becomes $25 million (the cost of a team with no stars) + $10 million (the cost of one star) + $10 million (the cost of the second star) + $10 million (the cost of the third star) + $11 million (the cost of the fourth star) + $12 million (the cost of the fifth star) = $78 million. Technically speaking, the salary column in Table 4.7 is not the long-run marginal cost of talent, or winning percent. If we measured talent in terms of winning percent, then the marginal cost of each unit of talent from the first star would be $10 million divided by the increase in winning percent— 129 that is, 429 minus 420—for nine more winning percent points. Marginal cost in this range would be about $1.1 million per winning percent point added. Because we’re talking more in terms of the actual way that talent is added to a team, player by player, then one simply adds up the salaries that produce any given level of winning in order to obtain the total talent cost of that level of winning percent. The example in Table 4.7 is an abstraction of the way talent is actually hired. Owners don’t simply line players up in a row, from worst to best at higher prices, and make their choices. Instead, talent is groomed in either the minor leagues or college or traded back and forth between teams. Plans to reach a particular level of winning can take time. The portrayal in Table 4.7 is an abstraction representing the thought process that owners go through when considering the level of talent they would want to hire. The real world, as always, is much messier. The total cost data in Table 4.7 are portrayed graphically in Figure 4.3. The 2003 total salary bill for the most probable 25-man rosters of MLB teams varied between $19.6 million and $149.7 million. Our talent cost function pretty much mirrors actual outcomes in MLB. COST COMPARISONS BETWEEN LOWER- AND HIGHER- QUALITY TEAMS To get a clear picture of the level of quality ultimately chosen by an owner, it is instructive to look at the short-run situations associated with each level of long-run total cost. Each point on the long-run total cost function is associated with a particular short-run production setting. This means that an owner will face a particular set of short-run cost functions, similar to the one shown for the Mariners in Figure 4.1, for any given level of winning that is chosen. Thus, one can think about the short-run as a particular choice of team quality. For a team to alter its quality, it is choosing to face higher fixed costs in its next short-run setting. We will use this simplification to talk about two kinds of teams. A lower-quality team has similar variable costs but much lower fixed costs (primarily on-field player and management talent) than does a higher-quality team. Of course, there are a few other long-run cost function elements beside player costs. These elements include other types of off- field talent that help determine the long-run winning chances for a team as well as the rest of the quality of the fan experience inside the stadium. According to the annual Forbes magazine report for 2007, MLB team total costs were between $92.4 million and $374.3 million, with an average of $166.6 million. In the more general consideration of the long run, the costs of “quality” both on the field and in the stadium would need to be added to the analysis. THE TENSION BETWEEN WINNING AND COSTS At this point, students typically ask me, “We’ve developed all of this structure, but what have we learned?” The answer is an important insight. As teams choose higher winning percents, 130 in the long run total costs rise. A look at Figure 4.3 shows they rise only gradually through a range of the winning percent at first, but eventually, they rise at an increasing rate. Remember that the argument for the long run is that costs eventually rise at an increasing rate due to diseconomies that seem reasonably attributed to management, either in the front office or on the field. This means that there is a tension between winning and costs. The more an owner wants to win, the more it’s going to cost. Perhaps because owners tend to be wealthy, fans, players, and on-field managers often miss this crucial limitation on winning. Hiring talent to put a team in position to win can raise costs by $40 million or more relative to a team with an average winning percent of 0.500. For example, suppose that a team owner determines that talent sufficient for 640 winning percent points must be hired in order to win the pennant (not unusual in the American League Eastern Division, for example). As shown in Table 4.7, the owner would have to hire seven stars at a total cost of about $105 million. On average, that’s about $173,841 per point. It may well be true that a wealthy owner would have that kind of money, but whether that level of talent is worth it to a profit-maximizing owner depends on the team’s revenue structure. If fans are willing to pay enough to cover the costs of hiring this level of talent, then it is worth it. But if the fans are not willing to pay, the owner will lose money. If the owner cares about the bottom line, don’t expect the team to win at that level. This can really grate on fans and on-field managers. Lou Piniella, manager of the Tampa Bay Devil Rays, voiced a common hope of fans and on-field managers, “It’s a Catch-22 situation. If you don’t win, people just (don’t come). But if you put a good team on the field and compete with any team that comes in here and win your share of ballgames, people will come out. It won’t happen overnight, but you’ll see a steady progression” (St. Petersburg Times, June 29, 2003, from their Web page, www.sptimes.com, accessed June 29, 2003). If Piniella is right, everything he says will follow. But if he is wrong, then following his prescription could lose the owner of the Devil Rays millions of dollars. If the owner doesn’t care about the bottom line and gets large amounts of satisfaction out of winning, then the talent may be hired even if the fans are unwilling to pay for it. Given that fans, players, and managers often butt heads with owners over their choices about winning, the evidence seems pretty clear that the bottom line does matter to nearly all owners. 131 http://www.sptimes.com http://www.sptimes.com LEARNING HIGHLIGHT: GERALD SCULLY (1941– 2009): PIONEERING SPORTS ECONOMIST Pioneering sports economist Gerald Scully was at the forefront of the study of sports economics, contributing to the very first book on sports economics, Roger Noll’s (1974) Government and the Sports Business. Scully analyzed the pro sports business for most of his career, and his numerous articles and two important books are obligatory reading for all sports economists. Although his most famous work was an analysis of player pay (Scully, 1974), where he estimated player contribution to team revenues prior to free agency, his work also covered nearly every aspect of the baseball business. Especially important for this chapter was his work on managerial efficiency, which he began with his colleague Phil Porter, but added to again on his own (Scully, 1989). Managerial efficiency is the analysis of how far a given manager’s actual winning outcomes are from the maximum winning possible observed among the history of mangers who had the same level of talent to work with. The lower the manager’s actual winning relative to the historical highs, the less efficient the manager is. Porter and Scully found that the most efficient managers, such as the legendary managers Earl Weaver, Sparky Anderson, and Walter Alston, set the bra for all others. Not surprisingly, managerial efficiency correlated quite highly with longevity. This leads to the notion that learning with a given team happens over time, illustrating that systematic improvements in managerial efficiency occur the longer a manager is with a team. Later, in The Business of Major League Baseball (Scully, 1989), Scully extended the idea of managerial efficiency to expansion teams. He contends that expansion teams start out very inefficient but improve substantially over their first 10 years. By the 10th year, Scully claims, expansion teams are about as efficient as any average team. Putting the idea of managerial efficiency in a general social science context, Scully (1994) later showed that MLB tenure is directly and significantly related to managerial efficiency. This casts doubt on the organization sociology view that managerial efficiency has more to do with the firm and market environment than with individual talent—at least in sports! Pioneering sports economist Gerald Scully [His photo.] 132 SECTION 6 Profit Analysis Generically, profits are defined as the difference between total revenue and total cost. We discussed the cost side earlier in this chapter. We found that team quality, using the winning percent proxy, is the owner’s long-run decision variable. We can imagine (or, if we’re lucky enough to have the data, calculate) the owner’s long-run cost of talent function. An example appears in Figure 4.3. Once the level of quality is chosen, the team will be in a short- run situation, with a specific set of short-run cost functions that depict TFC (primarily player compensation) and TVC (primarily team operations and player development). Output in the short run concerns the selling of seats at the stadium (and parking that goes along with it), other parts of the attendance experience (concessions and other types of shopping, restaurants), and games on television. In what follows, we’ll use gate attendance as our unit of measurement, recognizing that these other outputs are also important. Figure 4.4 shows how downward-sloping demand functions dictate marginal revenue and total revenue functions. One of the determinants of the shape and location of demand functions is fan preferences, especially for quality. When the owner chooses a higher level of team quality, fan demand increases and so do total revenues. However, quality is determined by the choice of winning percent—depending on its winning percent, the team will sell different amounts of attendance at different prices. Therefore, the particular set of demand, marginal revenue, and total revenue functions shown in Figure 4.4 would hold for a particular choice of winning percent. Team owners who maximize long-run profits seek to determine the highest level of profit that can be obtained at any winning percent they might choose. Thus, while each demand function gives the team’s short-run revenue potential, a look across all of these opportunities will allow the team owner to find the team’s long-run profit possibilities. All that the owner has to do is pick the winning percent with the highest profit. The result is maximum profit in the long run. It is easiest to derive the long-run profit function by comparing profits at two particular quality levels. SHORT-RUN PROFITS WHEN A TEAM CHOOSES LOWER QUALITY Let’s find the profit-maximizing level of output for a particular winning percent, or the team owners’ short-run profits at that winning percent, for a team considering a lower-quality choice. This decision is shown as WL in the left-hand side of Figure 4.5 (W stands for winning percent; subscript L denotes lower quality). Note that the SRTC function is for a 133 lower-quality team, as is the total revenue function. For example, a lower-quality level might be WL = 0.270. The owners know the least that can be spent on on-field and off- field talent according to the long-run total cost function. A representative SRTC function for this lower-quality choice appears in the left-hand side of Figure 4.5. On the demand side, quality is one of the primary preference determinants of the shape and location of the demand function for sports teams. If the team owners made the lower- quality choice, they would confront a given demand and its associated total revenue function. A hypothetical total revenue function for a lower-quality choice also appears in the left- hand side of Figure 4.5. The SRTC and total revenue functions are shown in the abstract, without any scale on the axes of the graphs. We would be able to put actual numbers on the axes only if we had data on revenue and cost over the possible levels of attendance in the short run. The short-run profit function simply plots short-run profits as the difference between short-run revenues and total costs at all different levels of attendance, A, where the L subscript signifies profits, revenues, and costs at the lower-quality choice. The profit function is shown in the lower portion of the left-hand side of Figure 4.5. The shape of the short-run profit function should make sense to you. For example, at attendance below A0L the team loses money because SRTC(A0L)L > R(A0L)L; the same is true to the
right of attendance level A1L. Between these attendance levels,
profits ?(A)L rise at first, reach a maximum, and then fall.
For a given level of quality, it is clear that choosing very high
attendance leads this team to lose the most money. The costs
incurred with ever-increasing attendance for a given team
quality eventually dwarf the revenues. A profit-maximizing
owner chooses the highest possible profit, given this lower
level of quality, which occurs at attendance level A*L.
Remember this combination for future reference: For a lower-
quality winning percent, WL, the profit-maximizing
attendance and profit combination is (A*L, ?*(A)L).
SHORT-RUN PROFITS WHEN A TEAM CHOOSES
HIGHER QUALITY
If we examine another possible quality choice for the owner,
we’ll see almost immediately how long-run profits come into
play in determining just how good a team will be. Let’s
suppose that the same team owner calculates profits if a
higher quality is chosen, denoted WH in the right side of
Figure 4.5 (with the H subscript standing for higher quality).
For example, the right side might represent a team that
chooses WH = 0.375.
Higher quality will alter both the costs and revenues from
attendance. On the one hand, TFC for a higher-quality team
are much larger because better on-field players and
management talent must be purchased. The result is that
134

SRTC are shifted to a higher level for a higher-quality choice.
On the other hand, the demand for this higher-quality team
shifts to the right relative to a lower-quality choice. Thus, the
revenue function for a higher-quality choice is also higher. In
the right-hand side of Figure 4.5, a new profit function is
derived for this possible higher-quality choice.
Again, a profit-maximizing owner would choose that level of
attendance for a team of higher quality with the highest short-
run profits. This is at attendance level A*H in the right side of
Figure 4.5. Remember this combination for future reference
as well: For a higher-quality winning percent, WH > WL, the
profit-maximizing attendance and profit combination is (A*H,
?*(A)H).
NOTES OF CAUTION
Two notes of caution are in order at this point. In the
hypothetical results of Figure 4.5, attendance is higher when
the owner chooses a higher-quality level, that is, A*H > A*L.
This seems to be an intuitive outcome. However, it needn’t be
so. For example, if revenues rose less than portrayed in the
right side of Figure 4.5, it is possible for attendance to be
lower even though quality rises.
The same caution should be applied to the profit outcome. In
Figure 4.5, profits are depicted as larger if the owner chooses
the higher-quality level, that is, ?*(A)H > ?*(A)L. Again, this
seems intuitive. However, profits could be higher for a lower-
quality choice. If revenues rise less than depicted in the right
side of Figure 4.5, profits could be lower even though quality
rises. Given the paucity of actual data on revenues and costs,
we are left with observed owner behavior and a bit of analysis
that you will see later in this chapter to settle this issue.
THE DOWNSIDE: TALENT DUMPING
The tension between winning and profits helps explain talent
dumping and why owners will continue to make trades or
sales of player talent that are destined to irritate the fans who
want a champion regardless of the owner’s bottom line.
The history of talent dumping goes way back, indicating that
the pursuit of profits has always dominated the thinking of
MLB owners. The “Curse of the Bambino” plagued the Boston
Red Sox from the time owner Harry Frazee sold Babe Ruth to
New York in 1920 in order to finance a Broadway play (he also
sold a bunch of other players and, in 1923, the team itself)
until the Red Sox’s World Series win in 2004. Boston didn’t
win a World Series over the intervening 84 years. Connie
Mack sold off his champion Philadelphia Athletics, one at a
time and in groups, because he wouldn’t meet the rising
salaries caused by the rival Federal League in 1915. Another
sell-off of the three-straight-pennant-winning Philadelphia
A’s by Mack occurred in 1931 due to declining revenues during
the Depression. Charlie O. Finley peddled most of the dynasty
Oakland A’s in the mid-1970s—they won big but couldn’t draw
fans. More recently, San Diego, Houston, and Florida all have
dumped talent. The reason in every case (except for Babe
135

Ruth) is clear. The cost of winning wasn’t worth the revenue
generated.
136

SECTION 7
Profits and Quality Choice
in the Long Run
The calculations in the preceding section cover only two of
many different levels of winning percent between zero and
one that team owners might choose. The same type of analysis
would hold for each potential team quality level. Each will
have its total revenue, short-run cost, and profit functions,
and each will yield a particular level of attendance that
maximizes profits in the short run.
A comparison across all short-run profit opportunities will
determine the best winning percent of all from the profit
perspective. In the long run, that’s the one a profit-
maximizing owner will choose. This choice can be directly at
odds with maximizing winning percent. This helps explain
how teams like MLB’s Milwaukee Brewers, the NBA’s
Memphis (previously Vancouver) Grizzlies, the NFL’s New
Orleans Saints, or the NHL’s New York Islanders have
survived quite nicely from the profit maximization perspective
by being lovable losers for so very long.
THE LONG-RUN PROFIT FUNCTION
Figure 4.6 shows the two results for the lower- and higher-
quality choices in Figure 4.5. Winning percent is on the x-axis
and profits on the y-axis showing the long-run trade-off
confronting this owner. In each case, there will also be an
associated short-run profit-maximizing output level. Let’s
suppose that we went through the same exercise, finding the
profit-maximizing result for all possible levels of quality
(winning percent). We would find the rest of the combinations
of winning percent and profit shown in Figure 4.6. This is the
long-run profit function confronting the team owner. Notice
that the relationships from the previous section are intact in
Figure 4.6: WH > WL and ?*(WH) > ?*(WL), but remember
this was by deliberate choice and that you should heed the
notes of caution in the previous section.
The long-run profit function in Figure 4.6 has a very
particular concave shape. For the long-run profit function to
be concave, fan willingness to pay for winning percent would
have to increase at a slower rate than costs. If this were the
case, then revenues would continue to rise, but eventually
costs would rise faster. If this relationship between increasing
revenues and costs holds, then profits will rise at first with
winning percent, and then eventually fall at higher winning
percent levels. In the figure, profits fall long before the team
gets anywhere close to winning all of its games. This is the
source of the fundamental tension between winning and
profits.
THE LONG-RUN PROFIT-MAXIMIZING LEVEL OF
TEAM QUALITY
137

An owner who maximizes profits will choose that level of
winning percent associated with the highest profits in the long
run. In Figure 4.6, the peak in profits occurs at a winning
percent denoted W*, and the best that the owner can do in the
long run is a profit level of ?(W*). For this particular owner,
the long-run outcome is a team that consistently strives for a
winning percent greater than 0.500. There are three
important observations from this outcome—profits can
constrain winning, profits influence competitive balance, and
small market teams are not necessarily economic losers.
PROFITS CAN CONSTRAIN WINNING
In the long run, if costs eventually rise faster than revenues,
then profit maximization constrains owners in their pursuit of
winning. This means that the owner will not hire talent in
order to win an outrageously large share of games unless that
is what fans will pay the most, in net, to see. Some owners
may be willing to violate profit maximization, but they may
pay dearly. Losses will mount, and other potential owners will
start to make overtures to buy the team. This is the
fundamental tension between owners, fans, players, and
managers. Owners who care about profits are constrained in
their pursuit of winning.
COMPETITIVE BALANCE
For our second insight, we return to the competitive balance
issue. The fact that profits constrain the pursuit of winning
can harm competitive balance on the field. When one team is
located where fans will pay far more for quality than the fans
of another team are willing to pay, then the higher-profit team
will win more, on average, over time. Not all owners will
choose to be lovable losers and some, with relatively larger
demand and large potential revenues, will choose to win the
most. As with revenue variation and competitive balance,
profit variation can harm balance.
This concept is shown in Figure 4.7. Here, the long-run
profit opportunities for two different owners are shown. They
are referred to as small market (S) and big market (B) owners.
The big market owner has greater profits through most of the
range of winning percent than does the small market owner.
Accordingly, the big market owner chooses the larger long-
run winning percent, W*B > W*S. As long as this relationship
between small and big market teams goes unchanged, the big
market team will dominate, competitively speaking.
Eventually, small market fans may revise their willingness to
pay downward over time. This observation relies on the idea
that big market teams actually do have higher profits.
However, remember the notes of caution from the previous
section before jumping to any conclusions.
SMALL MARKET DOES NOT NECESSARILY MEAN
ECONOMIC LOSSES
The competitive imbalance problem does not necessarily
mean that small market teams are not profitable. It’s just that
they may not be as profitable as their big market counterparts.
Don Fehr, Major League Baseball Player Association director,
has been quoted as saying, “This whole thing is not really an
138

issue of big market, small market. It’s larger revenue teams,
with smaller revenue teams complaining about not making as
much as their bigger partners” (Sporting News, August 24,
1992, p. 10). Note that the outcome in Figure 4.7 echoes these
sentiments. The small market owner still earns positive
economic profits, just not as much as the big market owner.
A big market owner can correctly calculate demand and
revenue, know costs, choose the level of talent to maximize
profits, and then have the whole thing fall apart. Team
chemistry, unexpected injuries to key players, off years in key
players’ lifetime performance, and just plain bad luck can all
wreck a perfectly rational economic plan. The same goes for
some small market teams. The owner can correctly calculate
small revenues, know their costs, choose the level of talent to
maximize profits, and then have the team do much better than
had been planned. On-field outcomes are uncertain in the
sports world. The analysis of profit-maximizing owners gives
insight only in terms of averages relative to this type of
uncertainty. There will be exceptions every year, which is
what makes sports fun. However, we would expect the
exceptions to even out over time. On average, big market
teams will have profit maximization occur at a higher level of
winning percent than small market teams.
Furthermore, the world is never set in stone; teams that once
were smaller-revenue teams can become larger-revenue teams
and vice versa. My favorite examples of the former are the
Atlanta Braves and Seattle Mariners. In 1990, the Braves
ranked 25th of 26 teams in terms of revenues. The Mariners
were dead last at 26th. By 2005, not only were both teams
closer to the top of the revenue heap (e.g., the Braves had
moved to 12th out of 30 teams), but the Mariners had actually
overtaken the Braves (Seattle was ranked 8th)! Of course,
nothing in this book is noted without purpose, and these
examples lead to a discussion of just when will new ownership
make a quality choice that differs from previous owners?
ALTERATIONS IN LONG-RUN PROFITS
Upon purchase of the Cincinnati Reds in 2006, new majority
owner Robert Castellini said (cincinati.reds.mlb.com, January
20, 2006), “If that’s all we have to offer, we shouldn’t be
hanging around. We’re going to give it 110 percent to put a
contender on the field.” This statement was made about a
team that had endured five consecutive losing seasons and
last appeared in the play-offs in 1995. The team subsequently
finished at below 0.500 for the 2006, 2007, and 2008 season
(and were playing 0.425 ball after 106 games in the 2009
season). One view is that a winning percent rebound takes
time. In this chapter, we’ve covered some theory and
observation leading to a much more fundamental question.
Why would a given team try to improve its quality? From the
economic perspective, the answer lies in the owner’s
perceptions of how profits might change in the future. Profit-
maximizing owners will beef up on talent if they believe that
doing so will raise revenues at a greater rate than the costs of
making it happen. An owner might expect that something
about demand has changed or is about to change. If that
139

figure:%23

figure:%23

change will increase revenues at a greater rate than would
previously have been the case relative to costs of winning,
then a higher winning percent will generate greater profits
than before.
Similar logic holds for the opposite scenario. If demand is
expected to decrease, then a lower winning percent will
maximize profits. If you are catching on to the power of the
insights gained from profit maximization, then this section’s
Learning Highlight: When Can a New Owner Change a
Team’s Fortunes? should make perfect sense to you.
Based on the Learning Highlight, we can see that only if the
last owner made horrible mistakes or has the same vision as
the new ownership but no money should we expect the new
owner to alter the level of team quality. Otherwise, the next
owner will run the team at pretty much the same level of
quality as chosen by the former owner. That would be the
profit-maximizing level of team quality.
140

LEARNING HIGHLIGHT: WHEN CAN A NEW OWNER
CHANGE A TEAM’S FORTUNES?
Based on profit maximization, should we expect a new owner
to behave any differently from previous owners? Let’s look at
a real-world example.
The title of a 1995 article in Baseball Weekly, when the Disney
corporation took over the helm of the Anaheim Angels, was
“Sale to Disney Spawns Anaheim Optimism.” Given the
massive resources at Disney’s command, the article predicted
that the Angels’ payroll would rise and the team’s chances for
consistent postseason play would improve. In addition,
because marketing is Disney’s cup of tea, it seemed reasonable
that the team’s budget for enhancing the fan experience
during the game would increase. In short, “Mickey Ball”
would be better and more fun than the brand of baseball
provided by the previous owners, the “Old Cowboy,” Gene
Autry, and his heirs.
Let’s look just at attendance impacts. In the first three years of
Mickey Mouse’s tenure with the Angels (1996–1998),
attendance averaged 2.0 million annually. In the three years
prior, it averaged 1.8 million—about a 15 percent average
increase. In those same three years prior to Disney’s purchase,
the team averaged 0.462, while in the three years after, 0.493,
for a 7 percent performance improvement. There was no
change in play-off appearances with Disney’s ownership. The
Angels were not in the play-offs in any of the six years.
However, Disney spent, on average, about 28 percent more on
players (in real terms) than Autry did in the three years prior.
Now, just in terms of attendance, was it worth an additional
$5.5 million in real terms to generate another 200,000 fans
each year? Because these figures translate into $27.50 in
spending (without including any of the variable costs) for each
of these additional fans at the gate, the answer is yes, if fans
spend more than that when they visit. The Fan Cost Index,
published annually by Team Marketing Reports, is a spending
index of admissions, concessions, parking, and memorabilia.
For 1998, the index averaged $28.33 per person for the
Angels. Therefore, Disney cleared $0.83 per fan. If variable
costs were less than that on a per fan basis (or about
$166,000 per game), then the bump in quality paid off. But
also remember that other revenues were produced as well.
Eventually, the team improved, winning the World Series in
2002. But Disney had been trying to sell it for a few years
prior to that and did sell the team to Arturo Moreno in 2003.
Mr. Moreno immediately invested in a higher-quality team,
landing free agent Vladimir Guerrero. Moreno clearly believed
that he saw something in the market for Angels’ success that
Disney had missed.
141

SECTION 8
Real-World Case Study:
The Seattle Supersonics
2000-2001 to 2004-2005
Many team owners and league commissioners claim large
losses for some of the teams in their leagues. Are these claims
believable? For example, in Chapter 1, you saw that franchise
prices and expansion fees for new entrants into sports leagues
have grown—sometimes astronomically. How can this be if
teams lose money? In this section, we’ll examine the economic
data of the Seattle Supersonics just prior to their sale to the
group that quickly moved them to Oklahoma City to become
the Thunder. Our aim is attempt to untangle the confusion
surrounding owners’ claims of massive financial losses. We’ll
also try to square our theory of profit maximization with the
real world of taxation.
THE SEATTLE SUPERSONICS ANNUAL OPERATIONS
Table 4.8 shows the annual operating revenues and expenses
of the Settle Supersonics (now the Oklahoma City Thunder)
for the 2000–2001 through 2004–2005 seasons (again, the
table is pretty cluttered, so only values adjusted to 2009
dollars are shown). The original information on the Sonics
was supplied to me by a reporter at the Seattle Times when
then owner Howard Shultz (of Starbucks Coffee fame and
fortune) was pursuing public-funding support for a new
arena. Again, while these data are just for a few years, they
comprise a large part of what precious little information exists
in detail about team revenues and expenses.
As discussed in Chapter 2, revenues come from the gate,
broadcast rights, arena arrangement, and royalties on league
properties. The first two dominate the Sonics’ revenues, but
they also made the play-offs in 2001–2002 and 2004–2005.
Note that getting to the first round and losing grossed the
Sonics about $2.5 million in 2001–2002 (the Sonics did not
make the play-offs in 2003–2004 but received a pittance in
sharing from the league). Getting deep into the play-offs is
quite lucrative, grossing the Sonics about $9 million when
they made it to the 2004–2005 conference semi-finals.
Operating expenses also follow the categories you would
expect. Far and away the largest expense is player
compensation. Although not shown separately in Table 4.8,
USAToday.com’s NBA Salary Database lists the Sonics’
payroll between $54 and $62 million over this period
(adjusted to 2009 dollars). Given these other data, it is clear
that salaries are included in “Team and Game Expenses.”
Teams also promote themselves, pay salaried employees,
cover interest expenses and their obligation to the league as
NBA members, and pay their taxes.
During his argument with local government over a new arena,
Mr. Schultz was quoted repeatedly that the team had large
losses over his entire period of ownership. Table 4.8
documents this, showing losses from operations of over $30
142

million (2004–2005) to nearly $60 million (2001–2002).
How in the world could this be true given that those were the
team’s two most successful seasons during his tenure as
owner? Better yet, if these types of losses really are correct,
how in the world did Schultz sell the team at the end of the
2005–2006 season to a syndicate headed by Clay Bennett for
$350 million, a 75 percent increase in just six years over his
original $200 million purchase price? That’s a cool 6.9
percent annual rate of growth adjusted for inflation on a team
that never shows positive net operating revenue and has
claimed operating losses between $24 million and $60
million.
ECONOMICS VERSUS ACCOUNTING
The answer lies in taking an economic view of team accounts
rather than the accounting view. The accounting information
goes to the Internal Revenue Service (IRS) and must satisfy
IRS rules. However, the question we’re addressing is the value
of owning this team. There are a host of reasons why the data
reported in Table 4.8 don’t tell the whole story of the value of
owning the Sonics.
RELATED BUSINESS OPPORTUNITY
First, there are other values to team ownership that will only
appear on another annual operations statement. When you
are the owner of the local team, doors to business
opportunities are opened that otherwise might not be. Much
as what happens on the golf course, business deals are made
and business connections are reinforced in the owner’s box
during and after basketball games. There is also the political
clout that comes with team ownership. All of these are values
that would occur only because of team ownership. However,
they do not appear on the team annual operations statement;
they are only found on other annual operations statements to
which we are not privy.
Raul Fernandez, 10 percent owner of the Washington Capitals
and part owner of the Washington Wizards, put it this way,
“It’s not just taking people to the luxury box, it’s taking them
to the locker room.Being able to share the ownership
experience with clients is extremely rare and a huge
competitive advantage in business” (Washington Post,
October 14, 2003, from their Web page,
washingtonpost.com).
COSTS THAT MAY NOT BE COSTS
An examination of the cost side also turns up some interesting
values to team ownership— not all costs are costs. Other
salaries for some teams include salaries to owners or
dependents. For example, during the court hearings on free
agency in the NFL in 1994, it came out that Norm Braman,
owner of the Philadelphia Eagles, paid himself a salary of $7
million ($11.4 million) in 1990 (Reingold, 1993). Similarly, the
Griffith family populated all of the main management
positions for the Minnesota Twins for many years. Without
doubt, services were provided in each case. But to the extent
that Braman’s self-collected salary was above the going rate or
the Griffiths were paid above the going rate, these expense
143

entries may be excess direct payments to owners (and their
family members).
There are other ways profits can be taken. Those same NFL
free agency trials revealed that Hugh Culverhouse, then owner
of the Tampa Bay Buccaneers, borrowed $22.7 million ($37
million) from the team in 1990 for investment and debt
reduction (Reingold, 1993). Some expenses can be direct
revenues to other activities of the owners themselves. For
example, suppose a member of an owner group is also a
principal in a law firm. That owner might “sell” the team the
firm’s legal services so that the cost of legal services (in the
“other general and administrative expenses” category)
becomes a revenue entry on the law firm’s annual operations
statements.
REVENUE SHIFTING AND TAX ADVANTAGES FROM
CROSS-OWNERSHIP
Another way that the value picture can be clouded concerns
revenue shifting and cross-ownership. Suppose a media
provider is a majority stockholder of a team (not the case for
the Sonics). The total taxes of the two entities may be lower if
the media provider, by virtue of its majority ownership of the
team, pays the team nothing for local broadcast rights. This
means that operating revenue is higher for the media provider
than it would be if it paid for the rights. However, if the media
provider’s revenues are already such that it shows losses, and
the additional revenues do not push net revenues to a positive
amount, then the media provider pays no taxes. In this
scenario, the team’s revenues fall by the amount of the unpaid
rights fee. This reduces the tax liability of the team. On net,
the total tax liability of the two entities is lower. Moreover, the
taxes that are avoided are just as easy to spend as an increase
in revenue. Note that this cannot occur if the media provider
wholly owns the team. In this case, the firm at large is the
taxpaying unit and there is no value to the shifting approach.
At the end of this section, the Learning Highlight: Wayne
Huizenga’s Fish Story details other aspects of cross-
ownership that allow a team to look like a financial loser
when, in reality, it generates tens of millions of dollars.
PASS-THROUGH FIRM ORGANIZATION AND PERSONAL
INCOME TAX SAVINGS
We should not forget that there is also a bottom line for the
team itself. If book profit is positive, the owners get that
amount, net of taxes. However, one of the most interesting
aspects of team ownership is, believe it or not, even if the
bottom line is negative, the value of owning the team can still
be millions of dollars. This was originally discovered by
Benjamin Okner (1974) 35 years ago and rigorously
demonstrated by Quirk and Fort (1992) over 15 years ago.
Suppose that net operating revenues are negative. In this case,
the owners pay no taxes on team operations. However, they
must eat the losses, right? Well, not completely, as long as
their team is organized as a pass-through firm for tax
purposes. Subchapter S corporations and partnerships are
pass-through forms of business organization that allow both
144

income and losses to carry over to individual 1040 forms
where personal income is reported to the IRS. The idea of the
pass-through originated in allowing small business owners to
pay typically much lower personal income taxes rather than
the corporate income tax rate on business profits. So if net
operating revenues are negative for the sports team, the losses
appear on the owners’ individual 1040 forms, and these losses
reduce the owners’ tax liability on other income unrelated to
the sports team. Thus, losses on the team shelter other
nonownership values. But from the economic perspective,
reduced taxes spend just like income.
Here’s how the tax shelter would go for Mr. Schultz in our
Sonics example. Undoubtedly, at his high level of earnings,
Mr. Schultz would pay the highest 33 percent marginal tax
rate. Summing the net losses of the Sonics at the bottom of
Table 4.8 yields about $217.9 million (again, adjusted to 2009
dollars). Passing this amount through to the 1040 Form
would generate a tax shelter on other earned income (as CEO
of Starbucks, for example) equal to 0.33 × 217.9 = $71.9
million. This tax savings spends just the same as income
earned.
ROSTER DEPRECIATION AND PERSONAL INCOME TAX
SAVINGS
Special tax status for sports team owners, in fact, do the tax
shelter one better. What if there were an acceptable
accounting practice that reduced the team’s income for tax
purposes but not in actuality? Then, reported net operating
losses wouldn’t really be lost, but taxes on personal 1040
forms could be reduced by the allowed loss anyway! There is
just such a special status, and we will explore how it works.
Amortization is the divvying up of the cost of a longer-term
project over a number of years. It covers intangibles rather
than the tangible depreciation of machinery. For example, a
firm might want to invest in a new piece of capital but not
have the money it needs. In this case, the firm saves up the
money over a number of periods, faithfully charging itself for
its future purchase. The opportunity cost of the money set
aside for these purposes is tax deductible.
Depreciation refers to the loss of productive capacity that
happens to capital assets, such as machines, buildings, and
equipment, as they age. In a sense, over a known period of
time, capital assets will wear out completely or be completely
depreciated. This loss in productive capacity is tax deductible.
Teams can own physical capital that depreciates, such as cars,
computers, and buildings. In that way, they are just like other
firms.
The acceptable tax practice that allows owners to reduce team
income for tax purposes, but not in actuality, concerns
another kind of depreciation. Bill Veeck, the famous baseball
owner and entrepreneur, successfully convinced the IRS that
the player roster is a piece of capital that wears out over time.
As such, he argued, this so-called roster depreciation should
be tax deductible. At the time of our Sonics example, when an
owner bought a team, the IRS allowed half of the purchase
145

price to be a physical capital asset, depreciable for up to five
years after the initial purchase (the rules have become much
more liberal lately, as covered in Chapter 12). Thus, in 2000,
the current owner bought the Sonics for $248 million
adjusted to 2009 dollars for comparisons to the values in
Table 4.8 ($200 million nominal). This means that they could
deduct half the purchase price, or $124 million, in equal
amounts for five years (straight-line depreciation is required
for intangible asssets), as lost productivity of the players they
bought. That’s about $24.8 million annually deducted from
net operating revenues. But it isn’t at all clear that these really
are losses to the team owner and, even if they are, subtracting
them is just bad accounting.
Here is why roster depreciation doesn’t really reduce income
but reduces taxes paid by owners: Admittedly, players “wear
out” eventually toward the end of their careers but that loss
accrues to players, not owners. Owners simply contract with
players for whatever value it is they can offer. If the owner
later prefers a “newer” bundle of skills, he or she simply writes
a different contract with a different player. So, the players
wear out, but it is they who bear the costs in terms of lower
return for their services. Owners do not bear the cost of this
human skill depreciation, so it is simply bad practice to allow
them any sort of “roster depreciation” allowance.
Roster depreciation also provides a second puzzle. Let’s play
along with this facetious logic and assume, for the moment,
that players do depreciate. Then they must either be
depreciated or count as a current business expense but not
both. But the IRS allows precisely this sort of double counting
to occur. Player compensation is carried in the expense
column, and the roster depreciation allowance is also
subtracted from net operating revenues.
Thus, allowing roster depreciation reduces owner income for
tax purposes but not in actuality. There really isn’t any loss
that the owner bears. If taxable income actually is generated
on the team, so that the pass-through to the owner’s 1040
form was positive, then roster depreciation at least produces a
tax reduction. If roster depreciation is actually large enough to
cause a loss (for tax purposes) on the team, then the loss
amount goes to the owner’s 1040 forms, producing an even
higher tax savings on other nonownership income. Pretty
slick.
Despite the fact that owners don’t actually bear the cost of
players’ skill depreciation and, even if they did, depreciating
and expensing represent double counting, the IRS continues
to allow this practice today. Truth really is stranger than
fiction, especially the truth of sports accounting. By looking
just a bit further into the Sonics data in Table 4.8, let’s see
how all this worked out for Mr. Schultz.
SPORTS ACCOUNTING PRACTICES AND THE SONICS’
ACCOUNTS
The lessons we just learned are that 1) losses are not as large
as owners report under allowable IRS rules for sports teams
and 2) there are substantial tax savings from these losses
146

anyway. The data in Table 4.8 cover all but the last year of Mr.
Schultz’ ownership of the Sonics and all of the years where he
could take the total $124 million roster depreciation allowed
by the IRS. This depreciation is part of the “Depreciation and
Amortization” category under “Costs and Expenses.” From
our first lesson, the total “Loss from Operations” over the five
seasons of $202.5 million is overstated by $124 million.
Adding the total roster depreciation back in yields a more
reasonable statement of loss calculated as follows: 202.5 –
124 = $78.5 million for five years, or about $15.7 million
annually (again, all of the figures here are adjusted for
inflation). Actual economic losses on the team were not nearly
as claimed.
But that is not the end of the story as we apply our second
lesson. After additional allowable adjustments (on asset sales,
interest rate swaps, and interest expenses), and including the
roster depreciation in losses from operations, total losses
claimed by the owner over the period in the table would be
$217.9 million (again, the last row in Table 4.8). As shown
earlier, passing these losses through to the 1040 forms filed
over the period, at a 33 percent marginal tax rate, is worth
$71.9 million in tax savings. If we count this tax shelter
against the more reasonable statement of losses, closer to
$78.5 million for five years, it’s all about a wash. The
remaining loss is $6.6 million over five years, or $1.3 million
annually.
From the perspective of the overall value of team ownership,
the following conclusion seems reasonable for Mr. Schultz’
Seattle Supersonics ownership episode. Just the bottom-line
value and tax shelter from owning the team cost him about
$1.3 million annually for five years. But owning the team also
generated all of the other possible values stated earlier in this
section—related business opportunity, costs that may not be
costs, and cross-ownership values. The simple question is
whether these other values of ownership were larger than $1.3
million annually. Given that Mr. Schultz actually earned about
6.9 percent annually, adjusted for inflation, after selling the
team six years later, our only conclusion is that the other
values of ownership more than covered the $1.3 million in
actual annual losses.
WHY DO OWNERS PLEAD POOR?
Most of this discussion concerns the difference between
allowed accounting practices and the economic value of team
ownership. The essential question about team ownership,
from the economic perspective, is whether all of the values of
ownership are enough to keep owners and their resources
engaged in team sports.
We can only assume that owners know that their actual
position is much better than their team accounts indicate
(otherwise, there are some very rich accountants out there!).
Economists have been pointing out this fact for decades.
Team values have risen quite handsomely, and this doesn’t
happen typically to assets that lose money over time. Just as
typically, very wealthy people do not line up to gain access to
147

an asset at a price of hundreds of millions if they anticipate
large losses.
Why do owners persist with their claims of losses? First, some
teams really may be suffering losses, especially small market
teams that no longer have a roster depreciation allowance (a
recent example is the bankruptcy of the Phoenix Coyotes in
the NHL). But there is another explanation. Owners may be
posturing, pleading poor, for strategic reasons. Showing the
public their IRS-allowable losses may be helpful to owners as
they deal with players at the bargaining table. If players can’t
figure out revenues or team profits, then they may be at a
bargaining disadvantage. Further, while players can hire
analysts to figure out the actual financial position of owners,
the general public cannot. Thus, claiming losses may also help
bolster owners’ positions as they bargain for new stadiums
with their current host cities; remember these data on the
Sonics came from just such a setting. If their claims of losing
tons of money are convincing, then chances for a stadium
subsidy or favorable lease contract may increase. We’ll return
to these issues in Chapters 10 and 11.
148

LEARNING HIGHLIGHT: WAYNE HUIZENGA’S FISH
STORY
The introductory story to this chapter lists the woes of former
Florida Marlins owner, Wayne Huizenga. Noted sports
economist Andrew Zimbalist (1998) couldn’t resist shooting
fish in a barrel. Recall that Huizenga claimed to have lost $30
million ($39.8 million) winning the World Series in 1997. But
looking at the data with an economist’s eye rather than an
accountant’s, Zimbalist notes that most of the losses actually
are nothing but an artifact of Huizenga’s ownership of both
the stadium that the Marlins play in and the team’s local
broadcast outlet.
On the stadium, Huizenga does not report luxury box or club
seat revenue as team revenue, presumably putting these
revenues under stadium operations instead. Zimbalist
estimates this value at around $16.5 million ($21.9 million).
Other revenues generated by the Marlins (through naming
rights, parking, signage, and major league merchandise) but
put under the stadium account Zimbalist estimates at around
another $13.9 million ($18.5 million). Zimbalist also notes
that concession revenue appears to be understated by about
$7.6 million ($10.1 million). In addition, Huizenga charges $5
million ($6.6 million) to cover stadium expenses. However,
because the stadium is shared with the NFL Dolphins, it
seems reasonable that no more than half of this amount
rightfully should be paid by the Marlins. Thus, on the
stadium, it is reasonable that about $38 million ($50.5
million) actually was generated by the Marlins and that costs
should be $2.5 million less ($3.3 million). On the stadium
alone, the $30 million ($39.8 million) in the red turns into
$15.5 million ($20.6 million) in the black.
But that’s before Huizenga’s cross-ownership of the Marlins’
media outlet is considered. Zimbalist reports that the media
outlet paid $2.1 million ($2.8 million) under market estimates
for the Marlin’s local broadcast rights. Estimates cited by
Zimbalist show that the value of the media provider rose from
$85 million ($112.9 million) to $125 million ($166 million).
The underpayment was worth $40 million ($53 million) to
the media provider, but Zimbalist sticks with the $2.1 million
($2.8 million). With a few other small discrepancies,
Zimbalist suggests that instead of a $30 million ($39.8
million) loss, an operating profit of $13.8 million ($18.3
million) was more likely. Zimbalist argues that the Marlins
were more profitable as losers in 1998 than they were as
World Series champions in 1997, even though they were quite
profitable that year as well. According to Zimbalist, Huizenga
hid all of this through cross-ownership of the stadium and the
media provider.
Wayne Huizenga claims to have lost $30 million ($39,8
million) on his Florida Marlins, but a certain economist
doesn’t agree. [Photo of Huizenga, Marlins backdrop.]
Source: Andrew Zimbalist, “Just Another Fish Story,” New
York Times, October 18, 1998. Online version (www.nyt.com).
149

http://www.nyt.com

http://www.nyt.com

SECTION 9
Chapter Recap
Most wealthy team owners have become wealthy by watching
the bottom line in all of their business endeavors. In this
chapter, we have assumed that profit maximization is the
guiding principle for team owners’ decisions concerning their
sports teams.
In the long run, all inputs in the sports production process are
variable. The most important variable in the long run is the
player roster. Regardless of the state of all other variables,
once the player roster is chosen, so is team quality.
Winning percent is the usual measure of quality. Once team
quality is fixed, the team is in the short run, economically
speaking. As a result, the short run is characterized by a fixed
amount of on- and off-field personnel. Once this is
determined, teams consider short-run outputs such as
attendance, concessions, parking, logo properties, and the
stadium experience for fans.
In the short run, fixed costs are pretty much identical to roster
costs, plus the costs of contractual front-office talent and any
contractual obligations associated with the stadium. Fixed
costs do not change with the level of output in the short run.
Variable inputs include team and stadium operations, player
development, and advertising and promotion. The costs
associated with variable inputs determine variable costs.
Variable costs increase at a decreasing rate over some initial
range of output but must increase at an increasing rate due to
diminishing returns to variable inputs.
For every quality choice, there will be a particular short-run
cost structure as well as a particular revenue structure
because quality is a primary determinant of fan demand that,
in turn, determines revenues. Profits are the difference
between revenues and costs. For each short-run situation,
there will be a maximum amount of profit that the owner
could choose. Long-run profit maximization requires the
owner to compare all of these possibilities and select the level
of quality with the highest profits. As long as costs of
increased quality eventually rise faster than revenues, there
will be one, and only one, highest profit quality level.
Sometimes great teams will falter, while weak teams will
surprise. However, typically and on average over the long
haul, teams with greater talent win the most. Owners wishing
to maximize profit will only choose higher quality if it pays off.
The highest level of talent is not necessarily concentrated
where profits are highest. This only happens if large-revenue
markets are also large-profit markets. However, as long as
there is a wide range of payoffs among teams, competitive
imbalance will be a problem. In addition, there is also tension
between some owners whose profit-maximizing level of
winning is low and their fans, players, and on-field managers.
150

How often have we heard the lament of long-suffering fans of
lovable losers? But as long as profits dictate talent choices and
some locations support only low levels of quality, there will
always be unhappy fans, players, and coaches.
In the real world of sports team ownership, the revenue side
of annual operations statements may not show all of the
team’s value. There can be related business opportunities for
sports owners; some expenses may not be costs at all; and
when there is cross-ownership of media providers and/or
stadiums, sports team owners can put revenues generated by
teams on different annual operations statements. Finally, the
ability to both expense and depreciate the team roster, a
perfectly legal sports accounting practice, can make a team
that is worth millions of dollars to its owner both in terms of
net operating revenue and personal tax reductions look like it
is losing money. Owners might find it advantageous to show a
loss when they confront players’ unions and when asking a
city for upward of a few hundred million dollars in stadium
subsidies.
151

SECTION 10
Key Terms and Concepts
You should have run into each of these in pop-ups in the text
of this chapter:
• Profit maximization hypothesis
• Short run
• Long run
• Short-run fixed inputs
• Short-run fixed costs
• Short-run variable inputs
• Short-run variable costs
• Short-run total costs
• Diminishing returns to variable inputs
• Long-run quality choice
• Long-run winning percent production function
• Limit to management ability in the long run
• Managerial efficiency
• Long-run total cost
• Tension between winning and costs
• Long-run profits
• Short-run profits
• Talent dumping
• Tension between winning and profits
• Profit variation can harm balance
• Revenue shifting and cross-ownership
• Pass-through firm
• Roster depreciation
152

SECTION 11
Review Questions
1. What is the profit maximization hypothesis? If teams earn
very low profits or lose money, does this mean the owner
does not maximize profit? Can owners who make
systematic mistakes survive in a league of profit-
maximizing owners? How?
2. Why is winning a useful way of describing sports team
output? What are the limitations of this abstraction?
3. Define the short run in economic terms. What do owners
sell in the short run? What is fixed in the short run? How
long does the short run last?
4. Define the long run in economic terms. What is long-run
output? What are the primary long-run decision variables
for sports team owners? If an owner alters the roster
during the season, when is the team back in the short run
again?
5. What are the short-run fixed inputs for a sports team?
The short-run variable inputs? What distinguishes the
team’s short-run fixed costs from the team’s short-run
variable costs?
6. Define short-run total costs. What type of expenditure
represents the largest element in short-run total costs?
Graphically, what is the vertical distance between short-
run total costs and short-run total variable costs equal to?
7. Why are the functions for short-run total costs and short-
run variable costs shaped as in Figure 4.1? What would
change to shift short-run total fixed costs up? What would
change to shift short-run variable costs up?
8. What is the main determinant of where teams end up in
the standings; that is, what determines whether they win
more or not? For our purposes, what is the definition of a
star player? Explain how star players contribute to a
roster that has no stars.
9. Describe the long-run winning percent production
function. What role does managerial efficiency play?
10. Define long-run total cost. What determines its shape in
Figure 4.3?
11. Define long-run profit. Define short-run profit. Why are
profits negative, then positive, and then negative again as
attendance increases?
12. Why is there conflict from the owner’s perspective
between winning and costs? What is talent dumping?
Why does it occur?
13. How do profits constrain winning? How can profit
variation harm competitive balance?
153

14. List the five elements that, from the economic
perspective, contribute to the value of owning a team but
do not appear on the team’s annual net operating revenue
report. You should be able to give an example for three of
the elements.
15. What is a pass-through firm organizational structure?
Why is such a structure valuable to sports team owners?
What is roster depreciation? Why is it valuable to a team
owner?
154

SECTION 12
Thought Problems
1. In the section on short-run costs, general and
administrative costs for the Seattle Mariners were
eventually characterized as fixed costs. How would you
determine where they truly belong?
2. In Figure 4.1, at an attendance of 2.9 million, what is the
dollar value of the vertical distance between the short-run
total cost function and the short-run variable cost
function? What does this dollar value represent? What is
the vertical distance equal to if attendance is 3.5 million?
Why?
3. In the long run, everything is variable. How can it be,
then, that there is some “limit to management ability”
that gives the long-run winning percent production
function its shape? Doesn’t this limit mean that
management ability is fixed? Explain.
4. Use a graph to explain what happens to short-run costs as
a team owner increases quality. What type of expenditure
is made by owners to make this shift occur?
5. Use a graph to explain what happens to revenues in the
short run when a team increases quality.
6. A team owner tries to decide quality level. By spending
another $12 million, the owner expects that the team will
finish one place higher. Does this mean that the owner
can raise the team’s place in the standings by one place
for each $12 million spent? Why or why not?
7. Graph a short-run situation where a big market team has
lower profits at its profit-maximizing level of attendance
than a small market team has at its profit-maximizing
level of attendance. In your graph, be sure to identify the
profit-maximizing attendance and profit in each case.
8. Using a carefully labeled graph of long-run profit
opportunities, answer the following:
a. What determines the shape of the profit function that
you drew in your graph?
b. Is this a long- or short-run graph? Explain.
c. With reference to different points on your graph in
part (a), explain why no profit-maximizing team will
ever want to win all of its games.
d. Draw the profit function for a team owner who would
rationally choose to win fewer than half the team’s
games in the long run.
155

9. Graph a small market team’s long-run profit function
when it is possible to earn profits in the long run. Suppose
that economic hard times befall the fans; incomes fall
dramatically. The result is that the owner can’t figure out
any way to break even. In your graph, show this new
profit outcome. What would you do if you were this
owner?
10. When a player says that an owner doesn’t want to win,
what is really being said?
11. Fans always crave a better team. Often they grow weary of
ownership that won’t give them one. If the team changes
owners, when will fan expectations of improvement be
met? Relate your answer to the Disney example in the
Learning Highlight: When Can a New Owner Change a
Team’s Fortunes?
12. What is the crucial characteristic of cross-ownership that
determines whether revenue shifting can occur? What is
the gain to cross-ownership to be had from revenue
shifting?
13. Suppose a team is organized as a pass-through firm.
Further, suppose that there are no physical depreciation
or roster depreciation or other long-term obligations. The
tax rates on both the team and the owner’s personal
income are 33 percent. Explain why the owner would be
indifferent between making $7.98 million on the team
before taxes and losing $16.2 million on the team after
taxes.
14. Give the arguments in support of roster depreciation.
Give the arguments against it. How do you come down on
the issue? Why?
15. How would you determine whether a given pro sports
team owner tries to maximize profits? Why would the
team’s annual net operating revenue statement not be
very helpful in your analysis?
156

SECTION 13
Advanced Problems
1. Examine the data in Table 4.4. Why did it take an
increasing amount of points at the margin to move up
through the standings (once out of the basement)? Now,
examine Table 4.5. The same pattern appears to hold over
a longer time period. Why is that?
2. Even though we gloss over the underlying production
function for winning in this chapter, it’s still fun to think
about. True or false, and explain your answer thoroughly:
The best football coach is the one who always chooses the
play that will gain the most yardage.
3. Why use the data across a few seasons in Table 4.5 rather
than the single-season data in Table 4.4 to ascertain
movements through the standings? What is gained by
examining the data over time?
4. Suppose you hear the following: “Arenas are
multipurpose buildings. If NBA and NHL owners could
just afford to own their own stadiums rather than rent
them, two things would happen. First, they’d be able to
get out from under high arena rental costs, reducing their
costs per game. Second, they’d gain control over the
revenues earned on other nonsports events. This would
improve their net revenue picture, and they’d be able to
buy better basketball or hockey talent.” There are three
things wrong with this statement. Identify and correct
each one of them.
5. “Our costs rise each year as they would in any household
or business. Like any NFL team we have operating costs
and player costs. We also are in a unique situation
because, unlike a lot of teams, we have a privately funded
stadium and there are costs associated with that”—Phil
Youtsey, Director of Ticket Sales, Carolina Panthers, on
increasing ticket prices by 2.5 percent for the 2002 season
(sportsbusinessnews.com, accessed March 25, 2001). Are
there really cost differences between owning a stadium
and renting a stadium? What are they?
6. Use this Table of Stars, Winning, and Salary to
answer the questions.
a. Graph the total cost of winning percent (from player
compensation only). Be sure to label the points in your
graph.
b. Is this the long-run or short-run total cost structure for
this team? Why?
c. What costs are missing from the complete picture of
the costs of winning?
157

figure:%23

figure:%23

d. How would inclusion of the costs you list in part (c)
alter your graph in part (a)? Can you envision how you
would draw a graph of this true picture of total costs?
7. A team owner learns from a research firm that the
demand for winning looks like this:

where P is ticket price, A is attendance, and W is the
quality of the team measured by winning percent. To
focus on the long run, the team owner sets marginal cost
equal to zero, but the owner knows that fixed costs
determine quality in the following way:

where W is winning percent. With this information,
answer the following:
a. What are the owner’s profits if a low-quality team is
chosen, say W = 0.350.
b. What are the owner’s profits if a high-quality team is
chosen, say W = 0.650?
c. Comparing your answers, aforementioned, should the
owner choose the high-quality or low-quality
alternative?
d. Graph the relationship between quality (remember,
0.000 < W < 1.000) and profits. What is the precise winning percent that earns the owner the greatest possible profit? 8. What dissatisfied you about the analysis in the Learning Highlight: When Can a New Owner Change a Team’s Fortunes? What additional data or analysis would make the example more satisfying? 9. Here’s a simplified team accounting problem (round your answer to whole numbers). Suppose an owner buys a football team for $200 million. The team is organized as a pass-through firm. The corporate and individual tax rates are equal at 33 percent. Physical depreciation is $4 million. The team has no other long-term obligation. In the first year of ownership, operating revenues minus expenses equal $16 million. Show that the team loses $8 million after taxes and all depreciation. Show that the actual value of the team to the owner, just from these calculations, is a positive $15 million. Finally, show that the value of roster depreciation to the owner was $7 million. (Hint: There are two positive components to this last part.) 10. What was Wayne Huizenga’s goal in rearranging revenues, as Zimbalist (1998) claims he did in the Learning Highlight: Wayne Huizenga’s Fish Story? How did all of the shenanigans cited by Zimbalist contribute to that goal? In particular, how did the revenue shifting due to Huizenga’s cross-ownership of the Marlins’ local cable 158 outlet support his goal? Did revenue shifting have any other advantages? Under what circumstances? 159 SECTION 14 References Berri, David J., Martin B. Schmidt, andStacey L. Brook. The Wages of Wins: Taking Measure of the Many Myths in Modern Sport. Palo Alto, CA: Stanford University Press, 2006. Kahn, Lawrence M. “Managerial Quality, Team Success, and Individual Player Performance in Major League Baseball,” Industrial and Labor Relations Review 46 (1993): 531–547. Klein, Gene, and David Fisher. First Down and a Billion: The Funny Business of Pro Football. New York: St. Martins Press, 1987. Noll, Roger. Government and the Sports Business. Washington, D.C.: Brookings Institution, 1974. Okner, Benjamin A. “Taxation and Sports Enterprises,” in Government and the Sports Business. Roger G. Noll, ed. Washington, D.C.: Brookings Institution, 1974. Porter, Philip K., and Gerald W. Scully. “Measuring Managerial Efficiency: The Case of Baseball,” Southern Economic Journal 48 (1982): 642–650. Quirk, James, and Rodney D. Fort. Pay Dirt: The Business of Professional Team Sports. Princeton, NJ: Princeton University Press, 1992. Reingold, Jennifer. “When Less Is More,” Financial World, May 25, 1993, 38. Scully, Gerald W. “Pay and Performance in Major League Baseball,” American Economic Review 64 (1974): 915–930. Scully, Gerald W. The Business of Major League Baseball. Chicago, IL: University of Chicago Press, 1989. Scully, Gerald W. “Managerial Efficiency and Survivability in Professional Team Sports,” Managerial and Decision Economics 15 (1994): 403–411. Singell, Larry D., Jr. “Managers, Specific Human Capital, and Firm Productivity in Major League Baseball,” Atlantic Economic Journal 21 (1993): 47–59. Zimbalist, Andrew. “Just Another Fish Story,” New York Times, Sunday, October 18, 1998. Online version (www.nyt.com, accessed October 18, 1998). 160 http://www.nyt.com http://www.nyt.com SECTION 15 Suggestions for Further Reading Surely, I can come up with some. 161 CHAPTER 5 Sports Market Outcomes, Part I: Leagues, Team Location, Expansion, and Negotiations Free market economics is the process of driving enterprises out of business. Sports league economics is the process of keeping enterprises in business on an equal basis. There is nothing like a sports league. Nothing. —Paul Tagliabue, Previous NFL Commissioner Sports Illustrated, September 10, 1990. CHAPTER OBJECTIVES After reading this chapter, you should be able to: • Understand that leagues facilitate play on the field and maintain the business structure of sports. • Explain how the business structure of sports leagues establishes and maintains territorial exclusivity for member teams. • Understand the financial and strategic components of expansion and team relocation choices by the league. • Understand the role that rival leagues have played in the formation of today’s modern leagues. • Discuss how owners act in joint ventures through their leagues, including TV contract negotiations, labor negotiations, and dealings with their host cities. SECTION 1 Introduction Commissioner Tagliabue could not have been more correct about there being nothing like a sports league. For example, when MLB’s Brooklyn Dodgers and New York Giants left for California in 1957, they left a big hole in the fans’ hearts. In 1959, after failing to convince either an existing MLB team to move or MLB to expand, William Shea decided to form the new Continental League with Branch Rickey, perhaps the most famous baseball businessman in history, as president. However, Shea and Rickey couldn’t get the existing American League and National League teams to agree to Continental League team locations under MLB rules for a new league to join the existing American League and National League, and the Continental League folded in 1960. Not two months later, the National League announced two expansion teams for Houston and, you guessed it, New York. The American League agreed to expand to Los Angeles and to the just-vacated Washington, D.C., market. All of the MLB owners had joined forces and determined that there was no room for Continental League teams in cities that they then immediately occupied after the Continental League folded. In any other industry, this type of behavior by rival firms would at least have drawn antitrust scrutiny. Leagues are also “different” in their ability to coordinate the economic activity of separate franchise owners to enhance value for all. In testimony before Congress in 1981 (NFL Report, 1999, p. 1), NFL founder and Chicago Bears icon George Halas described the birth of the NFL. Each of the original 12 franchises was priced at $100, but none of the owners actually made the original payment! Obligations of $1,200 were all that President Jim Thorpe had to go on at the inception of the league. Halas’ is a quaint story of the humble beginnings of the NFL. But the economic result is anything but humble. Revenue estimates for the league now hover around $7 billion and Stephen Ross’s purchase of up to 95 percent of the Miami Dolphins (who also own their own stadium), culminated in 2009, has been calculated to cost close to $1 billion by Forbes magazine. In this chapter, we investigate the behavior of team owners as members of leagues. We will see that owners seek to enhance their individual welfare through coordinated league action and that leagues set the stage for play on the field as well as the business structure of member teams. We will also look at how leagues establish and protect the team territories of owners and negotiate for member owners in a more profitable way than owners would be able to individually. 163 SECTION 2 Why Leagues? Making Play and Profits Leagues enable owners to pursue economic goals and objectives that they cannot pursue as successfully acting alone, such as setting a season schedule, organizing championships, and implementing rule changes. Coordinated league activity makes league play happen, but it also provides owners with many profit opportunities off the field. We cover the general idea behind such cooperation by team owners in this section, but the actual elements of that cooperation receive separate attention in subsequent sections of this chapter. SINGLE-ENTITY COOPERATION: MAKING LEAGUE PLAY HAPPEN Some cooperative actions among teams must happen for league play to occur at all. This type of activity is called single- entity cooperation. Single-entity cooperation defines the actions that owners must take to make league play happen in the first place—setting schedules, the rules of play, and the structure of championships. However, even at this seemingly innocuous level of creating league play, economic issues arise. SETTING THE SCHEDULE Setting the schedule is the most basic form of single-entity cooperation. If team owners cannot cooperate with each other to set a schedule, then league play, by definition, cannot occur. Even this type of single-entity cooperation has economic impacts. Again in his congressional testimony (NFL Report, 1999, p. 2), George Halas emphasized that balanced scheduling, home and away, was essential to the success of the league. Originally, all owners wanted to play the biggest drawing teams (home and away), but Halas convinced them that the season would be more interesting to fans, and stabilize more teams financially, with such a balanced scheduling approach. Halas’ wisdom here is that leagues need to act in their single- entity capacity to include all teams in the schedule. If only larger teams played each other, smaller-market teams would fail economically. In trying to generate widespread fan interest and league growth, Halas understood that stronger teams must play weaker teams as well as other strong teams in order to cultivate broad fan interest. This broad fan interest is especially important for championship play at the end of the season. Another important element in setting the schedule is establishing season length. There is nothing magical about the length of seasons except that it helps determine profits from the regular season and the play-offs. Season lengths certainly have changed over time. The season length in the American 164 League increased from 154 games to 162 games in 1961, a 5 percent increase (the National League followed suit the next year). The most recent change in the NFL season (there have been many over the years) was from 14 to 16 games after 1976, a 14 percent increase. SETTING THE RULES If teams play under different rules, they aren’t playing the same game. In addition, there must be officials and appeals for the sake of fairness. Once again, however, even this basic single-entity determination has economic elements. Rule changes alter the balance between offense and defense in producing winning margins. In turn, this changes the pattern of winning between teams, which is what fans pay to see. Sports economics pioneer Gerald Scully (1989) details the economic implications of changes in playing rules. For example, narrowing the strike zone would be expected to favor hitting relative to pitching. In Table 5.1, the “predicted” column shows what Scully predicted would happen to batting average and earned run average due to narrowing the strike zone in 1969. Batting averages should rise because a good pitch is easier to detect in a smaller zone. Pitchers should fare worse, with a higher earned run average. As Table 5.1 shows, both of these results did occur in 1969. The designated hitter rule was another important change in the American League, relative to the National League. In the American League, teams are allowed to designate a hitter to take the pitcher’s place in the lineup. This means that there is one more skilled batter in American League lineups than National League lineups. Scully (1989) points out that measures of hitting, especially slugging average, increased significantly in the American League over the National League after the designated hitter rule was implemented. What do all of these rule changes have to do with economics? In narrowing the strike zone or adopting the designated hitter, the offense was favored because that’s what owners thought fans would want and pay to see. If fans get bored with the length of games, shorten them. If fans want more action, institute more offense if it is affordable relative to the value created. This is surely what the NHL decided when it changed the final standings point system so that both teams get 1 point when a game goes to overtime, and the winner gets an additional point, starting with the 1999–2000 season. While more games should go into overtime with this change (you still get a point for tying rather than losing!), scoring during overtime should rise since there is the additional overtime point. Banerjee, Swinnen, and Weersink (2007) document that this is precisely what occurred under the altered point assignment. COOPERATION AND CHAMPIONSHIPS The ultimate indicator of fan demand for winning is the crowning of a league champion. Therefore, determining a champion is very economically valuable. However, the way championships are structured also produces economic incentives for team owners. Relative to a league that just 165 crowns the team with the highest winning percent as its champion, adding play-offs does two things: (1) It extends the season and fan interest for a few teams, and (2) it reduces the returns to buying talent. James Quirk and I (Fort and Quirk, 1995) found that play-offs reduce the chances that the team with the highest winning percent will become the eventual league champion, even though those chances remain higher than 50/50. This means that the expected value of talent falls, too, causing owners to choose lower talent levels. One implication of this is that owners have a greatly reduced incentive to hoard talent to ensure the highest winning percent relative to the rest of the teams in the league. Finally, the number of play-off games also is a choice variable for leagues. For example, prior to 1968, there were no play- offs in MLB. Indeed, there were no divisions in either the American League or National League, so the winners of each league just met in the World Series. The American League and National League championship series started after divisions were created in 1968. With the 1992 expansion, MLB added another round of play-offs after three divisions were created in each league. Once again, these rounds generate additional revenues for a few owners and impact the talent choices of all owners in the league. Similar increases in the number of championship rounds have been adopted by the other major professional leagues over time as well. JOINT VENTURE COOPERATION: THE ECONOMICS OF LEAGUE BEHAVIOR Walter Neale (1964) was the first to recognize the important relationship between single-entity action and economic outcomes, naming it the “peculiar economics” of team sports. But this peculiar economics only begins with the single-entity actions of owners in leagues. Once owners act together in pro leagues to set the stage for competition on the field, they may also act together to raise profits for member owners. All cooperative actions that do not make play happen are called joint ventures. In this section, we will set up the idea of joint ventures, the main categories of which are covered later in the chapter. In a joint venture, all owners in the league surrender part of their autonomy to allow the league to act on their behalf. However, the objectives of joint ventures can also be pursued individually by owners rather than through the league. Given that, economic intuition suggests that joint ventures are cooperative acts aimed at increasing profits relative to acting individually. Single-entity action helps create efficient schedules, rules, and championships. However, since the time of Adam Smith, we’ve known that joint venture cooperation can facilitate market power and its inefficiency (Wealth of Nations, Book I, Chapter X), “People of the same trade seldom meet together ... but the conversation ends in ... some contrivance to raise prices ... [Government] ought to do nothing to facilitate such assemblies, much less 166 render them necessary.” Market power doesn’t have to be the result of joint venture actions; it just ends up that way. Owners really do control their leagues. If leagues do not make owners better off than if they were acting alone, the owners will simply leave the league. A variation on this theme is secession from the current league and creation of another. This doesn’t happen often because leagues are good at doing what owners require, but it does happen. In the 1940s, the National Basketball League (NBL) faced a rival, the Basketball Association of America (BAA) (Quirk and Fort, 1992). The BAA, composed primarily of arena owners, had a lock on the largest and most valuable venues. Defection to the BAA by NBL teams because their own league could not secure these profitable venues led to the demise of the NBL. After the 1947–1948 season, the four strongest NBL franchises went over to the BAA. After the next season, six more NBL teams defected. The NBL died because it hadn’t satisfied one of the most basic requirements for its member owners, namely, a chance to play in the most lucrative venues. The Learning Highlight: Challenging European Football’s Power Structure at the end of the section presents an international case where a sports organization is failing to give its members what they want most and the likely repercussions. In the following sections, joint venture activity is divided into three main categories: territory definition and protection; expansion and team relocation; and league-level negotiations with TV, players, and host cities. Joint action is not required for any of these activities. Acting together on these issues must simply be more valuable to owners than acting individually would be. 167 LEARNING HIGHLIGHT: CHALLENGING EUROPEAN FOOTBALL’S POWER STRUCTURE The Union of European Football Associations (UEFA) organizes the European football (soccer) championship at the level just below World Cup competition. Essentially, the champions from each European country meet to decide the champion of Europe and the tournament, fitting enough, results in the awarding of the “Champions Cup.” The prize to the participants is the chance at becoming the European champion, and a rather substantial payday from the TV revenues for the tournament. An important feature of these payments is that they are larger for the finalists than they are for teams that do not make it past the early rounds. Herein lies a problem for UEFA. Smaller countries, such as Scotland, Belgium, and Luxembourg, usually make early exits from UEFA cup competition. However, the teams in these smaller countries are media dynamite in their own markets. This has recently led to a movement by national European sports organizations, at the behest of their most winning teams, for a “small country” European championship. Of course, such defection would reduce the value of the UEFA tournament. UEFA responded that any teams that participate in such a championship will be banned from UEFA cup play and, quite possibly, World Cup consideration. One would think that this ultimatum would end the quest for a small-country championship. However, the smaller countries responded that they would push forward anyway. One of two things must be true: 1. This group of small-country malcontents may just be pushing UEFA’s hand to see if the threat will actually be carried out. 2. It may actually be worth enough to the small-country teams, in terms of added TV revenue, to break off and start their own championship round. Either way, UEFA is not meeting the needs of these small- country teams; therefore, these teams are seeking a structure that will. Teams from small countries are pressing for their own UEFA cup to capture a larger return fro postseason play. [Soccer picture.] Source: Sports Business Journal, December 25, 2000, p. 33. 168 SECTION 3 Territory Protection and Definition The first joint venture activity with implications for other teams, other leagues, and fans is the definition and protection of individual owner territory. Through franchise agreements, leagues have been allowed to grant exclusive territories to their members. This means that the league is allowed to define a territory and ensure that no other members locate within that territory. The result is that exclusive territories are the primary determinant of the revenue structure that any owner will enjoy. Protection of that revenue structure is a fundamental reason to join a league in the first place. Let’s examine how leagues perform this essential task and its impacts. EXCLUSIVE TERRITORIES AND MARKET POWER How many times have you heard it said that in the hearts of fans there is no substitute for the home team? Packers fans are not Vikings fans, and Yankees fans couldn’t care less about the rest of MLB except insofar as they are the competition. Leagues rely on this fan identification when setting exclusive territories for member teams. FRANCHISE AGREEMENTS Leagues manage exclusive territories through franchise agreements with their member owners. A franchise agreement is a contract between the league and an owner that clearly specifies what it means to own a pro sports team. The owner names the franchise, enters the season schedule and play-off schedule, agrees to abide by the league’s operating rules, and eventually shares in leaguewide broadcasting revenues and expansion fees. The franchise agreement also specifies the responsibilities of the league to enforce operating rules and to protect the franchise territory from other league members. The law of the land, enforced by the courts, endows leagues (even though each league really is just a collection of owners who benefit from joint ventures) with the rights to their franchises, logos, and properties. Leagues, in turn, license logo rights and other properties to various sellers, and they sell franchises to owners. This ability to define territories has survived over time subject to judicial review under the antitrust laws (see Chapters 8 and 12 for more on antitrust). The rules in every pro sports league strictly govern the ability of any member of the league to encroach on existing territories. Teams cannot simply move to a new territory without careful review and approval by their league according to its bylaws. ECONOMIC IMPACTS OF EXCLUSIVE TERRITORIES Nothing about franchise arrangements necessarily restricts competition. For example, when McDonald’s grants a franchise, there is plenty of other competition from other fast- 169 service providers (e.g., Burger King). Indeed, the competitively determined customer base acceptable to potential franchise buyers often is such that quite a few McDonald’s restaurants are located in the same city. This situation is not often found in pro sports leagues. First, leagues themselves only put more than one team in a very few of the largest markets in the country. And we’ve already discussed how multiple teams in the same area are not necessarily in the same market! Second, once a league is in place, competition from competing leagues is either squelched or reduced in such a way that no meaningful economic competition to the dominant league’s teams ever will be forthcoming. Finally, and crucially, remember that all of the individual McDonald’s owners do not comprise the decision- making body for that corporation as owners do in their respective leagues. Economically speaking, careful management of exclusive geographic territories by leagues provides owners with market power. Once exclusive territories are created and protected, there literally is no substitute team, in that sport, in that area for fans. There are other entertainment substitutes, but not another team in the same sport. From an economic perspective, this managed absence of substitutes allows individual teams to maximize profits as monopolies in their sport in their territory (although they still compete with entertainment providers in general). You know from your general economics training that firms with market power reduce output and raise prices relative to a more economically competitive outcome. You also know that this type of firm earns positive economic profits. QUANTITY RESTRICTIONS Leagues impose two types of quantity restrictions. One that we have already discussed is season length. The second and more important output restriction in sports follows directly from the maintenance of exclusive territories. Owners acting through their leagues limit the number of teams in their league by choice rather than through the forces of competition. There are two important indicators that the number of teams is smaller than a more economically competitive sports world would give to fans. First, rival leagues do form occasionally. This indicates that the previous league size was too small in the eyes of the fans of the rival league. Second, every time a league announces that it plans to expand, a long line of candidate-owners forms in the hope of becoming the newest addition to a league. Wealthy people typically do not line up to get assets that generate low returns. This suggests that the expansion team will be economically viable or, in turn, that the number of teams prior to expansion was smaller than the set of possible markets could support. Because the line is long, economic intuition suggests that savvy businesspeople would snap up more than just the expansion team(s) being offered. Either way, the number of teams, even after expansion, is smaller than competition would provide. 170 HIGH PRICES The remaining effects of the market power generated by exclusive territories can be seen in sports pricing and profits. In Chapter 2, we saw how sports teams facing downward sloping demand functions have market power. One aim by owners in this situation is to employ a variety of price discrimination mechanisms to increase fan expenditures. The cost of providing on-field competition does not vary by time of day, day of the week, or the age of fans, but fans’ willingness to pay does vary along each of these dimensions. Profits are higher when owners can price discriminate and owners can only price discriminate if they have market power. In Chapter 4, we also saw how profits can be positive in the long run for sports teams. Territorial exclusivity, controlled by leagues, maintains the market power of teams in a given location, leading to all of these pricing outcomes. 171 SECTION 4 Expansion and Relocation: Finance Leagues expand and member owners relocate their teams for two reasons. First, there can be money in it. League expansion, increasing the number of teams in the league, pays because the franchise right is valuable to potential owners. Team relocation to a more profitable location also is good for the owners who move their teams. Other teams in the league may earn any spillover value in the national TV contract. In addition, expansion and relocation also protect existing owners from outside competition. As some cities grow and prosper and others fade economically, the most valuable locations for pro sports franchises change. New York Giants owner Horace Stoneham knew this only too well when he moved his team to San Francisco with this parting shot, “Tell the kids I haven’t seen their fathers at the ballpark lately” (Sporting News, October 5, 1992). Leaving some viable locations without teams enhances the bargaining power of existing owners with their current host cities, but if an existing league is slow to move, a rival league may move into these new economic centers. Expansion and relocation can preclude this possibility. Let’s examine these components of a league’s expansion and relocation decisions. EXPANSION AND RELOCATION: THE DIRECT FINANCIAL COMPONENT Suppose an existing league is considering a candidate for expansion or relocation. We would expect the league to approve the expansion or move if, on net, the current team owners in the league will be better off because of the expansion. Selling the franchise right to an owner can generate hundreds of millions of dollars for current owners. The important dollar elements include the expected ownership value of the expansion team, gate impacts for member teams, and the impacts of expansion on league television rights values. Although this is just as true for a team move, it is easiest to see with an expansion, so let’s stick with that example. NET PRESENT VALUE OF THE EXPANSION TEAM Members of the current league, really just the current team owners, decide the expansion fee to charge. Actually, they are trying to estimate the net value of ownership that we just covered in Chapter 4. Thus, the current owners must estimate revenue sources in the expansion city (gate, television, and venue) and costs (operations, player costs, and the owner’s opportunity cost). In addition, there will be values to any related business operations, “costs” that are actually profit- taking, cross-ownership tax advantages, pass-through tax advantages, and shares of any future league expansions. These can be summarized as the expected net present value of expansion team ownership, NPVO: 172 where “O” denotes ownership of the expansion team; V is the overall value of ownership; C is cost of ownership; t indexes time into the future from the base year “zero” to the end of the planning horizon, T; and r is the interest rate. The t subscripts on the right-hand side of the NPVO expression make it clear that value, costs (including the owner’s opportunity cost), and the interest rate all vary over time. To the extent that the league can make such an estimate, the expansion fee will include the chance to capture. But there are other financial impacts from expansion that owners must consider and that are not captured by NPVO. OTHER EXPECTED LEAGUE IMPACTS Expected local revenue impacts (gate, concessions, parking, and in some cases, local television) for member owners represent another element in the expansion fee consideration. Typically, expansion teams are not very successful at first, as shown in Table 5.2. On average, in MLB, it takes an expansion team almost seven and one-half years to win more than half its games. It takes between five and six years in the other three pro leagues. There are a few bright spots, such as the recent and successful Arizona Diamondbacks in MLB and the Jacksonville Jaguars in the NFL, but on average, expansion teams in MLB barely win 40 percent of their games over their first five years. Because fans love a winner, all else constant, this means revenues that depend on the quality of opponents will be lower when teams play expansion clubs. Expansion clubs simply tend not to be very good, and fans will be less inclined to pay as much to watch them. In addition, there may be expected national TV impacts, especially important in the NFL where there is only a national contract. Here we must be careful to distinguish between expansion and relocation impacts. Expansion impacts would be expected to be positive because adding a team expands the total market for the league’s games. But some expansion candidates will contribute more to the value of the national television contract than others, depending on those elements important to sports advertisers. As discussed in Chapter 3, advertisers care about the size of target demographic groups and how often these groups are expected to watch the sport in the expansion location. However, this is a consideration at the margin. Fans in every city watch sports. Granting a franchise to a potential owner in a particular city will increase interest in that city, but it is unlikely that everybody in that location is just going to stop watching if they do not get an expansion team. Thus, the current owners in the league compare the additional value to the national TV package between competing franchise locations. This was particularly important in the most recent NFL expansion to Houston, detailed in a subsequent section of this chapter. Relocation, in comparison, may or may not enhance the value of the national contract. It all depends on the value of the audience in the original and new locations. This can make 173 relocation a vexing problem for leagues. Owners would only consider relocating their team if it were in their personal best economic interest. But impacts on the national contract could be harmful to the league as a whole. It becomes a difficult league decision process in this case. Because these local and national TV impacts are an identifiable estimated dollar amount, they can be included in the expansion fee consideration process as the change in the net present value to existing owners, ?NPVn: where ? denotes change, and there are i = 1,, n total current owners. ?Nt denotes additions to the national TV contract in each period t, and Lit are the local costs at one of the current locations in each period t. The rest of the variables and notation are as in the calculation of the net present value of the expansion team. For what follows, it is essential to note that the sign of ?NPVn depends on whether the discounted additions to the national TV contract exceed the discounted local costs. THE EXPANSION FEE It is now easy to see how the expansion fee would be determined, in general. The league would like to charge a new expansion owner the net present value of ownership. In addition, if the change in value to current owners is positive, they would like to keep it. But if the change in value to current owners is negative, they would like to recoup it from the new owner. For our first case, suppose that the impact of expansion on national TV is positive and outweighs the impact of expansion on local revenues so that ?NPVn > 0. The
current group of owners would then like to charge an
expansion fee F = NPVO and simply pocket ?NPVn > 0 as it
accrues over time. Remember, the owner candidate’s share of
the national contract is already part of NPVO.
However, now think about the value from the potential
buyer’s perspective. The buyer knows that the value of
ownership is NPVO. But the buyer also knows they will be
generating ?NPVn > 0 for the current group of owners. As a
result, the buyer knows that the least the league would accept
would be F = NPVO − ?NPVn; something is better than
nothing, and the owners know the value of ownership. The
actual outcome would depend on the level of competition over
expansion franchises. Suppose there were only one potential
buyer. With a strong bargaining position, this single buyer
might be able to bargain the price down to F = NPVO −
?NPVn. More realistically, and more consistent with actual
outcomes, there are many buyers and heavy competition for
the expansion franchise. In this case, competition over the
franchise should allow the league to find a buyer willing to
accept the higher franchise price F = NPVO. The league would
be able to keep the positive value of expansion and extract the
value of ownership from the competitive buyers.
174

Another very interesting case does exist. Suppose that ?NPVn
< 0, instead of producing a net positive value for current owners in the league. In this case, owners will still consider expansion as long as NPVO > |?NPVn|. Indeed, the current
group of owners would love to be able to set franchise price at
F = NPVO + |?NPVn| and recoup the amount that local losses
exceed the added value of the national contract.
But this raises the very interesting question, what type of
potential buyer would ever pay more than the value of
franchise ownership? There are a couple of possibilities. First,
and consistent with profit maximization, the potential buyer
may think the estimate made by the current group of owners
is less than the true underlying value over time. If so, this
buyer would pay more than for the expansion franchise.
However, it would be sheer coincidence if the potential
buyer’s estimate of the mistake made by the current league
equaled |?NPVn|. A second possibility that we must admit is
that the owner may not seek to maximize profits. If there are
consumption benefits, then this type of owner would pay
more than the net present value of ownership as we’ve set it
up here. If a truly wealthy potential expansion owner doesn’t
care about the bottom line, the only limit on the price is the
marginal value of consumption from team ownership to that
owner.
The upshot of all this, given that competition among potential
buyers usually is heated, is that the most likely expansion
franchise price is F = NPVO. The current group of owners
would be expected to extract the value of ownership from
potential competitive buyers and, if the other changes in value
to current owners are positive, they just keep them. If those
other values are negative, it’s too bad for the group of current
owners because it is unlikely that 1) a new expansion owner
should have any different estimate of NPVO or 2) violate a
sound investment approach. Note also that the expansion fee
will vary across all of the locations under consideration during
any expansion episode. Large-revenue markets would have
higher expected profits and less negative impact on local
revenues than small-revenue markets. Using B for a potential
big-revenue-market location and S for a potential small-
revenue-market location, all else constant we would expect FB
> FS.
A FEW PRACTICAL CONSIDERATIONS
There are additional practical considerations about the
expansion fee. First, because fees tend to run from tens to
hundreds of millions of dollars, they typically are paid to the
league on an installment plan. Second, one would suspect that
the length of time over which negative local revenue impacts
occur would not be as long as the horizon used by member
teams to estimate expected profits. After all, even expansion
teams eventually got close to the league’s average winning
percent after about seven and one-half years.
A second practical consideration concerns owner viability.
Suppose that two potential owners represent the same
financial value of expansion for the league, but one is riskier
in terms of making expansion fee payments. This might lead
175

the league to reject one seemingly similar potential owner in
favor of another on owner-viability grounds alone. Even in
this era of franchise-free agency, the leagues prefer stability.
Demand for team sports requires time to mature. Mature
demand, with loyal fans, usually is the type that generates the
most revenue in the long run.
Two other values to expansion do not vary across potential
expansion owners. First, under league bylaws governing
expansion, the new team(s) must draft a specific number of
players from existing teams. This is called an expansion draft.
Of course, the rules also allow existing owners to protect
nearly all players of any real value. Thus, the players eligible
for the expansion draft typically are high-priced players at or
nearing the end of their careers and in the final years of long-
term contracts. As always, there may be a few bargains in an
uncertain world, but by and large, we would expect players in
the expansion draft to be overpriced.
The evidence on this is pretty clear. In the 1992 MLB
expansion, neither the Colorado Rockies nor the Florida
Marlins drafted a single eligible player with contracts over $2
million ($3 million). They chose much cheaper, younger
prospects and filled the rest of their lineups with free agents
and trades. Apparently, the expansion teams found all of the
older expansion-draft-eligible players to be overpriced and
did not take a single one of them at the going rate.
Finally, sharing of national TV contract revenues can be
postponed for new expansion team owners. In their 1992
expansion franchise agreements, Colorado and Florida agreed
to receive no share of MLB’s national TV contract until 1994.
Alas for them, just as they were about to receive their first
share of the national TV contract, the MLB strike occurred.
Large portions of the national TV contract were not collected,
a bitter pill for the Rockies and Marlins, who had waited two
years for their shares.
PROFIT EXTRACTION AND OWNER BEHAVIOR
The logic behind the determination of expansion fees raises
an interesting issue. The expansion fee includes the expected
net ownership value of the expansion team. In the 1990s,
these fees averaged close to $200 million in the NFL, $100
million in MLB, $82 million in the NBA, and $36 million in
the NHL. They have grown at extraordinary rates in the last
few decades (as high as 70 percent annually in the NBA). To
the extent that league estimates are close to the true value of
ownership, existing owners extract the largest possible
payment with just enough left to cover the expansion owner’s
opportunity costs. This means that one of the expansion
owner’s costs is the imputed expansion fee; owners must
include covering the expansion fee in their fixed costs. Only
unexpected changes alter the outcome. For example, if
stadium needs were unknown at the time the expansion was
granted and the new owner discovers that the local host city
will pay a generous stadium subsidy, then the new owner may
earn profits that the existing owners did not expect at the time
of expansion.
176

This extraction of value by the existing league owners forces
the expansion franchise owner to exploit market power just to
break even economically. If the new owner does not fully
exploit market power but the league included profits from
that type of behavior in the expansion fee, then the owner will
lose money. Collected profits over time will not cover the
imputed expansion fee. To the extent that the new owner is
able to exploit market power better than league owners
anticipated at the time of expansion, profits over and above
the expansion fee can be had. We expect owners to maximize
profits, and the expansion fee is a cost that must be covered
by the full exploitation of their market power position.
177

SECTION 5
Expansion and Relocation:
Strategic Behavior
The foregoing discussion considered only the financial value
of expansion to member owners. The introduction to this
section mentioned that there are also two strategic issues
associated with expansion and relocation. Expansion and
relocation can be carefully managed to enhance the
bargaining power of existing owners with their host cities.
Expansion and relocation can also be used to preclude
competition from existing or potential rival leagues. The
money issues just discussed, plus consideration of these
strategic values, give the complete account of the values that
go into the expansion decision.
EXPANSION, RELOCATION, AND OWNERS’
BARGAINING POWER
Leaving economically viable team locations empty is valuable
to current members of the league in bargaining with their host
cities. If the host city, county, or state fails to meet an owner’s
subsidy demands, the owner can simply threaten to move to
the open viable location. If elected officials do not want to lose
the team, the owner’s upper hand can be worth millions of
dollars. The strategy here concerns the preservation of
believable threat locations. A believable threat location is an
alternative location for the team that politicians in the current
host city would believe is truly a viable location.
To see just how important this threat value is, consider the
case of Tampa Bay/St. Petersburg during the 1990s. Despite
spending nearly $200 million on the Suncoast Dome stadium
(the forerunner of today’s Tropicana Field) to prove that they
were worthy, owner-hopefuls in Tampa Bay/St. Petersburg
did not get an MLB expansion team in 1993. Owners in
Denver and Miami received the teams for that expansion
round. The MLB owner-hopefuls in Tampa Bay/St.
Petersburg let it be known that they would be willing to host
an expansion team or buy an existing team.
Immediately following this failed attempt, in rapid succession,
the Chicago White Sox, San Francisco Giants, and Seattle
Mariners used Tampa Bay/St. Petersburg as a believable
threat location against their host cities. The White Sox
threatened to move to Tampa Bay/St. Petersburg, with its
“ready, willing, and able” empty Suncoast Dome unless they
received a new publicly funded ballpark. The hopefuls in
Florida printed up Tampa Bay White Sox tickets and mocked
up a few hats, joyful at the thought of a team. But with
extraordinary participation by then Illinois governor James R.
Thompson, at midnight of the last day of the session, the
legislature capitulated, and the financing groundwork was laid
for the new Comiskey Park in Chicago (now named U.S.
Cellular Field).
178

The Giants, after losing a few referendum votes for new
stadiums in San Francisco and the surrounding area, also
invoked the Tampa Bay/St. Petersburg ploy. Tampa Bay
Giants hats were mocked up, and season tickets were printed.
After a cooling-off period, commanded by then National
League president Bill White, the team was sold and kept in
San Francisco. The new owner immediately began to press for
a new ballpark and, eventually, significant infrastructure
support for the privately funded PacBell Park (now AT&T
Park) was granted by the city.
Last, but not least, the owner of the Mariners threatened to
sell the team to the Tampa Bay/St. Petersburg hopefuls in
order to raise enough money to pay the interest on loans to
keep his local radio network financially afloat. Once again,
there was much rejoicing in Florida. At the last minute, a new
ownership group kept the Mariners in town and immediately
began pressing for a new stadium. Shortly thereafter, the state
legislature produced a funding package for Safeco Field.
Clearly, as an open, believable threat location, Tampa Bay was
valuable to the owners of the White Sox, Giants, and
Mariners. Hundreds of millions of public dollars were
allocated to keeping these teams in their cities. Heavy
bargaining leverage was assured for three member owners
when MLB expanded to Miami and Denver rather than
Tampa Bay. Eventually, the value of that leverage must have
declined because an owner group in Tampa Bay did get a team
in the 1998 MLB expansion. At the end of the section, the
Learning Highlight: Keeping the Oilers in Edmonton is quite
instructive on the value of believable threat locations in
another league, the NHL.
EXPANSION, RELOCATION, AND PRECLUDING
COMPETITION FROM RIVALS
Maintaining believable threat locations is valuable to existing
owners. However, there is a downside to this practice. By
leaving viable locations empty, the league puts out the
welcome mat to the possibility of competition from rival
leagues. As we will see, rival leagues have actually formed
around just one or two teams in large-revenue markets also
occupied by another league’s team. The rival league gains
initial access in one or two large-revenue markets and puts
the rest of its teams in smaller markets uncovered by the
dominant league. The result is competitive economic pressure
on the dominant league. With competition, some fans shift to
the rival league. Demand and revenues decline for teams in
the dominant league. In addition, the rival league can begin
bidding up the price of players so that costs rise for the
dominant league. When prices fall and costs rise, profits must
also fall.
Expanding the league to include the viable location or moving
an existing team to that location should reduce the
attractiveness of formation of a rival league. Strategic
expansion and relocation can preclude rival competition. The
same would be true of putting another team in a market large
enough to support two teams. If the rival league cannot gain
access to some large-revenue-market areas, it is unlikely to
179

form. Table 5.3 shows how well pro sports leagues have done
at keeping a presence in the top 30 population centers. MLB
always has been best at covering the top 30 (Fort and Quirk,
1995), whereas some cities are just not basketball or hockey
towns, as the NBA and NHL have the worst coverage of top 30
population centers. As you can see in Table 5.3, the NFL also
has plenty of room to expand.
Let’s look at an example of strategic relocation. Up until the
1950s, the majors had simply ignored moving west of the
Mississippi River. However, population grew steadily in the
West, especially in California. The Pacific Coast League (PCL)
thrived with teams in Los Angeles, San Francisco, Vancouver
(BC), San Diego, Hollywood, Seattle, Portland, and
Sacramento. During U.S. House of Representatives antitrust
hearings in the early 1950s, members of the Celler Committee
(named for its chairman, Emanuel Celler of New York)
scolded MLB for its reluctance to move west and told MLB to
design a procedure that would show how a rival league could
obtain officially sanctioned MLB status. MLB responded with
a procedure and a set of requirements, based on attendance,
that nearly none of its own teams could meet (the same rules
that the Continental League tried to use in the example at the
beginning of the chapter).
The PCL was so prosperous that its principals actually tried to
gain official recognition as a third major league under these
guidelines. However, the clearly unreasonable requirements
stalled the PCL’s campaign for major league status. In the
meantime, MLB’s Brooklyn Dodgers and New York Giants
moved west in 1957, occupying what were clearly the anchor
cities for the PCL, Los Angeles and San Francisco. With this
ready-made historical MLB rivalry now in their area, baseball
fans shifted loyalty away from their old PCL teams and toward
the MLB’s new Los Angeles Dodgers and San Francisco
Giants.
Shortly after the Dodgers and Giants appeared, the PCL was
almost entirely out of California, with only the Padres
remaining in San Diego. The other California teams moved to
Phoenix, Salt Lake City, and Spokane. After losing their
anchor cities, the PCL was relegated to AAA minor league
status. Indeed, most of the remaining teams in the AAA
version of the PCL were either purchased by MLB teams or
entered into close contractual relations with MLB teams.
Clearly, the relocation of the Dodgers and Giants ended all
chances, however unreasonable, for the PCL to obtain major
league stature. It also ended the chances for any of the PCL’s
franchises to individually join MLB, a typical practice toward
other rival leagues. If there’s one thing we know, it is that in
every pro sport, even though rivals sometimes appear, the
single dominant league outcome always prevails. MLB, the
NBA, NFL, NHL, and even the WNBA have all confronted
rival leagues, and one dominant league always is the result.
THE 2000 NFL EXPANSION
The 2000 NFL expansion into Houston serves to illustrate
nearly all of the elements in the decision by league owners to
180

expand. The decision boiled down to two locations that had
recently lost their NFL teams, Los Angeles (which had lost the
Rams to St. Louis and the Raiders to Oakland) and Houston
(which had lost the Oilers to Nashville, later renamed the
Tennessee Titans).
The Houston owners group raised the $700 million ($867
million) expansion fee. In addition, it was a top 10 population
center and a large TV market. The owner group was
financially solid, and Houston taxpayers had committed to a
new stadium. Added to the expansion fee, the stadium
commitment raised the dollar value of the amount offered by
the Houston owner group to nearly $1 billion ($1.24 billion).
In addition, Houston had a team before, and the fans were
hungry for another. Sports Illustrated reported that $48
million ($59 million) in PSL revenues were collected by the
expansion Houston club in just 21 days (January 16, 2000, p.
23). Los Angeles, also with a viable owner group but with
many unresolved stadium issues, relied on its strength as an
economic and TV powerhouse but lost its bid for a team. The
Houston Texans began play in September 2002.
This is a marvelous example of how a league evaluates the
value of TV markets at the margin, as described earlier in this
chapter. According to the Los Angeles Times (June 6, 2003),
the NFL enjoyed a 9.5 rating on broadcast television in Los
Angeles in 2002, a season when there were no NFL teams in
the area. This rating was 58 percent better than that enjoyed
by the NBA Lakers, 188 percent better than the MLB Dodgers,
and a whopping 764 percent better than the NBA Kings.
Overall, the Los Angeles rating was better than New York’s 9.3
with two NFL teams. At the margin, it is entirely possible that
ratings in Houston were better, given the presence of a team
there, than the ratings would have been in Los Angeles had
that owner group received the team.
In addition to all of the other comparisons, strategically there
was quite a bit of value to NFL owners in keeping the Los
Angeles market open. It is a completely believable threat
location for other teams in the league to use against their
hosts. Of course, the NFL owners also run the risk associated
with leaving Los Angeles open. Such a location probably could
hold two teams, easily. Rival leagues have certainly started
with less, a topic to which we now turn.
HISTORY OF RIVAL LEAGUES
The brand new United Football League began play in October,
2009 (www.ufl-football.com). Its stated “game plan” is to
challenge the NFL’s lock on major league football, and it has
carried out its plan, so far, in a way similar to many rivals in
the past. Some modicum of success rest on finding an
underserved set of smaller-revenue locations, plus at least one
team in at least one megalopolis market, and, finally, some
form of national TV coverage. The UFL has chosen Las Vegas
and Orlando for the smaller-revenue markets, with ambitions
in Hartford and Sacramento. The larger-revenue markets
currently are New York and San Francisco, but the alternate
location is (you might have already guessed) Los Angeles.
However, so far, only Premium TV has picked up UFL games
181

http://www.ufl-football.com

http://www.ufl-football.com

(eight games on Versus and five on HDNet). It is unclear
whether they will land a national TV contract. And given the
history of rival leagues, the UFL’s prospects are not great.
The history of rival leagues is too long to include in this book.
(The interested reader can find a pretty complete account in
Quirk and Fort [1992, 1999]) Although the current structure
of leagues in professional sports has been fairly stable for
nearly 20 years, rival leagues have formed and have been
successful. Table 5.4 lists rivals for the four major U.S. team
sports.
How these rival leagues formed is quite clear. First, the
artificial restriction on the number of teams by leagues often
leaves room for a rival league to put a team in the largest
revenue markets. How do you convince the sole occupant of a
market with vast revenues to act in the interests of the league
and accept another close neighbor? Expansion into an existing
team’s franchise area requires unanimous consent, typically.
If that team cannot be compensated or the behavior of
existing teams is to not pay compensation to the current
single-occupant owner even if they could, then a protective
expansion can be vetoed. If the league fails to get that second
team in place, then the door is open to a rival league in a large
market.
A classic example occurred in 1960. To that time, the NFL left
the pro-football hungry fans in Dallas without a team. The
upstart AFL (version IV) made it clear that one of their
original teams would be in Dallas. True to its word, the AFL
began play in 1960, and one of its original teams was the
Dallas Texans. Interestingly enough, the NFL also granted an
expansion franchise that began play as the Cowboys in that
same year. The battle for Texas was on. Eventually, the NFL
brand proved too strong, and the AFL Texans moved on to
become the Kansas City Chiefs. But clearly, the AFL gained
substantially from having one of its original teams in such a
strong market.
The second part of the answer has to do with pushing a good
thing too far. An open viable market provides valuable
bargaining leverage to member owners. However, if you leave
a plum hanging on the tree for too long, a rival league is quite
likely to pick it. Any type of miscalculation on the part of the
league about how long it can leave a viable location open can
lead to the formation of a rival league. Indeed, we may be
about to see if the NFL has erred in leaving the Los Angeles
market empty for so long. The upstart UFL clearly has its eyes
on that prize.
The case of the PCL and MLB is just one example of how
relocation protects against rival leagues. Leagues have always
moved their teams around quite a bit, except for the last 30
years in MLB. Table 5.5 shows team moves by league. No
league beats the NBA in team moves, either by the number of
examples or the number of times a given team has moved.
Two NBA teams have been in four different cities, and there
are four three-city teams. Although teams typically move less
in the NHL, we do have the very interesting example of the
Dallas Stars. If you count the merging of the original
182

Minnesota North Stars with the Cleveland Barons, that team
has moved five times. All of this movement should be to
higher-valued locations, helping to preclude the formation of
rival leagues.
One observation about the history of rival leagues that must
be emphasized is that they simply do not survive. The single
dominant league outcome has always prevailed. As shown in
Table 5.4, partial mergers and assimilations characterize
nearly every rival league episode. Mergers and assimilations
typically occurred by offering the owners of the most
successful teams in the rival league very inexpensive
franchises in the dominant league. Those owners then just
moved their entire team, lock, stock, and barrel, into the
dominant league. Sometimes assimilations happened by
melding some teams in the rival league with dominant league
teams. Let’s examine this single dominant league outcome in
more detail.
THE SINGLE DOMINANT LEAGUE AND GAME THEORY
The earlier section on the history of rival leagues makes it
clear that dominant leagues have confronted competition. But
why haven’t any of these rival leagues been able to make a go
of it over the long haul? The first part of the explanation
involves a simple game theory explanation that suggests rival
leagues that might be operating on a shoestring may not be
able to survive the competitive struggle for existence (Fort
and Quirk, 1997). The rest of the explanation concerns lack of
enforcement of laws designed to foster economic competition.
Let’s begin with the game theory portion of the explanation.
Suppose two rival leagues, each with two strategy choices,
spend high on talent and become a “major” league in the eyes
of fans or spend low on talent and remain a minor league.
This idea focuses on the idea from Chapter 2 that fans care
about the absolute level of play in their identification of what
it means to be a major league. The matrix of outcomes is in
Figure 5.1.
Let’s examine choices by League 1 in Figure 5.1. If League 2
spends high, then the best that League 1 can do in response is
to also spend high (if League 1 spends low, League 2 will be
perceived as a major league in the eyes of the fans). If League
2 spends low, then the best that League 1 can do is, again, to
spend high (and become the only major league in the eyes of
fans). Because the rewards are symmetric, we would expect
League 2 to always choose to spend high as well. Given this
dominant strategy for each league, the result is that both
spend high and are seen as equal in the eyes of fans, and we
have rival leagues.
The question from the perspective of the single dominant
league outcome is whether this equilibrium can survive.
Because both leagues are spending high, but the rival
probably doesn’t have teams located in as many large-revenue
markets, the rival league could fail. Competition may not be
self-sustaining if the returns are too low for the rival league.
183

Owners in a league can use revenue sharing, the draft, luxury
taxes, or salary caps (all covered in the next chapter) to
transfer money among themselves and enhance survival.
Alternatively, they can use scheduling penalties and play-offs
that reduce the certainty of pay-offs to talent stocking against
the most powerful teams. However, implementing these
mechanisms between leagues is more difficult and possibly
illegal. So rivals may fail.
This idea also leads to some speculation about whether the
dominant league can intentionally kill a rival by bidding up
the price of talent. If the dominant league, with a huge war
chest from past and future expected profits, starts to bid up
the price of talent against a rival, then the rival may not
survive. If the rival league is not as profitable or it was poorly
funded at the outset, bidding up talent prices could be the
end. Although there is no definitive proof of intentional
strategy, this sounds an awful lot like what might have
happened to African American baseball leagues (AABLs) after
integrating with MLB. This is covered in the bonus Learning
Highlight: African American Baseball Leagues as an MLB
Rival at the end of the section.
James Quirk and I (1999) sum up the single dominant league
outcome in this way:
The profit potential of additional franchises in the
megalopolis cities has been one factor driving the entry
of rival leagues in the past, and might in the future as
well. Still, all in all, it appears that leagues have
managed to expand sufficiently to deter entry while still
preserving enough vacant sites to make move threats
believable, which is bad news, of course, for fans and
taxpayers. (p. 136)
ENFORCEMENT OF THE ANTITRUST LAWS
One last observation is that owners, acting through leagues,
do not have to be allowed to behave this way. Antitrust laws
are over 100 years old and are designed to handle precisely
these problems. Indeed, it has often been the case, from the
famous Standard Oil of Ohio case on down to the recent
Microsoft case, that dominant firm outcomes have drawn the
scrutiny of regulators and the courts. However, the courts and
Congress have not acted to stop the single dominant league
outcome in pro sports. We’ll talk about this in Chapter 12.
ECONOMIC IMPACTS OF LEAGUE EXPANSION AND
RELOCATION CHOICES
The upshot of this careful management of location by and for
the benefit of member owners is to keep the price of
franchises quite high. As fan demand for the league’s product
increases over time, revenue growth is channeled to existing
teams. This raises franchise prices in a way that would not
happen if new teams were formed to satisfy increasing fan
demand. If new teams were formed, some new owners would
earn part of that increase in revenues. So, from the league’s
perspective, only occasionally is expansion the better way to
increase the value of the league enterprise, and one would
184

have to say that leagues have cultivated this value expertly.
Refer back to Table 1.3 and refresh your memory on just how
much the very judicious use of expansion has led to dramatic
growth in the value of expansion franchises over time.
In addition, leagues exercise caution in this choice because it
can be very difficult to undo expansion even if the economic
situation warrants it. MLB made a big noise in 2001 about
trying to reduce the number of teams in the league by two but
never followed through. Some argue that the NHL has allowed
expansion to get out of hand and may face the prospect of
league contraction in the near future. Yet another reason for
caution is whenever there is an expansion decision in
professional sports, government scrutiny is brought to bear on
the league decision process. Nothing is surer to generate
House and Senate hearings on the market power position of
pro sports leagues than expansion. Representatives all want
the expansion franchise to go to the group in their political
jurisdiction. It should come as no surprise that sports leagues
would rather not have their exercise of market power under
the congressional microscope.
185

LEARNING HIGHLIGHT: KEEPING THE OILERS IN
EDMONTON
Once the team of the “Great One,” Wayne Gretzky, and a five-
time Stanley Cup champion, the NHL Oilers needed a neat hat
trick to remain in Edmonton in 1997. The example also ties
together lessons from Chapter 4 on sports accounting and the
value of team relocation or, more technically in this case, the
battle against relocation.
Then owner Peter Pocklington claimed that the Oilers had lost
$60.75 million since 1990 (all values in this learning highlight
are converted to US$ at the then prevailing rates). This was
despite the fact that he had used the threat of moving to a
believable alternative location in 1993 to get federal,
provincial, and local governments to remodel Northlands
Coliseum (currently Skyreach Centre) and Pocklington’s
minor league baseball field. The public contribution was about
$20.25 million. In addition, Pocklington reportedly was $162
million short on taxes owed to the province of Alberta. He put
the Oilers up for sale asking $85 million.
But the federal, provincial, and local governments that helped
rebuild Skyreach Centre had exhibited a modicum of
foresight. A clause in the remodeling agreement (good until
2004) stated that if a local group could muster $70 million,
then the team was theirs. The timing of the $70 million went
like this. First, a $5 million nonrefundable payment had to be
made within 30 days of any offer made for the team. An
acceptable payment plan for the remaining $65 million had to
be reached in 2 months after the $5 million payment. This
clause would play a crucial role as events unfolded.
Enter Cal Nichols, president of the Edmonton Investor Group
(EIG). Nichols made his fame with a string of gas stations in
western Canada and had been instrumental in helping to keep
the Oilers in Edmonton in the past. He began to rally the
troops, and EIG started learning some interesting sports
accounting lessons covered in Chapter 4. Remember, one
lesson from Chapter 4 was that not all is as it seems on sports
annual operations reports. Werner Baum, an eventual
investor in EIG and former accountant for the Oilers, went on
the record saying, “Those numbers [Pocklington’s claimed
losses], I knew, weren’t anywhere near what the actual
situation was.” Baum claimed that Pocklington was including
accounts not related to the hockey team in order to paint a
bleak picture for those who wanted the team to stay in
Edmonton. Apparently Pocklington hoped that frightening
away smaller local buyers with scary stories of mounting
massive losses would allow a bigger offer than $70 million
from an outside buyer who would probably move the team to
the United States.
EIG also discovered that the Canadian tax authorities
permitted a roster depreciation allowance that was even more
generous than the one for U.S. team owners discussed in
Chapter 4. In Canada, 60 percent of the purchase price could
be declared player roster depreciation over a reasonable
period, a sizable tax shelter for a pass-through firm. Cal
Nichols noted, “If we did nothing more than break even over
186

the first four years, at least we would be in a position where
we could get almost 100 percent of the investment as a tax
write-off against other income and still have the paper, the
shares of the Oilers.”
In the meantime, that bigger outside offer was made by NBA
Houston Rockets owner Les Alexander. He placed his $82.5
million bid in early 1998. EIG knew it was coming, but unless
they could come up with the $70 million, it looked like a
pretty good bet that the Houston Oilers would resurface but in
the NHL as another occupant of Houston’s arena. Through
the force of Baum’s arguments that losses were not as they
seemed, and the lure of lucrative tax advantages, EIG made
the $5 million deposit in the nick of time. They used the
remaining two months to try to get part of the remaining $65
million from local government officials.
EIG tried to pressure the city council for $14.2 million and
part ownership but failed. They did manage to obtain a rent
subsidy of $2.4 million annually through 2004. The mayor of
Alberta suggested that this was a no-brainer given the huge
economic spin-off to the city area generated by the team. (Stay
tuned for more on the economic value of sports teams to cities
in Chapter 10.)
At the time, some questioned if EIG would keep the team in
Edmonton after the restrictions on outside team sales ended
in 2004. After all, there was a genuine offer of $82.5 million
for the Oilers (similar to the $80 million paid for the San Jose
Sharks in 2002). When the tax breaks that fueled most of the
private support ran out, coincident with the 2004 release
date, would EIG just cash in on the additional $12.5 million?
Actually, it worked out better than that for EIG. Daryl Katz,
Chairman and CEO of the Katz Group (owner of one of the
largest drug store chains in North America), bought the Oilers
outright for about $200 million ($199.3 million) in 2008. He
immediately began a campaign for a new arena.
Source: Inspired by CAmagazine, October 1999, pp. 24–32.
187

BONUS LEARNING HIGHLIGHT: AFRICAN AMERICAN
BASEBALL LEAGUES AS AN MLB RIVAL
A reasonable argument can be made that AABLs were killed
by the integration of MLB. Not to downplay the significant
historical role of the integration of MLB, but the shenanigans
killed a pretty successful entertainment enterprise, namely,
AABLs. Joel Maxcy and I (Fort and Maxcy, 2001) show a
number of interesting things about the relationship between
MLB and AABLs. AABLs were profitable and viable. Owners
did well, and star players in the AABLs were paid on a par
with their white MLB counterparts. There is some evidence
that AABLs were a thorn in the side of major league owners
from the attendance perspective.
Integration, after Jackie Robinson joined the Brooklyn
Dodgers in 1947, occurred in a brutal fashion from the
perspective of AABL owners. Although a few owners like Bill
Veeck did buy out contracts, typically MLB owners raided the
AABLs relentlessly for talent, at most paying pennies on the
dollar. Unlike most rival leagues, AABLs were precluded from
responding by just bidding up the price of talent because MLB
stars did not want to play in AABLs (although there were a
few MLB stars that did tour with AABLs in the off-season).
Branch Rickey, president and general manager of the
Brooklyn Dodgers, who signed Jackie Robinson, simply called
AABLs a “racket” to justify paying no compensation for
players in that league. Finally, the minor leagues were locked
up by MLB, typically through ownership of minor league
teams by MLB teams, so the few remaining AABL owners
really had nowhere to turn for organized league play.
The result? Unlike other rival leagues, AABLs received no
cheap offers to join MLB, and a profitable league endeavor
simply died on the vine within three years of integration, by
about 1950. Literally hundreds of professional baseball
players were precluded from earning a living at their sport,
and countless hundreds of thousands of fans lost their only
access to live major league baseball. Almost. MLB did expand
into the previous AABL strongholds of Kansas City and
Baltimore within a few years of the demise of the AABLs.
However, even across color lines, the relentless single
dominant league outcome prevailed.
MLB integrated with Jackie Robinson in 1947, but the
aftermath was the end of a proud legacy of African American
baseball leagues. [Black Dodgers with Leo Durocher.]
Source: Fort and Maxcy (2001).
188

SECTION 6
Negotiations
The impacts of joint venture behavior also occur as leagues
negotiate with media providers, players, and their host cities.
The most striking observation about these joint venture
negotiations is the exercise of market power. Because
individual owners could handle these negotiations, economic
intuition leads us to suspect that leagues produce better
results for owners than they could get on their own. Let’s
examine the three joint venture negotiation settings:
broadcasting, players, and host cities.
JOINT VENTURE: BROADCAST RIGHTS
League negotiations with media providers are a joint venture
that could be done individually by team owners. Indeed, in all
leagues except the NFL, which only offers national rights,
individual teams negotiate their own local TV agreements.
Owners act together through their leagues to confront media
providers (network, cable, and dish TV) with a league
monopoly on national broadcast rights. If media providers
want to broadcast a major league sport, they have no
alternative but to deal with the league as a whole. This leads to
value extraction by the major leagues through national TV
contracts. The profits flow primarily from advertisers to the
teams in sports leagues through media providers.
JOINT VENTURE: PLAYERS
Nothing requires team owners to deal as leagues with players.
In the fast-service industry, for instance, the McDonald’s
corporation does not confront organized labor in particular
states through joint venture activity on behalf of all of its
franchisees. Indeed, sports leagues do not centrally negotiate
pay scales through their leagues, instead reserving individual
contracting rights for every owner. So, once again, we are led
to conclude economically that the joint venture provides a
better outcome for owners than acting individually toward
players’ labor unions.
Labor relations are such an important topic that they deserve
their own chapter, Chapter 9. However, in a nutshell, the
object of negotiations with players is for owners to try to keep
as much of the revenues as they can while lowering costs as
much as possible. Player payments are the largest cost to
owners, and one aim of negotiations is to reduce this cost as
much as possible. Players also produce value at the gate and
on TV, and owners would like to keep as much of that revenue
as they possibly can. Historically, by sticking together in
negotiations as a joint venture, pro owners have been very
successful at keeping the vast majority of the economic value
produced by players. This was especially true prior to free
agency, as we will see in Chapter 8.
189

JOINT VENTURE: HOST CITIES
Finally, leagues produce an especially controversial joint
venture outcome in the dealings of owner members with host
cities. We’ll devote Chapters 10 and 11 to the relationship
between teams and their hosts, but they are so much in the
news that you probably already have some understanding of
stadium issues. There also were hints about it in the Learning
Highlight: Keeping the Oilers in Edmonton.
Teams put demands on their host cities for improved lease
terms, stadium renovation, and even publicly funded new
stadiums and arenas. As state and local governments are
understandably reluctant to subsidize sports enterprises
generating hundreds of millions of dollars in revenue, the
teams threaten to leave for one of the viable locations their
league may have kept open. Because owners, acting as
leagues, have controlled the location of teams so well, the city
is at a bargaining disadvantage. Hosts are unable to appeal to
some other team to replace their current team. Invariably,
then, it is an all-or-nothing proposition, and cities usually (but
not always) give in. It should come as no surprise that teams
make out handsomely.
We have covered all of the ways that leagues deal with the
external environment confronting their member team owners.
In the next chapter, we move on to probably the most
important issue that owners must deal with among
themselves, namely, competitive balance.
190

SECTION 7
Chapter Recap
Leagues facilitate play on the field and the business structure
of sports. Coordinated behavior through leagues provides
teams with opportunities they could not exploit acting alone.
Teams act together but not in an anticompetitive fashion as a
single entity when leagues determine play on the field. This
single-entity behavior has economic consequences through
scheduling and rule changes.
Nearly all of the major economic impacts on owners occur
through league joint ventures. Because owners could perform
most of these same functions individually, economic intuition
suggests that joint ventures generate greater profits. Owners
maintain exclusive territories through league cooperation.
This is the fundamental source of team market power.
Expansion and relocation are important joint ventures used
by leagues to maintain market power for existing teams.
Leagues decide on team location based on a few important
factors. The financial value of expansion is important, but so
is the trade-off between leaving a location open for its
believable relocation threat value against existing host cities
and the chance that rival leagues will occupy viable locations.
An especially troublesome issue is expansion and relocation
when a location will economically support more than one
team. If the single team in such an area is protected, the
league is less profitable than it could be, and a rival league is
offered a location where it can establish a beachhead.
All major sports leagues face rivals, but the single dominant
league outcome always prevails. For fans, the result is always
the same. They get less of a sport, in fewer locations, at a
higher price as a result of the careful management of team
locations by leagues. How is it that the single dominant league
outcome remains so pervasive over time? Simple game theory
shows that rival leagues can fail if, in the face of stiff
competition over player talent, the resources available to the
rival are insufficient. In addition, the dominant league might
just be able to bid up the price of talent enough that rivals
must bow out.
There are other joint ventures as well. Leagues typically
negotiate TV deals, bargain with players individually and with
unions, and set the bargaining position of teams relative to
their host cities. In all of these situations, the league provides
a superior outcome for teams than they could obtain acting
alone.
191

SECTION 8
Key Terms and Concepts
You should have run into each of these in pop-ups in the text
of this chapter:
• Single-entity cooperation
• Joint ventures
• Exclusive territories
• Franchise agreements
• Quantity restrictions
• League expansion
• Team relocation
• Ownership value
• Expected local revenue impacts
• Expected national TV impacts
• Expansion fee
• Owner viability
• Expansion draft
• Imputed expansion fee
• Believable threat locations
• Rival leagues
• Strategic expansion and relocation
• Single dominant league
• Joint venture negotiations
192

SECTION 9
Review Questions
1. What is single-entity cooperation? How is it essential to
league play?
2. How can single-entity cooperation have economic
impacts on member owners through scheduling? Through
the setting of rules? Through setting up championships?
3. What is joint venture cooperation? What distinguishes it
from single-entity cooperation?
4. Is cooperation required for owners to do the things they
do as joint ventures? Give an example.
5. Explain territory definition and protection. What do
member-owners get through the definition and protection
of territory by their league?
6. What is a franchise agreement? Why is it necessary?
7. Describe the impact of exclusive territories on team
quantity and price.
8. Define the net present value of an expansion team to a
potential buyer. Define other expected league impacts due
to expansion. Explain what is meant by national TV
impacts at the margin due to expansion.
9. What determines the expansion fee?
10. What is an imputed expansion fee? Is it a fixed or variable
cost?
11. How do expansion and relocation help owners in their
negotiations with host cities? Discuss the role of
believable threat locations in your answer.
12. What is a rival league? How do expansion and relocation
help owners to preclude rivals?
13. List the economic impacts of strategic expansion and
relocation choices by leagues.
14. List the three types of joint venture negotiations done by
owners acting through leagues.
15. Is cooperation required for such negotiation to occur? If
so, why? If not, then why do owners negotiate through
joint venture activities?
193

SECTION 10
Thought Problems
1. In Table 5.1, explain the pattern of predicted impacts for
widening the strike zone in 1963. After that change, which
one would you expect to enjoy an increase in value,
relatively speaking, hitters or pitchers? Explain.
2. What does a new owner gain when purchasing a franchise
from another (remember the lessons of Chapter 4 in
terms of the values of ownership)? What is the most that
anybody would pay in order to have a pro sports
franchise?
3. What essential service did the NBL fail to provide its
teams that led them to jump to the BAA? Why didn’t the
NBL just respond to these owner needs? What does this
say about the ability of rival leagues to form?
4. Why does market power result from territory definition
and protection by owners acting as leagues? Provide a
nonsport example where such protection does not
generate market power. What is the crucial difference
between your example and a pro sports league?
5. A league is considering expansion into two different
markets, A and B. Suppose all else is constant in these
two markets except for each of the following differences,
considered one at a time. For each difference, what
happens to the expansion fee in Market A relative to
Market B?
a. Market A is a larger revenue market than Market B.
b. Population growth is greater in Market A than in
Market B.
c. Currently, city politicians and voters are hostile to any
consideration of subsidies for a sports team in Market
A, but not so in Market B.
d. Interest rates are higher in Market A than in Market B.
6. What factors determine the difference between the
expected financial value of expansion and the franchise
fee? Is it possible that an expansion decision could rest on
one of these factors rather than on the expansion fee
itself?
7. Define the net present value of ownership. Explain fully,
including a definition of all terms in your expression for
this value. Two things must be true for owners to both
extract an expansion fee equal to your stated value of
ownership and simply pocket the other changes in values
to current owners. What are these two things? Again,
explain fully.
194

8. What must be true about other league impacts due to
expansion for owners to wish they could charge a
franchise fee, ? Explain fully, including a definition of all
terms in this expression. What could possibly lead a
potential team owner to pay such a franchise fee in excess
of the league owners’ estimate of the net present value of
owning the expansion team?
9. List all of the differences in the current league owners’
estimate of the net present value of owning an expansion
team versus selling an existing franchise and relocating it
to another city. Also list the differences in the other
league impacts that would occur due to relocation as
opposed to expansion.
10. Ultimately, what is the most likely value of the expansion
fee, F, in actual practice? Why?
11. Owners are often criticized as rapacious monopolists.
Defend their behavior in terms of the expansion or
franchise fee that they paid for their teams.
12. Suppose you are a league consultant. What guiding
principles would you use to help plan the league’s future
expansion and relocation decisions? Why is a presence in
the top 30 cities so important?
13. The Los Angeles market is an important market for any
league. Explain how a complete comparison of the Los
Angeles and Houston markets led the NFL to grant an
expansion franchise to Houston rather than Los Angeles.
Be sure to include in your explanation the information in
the text on television ratings in Los Angeles compared to
other cities in 2002.
14. Why is the NFL so much worse than MLB at covering the
top 30 population centers? Does this mean they run a
greater risk of a rival league? Why or why not? In your
answer, be sure to present how a rival league can even get
started in the first place (remember, there are two
components).
15. Suppose that joint venture negotiations were not allowed
on the part of owners. Explain the expected outcome on
the value of TV contracts for the league members.
195

SECTION 11
Advanced Problems
1. There are official rule changes made by agreement among
owners (some require the consent of players as well). But
there are changes also in the way that rules are enforced.
Using Scully’s (1989) ideas concerning the reasons for
rule changes, why would owners alter the enforcement of
particular rules. Pick an example in a single pro sport and
apply that logic.
2. As detailed in the Learning Highlight: Challenging
European Football’s Power Structure, why would UEFA
fail to give small-country teams their own championship
round? Think along the lines of giving members what
they want.
3. It seems these days that everybody wants a pro sports
team. But, almost universally, expansion teams are weak
competitors and stay that way for quite some time. For
example, Quirk and Fort (1992, pp. 251–252) offer the
following Evidence on the Number of Years for
Expansion Teams to Reach 0.500 over the period
1900 to 1989:
a. Given the bleak prospects for an expansion team on
average in any league, how would you explain the high
demand for new teams on the part of the residents of
some cities?
b. Given the data, what do you think expansion does to
the overall quality of league play from the fans’
perspective? How does this enter into the
determination of the expansion fee by a league?
c. What factors would explain the difference between the
shortest time to 0.500 and the longest time to 0.500 in
each league? (Hint: Think about the lessons from
Chapter 2.)
4. When free agency was introduced in the NFL after the
1992 season, some argued that expansion teams would
have an easier time competing than in the past because
new owners could just buy great talent and enter the
league with a running start. Using the data in Table 5.2,
would you have made the same prediction? (Hint:
Remember that free agency started in 1976 in MLB.
Compare expansion team’s success before and after that
year.) Derive the same information shown in Table 5.2 for
NFL expansion, and see if your prediction would have
been correct. Explain this outcome in terms of the
optimal choice of talent for expansion teams.
5. What possible economic justification was there for MLB’s
expansion Colorado Rockies and Florida Marlins to agree
to forgo a share of the national TV contract for their first 2
196

years of existence? Explain in terms of both the current
league owners’ estimate of the net present value of the
expansion teams to the new owners and in terms of the
other expected league impacts due to expansion.
6. We saw in Chapter 2 that sports team owners use price
discrimination to raise their profits above the level that
could be made by charging fans a single price. This is
especially true of so-called quality game pricing where
some teams now charge more when a popular opponent
comes to town or late in the season when play-offs are
being decided. Critics have described these price
discrimination mechanisms as “gravy” on top of the usual
profits that could be made. The critics suggest that
owners could do quite nicely without them and garner
goodwill from fans. Use the logic of imputed expansion
fees to show that profits from price discrimination are not
just extra, over and above the profit that could be made
without price discrimination.
7. Refer to the Learning Highlight: Keeping the Oilers in
Edmonton.
a. Why did the Canadian government officials mentioned
in the highlight impose the rule that local owners had
a chance to buy the team at what would surely be an
artificially low amount relative to prices that typically
come out of the market for NHL franchises? What was
Peter Pocklington’s response when he put the team up
for sale in 1997?
b. How did the accounting lessons from Chapter 2 figure
into EIG’s success at generating the required $70
million payment?
c. What was the source of a potential windfall gain for
EIG if it did sell the Oilers after 2004? Would you have
sold the Oilers in 2004 if you were EIG? Why or why
not?
8. What led the NFL and the AFL (version IV) both to put
teams in Dallas, Texas, in 1960? Be careful to distinguish
the motivation of the NFL from that of the AFL.
Ultimately, the AFL Texans left to become the Kansas City
Chiefs. Does that necessarily mean that the NFL won in
the battle for territorial rights against the AFL in this
case? Why or why not?
9. Al Davis, now owner of the Oakland Raiders, was
commissioner of the AFL (version IV) at the time of the
merger between the NFL and the AFL in 1969. He urged
the AFL owners not to merge, arguing that in a very short
time the AFL would become the dominant professional
football league. Using a version of Figure 5.1, what was
Davis thinking in terms of the rival league talent
dilemma?
10. Recount the death of the PCL. How is it related to the end
of AABLs in the Learning Highlight: African American
Baseball Leagues as an MLB Rival? And how is the death
of both leagues related to game theory and the single
197

dominant league given in the text? Explain using Figure
5.1.
198

SECTION 12
References
Banerjee, A., Swinnen, J., and Weersink, A. “Skating on Thin
Ice: Rul Changes and Team Strategies in the NHL,” Canadian
Journal of Economics 40 (2007): 493–514.
Fort, Rodney, and Joel Maxcy. “The Demise of African-
American Baseball Leagues: A Rival League Explanation,”
Journal of Sports Economics 2 (2001): 35–49.
Fort, Rodney, and James Quirk. “Cross-Subsidization,
Incentives, and Outcomes in Professional Team Sports
Leagues,” Journal of Economic Literature 23 (1995): 1265–
1299.
Fort, Rodney, and James Quirk. “Introducing a Competitive
Economic Environment into Professional Sports,” in
Advances in the Economics of Sports, vol. 2, ed. Wallace
Hendricks. Greenwich, CT: JAI Press, 1997.
Neale, Walter C. “The Peculiar Economics of Professional
Sports,” Quarterly Journal of Economics 78 (1964): 1–14.
Quirk, James, and Rodney D. Fort. Pay Dirt: The Business of
Pro Team Sports. Princeton, NJ: Princeton University Press,
1992.
Quirk, James, and Rodney D. Fort. Hardball: The Abuse of
Power in Pro Team Sports. Princeton, NJ: Princeton
University Press, 1999.
Scully, Gerald. The Business of Major League Baseball.
Chicago, IL: University of Chicago Press, 1989.
199

SECTION 13
Suggestions for Further
Reading
Surely, I can come up with some.
200

CHAPTER 6
Sports Market
Outcomes, Part II:
Leagues and
Competitive Balance
It’s too late merely to view with alarm the mess into
which baseball has managed to get itself. The time
has come to sound the alarm before the national
pastime, as it still calls itself, collapses under the
weight of its own archaic rules, mismanagement,
lack of leadership, greed, and plain stupidity … The
symptoms of near disaster are plain enough: The
Yankees make an almost annual farce of the
American League pennant race—a most unhealthy
condition. Interest in big-league baseball is on the
downgrade.
—Bill Veeck
As told to Murray Robinson, “I Know Who’s Killing Baseball,”
article 1 in a series for the Hearst newspapers.
Hearings Before the Subcommittee on Antitrust and
Monopoly, Committee on the Judiciary, U.S. Senate,
85th Congress, 2nd Session, July 1958, p. 717.
CHAPTER OBJECTIVES
After reading this chapter, you should be able to:
• Use the two-team diagram of a pro sports league
to determine equilibrium winning percents and
payrolls.
• Understand that the cause of winning imbalance
and payroll imbalance is the underlying
imbalance in revenue potential across the league.
• Repeat the lessons of the historical data on
competitive imbalance in pro sports leagues.
• Understand whether the various mechanisms
touted to enhance competitive balance—such as
revenue sharing, the draft, luxury taxes, and
salary caps—actually will accomplish that task.
• See why it is imperative that leagues be able to
discipline member teams to follow league
policies, especially those affecting competitive
imbalance.

SECTION 1
Introduction
Perhaps you were surprised to see that this statement was
made in 1958. It sounds almost exactly like the lament you
could find in the press or popular accounts of the problems in
pro sports, such as Bob Costas’s impassioned assessment of
MLB in his book Fair Ball. In this chapter, we address the
primary internal issue confronting owners in a pro sports
league, namely, competitive imbalance. We will see how, at
times, individual owner actions can be detrimental to overall
league welfare, spurring pro leagues to develop mechanisms
ostensibly designed to enhance competitive balance. But, as
with most things, we will also see that things are not always as
they are touted to be. Some of these mechanisms actually do
not affect competitive balance at all. Instead, they reallocate
sports revenues from some owners to others and from players
to owners.
202

SECTION 2
Competitive Imbalance
So far, it looks like the world is a sports league’s oyster.
Exclusive territories are carefully maintained to the dramatic
economic benefit of the current group of owners. One of the
downsides to territorial exclusivity is the possibility of rival
leagues. However, the maintenance of exclusive territories
also has another potential downside, called competitive
imbalance. A league suffers from competitive imbalance if,
year after year, there are very strong and very weak teams,
and the mix of these strong and weak teams doesn’t change.
The danger is that competitive imbalance, if bad enough, can
reduce fan interest in the league product and, eventually,
individual team profits. This potential danger follows from
Rottenberg’s uncertainty-of-outcome hypothesis introduced
in Chapter 2. Under that hypothesis, fans prefer closer games
to blowouts and a changing mix of teams in postseason play.
If the hypothesis holds for fans, competitive imbalance would
drive fans of losing teams that nearly never appear in
postseason play away from watching home games. Those
teams would become economically unviable, owners would
get out of the game, and the league would not be able to find
replacement owners. The spillover losses to the remaining
owners would come in the form of lost shared revenue and a
reduction in the total fan base that would watch postseason
play. In short, the entire league would suffer economically.
This type of imbalance is easy to spot, especially in baseball.
The year 1993 was a particularly interesting year in this
regard. First, it was the last time that the 2008 World Series
Champions appeared in the World Series—but the Phillies
were beaten then by the Blue Jays. Interestingly, that was also
the last time the Blue Jays appeared in the postseason! Their
misfortune is to be in the same division as the Yankees, and as
such, 2008 marked the end of the 14-year string of Yankee
play-off appearances that began after the Blue Jays won the
1993 World Series. And this is precisely the type of problem
we are discussing here. What problems do strings like the
Phillies’ and the Jays’ futility (actually, the Phillies also made
the 2007 play-offs) and Yankees’ success cause to fans and the
league? The answer for MLB was captured cogently as follows
by the Blue Ribbon Panel Report (Levin, Mitchell, Volcker,
and Will, 2000):
While most fans do not demand or expect that their team
will reach postseason play each year, some have ample
reason to believe that the club they root for will remain
chronically uncompetitive. Because revenue Quartile III
and IV clubs have not been winners and have barely been
participants in the postseason for the past 5 years, many
fans have come to believe that it is unlikely these clubs
will reverse that fate in the next few years. The presence
in the game of clubs, perhaps a majority, that are
chronically uncompetitive, alongside clubs that routinely
203

dominate the postseason, undermines the public interest
and confidence in the sport. (p. 42)
Lest we think it’s just about baseball, Mike Brown, owner of
the NFL Bengals, and Ralph Wilson, owner of the NFL Bills,
are recent vocal critics of growing disparity in the revenues in
their league, too.
But whether competitive imbalance is a problem begs an
examination of competitive imbalance in the first place. For
the rest of the chapter, let’s examine three things. What
causes competitive imbalance in the first place? What do the
data on competitive imbalance tell us (has balance worsened
over time in pro sports)? Finally, how have pro sports leagues
tried to handle competitive imbalance, and what other
techniques could be used to enhance league balance?
THE THEORY BEHIND COMPETITIVE IMBALANCE
Revenue disparity drives competitive imbalance in all leagues.
Revenue disparity is guaranteed once owners are granted
exclusive geographic franchises because different franchise
areas have different drawing power. This generates different
revenues and abilities to buy the talent required to win.
Subject to the usual caveats concerning uncertainty, higher
revenues yield better teams on average. Exclusive territories
do provide market power positions for owners, but they lead
to competitive imbalance.
GRAPHICAL ANALYSIS OF LEAGUE WINNING
PERCENTS
It is easy to show the winning percent outcome in a league
with a simplified graphical analysis of a two-team league. In
Chapter 4, remember that this is the owner’s long run choice
of team quality on the field. Figure 6.1 shows the marginal
revenue functions of two owners. Following our simplification
from Chapter 4 that owners produce winning for fans, and the
Rottenberg idea that fans care about the relative fortunes of
their team, the x-axis measures winning percent from 0 to 1. If
we measure talent in units that it takes to produce one more
unit of winning percent, then the x-axis also measures the
amount of talent hired by each owner. We’ll assume that
talent is measured in these units. Finally, all models have
limitations. One of the most important limitations of the
model in Figure 6.1 is that talent supply is fixed—one team
increases its winning percent only by buying talent from the
other team. Thus, this model “works” for MLB, the NBA, the
NFL, and the NHL since, once the season starts, all of the
talent that can be had is now in the league. The model would
be much less useful in world leagues where, even when the
season is underway, teams in one country can buy talent from
teams in another.
Both a larger-revenue market owner and a smaller-revenue
market owner are shown in Figure 6.1. Remember, the only
way it makes sense to talk about larger-market and smaller-
market owners is in terms of the revenues that they can
generate in their respective locations. MRL(W) is the marginal
204

revenue from winning percent for the larger-revenue market
owner and MRS(W) is the marginal revenue from winning
percent for the smaller-revenue market owner. The idea of
smaller- and larger-revenue markets is purposely exaggerated
in Figure 6.1 to make the point clearly. Every level of winning
is more valuable to the larger-revenue market owner than it is
to the smaller-market owner; that is, for all levels of winning
percent, W. For example, at W = 0.250 (the teams were to win
one-fourth of their games) MRL(0.250) > MRS(0.250), and
the same would be true for every other level of winning
percent.
Figure 6.2 is a useful device in this simplified, two-team
world. It shows exactly the same information that was
contained in Figure 6.1, but it makes it easy to present the
equilibrium winning percents for the two owners. Winning for
the larger-market team is measured from zero to one on the x-
axis from left to right. Winning for the smaller-market team is
measured from zero to one on the x-axis from right to left. The
two marginal revenue functions, MRL(W) and MRS(W), show
precisely the same relationship between winning percent and
marginal revenue as before; they are just relocated with
respect to their new graphical origins. The origin for the
smaller-revenue market team is in the lower-right corner, and
the origin for the larger-revenue market team is in the lower-
left corner.
We will assume that each of the owners in our two-team
league must compete for talent. We will also need to
remember a useful fact: The sum of the winning percents in
any league must equal half the number of teams. The average
winning percent in any league is 0.500. The formula for the
average is:

where n is the number of teams, and Wi is the winning
percent of team i. Multiplying both sides by the number of
teams, n, gives the sum of winning percents across teams
equal to 0.500 × n; that is, the sum of the winning percents is
equal to half the number of teams. Thus, in this two-team
world, the sum of winning percents must be equal to one. This
means that WL + WS = 1 everywhere along the x-axis, or that
WL = 1 − WS. With these preliminaries out of the way, we’re
ready to examine winning percent outcomes.
EQUILIBRIUM WINNING PERCENTS
Suppose that talent has been chosen so that the winning
percent for the larger-market team is W1L. Because the sum of
winning percents must equal one, we know that W1L = 1 − W1S,
as shown in Figure 6.2. However, this cannot be an
equilibrium winning percent outcome. At this outcome, we
know that MR1L > MR1S. This means that the larger-market
owner would make more money than the smaller-market
owner on the last unit of talent hired by the smaller-market
owner. The most that the smaller-market owner will pay is
MR1S, but the larger-market owner can afford to pay up to
MR1L > MR1S. Thus, the larger-market owner can make an
205

offer to the player supplying that last unit of talent that is
greater than the amount the smaller-market owner will pay.
That unit of talent would move to the larger-market team, and
the outcome would be an increase in the winning percent of
the larger-market team. This would be true as long as the
teams were to the left of the intersection of the marginal
revenue functions.
The same logic, but in the opposite direction, would hold if
talent is chosen so that MRS(W) > MRL(W). In that case, the
smaller-market owner would bid talent away from the larger-
market owner. This may seem odd at first since every unit of
talent is worth more to the larger-market owner. But
remember we’re comparing between the two at the margin.
Once the larger-revenue owner has a high enough level of
talent, the very next unit under consideration could be worth
more to the smaller-revenue owner. This would be true for all
talent combinations to the right of the intersection of the
marginal revenue functions.
There is only one place where the two owners no longer try to
bid talent away from each other. That allocation will be the
winning percent equilibrium. In equilibrium, it must be the
case that MRS(W) = MRL(W), that is, at the intersection of the
two marginal revenue functions. This is shown in Figure 6.3
at MR*S(W) = MR*L(W) so that W*L = 1 − W*S.
IMPORTANT FEATURES OF THE COMPETITIVE
EQUILIBRIUM
There are three important features in the equilibrium winning
percent outcome shown in Figure 6.3:
1. Marginal revenues are equal across teams, that is,
MR*S(W) = MR*L(W).
2. Revenue imbalance causes competitive imbalance.
3. Revenue imbalance causes payroll imbalance.
The first feature describes equilibrium and provides an
important observation. While the total amount of talent hired
by the larger-revenue owner is greater, the marginal value to
both owners of another unit of talent is equal. The important
observation is that there is a limit on talent purchases by the
larger-revenue market owner. The larger-revenue market
owner will not buy all the great talent because, eventually, the
marginal unit of talent is more valuable to the smaller-
revenue owner.
This first feature of equilibrium winning percent also tells us
the price of talent. Both owners will confront the same price
of talent, and that price equals marginal revenue for both
teams. Let P denote the price of one unit of talent. Remember,
we measure talent in terms of units that generate one more
unit of winning percent So, in equilibrium, either owner
would have to pay P = MR*S(W) = MR*L(W) in order to obtain
another unit of talent, as depicted in Figure 6.3.
The second feature of the competitive talent equilibrium is
that the larger-revenue market team has a higher winning
206

percent than the smaller-revenue market team. This is also
easy to see in Figure 6.3 because equilibrium is to the right of
0.500 for the larger-revenue market team. This only makes
sense; every single unit of talent hired produced winning that
was more valuable to the larger-market owner. The larger-
revenue owner therefore hires more talent and wins more.
Don’t lose sight of this important distinction. A common
statement about imbalance is that larger-market owners buy
all of the great talent because they have such high revenues.
This actually clouds the true line of causality. Profit-
maximizing owners evaluate the revenue potential of their
market. If that market will pay the most for a high-quality
team, the owner buys the talent to make it so. Only then can
the owner collect the high level of revenues from fans at the
gate and from TV rights fees. As long as there are markets that
generate different amounts of revenue for different quality
teams, there will be owners who choose talent so as to
generate competitive imbalance.
The final important feature of the equilibrium winning
percent outcome can be seen as follows. It is easy to see how
much each owner pays, in total, for the winning percent that
they choose using Figure 6.3. The bill for the larger-market
owner is the left-hand shaded rectangle, calculated as P ×
W*L; that is, the price of each unit of winning multiplied by
the number of units of winning. This amount also represents
the payment to talent because we have measured talent in
units that produce each unit of winning percent. Therefore,
the shaded rectangle also represents the total payment to
talent hired by the larger-revenue market owner. Similarly,
the right-hand shaded rectangle is the smaller-market owner’s
talent bill, calculated as P × W*S. The conclusion is that the
larger-revenue market owner spends more on talent than does
the smaller-revenue market owner. So, along with competitive
imbalance, there will be payroll imbalance as long as there are
larger- and smaller-revenue market areas. Again, don’t lose
sight of this line of causality. The existence of market areas
that generate different amounts of revenue for different
quality teams drives both competitive imbalance and payroll
imbalance.
Note that this equilibrium outcome suggests that the larger
the revenue imbalance, the larger the competitive imbalance.
Suppose that MRL(W) shifts to the right, an increase in
revenue imbalance. Perhaps incomes in the larger-revenue
market increase or population rises. You should be able to
convince yourself that the following will occur. First, the price
of a unit of talent will increase. Second, talent spending by the
larger-revenue market owner will increase; they buy more
talent at the new higher price. Third, because the price of
talent is rising in the inelastic portion of the smaller-revenue
owner’s marginal revenue function, spending by the smaller-
revenue market owner also rises. Finally, the winning percent
of the larger-revenue market team increases.
You have just seen the usefulness of the two-team league
model—characterizing equilibrium and making predictions
about what happens in this equilibrium model when
something changes. Here is a real-world example. The NHL’s
207

Colorado Rockies left Denver and moved to New Jersey to
become the Devils in 1981. This was just three years after both
the Denver Nuggets and the New Jersey Nets joined the NBA
(following the formal merger between the NBA and the ABA
described in the last chapter). Note that at the time this left
the Denver Nuggets in sole possession of the Denver pro
sports market (the Rockies baseball team had not yet been
created). To this day, by the way, the Nuggets are in the
bottom half of the NBA revenue hierarchy. What does our
two-team league model suggest about the impacts of such a
move by the NHL Rockies?
The best method to use has four steps: Start from the
equilibrium before the move, model the impacts of the
change, find the new equilibrium, and finally draw
conclusions. So let’s start with Figure 6.3 as our equilibrium
before the move and think of the Nuggets as the smaller-
revenue team. Step 1: We have all of the equilibrium
characteristics just listed—the larger-revenue team has a
higher winning percent and a higher payroll than the Nuggets
in order to actually collect along its greater marginal revenue
function. Figure 6.4 demonstrates the change that occurs
when the Rockies left for New Jersey. Step 2: The Nuggets
marginal revenue function increases from MRS to MR′S; some
fans of the Rockies probably will turn into fans of the Nuggets
and bring their pocketbooks with them. The new equilibrium
is also identified in Figure 6.4. Step 3: W′S > W*S, W′L < W*L, and the price of talent rises from P to P′. And that just leaves us to draw conclusions. First, think about the Nuggets. You should be able to see that the Nuggets owner was clearly better off after the Rockies left. The team improved in quality (W′S > W*S) and, despite having a higher
payroll (P′W′S > PW*S), their net value of winning increased!
This last situation is so because the net value of winning after
the Rockies left is ?abP′, which is greater than the net value of
winning before, namely, ?cdP. This is true because distance ac
is larger than distance P′P and distance ed is positive. Turning
to the larger-revenue owner, quality decreased (W′L < W*L) and payroll increased (point b is on the inelastic portion of MRL so an increase in price increases spending on talent). This owner must be worse off since they are paying more, winning less, and the value of winning has not changed. And it’s true since the net value of winning has decreased by area P′bdP. Finally, players are clearly better off—they all are still hired, and the price of talent has risen! While a simple model like Figure 6.3 helps us understand equilibrium, it is also insightful about changes that occur for the league. 208 SECTION 3 The Data on Competitive Balance The theory informs us that there will be competitive imbalance and payroll imbalance as long as owners have unequal earning power and that competitive imbalance worsens, the more imbalanced revenues become. Before we turn to the remedies that have been suggested for this competitive imbalance, let’s look at the data of actual competitive balance outcomes in pro sports leagues. We touched on revenue imbalance in Chapters 2 and 3. Here, we extend the idea by looking at imbalances in revenues, payrolls, winning percents, and championships. But first we must specify just how we’re going to measure competitive imbalance. COMPETITIVE IMBALANCE MEASUREMENTS We will use three measures of competitive imbalance in pro sports leagues. A brief overview of these measurements is given here (your instructor may point you to more in-depth treatments as needed). We will use the Gini coefficient to measure revenue and payroll inequality. You may have seen the Gini coefficient used to describe income inequality in your basic economics course. The Gini coefficient lies between zero and one; the larger the Gini coefficient, the greater the inequality in revenues or payrolls between owners in a league. For example, if the Gini coefficient were equal to one, a single owner would have literally all of the revenue in a league. In practice, this won’t happen, but such an outcome is the theoretical limit on revenue inequality. The Gini coefficient proves problematic in measuring the inequality in winning (see Utt and Fort, 2002, for a complete discussion). If there are more than two teams in a league, it is theoretically impossible for one owner to have all of the wins due to the zero-sum nature of league play. For example, in a three-team league where each team plays the other only once, there are three total games. But the most that a single team can win is two games. So for winning percent, we turn to another measure. The standard deviation of winning percent is a statistical measure of the spread of winning percent around its league average value of 0.500. The smaller the standard deviation, the lower the spread and the closer should be the winning outcomes. If we check the behavior of the standard deviation of winning percent over time, any increase represents growing competitive imbalance, whereas any decrease means the league has become more balanced over time. There is one added complication. The standard deviation is sensitive to season length. For example, think of a two-team league playing a 10-game schedule. One team wins nine games, and the other team only 209 wins one. The standard deviation of winning percent is 0.566. Now suppose the two teams play a 12-game schedule. The same team wins only one game and the other team wins 11. Now the standard deviation is 0.589 even though it is reasonable to say the imbalance is the same. The following invention controls this characteristic of the standard deviation and provides a tidy comparison of winning percents over time. The invention is the standard deviation of winning percent in a theoretically perfectly balanced league. One definition of “perfectly balanced” is that the probability any team in the league beats another literally is 0.5 (the proverbial “on any given day any team can beat any other” actually is true here). It ends up (an easy presentation is in Fort and Quirk, 1995) that the standard deviation of winning percent in this version of a perfectly balanced league is , where m is the length of the season measured in games. If we divide the actual standard deviation by the standard deviation for a perfectly balanced league, we have controlled for any changes in season length since the denominator contains . This standard deviation ratio of actual to perfectly balanced standard deviations is greater than or equal to one. The closer the measure is to one, the more balanced the league is. We will use this ratio to compare winning percents over time in pro sports leagues. Our final measure of competitive imbalance concerns the postseason. Suppose a division in a league has 15 teams. If play-off appearances were equally likely, we’d expect to see each team win the division once every 15 years. Naturally, then, a measure of postseason balance is the average number of years between championships for each of the teams in the league. This measure has the added advantage of controlling for the fact that some teams that were in the league longer had a greater number of chances to get to the postseason. We could calculate the Gini coefficient (for an example, see Quirk and Fort, 1992), but this one is easier and insightful enough for our purposes. And off we go. REVENUE IMBALANCE Table 6.1 shows Gini coefficients for revenue imbalance in the four pro sports leagues. Because league choices on team location help determine revenues, the data are shown at the league level. The revenue data used to calculate these coefficients came from congressional hearings in the 1950s, from Professor Gerald Scully for the some of the 1980s in MLB, and more recently from reports in Financial World and Forbes magazines. We need to be clear about this from the outset: Different analysts have different feelings about the veracity of the data behind this analysis. For example, each year, team owners dispute the Forbes reports. Also note that there weren’t enough data to yield averages for some decades. However, these data are all we have, so let’s see what they suggest about revenue inequality. First, we’ll examine each league individually. In years where we have data across all four leagues, MLB typically battles the NHL and the NBA as the league with the worst revenue imbalance. However, except for a backslide in 210 the 1950s, revenue balance has steadily improved in MLB. A look at the past 20 years shows revenue imbalance on a roller coaster ride, rising after the strike of 1994–1995 and falling again into the 2000s. In summary, as measured by the Gini coefficient, revenues have become more balanced by about 37 percent in MLB since the 1950s. In the NBA, revenue balance has also improved steadily since the 1950s. While it suffered the worst revenue imbalance of all leagues in the 1950s, only the NFL now exhibits more revenue balance than basketball. Indeed, NBA revenues have become more balanced by about 40 percent since the 1950s. The NFL is always the most balanced league in terms of revenues. This is no surprise given that the team owners in the league share nearly all of its revenues. The league enjoyed a 55 percent improvement in revenue balance since the 1950s, and balance has been steadily improving through the 2000s. Finally, we come to hockey. The NHL has shown steadily improved revenue balance over the last 20 years in Table 6.1. Revenue imbalance increased in the 1990s, relative to the long-ago 1950s, but has improved by 21 percent over the last two decades. However, the NHL is now tied with MLB as the league with the most revenue imbalance among all pro sports leagues. Overall, suppose we think of the four major leagues as a “sports economy” of sorts. On average, across the decades where there are data for all four sports, revenue balance has improved without fail since the 1950s. Indeed, as measured by the Gini coefficient, there has been a 39 percent improvement in revenue balance in this sports economy. There are two general reasons why this may have occurred. On the one hand, leagues have enacted policies designed to enhance revenue balance. But this has always been the case in the NFL, and that league still shows steady improvement. Other underlying factors have to do with the source of revenues in the first place, namely, fan willingness to pay. As population centers grow larger and perhaps richer, one would expect the willingness to pay by fans to also begin to equate between team locations. But only future research will answer this question. PAYROLL IMBALANCE Gini coefficients for payroll imbalance, again at the leaguewide level where these amounts are determined by competition in the market for talent, are shown in Table 6.2. These data also are from congressional hearings, popular media accounts, and open to the same type of disagreement noted earlier for revenues. But they are what we have, and here is what the data appear to tell us about payroll imbalance. The table shows both opening day (OD) payrolls and end-of-season (EOS) payrolls. The issue of which one is more relevant depends on the question addressed. The discussion hereafter uses EOS payroll imbalance since that reflects the eventual unfolding of the long-run quality plans of owners (as discussed in Chapter 4). In addition, EOS payrolls have received less attention generally speaking. 211 MLB has always had the most payroll imbalance of all leagues. Payroll imbalance has worsened about 42 percent in baseball since the 1950s. Only the NHL shows more growth in payroll imbalance than baseball. Interestingly, the imbalance has been worse in the American League (AL) than in the National League (NL) in every tabled decade. Through the 1980s, the NBA was second to MLB in payroll imbalance. However, it has been overtaken by the NHL for the last 20 years. The level of payroll balance in basketball remained unchanged to the end of the 1980s. Despite a bit of an improvement in the 1990s, it has suffered a relapse in the 2000s, and despite this backslide, the NBA is the only league to show an improvement overall; payroll balance has improved about 5 percent. Football has always been the most balanced league in terms of payrolls. Overall, payroll imbalance worsened in the NFL, but over the last three decades, it has improved steadily (9 percent). Payroll balance clearly worsened, comparing the 1980s to the 1950s. This is curious since the league revenue sharing was the same over this extended period. As expected, payroll balance improved with the implementation of the league payroll cap in the mid-1990s, but this raises another curiosity. Payroll balance has continued to improve through to the present even though the league’s sharing and cap system has remained unchanged over these decades. Finally, we turn to hockey. First, as already noted, the NHL overtook the NBA as the second-worse payroll imbalance league in the last two decades. Second, the NHL has shown the most dramatic decline in payroll balance overall (50 percent since the 1950s). However, things may be looking up on this front since payroll balance has improved 58 percent in the postlockout period relative to the level of imbalance prior to the lockout. The league payroll cap may be responsible, but this will be a more statistically comfortable judgment only after more time passes and more data are generated. Taking our sports economy viewpoint, payroll imbalance has been a roller-coaster ride in pro sports overall, with the level in the 2000s about what it was in the 1950s. As measured by Gini coefficients, the bright spot is that payroll imbalance has improved over the last two decades by 7 percent. We are left with the conclusion that there must be a significant and growing gap in the value of talent between smaller- and larger-revenue markets in all pro sports leagues with the possible exception of the NBA. WINNING PERCENT IMBALANCE Recall that our measure of winning percent imbalance is the ratio of the actual standard deviation to the standard deviation of a perfectly balanced version of the league. Here, for MLB, we must be careful to calculate this measure only over the games where it is created. Because there were no games between the AL and NL until very recently, and few of them even now with “interleague play,” we will calculate our ratio measure separately for the AL and NL. For the other three pro sports leagues, play occurs between nearly all teams 212 in each league, so the measure is calculated leaguewide. In addition, the NHL uses a point system to determine final standings and entry into the play-off. Here, we convert to winning percent for comparison purposes (our standard deviation ratio measure can also be calculated for a point system, but that is also more complicated than we need for our purposes here). Finally, while all results are shown in Table 6.3, the discussion that follows focuses on the decades where data exist for all four leagues, that is, from the 1940s on. Remember that there need be no apology for the source of these data—anybody can find game outcomes on a daily basis. Turning first to MLB, only the NFL has had more balanced play over the season than baseball. In MLB for both the AL and NL, winning percent imbalance has fallen significantly (10 percent in the AL and 33 percent in the NL). But look carefully at Table 6.3 and you’ll notice that winning percent imbalance is creeping upward in the AL (19 percent comparing the last two decades). Indeed, in the AL, this measure of imbalance has climbed back to its pre-1970s level, greater than 2.0. Although the AL has more often than not been more imbalanced than the NL, it is 22 percent more so over the last decade. Since it began play in 1947, winning percent imbalance in the NBA has remained virtually unchanged compared to the present day. The NBA has been the most unbalanced league in terms of winning percent in every decade except the 1950s, when the Yankees reigned supreme in MLB’s AL, and in the 1970s when it placed second to the NHL. The 1990s showed the worst balance for the NBA, but the most recent decade looks more like typical historical levels for basketball. Winning percent imbalance in the NFL is completely consistent with our earlier observations about revenue and payroll imbalance for the league. Except for the 1940s, the NFL has had the most balanced winning percents among teams of all pro sports leagues. On top of that, winning percent imbalance has declined 16 percent since the 1940s. Notice that no substantial gains were associated with the institution of the salary cap in 1994. The gains in balance occurred after extensive revenue sharing was already in force. This suggests that the revenue base of NFL markets has been gradually equalizing over time. Most recently, imbalance over the season of play is the worst it has been in the NFL since the 1970s. Our last look at the NHL shows it to be quite unbalanced as measured by winning percents up to the 1970s. Since then, there has been improvement in every decade. Indeed, measured overall from the 1940s, balance over the season in hockey improved 7 percent. However, as has been true since the 1980s, only the NBA is less balanced over its regular season of play than the NHL. We close out our investigation with a quick look at winning percent imbalance across the sports economy. Since the 1940s, when all four leagues first existed simultaneously, winning percent imbalance has declined 13 percent. However, balance took a roller-coaster ride until the end of the 1970s. From the 1980s on, the standard deviation ratio has settled just below 2.0. The forces 213 determining the distribution of team quality and the resulting outcomes over the regular season have not changed much in any given decade since the 1940s, and not at all over the last three decades. CHAMPIONSHIP IMBALANCE Our investigation of imbalance thus far has concerned outcomes during a given season—revenues are tallied at the end of the season, payrolls are set prior to the start of a season (and their level changes to the end of the season), and winning percents are calculated after the last regular season game is played. But championship imbalance, or postseason imbalance, also matters to fans. It’s not just whether play during the season was competitively balanced; it also matters which teams make it into the play-offs. We have chosen a simple way to analyze the data on championships—years between titles. Table 6.4 shows years per title for AL and NL champions that meet in the World Series in MLB, and conference champions for the other three pro sports leagues (conference champions meet in the NBA Finals, the NFL Super Bowl, and play for the Stanley Cup in the NHL). The entries are restricted to relatively more successful teams with less than 10 years between championships. The data strongly indicate that titles are unequally distributed and, typically, go in favor of larger- revenue market teams. Teams in New York and Los Angeles have the lowest years per title in all leagues except hockey, and the larger-revenue market Canadian cities of Ottawa and Montreal dominate that league. It’s always a judgment call as to which teams fall into the larger-revenue and smaller-revenue markets, but here goes. Only remotely resembling smaller-revenue markets are Florida and Oakland in MLB; Syracuse, San Antonio, and Rochester in the NBA; Oakland and Buffalo in the NFL; and Edmonton, Hamilton, and Carolina in the NHL. Many on this list of smaller-revenue successes are also simply historical artifacts—Syracuse and Rochester no longer exist (NBA), Oakland I could be termed a different team than the current incarnation of the Raiders (NFL), and Hamilton also no longer exists (NHL). The teams that have larger revenues can change over time, as with the Atlanta Braves and Seattle Mariners in MLB. But larger-revenue market teams simply always have dominated the play-offs in pro sports leagues. BACK TO THE THEORY Our simple two-team theory suggests that revenues should be highly positively related to both payrolls and winning percent. Further, if revenue imbalance increases, so should payroll imbalance and competitive imbalance. We can appeal to simple general statistics to get a basic understanding of the power of these relationships in actual practice. A simple correlation statistic shows how two variables are related. Correlation coefficients are between zero and one, and the closer the coefficient is to one, the stronger the 214 relationship between the two variables under consideration. A negative correlation shows that variables move apart, whereas a positive correlation shows that they move together. We’ll examine just MLB here and leave an examination of the rest of the leagues to a project your instructor can assign. In MLB, shown in Table 6.5, the correlation between revenues and payrolls always has been quite strong, and this is especially so for the current decade of the 2000s. This strong, positive result supports the theory. Owners evaluate the revenue potential in their market and hire talent accordingly. The correlation between revenues and winning percent also follows the theory. The correlation is always positive and was quite strong through the 1950s. However, the correlation is much weaker over the past two decades. The correlation between payroll and winning follows exactly the same pattern as the correlation between revenue and winning —positive as the theory suggests but much more weakly so in recent decades. One summary observation from Table 6.5 is that it appears to be much easier to turn eventual revenue collections into payroll than it is to turn that money into a winning outcome in the 1990s and 2000s. Just why this is so recently, but not in the earlier decades through the 1950s is an interesting question. One explanation is that, despite increased payroll imbalance in MLB over time (Table 6.2), competitive balance over the season has improved for the most part (Table 6.3). Perhaps this is due to improvements in the overall level of player talent combined with improvements in on-field management. Another possibility is that the steps MLB has taken to reduce the impact of revenue disparity may be having some impact. But more discussion about that in the rest of the chapter. Finally, the theory also receives some support since competitive imbalance increases as revenue imbalance increases. From the data in Tables 6.1 and 6.3, the correlation coefficients for the 1990s were 0.568 in the AL and –0.103 in the NL. The former is reasonably strong and adheres to the theory. In the NL, however, the result is much closer to zero. But a look inside the decade reveals something quite interesting. In the NL, while the overall correlation is essentially zero, it was 0.936 in the years after the strike (1996–1999). For the 2000s, the correlations are 0.166 and 0.493 in the AL and NL, respectively, and at least both are positive as the theory suggests. Even with such a simple statistic, it appears the simple two- team league theory has strong explanatory power. This is nothing new, however. Scully (1989) states the conclusion of his analysis of the relationship between team expenditures, revenues, and winning for the early 1980s: “[W]hat can be concluded is that both club revenues and costs are positively related to team quality and that they tend to increase at about the same rate” (p. 126). The lesson is clear. Despite arguments in the popular press and by owners that the problems plaguing sports leagues are due to payroll imbalance, it is actually the unequal revenue potential in different geographic locations that drives both payroll imbalance and competitive 215 imbalance. As we saw in the last chapter, this unequal revenue potential is in part due to the careful maintenance of territorial exclusivity by owners acting as leagues. One crucial observation remains about the data on competitive balance. It is limited to a strict “tracking” exercise of how it has behaved over time. But this is not all that matters about competitive balance. Ultimately, how fans respond to competitive imbalance will determine whether it is a problem for leagues or not; it is the role of outcome uncertainty in demand, as in Chapter 2, that decides the issue. Our general verdict is that balance over the season has improved in all sports except the NBA, where it has probably remained about the same, although this measure has been pretty stagnant since the 1980s when averaged across all four sports. Further, across all four major leagues, it would be difficult to make the case that the postseason has become any less balanced over time; it has always been so. But if fans would show up in larger numbers, or spend more on direct subscriptions and league merchandise, if balance were improved, then leagues still would have to deal with the issue. Professor Stefan Szymanski (2003), a preeminent sports economist in England, has suggested that there is little evidence of any fan responsiveness to competitive balance. I take a softer line (Fort, 2006)—while there has been a fair amount of investigation into fan response to competitive imbalance, we are far from a definitive answer. In any event, what is clearly true is that leagues often cite competitive balance problems as the reason for particular interventions in the business of sports. So let’s turn to a discussion of these sports business devices. 216 SECTION 4 Remedies for Competitive Imbalance: Revenue Sharing Feelings are strongly mixed among analysts about whether the preceding evidence shows that there is a competitive balance “problem” in any pro sports league because research on just how imbalanced leagues ought to be is in its infancy. The issue is further complicated from what we learned in Chapter 2. Even if balance had improved over the decades, fans could be simply more dissatisfied than ever with the current level of imbalance they see. In addition, you know from reading the last chapter that competitive imbalance is too large from the perspective of general sports fans (not just the ones who have a team now). This follows from the fact that owners, acting together through leagues, artificially restrict the number of teams fans get to enjoy. Regardless of the current measurement of competitive imbalance, there always is the market power problem to consider. In any event, let’s have a look at the more familiar mechanisms pro sports leagues have implemented to ostensibly reduce competitive imbalance— national TV revenue sharing, local revenue sharing (gate, attendance related, and local TV), luxury taxes, the draft, and salary caps. A rather unusual alternative, not tried since the NL contracted from 12 to 8 teams at the end of the 1899 season, is described in the Learning Highlight: Shrinking MLB at the end of this section. NATIONAL TV REVENUE SHARING In Chapter 3, we saw that all leagues sell some portion of their games through a national broadcast contract. Networks (ABC, CBS, ESPN, NBC, TBS, and TNT) get games of broad appeal through these contracts and send them over the air or through cable redistribution. It is easy to see that this type of national revenue sharing can have no impact on competitive balance using one of the most basic of economic insights. If cost or reward is not tied to the actual action an economic agent chooses, then the cost or reward cannot change that choice. If the speed limit were enforced by the honor system, drivers would speed as if there were no limit (some will drive the limit anyway, others not). But rewarding good drivers (as insurance companies do) and punishing speeders (as law enforcement does) changes behavior. The same is true of sharing the national TV contract. All teams get an equal share of the national TV contract in all leagues simply by virtue of being a league member. In this case, because owners’ shares of that contract do not depend on how well their team performs, there is no reason at all to expect national TV contract shares to alter team quality choices by owners. LOCAL REVENUE SHARING 217 Local revenue sharing is touted as a solution to competitive imbalance. Local revenues are generated by attendance at the gate, through attendance-related sales like concessions and parking, and through the sale of local TV contracts (Chapter 3). The NBA is the only league that does not share any local revenue (but it does share some other revenues as noted later in this chapter). As the oldest league, MLB had the first revenue-sharing formula—the visitor simply received a nickel per ticket. That evolved into an equal-proportion gate revenue sharing system, where the home team kept approximately 80 percent of the gate and the visitor received the rest. Since 1996, MLB shares local revenues under a pooled revenue-sharing system. Owners contribute an agreed-upon percentage (between 30 and 40 percent depending on the year) of all local revenues into a pool, and then all owners get back an equal share from the pool. Clearly, this system redistributes revenue from larger-revenue owners to smaller-revenue owners. Larger- revenue owners put in more than they get back, and smaller- revenue owners get back more than they put in. Local sharing by NFL owners always has only included gate revenues. Prior to 2001, they practiced equal-proportion gate- only sharing. The home teams kept 60 percent of the gate. Since 2002, their system also is a pooled revenue-sharing arrangement. The home team still keeps its historical 60 percent. The rest goes into the pool with equal shares to each owner. This NFL system still only includes gate revenues. The NHL has the newest revenue-sharing arrangement involving all local revenues, but the formula is quite complex. Up to the first 25 percent of the redistribution comes from portions of the larger-revenue owners’ shares of the national contract (not really local revenues, but shared from one team owner to others). Up to one-third comes from shares of the top 10 revenue owners’ shares of a league-enforced savings account. The rest comes from specifically stated parts of regular season and play-off revenues of the top 10 largest revenue owners. We can use our simple two-team league model to investigate the impacts of local revenue-sharing devices on competitive balance, the price of talent, and the welfare of owners and players. We’ll do this for equal-proportion gate sharing (the original MLB and NFL approaches) and pooled gate sharing (currently for the NFL). After that, we’ll take a brief look at the data on owner impacts and competitive balance in the NFL. Then, we’ll move on to analyze pooled revenue sharing extended to all local revenues including local TV (currently the practice in MLB), and do a similar look at the data for owners and competitive balance in MLB. The complexity of the NHL approach would require more space than I care to devote to it. GRAPHICAL ANALYSIS OF EQUAL-PROPORTION GATE REVENUE SHARING Under this form of sharing, teams only keep a portion of their home gate, ?, where 0 ≤ ? ≤ 1. If ? = 0, the home team keeps 218 none of its own gate revenue. If ? = 1, the home team shares nothing. For example, in the NFL prior to 2001, ? = 0.6. In MLB prior to 1996, it was ? = 0.8. Figure 6.5 follows the step-by-step assessment of the impact of a change on the two-team equilibrium. While teams have other revenue sources, none of them are affected by equal- proportion gate sharing, so we focus just on gate revenue. As always, begin the assessment from an original equilibrium. Each owner considers the marginal gate value of winning, denoted MRGL(W) and MRGS(W) for the larger- and smaller- revenue owners, respectively, (the “G” makes it clear we are only considering gate). As discussed around Figure 6.3, the definition of equilibrium is MRGL = MRGS. In this equilibrium, W*L = 1 − W*S and W*L > W*S in Figure 6.5. Next,
we need to show how equal-proportion gate revenue sharing
changes the equilibrium. When each team takes this form of
revenue sharing into account, it alters the calculation of the
marginal value of winning. For the larger-revenue owner, that
calculation is ?MRGL − (1 − ?)MRGS (this is a difference,
rather than a sum, due to the limitation in our model where
talent comes to one team only at the expense of the other). For
the smaller-revenue owner, ?MRGS − (1 − ?)MRGL is the
marginal revenue calculation. These calculations of marginal
value of winning under the sharing arrangement give the new
marginal revenue functions depicted in Figure 6.5.
The next step is to determine the new equilibrium result. It is
still the case that equilibrium can only occur when marginal
revenues are equal to each other. A little algebra shows the
following. First, set the new perceptions of marginal revenue
equal to each other as required by the definition of
equilibrium:

If we collect all terms with on the left and all terms with on
the right, the result is as follows:

Cancelling like terms on each side of the equal sign gives
MRGL = MRGS. And here’s the crucial point: MRGL = MRGS is
the equilibrium condition whether there is equal-proportion
sharing or not! This means that the amount of winning
produced by each team is the same with and without equal-
proportion gate revenue sharing, that is, W*L = 1 − W*S. The
only difference in the equal-proportion gate revenue sharing
equilibrium is that the price of talent falls from P to P′. This
result is depicted in Figure 6.5. The marginal gate value of
winning must shift down parallel and equally for both owners
in order to yield the same winning percent combination.
But let’s not forget our last step, assessing how changes in
equilibrium impact owners and players. First, clearly players
are worse off. Talent is distributed between the two teams
exactly as before the sharing arrangement, but the larger-
revenue owner’s payroll falls by PbcP′ and the smaller-
revenue owner’s payroll falls by bPP′c. League payroll falls,
and players earn less.
219

Turning to owners, Figure 6.5 shows that each team owner is
actually better off than before sharing! The net total value of
winning at the gate for the larger-revenue owner was ?abP
prior to sharing and ?dcP′ after. But these two triangles are
exactly the same size. The entire shift downward actually
comes out of pay to players since abcd = PbcP′! Similarly, for
the smaller-revenue owner, the net total value of winning at
the gate also remains the same (?fbP = ?gcP′, and players
cover the shift since fbcg = PbcP′). Since payroll for each
owner falls by exactly the amount of their revenue sharing,
each has the same possible revenue from winning but pays
less to pursue it.
But that isn’t the end of it from the owners’ perspective. Out
of the reduced payrolls that actually fund the revenue sharing,
the smaller-revenue owner ends up with the largest share. The
amount of revenue shared by the larger-revenue owner is the
area of the parallelogram abcd. The amount of revenue shared
by the small–revenue market owner is parallelogram fbcg.
The difference between the two is the net gain to the smaller-
revenue market owner. Since abcd > fbcg, the smaller-revenue
owner receives the larger share. But it is crucial to recognize
that the source of the shared revenues is actually the
reduction in payroll to players.
Thus, even though wealth is transferred to the smaller-
revenue owner (because the shared base of total revenue is
larger from the larger-revenue market owner), the talent
choice is the same for both owners and so is the net total value
of winning. Equal-proportion gate revenue sharing has no
impact at all on competitive balance. It is interesting to note
that our view of pro sports team owners as profit maximizers
is truly important for this observation about equal-proportion
gate revenue sharing. In a league of winning-percent-
maximizing teams, gate revenue sharing can improve
competitive balance, as shown by Kesenne (1996, 1999,
2000). Further, Szymanski (2004) has shown that revenue
sharing also changes balance if one examines open rather
than closed leagues.
ANALYSIS OF POOLED GATE REVENUE SHARING
Under this form of sharing, teams keep some percentage ?,
put the rest into the pool, and all get back an equal share of
the pool. For our two-team league, the larger-revenue owner
contributes (1 − ?)MRGL, the smaller-revenue owner
contributes (1 − ?)MRGS. Each owner receives the equal share
[(1 − ?)/2](MRGL + MRGS), that is, each share equals half the
pool. In this case, we don’t really need a graph in order to gain
our insights, but you could surely produce one similar to
Figure 6.5.
Following our steps, in original equilibrium before sharing,
each starts out as usual: MRGL(W) and MRGS(W) represent
their marginal gate revenues from winning and the initial
equilibrium has MRGL = MRGS with W*L = 1 − W*S and W*L >
W*S. Next, we need to think about how pooled gate revenue
sharing changes the equilibrium. Taking pooled sharing into
account, the marginal value of winning for the larger-revenue
owner becomes as follows:
220

The second parenthetical is a difference, rather than a sum,
again because talent comes to one team only at the expense of
the other. For the smaller-revenue owner, the marginal value
of winning in the presence of pooled gate revenue sharing is
as follows:

The new equilibrium still occurs where the marginal values of
winning intersect:

Collecting like terms for MRGL on the left and MRGS on the
right, and then cancelling common terms on the left and right
side of the equal sign, gives us MRGL = MRGS. Just as in the
case of equal-proportion gate revenue sharing, MRGL = MRGS
is the equilibrium condition whether there is pooled revenue
sharing or not! This means that the amount of winning
produced by each team is the same with and without pooled
revenue sharing, W*L = 1 − W*S.
The only difference between pooled gate revenue sharing and
no revenue sharing is that the price of talent falls. It is easy to
see that the price of talent does not fall as far for pooled gate
sharing as it did for equal-proportion gate revenue sharing.
Suppose ? = ? so that the home team’s share under equal-
proportion gate revenue sharing is the same as under pooled
gate revenue sharing. This assumption allows us to isolate just
the impacts of the other elements in the shift. With ? = ?, we
can write the marginal value of winning under pooled gate
revenue sharing for this owner as follows:

To determine the size of the shift in marginal value of
winning, we want to compare this expression to the marginal
value of winning under equal-proportion gate revenue
sharing, that is as follows:

The difference between them is:

Thus, the marginal gate value of winning cannot shift down as
far under pooled sharing. Since the shift is again going to be
parallel and equal for both owners, the same is true for the
smaller-revenue owner.
The upshot of all of this is as follows. First, competitive
balance is the same under pooled gate revenue sharing and
221

equal-proportion gate revenue sharing as it is without any
revenue sharing at all. Qualitatively, since the shift downward
in marginal gate value of winning is again parallel and equal
for both owners, the directions of the impacts on them are the
same as under equal-proportion gate revenue sharing. The
total net value of winning remains the same for both owners,
the revenue that they share actually comes from reduced
payroll, and the share is larger for the smaller-revenue owner
than for the larger-revenue owner. But quantitatively, since
the shift is smaller, the impacts on owners will be smaller
under pooled gate revenue sharing. Let’s see how these
predictions stand up to a bit of data.
GATE REVENUE SHARING CHANGES IN THE NFL
The NFL changed its revenue sharing plan for the 2001
season from equal-proportion gate revenue sharing to pooled
gate revenue sharing. Forty percent of all gate revenue goes
into a pot that is shared equally by all teams. The 2003 Forbes
NFL Team Valuations include gate revenue for the 2002
season (unfortunately, gate was not reported separately until
this report). At that time, higher-gate teams averaged about
$68.3 million, lower-gate teams about $24.7 million, and the
median-gate teams around $33.0 million. Let’s suppose one
of each of these teams is in the league and focus on the change
this will have on owners.
The calculations are shown in Table 6.6. Before the change,
the owner of the high-gate team sent 40 percent of its
revenues, or $20.2 million, to the other owners and received
half of the 40 percent that each of the other owners
contributed, or $11.6 million. On net, the high-gate owner
made a net contribution of $15.7 million. By the same steps,
the owner of the median team sent $13.2 million to the other
two and received $18.6 million for a net income of $5.4
million. The owner of the low-gate team is also a beneficiary.
That owner sends $9.9 million into the pot and gets back
$20.2 million for a net income of $10.3 million. This seems
logical because the order of the net outcome varies inversely
with gate; the owner of the lowest-gate team gets paid and the
owner of the median-gate team receives less and the owner of
the high-gate team actually pays, on net.
One problem with sharing so much revenue is that some
owners might decide to scrimp on players because they are
going to receive the same share of league revenues. There is
an incentive to field a weaker team than the market will
support and let the other owners field good teams, generating
large revenues. The result is that the weaker team’s share of
revenue remains pretty much the same, but the team will have
an even larger net financial result because costs have been
lowered. Even 12 years ago, this was well recognized. Sports
writer Paul Attner (Sporting News, March 8, 1993) it this way:
The NFLPA [Players Association] has maintained since
the days of former executive director Ed Garvey that all
of the league’s revenue sharing has stripped owners of
the incentive to spend what it takes to build winners.
Nonsense, reply the owners, who claim they were just
222

practicing fiscal responsibility in their efforts to keep
salaries under control (p. 20).
This problem of free riding on other owners is one of the
reasons there is no corporate ownership in the NFL. Because
owners are individually responsible for their team’s bottom
line, rather than just passing off the costs to operating
revenues for a corporate owner, the incentive for other owners
to monitor that behavior is greater. Fleisher, Shughart, and
Tollison (1989) investigated this idea and found that teams in
the hands of fewer owners had a greater degree of revenue
sharing. Let’s see if the new NFL’s sharing strategy helps.
IMPACTS OF THE NEW NFL ARRANGEMENT ON
OWNERS
Because the gate revenue sharing percentages have not
changed under the new straight-pool sharing arrangement, all
three team owners contribute the same amounts as before. As
shown in Table 6.6, the total in the pool is $50.4 million. The
new plan has each team owner receiving an equal share, or
about $16.8 million. Under this scheme, the high-gate team
owner remains a net contributor at $10.5 million. The owner
of the median-gate team now earns a lower net payment of
$3.6 million on net. The owner of the low-gate team remains a
beneficiary, receiving a net of $6.9 million.
Notice what has changed. First, the net payment to the owners
of the low-gate and median-gate teams has fallen by 33
percent. Of course, this means that the high-revenue team
owner’s net contribution fell by 33 percent. The result should
reduce the incentive for the low-gate owner to scrimp on
talent.
IMPACTS OF THE NEW NFL ARRANGEMENT ON
COMPETITIVE BALANCE
Alterations in balance during the season can be seen returning
to Table 6.3. Essentially, the move to pooled sharing had
impacts on balance revealed by the average of our standard
deviation ratio over the 2000s compared to the 1990s. We see
that really nothing much happened, as our graphical
investigation suggests. Over the 1990s, our measure is 1.51,
and over the 2000s, it is 1.56 (arguably, an insignificant 3
percent difference). The theoretical implications from Figure
6.5 seem quite useful both for the impacts on owners and on
competitive balance.
ANALYSIS OF AGGREGATED POOLED REVENUE
SHARING
Finally, we come to the form of pooled revenue sharing
currently practiced in MLB. Essentially, all that happens is
that we must add the rest of local revenues, in addition to gate
revenue, to the pooling. Having already analyzed gate-only
pooled revenue sharing, extending the analysis to include
other local revenue is as easy as increasing the size of the pie.
Let’s have owners keep the same percentage ?, but add local
TV revenue into the pool. Denote the marginal TV value of
winning by MRTVL and MRTVS for the larger- and smaller-
223

revenue owners, respectively. Now, the larger-revenue owner
contributes (1 − ?)(MRGL + MRTVL) and the smaller-revenue
owner (1 − ?)(MRGS + MRTVS) and each receives half the
pool back as its equal share, that is:

Again, we don’t really need to appeal to graphical analysis to
gain our insights.
In the equilibrium without any sharing, MRGL(W) +
MRTVL(W) and MRGS(W) + MRTVS(W) represent marginal
local revenues from winning. Without any revenue sharing,
equilibrium is determined by W*L = 1 − W*S and W*L > W*S.
Taking the addition of local TV revenues into account, the
marginal local value of winning for the larger-revenue owner
becomes as follows:

The second parenthetical is a difference, rather than a sum,
again because talent comes to one team only at the expense of
the other. For the smaller-revenue owner, the marginal local
value of winning in the presence of pooled gate revenue
sharing is as follows:

The new equilibrium still occurs where the marginal local
values of winning intersect. Setting these two marginal value
expressions equal and collecting terms yield the following:

Cancelling like terms on the left and right side of the equal
sign gives us MRGL(W) + MRTVL(W) = MRGS(W) +
MRTVS(W) as our equilibrium condition. But this is precisely
the equilibrium condition without pooled local revenue
sharing! This means that the amount of winning produced by
each team is the same with and without pooled local revenue
sharing, W*L = 1 − W*S, with W*L > W*S. Finally, as with all
revenue-sharing arrangements, the price of talent does fall.
For our last comparison, recall that MLB switched from
equal-proportion gate revenue sharing to pooled local sharing
in 1996. Since both mechanisms have the same equilibrium
distribution of talent, again the qualitative impacts on owners
and players are the same—payroll paid to players falls, payroll
savings represent the amount of revenue to be shared, and the
smaller-revenue owner gets a larger portion. Quantitatively,
we wish to know which is the larger shift determined by the
following comparison (pooled local sharing and equal-
proportion gate sharing, respectively, with ? = ?):
224

and

The difference between them is:

Technically, then, the shift is smaller as long as MRGL +
MRTVL + MRGS > MRTVS. While this seems likely,
theoretically, we must allow for the case where the shift is
even larger if the marginal local TV value of winning swamps
all other marginal values of winning in both markets.
The upshot of all of this is as follows. First, competitive
balance is the same under pooled revenue sharing extended to
local TV, pooled gate revenue sharing, and equal-proportion
gate revenue sharing as it is without any revenue sharing at
all. Qualitatively, again, the direction of the impacts on
players and owners is the same. The total net value of winning
remains the same for both owners, the revenue that they
share actually comes from reduced payroll, and the share is
larger for the smaller-revenue owner than for the larger-
revenue owner. But quantitatively, it seems most likely that
the impacts on owners will be smaller under pooled local
revenue sharing extended to local TV. Not as much revenue
goes from players to owners and the size of the gain by the
smaller-revenue owner will be smaller than under equal-
proportion gate revenue sharing.
LOCAL REVENUE-SHARING CHANGES IN MLB
MLB abandoned its old 80/20 equal-proportion gate revenue
sharing plan during the negotiations that ended the strike of
1994–1995. Since 1996, MLB owners have practiced pooled
local (gate, concessions, parking, and local TV) revenue
sharing. From 1996 to 2001, about 17 percent of all local
revenues, net of stadium charges, were shared under a “split-
pool” arrangement. Seventy-five percent of the pool was
shared equally among all owners. The remaining 25 percent
was shared equally only among owners with below league
average total revenues. From 2002 on, 34 percent of all local
revenues, net of stadium charges, were shared equally by all
owners. In either case, the idea was to put more money in the
pockets of weaker owners in order to give them a better
chance to compete for talent. Let’s see how it shakes out.
IMPACTS OF POOLED SHARING ON OWNERS
The data on MLB gate revenues only go back to 2001. These
are precisely the data that are appropriate for a comparison
between the original “split-pool” and current pooled local
revenue-sharing arrangements. We’ll go ahead and compare
to the equal-proportion gate revenue sharing period using the
same data, despite that fact that it ended much earlier in
225

1996. Again, let’s adopt a comparison as if there were one
high-revenue owner, one median-revenue owner, and one
low-revenue owner. The calculations are in Table 6.7 based
on the 2001 MLB Commissioner’s public report of team
operations across MLB. To determine the outcome under the
oldest (approximately) 80/20 equal-proportion gate-sharing
arrangement, the data in Table 6.7 indicate that gate revenues
are around $88.1 million for the highest-revenue owners,
$48.6 million for the median-revenue owners, and $15.7
million for the lowest-revenue owners. The highest-revenue
owner would have contributed 20 percent of gate revenues, or
$17.6 million, to the other owners and received half of the 20
percent that each of the other owners contributed, or $6.4
million. That makes the net contribution $11.2 million. The
owner of the median-revenue team would have sent 9.7
million to the other two and received $10.4 million for a net
gain of about $700,000. The owner of the lowest-revenue
team would have paid $3.1 million and got back $13.7 million
for a net gain of $10.5 million. Once again, this seems logical.
The owner of the weakest-revenue team is paid the most out
of the reduction in pay to players, and the owner of the
highest-revenue team is paid the least.
Under the local revenue-sharing arrangement in place for the
1996–2002 seasons, each team contributed 17 percent of all
local revenues net of expenses for ballpark financing.
Unfortunately, the data in the 2001 Commissioner’s report do
not list ballpark financing expense, so we’ll have to leave those
out of the rest of the discussion. The split-pool sharing
arrangement described above generated the results in the
center panel of Table 6.7. The Forbes report suggests that the
largest-market teams had local revenues of about $184.7
million, whereas the median- and smallest-revenue teams had
local revenues of $94.6 million and $29.3 million,
respectively. The total in the pool would have been $52.5
million (17 percent of summed local revenues across all
teams), about 72 percent more than was shared under the old
80/20 plan. The new plan has each team owner receiving a
share of about $13.1 million, that is, (0.75 × 52.5)/3. In
addition, in our three-team illustration, the lowest-revenue
owner would also receive the entire remaining 25 percent, an
additional $13.1 million. Under this scheme, the largest-
revenue owner’s net position is worse by 63 percent, and the
median-revenue owner now receives less than their
contribution to the revenue sharing pool (but remember, the
net still all came out of payroll reduction). Finally, the lowest-
revenue owner’s net receipt doubles!
Let’s move on to the pooled local revenue-sharing
arrangement in place in MLB since 2002. Sharing was
increased to 34 percent of all local revenues, net of ballpark
expenses, and all owners receive an equal share. Cutting
straight to the chase, the last column in the last panel of Table
6.7 makes it clear that even though the total amount going
into the pool doubles, the net position of the highest-revenue
owner falls again (by another 52 percent), the median-
revenue owner returns to being a net beneficiary, and the net
receipt for the lowest-revenue owner increases yet again
226

(about 18 percent). Clearly, the latest pooled sharing approach
mostly redresses the impact on the median owner at the
expense of the largest-revenue owner.
IMPACTS OF POOLED SHARING ON COMPETITIVE
BALANCE
The data in Table 6.8 show our standard deviation ratio
bracketing 5 years before and 5 years after the introduction of
each pooled revenue approach in MLB. The first comparison
is the 5-year period under the old 80/20 equal-proportion
gate revenue sharing period prior to the adoption of “split-
pool” sharing. As a matter of computation by our standard
deviation ratio measure, competitive balance worsened by
about 5 percent in the AL and 1 percent in the NL. While one
might quibble, these changes are close to zero as suggested by
the theoretical treatment we just covered. The second
comparison is between split-pool sharing and the current
pooled sharing arrangement. In this case, competitive balance
worsened by 20 percent in the AL and improved by 5 percent
in the NL. The latter result squares with our theoretical
prediction, but the AL result does not; a 20 percent change is
difficult to ignore. It also is the case that while more and
different teams appear in the play-offs, they do not make it
very far compared to larger-revenue teams (with the notable
exception of the Tampa Bay Rays in 2008).
This lack of increased balance has been the topic of recent
discussion in the press. Critics suggest that smaller-revenue
owners are not spending their revenue-sharing receipts to
improve their teams. For example, economist Andrew
Zimbalist states that it is “silly” to think that these teams are
putting all of their money into player development (Pittsburgh
Tribune-Review, www.PittsburghLive.com, June 18, 2006). In
the same article, MLB responds that they are required under
the collective bargaining agreement with players to enforce
exactly this disposition of net sharing to smaller-revenue
owners. Rob Manfred, MLB chief labor executive, and Kevin
McClatchy, managing general partner of the Pirates, point out
that the players’ union has never complained and every
recipient club was living up to the agreement to spend on
team quality.
Our theory tells us that the critics are missing the mark; there
is no reason to suspect that either larger- or smaller-revenue
owners will spend their share of the pool on talent in the first
place! Pooled sharing simply reallocates money from players
to owners with a larger share going to smaller-revenue
owners. A bit of data on net receipts and payrolls do not
support the critics either. Table 6.9 shows how payrolls have
changed relative to net revenue-sharing receipts for a few
years where data are available during the current pooled
sharing approach in MLB. While it cannot be because of
pooled revenue sharing, the vast majority of net recipients do
indeed increase their payrolls, and the average change in
payroll among net recipients is greater than zero (although
not by much in 2005).
This suggests that critics should be looking elsewhere both for
an explanation of owner spending on payroll and for the
227

http://www.PittsburghLive.com

http://www.PittsburghLive.com

decline in balance in the AL. Table 6.1 gives an example of
where to look—revenue imbalance worsened in MLB over the
last three decades, and more so in the AL. Since pooled
sharing cannot fix this problem, theoretically, and has not
fixed the problem, according to the data, then some other
mechanism would need to be employed if the league truly is
interested in improved balance. For example, as mentioned
earlier, the commissioner keeps a discretionary pool from
revenue sharing to reward teams that improve. Since
payments are directed at performance, this type of
reallocation would be predicted to change talent choice. And
so can another approach—taxing talent choices instead of
rewarding them—to which we now turn.
228

LEARNING HIGHLIGHT: SHRINKING MLB
Only once has a league reduced the number of its member
teams. That was in the NL in 1900, a few years before the
formation of the modern MLB, when the original Baltimore
Orioles, Cleveland Spiders, Louisville Colonels, and original
Washington Senators were absorbed by other teams in the
league. More typically, a merger between rival leagues results
in a larger single league than that existed prior to the rivalry.
Some of the rival league teams disappear after the merger, but
a few join the dominant league.
Not long ago, talk of shrinking MLB arose. Jerry McMorris,
owner of the Colorado Rockies, raised the question during an
owners’ meeting in 1999, “What would happen if instead of
relocating and expanding, we considered consolidating
teams?” (Sports Illustrated, August 20, 1999, p. 65). The
particular targets at that time were teams in financial trouble,
Montreal, Minnesota, Kansas City, and Oakland. The threat to
shrink MLB earned some credence when the league recently
killed the sale of some of these teams to what appeared to be
reasonable owner groups.
Why shrink MLB? Shrinking the number of teams spreads
shared revenues over fewer teams, raising the amount per
team. Presumably, with increased shared revenues, those
teams would become stronger. In addition, quality of play
would improve with talent concentrated in fewer teams. Not
surprisingly, the idea has not been brought up that closing
down these existing franchises would open up new believable
threat locations for the remaining MLB owners.
However, there are quite a few arguments against shrinking
the league. Suppose that the idea of shrinkage had occurred
10 years ago. Likely candidates at that time would have
included Cleveland, Atlanta, and Seattle, which are all now
play-off-contending teams. It also is not at all clear that MLB
can contract under the antitrust laws. Even if it can, other
obstacles lie in the way. Eliminating teams would open a few
more places for potential rival leagues to add to their list of
possible cities. In addition, the players would be unlikely to sit
idly by while upward of 180 jobs were eliminated. Finally,
senators and members of Congress representing the fans who
lose their teams could easily tie MLB up in hearings for years.
Eliminating MLB teams will do two things. First, it will open
viable threat locations for the remaining owners to use against
their host cities. Second, it directs revenue increases in the
future to the smaller number of remaining owners. Some fans
and observers like the idea because it will “raise quality.” Tell
that to the fans in Oakland and Kansas City who would have
ended up without any major league baseball to enjoy of any
quality level at all.
If MLB contraction occurred as planned, the Twins would
have been history, and it could have been many years before
Minnesota fans saw the likes of Brad Radke again. [Phot of
Radke.]
Source: Sports Illustrated, August 20, 1999, p. 65.
229

SECTION 5
The Luxury Tax, the Draft,
and Payroll Caps
The so-called payroll luxury tax is used only in MLB where it
is referred to as the competitive balance tax (the penalty for
violating salary caps in the NBA, NFL, and NHL are
sometimes erroneously referred to by the same term). A true
luxury tax is a progressive charge on the amount of talent
hired. Essentially, the league and the union establish spending
thresholds (starting at $148 million in 2007, growing to $178
million by 2011), and teams are free to spend as they choose
on talent. However, if they spend more than the thresholds,
they pay a progressively larger tax rate on spending past each
threshold (eventually up to 22.5 percent for first-time
violators, 30 percent for second-time violators, and 40
percent for third-time violators). By taxing talent
differentially based on how much is purchased, competitive
balance definitely will be altered.
THE LUXURY TAX
The essential elements of the luxury tax outcome are in
Figure 6.6. In a two-team league, an effective luxury tax
would simply tax the larger-revenue owner’s talent choice.
This abstraction really does no harm because the effective
talent spending level in the real-world version typically is
chosen to affect only the few largest spenders. Let the tax on
the larger-revenue owner be equal to t1. This has the effect of
making each unit of winning percent worth (1 – t1) MRL(WL)
to the larger-revenue market owner. The dotted-line segment
in Figure 6.6 shows this impact for t1 = 0.5, a 50 percent tax
on spending (just because it is easiest to portray).
The imposition of the luxury tax reduces the larger-revenue
market owner’s profit-maximizing winning percent to WtL < W*L (equilibrium still must occur at the intersection of the larger- and smaller-revenue owner marginal revenues). Because the tax reduces the net value of talent at the margin and only for one of the teams, it also alters the amount of talent chosen by both teams; the smaller-market owner has an increased winning percent. The price of talent falls, the smaller-market owner purchases more talent, and the sum of the two winning percents must equal one. Note, too, that payments to talent are now smaller shaded rectangles because the price of talent has fallen in equilibrium to P′. When the net value of hiring talent falls, less of it is used and payment falls. Thus, larger-revenue market owners and players share the burden of the tax. Total tax revenues collected from the larger-revenue owner will be equal to t1 × (P′ × W*L), that is, the tax rate applied to that owner’s total expenditure on talent. In the real-world case of the MLB luxury tax, the 2007 labor agreement between the owners and the players dictates that 75 percent of the tax revenues go to player benefits, and the remaining 25 230 percent to the industry growth fund (a public relations drive to “grow” the fan base) and clearing disputed player contracts. It is extremely important to note the following. The luxury tax will reduce competitive imbalance, but what are the chances that it can reduce imbalance considerably? Refer back to Figure 6.6. For a tax of 50 percent, the reduction in competitive imbalance was quite small. If you get out your ruler, the gain in winning percent for the smaller-revenue team was about 0.045. The limiting factor is, as always, how much more talent the smaller-market owner will buy when the price of talent falls. Indeed, the 22.5 percent rate applied to first offenders seems unlikely to have much impact at all. Even at very prohibitive tax rates, like the 40 percent rate for third offenders, the gains in competitive balance may not be all that large. This may help to explain why the luxury tax in place in MLB has not stemmed the tide of growing competitive imbalance. THE DRAFT Can we add the reverse-order-of-finish draft to local TV revenue sharing as a mechanism that might increase competitive balance? Under such a draft, the worst teams from the preceding season get first choice of the new talent coming into the league. This is highly touted as a remedy to competitive imbalance. After all, poor teams draft the better players and should be able to close the gap with richer teams, right? Nearly all sports economists would say no. We are going to save a full-blown treatment of the draft for Chapter 8, but the idea is that talent always moves to its highest valued use in the league. Therefore, weak teams draft strong talent, but stronger teams in larger-revenue markets then pay to move that talent toward them. In this way, the talent ends up in the same place as it did without a draft. The only differences are as follows: • The draft reduces the amount that players are paid up front to join a given team because they must play for the owner that drafted them (again, details are in Chapter 8). • Weaker teams get paid by larger-revenue market teams for the talent they bring into the league through the draft. The data tend to support the view that reverse-order-of-finish drafts do not alter competitive balance. Table 6.10 shows a before-and-after look at the NFL and MLB drafts. For the NFL, our standard deviation ratio measure of competitive balance increases 7.2 percent from the before period (Period 2, in Table 6.10) to the after period (Period 3, in Table 6.10). One might be inclined to think this means balance worsened with the draft. But this really isn’t much of a change, especially if we look at an additional period prior to the draft (Period 1, in Table 6.10). With this additional period, you can see that the average change yields an improvement of 5.5 percent in competitive balance. Once again, this isn’t much of a change and provides evidence that the draft has no effect on competitive balance. 231 Turning to baseball, Table 6.10 shows that our measure of competitive balance decreases 18.9 percent from the before period (Period 2, in Table 6.10) to the after period (Period 3, in Table 6.10) in the AL. The NL statistic also decreases by 14.5 percent. These changes seem large enough to suggest that there were significant improvements in competitive balance due to the draft in baseball. Once again, however, let’s look at one additional period prior to the draft. There is a downward trend in our statistic measuring competitive balance that had nothing to do with the draft. In the NL, the decline before and after the draft is only 2.5 percent greater than the trend, essentially no change. But in the AL, the decline before and after the draft is still 8.4 percent greater than the trend. Reasonably, the AL result is worth pondering because it appears to refute the idea that the draft has no impact on competitive balance. One explanation is that there may have been something else changing at the same time (always a possibility when you look at very simple statistical comparisons like averages). Another is suggested by the CBS ownership of the Yankees incident detailed in Chapter 3. This period coincides with the imposition of the draft and was an uncharacteristically weak time for the Yankees. That alone may have been enough to make it appear that the draft was effective at redistributing talent in the AL. So, although these draft cross-subsidies share the wealth created by players with lower-revenue teams, they do not generally lead to any change in competitive balance (although a more detailed analysis of the AL after the draft would prove interesting). Players end up distributed through the league roughly as they would without the draft. PAYROLL CAPS “Salary caps” in North American pro sports leagues are poorly named. Actually, they are the result of revenue sharing between owners and players and should more properly be named “revenue sharing payroll caps.” In fact, this type of cap on payrolls does not include any cap on individual player salaries (although those do exist for some players in some leagues). First, during collective bargaining (Chapter 9), owners and players agree upon the portion of leaguewide revenues that they will share. In the NFL, this amount is called defined gross revenues (DGR). In the NBA and NHL, the revenues shared between owners and players are called business-related income (BRI). Then they agree upon the split between owners and players. In the 2006 extension agreed upon between NFL owners and players, the share for players is destined to increase from 57.5 percent to 58 percent by 2011. The minimum is 50 percent, and owners must pay players any amount below that. Let’s label the players’ share s. In a league of n teams, the following formula determines the cap on owner spending on talent, C (we use DGR, here, but BRI would do just as well): Thus, the “salary cap” is actually a cap on payroll spending that is equal across all teams. The hope is that, by equalizing 232 talent spending, team quality also will equalize across teams and enhance competitive balance. The salary cap was saved for last because it provides an introduction to the other problems confronting leagues. The cap, like the luxury tax, can enhance a league’s competitive balance outcome. In fact, if properly designed and enforced, these three can eliminate the impacts of revenue imbalances completely. GRAPHICAL ANALYSIS OF SALARY CAPS Figure 6.7 shows the intended impact of the salary cap’s equal spending limitation. Since no league actually has chosen such a “hard cap,” this presentation actually helps us understand the actual choices by leagues, detailed shortly. For our diagrammatic exposition, the hard cap is shown on a per unit talent basis as the line CC below the competitive price P*. Equal spending, guaranteed by the presence of a minimum close to the cap, means an equal outcome, thus, WCL = WCS = 0.500. If this outcome lasts, the sizes of the shaded rectangles that measure team spending are the same. There is some careful wording in the preceding paragraphs— intended impact, if this outcome lasts, spending should be equalized. These qualifiers were chosen for two reasons— actual salary caps are subject to cheating and they have been “soft” rather than “hard” by design. Given the cap CC in Figure 6.7, the reason for cheating is easy to see. Both owners would rather not be at a winning percent of 0.500. The smaller-revenue market owner would rather choose an amount of talent W*S, where C = MRS and a winning percent lower than 0.500. This is because C > MRS for all units
beyond W′S. Thus, to accomplish its competitive balance
goals, the payroll cap must also be a payroll floor. Both the
NBA and NFL cap agreements have very specific language
dictating a minimum amount of spending very close to the cap
amount. Of course, this minimum requirement only matters if
the cap is chosen so that C > MRS.
In addition, the larger-revenue market owner would be happy
to buy the talent that the smaller-revenue market owner
wishes to relinquish because MRL > C on each unit between
0.500 and W′S. When owners on both sides of the cap wish to
cheat, one would expect them to try.
Thus, the salary cap presents the league with a monitoring
and enforcement problem. All of this creates a need for “cap
police.” The 0.500 equal winning percent outcome cannot be
maintained without league monitoring and enforcement. In
both the NBA and NFL, this job has been given to the
commissioner’s office. The commissioner is allowed to punish
cap violations with monetary fines, voided contracts, lost draft
picks, and suspension of team officers involved in the
violation.
In addition, fairness clauses in cap agreements allow some
cap violations. It is typical to allow some cap overruns when
owners attempt to retain their most treasured free-agent
players, for example. In cap language, caps that have more of
233

these exceptions are referred to as “soft” caps, and caps that
have fewer exceptions are called “hard” caps.
ANALYSIS OF CAP EFFECTIVENESS
Given the enforcement problem posed by caps, as well as the
fairness clauses that make it legal for some owners to spend
more than the cap, it is natural enough to check to see if
spending actually is anywhere near the cap. Table 6.11 shows
caps and payrolls for the NBA, NFL, and NHL. The table
makes it clear that even the average NBA and NFL payroll has
been larger than the cap almost every year of its existence
(Staudohar [1999] was the first to notice this). Of course,
because this is an average, there are also teams far below and
far above the cap. In the few years of its existence, average
payrolls have all been below the cap.
James Quirk and I (Fort and Quirk, 1995) originally showed
almost 15 years ago that it is only by happy coincidence that
any NBA owner actually spends anywhere near the cap in any
given year. Larger-revenue market owners typically violate it
by as much as 25 percent. Table 6.12 shows that nothing
much has changed since our initial analysis. Fully 93 percent
of NBA teams are recently over the cap with not a single
larger-revenue team at or below the cap and the largest-
revenue New York Knicks lead the way. In the NFL, a much
lower 50 percent of teams are recently over the cap. As one
would expect given the breadth of revenue sharing, even the
larger-revenue Washington Redskins are below the cap. The
NHL is again quite different—only 20 percent of teams are
over the cap, but it is also true that all are larger-revenue
market teams.
The NFL and NBA caps are not met, even on average. This
means that those caps cannot be generating equal playing
balance, but they may be making improvements. It is hard to
tell in the NFL, because it is the most balanced of all the
leagues. The NBA, on the other hand, is not balanced. It is not
clear why caps are ineffective. As mentioned earlier, it is
partly due to the fairness clauses that allow some teams to
spend over the cap. However, cheating on salary caps cannot
be ruled out as part of the explanation for both the NBA and
the NFL. The league offices, charged with enforcing the caps,
may not have been watchful for violations. Or league offices
may simply hve found it impossible to actually impose caps
because the largest-revenue owners want to violate the cap
away. The NHL cap appears to have a better chance of
improving balance, but there is little data yet to investigate in
any detail.
Interestingly, the point of the 1998 NBA owners’ lockout of
the players was to gain a hard cap, which they did through
collective bargaining that year. The little data that we have
indicate that it still isn’t working. Both the 1998 and 1999
seasons have mean salaries well in excess of the cap, and the
data in Table 6.12 indicate that larger-revenue teams remain
the primary violators. It appears the league has not come to
grips with the problem it sought to solve with the lockout.
234

In summary, neither any form of revenue sharing currently in
use by any league nor the reverse-order-of-finish draft will
improve competitive balance. The luxury tax and salary caps
have the potential to improve balance. All of the mechanisms
discussed so far have one thing in common. Whether or not
theory suggests that they will reduce competitive imbalance,
they all will reduce payment to players. Small wonder then
that each of these mechanisms has been a bone of contention
in labor negotiations between owners and players, as we’ll see
in Chapter 9.
There are two other ways to reduce competitive imbalance in
pro sports leagues. The historically unique contraction
approach of 1900 in last section’s Learning Highlight:
Shrinking MLB is one. Not a single pro sports league has ever
used the other, harnessing the power of economic competition
to reduce competitive imbalance. Why this method never will
be used is discussed in this section’s Learning Highlight:
Competitive Imbalance and the Power of Economic
Competition.
235

LEARNING HIGHLIGHT: COMPETITIVE IMBALANCE
AND THE POWER OF ECONOMIC COMPETITION
Theory suggests that the root cause of any competitive
imbalance problems is that some owners are led to choose
greater talent levels than others in order to collect the vastly
larger revenues offered in their markets.
Local revenue sharing, luxury taxes, and salary caps all aim to
reduce the marginal value of talent to owners, holding
constant the fact that leagues have maintained exclusive
territories of vastly differing revenue potential. But territorial
exclusivity is the domain of the league, and there is a perfectly
direct mechanism to address the problem at its root cause.
Just put more teams in the largest-revenue “megamarkets.”
This approach and its historical origins are discussed in great
detail in my book with James Quirk, Hardball: The Abuse of
Power in Pro Team Sports (Quirk and Fort, 1999). Suppose
that only one team occupies the market with the greatest
revenue potential. The owner creates the league’s highest-
quality team to collect that revenue, and competitive
imbalance ensues. If the league allowed another franchise in
the same location, competition should reduce the current
owner’s advantage. Fans now have more substitutes, and the
demand for the current team’s product shifts left. So does the
marginal revenue function for the current owner. As a result,
that owner chooses a lower level of talent, and competitive
imbalance is eased.
Clearly, owners understand the forces of this type of
competition. After all, it is the very essence of their joint
venture activity to keep those forces at bay. The careful
management of team location both protects the territory
exclusivity of the larger-revenue market owners and impedes
any possible rival leagues from putting franchises in the most
valuable territories.
This also makes it easy to see why leagues have never
harnessed economic competition to reduce competitive
imbalance on the field. Some owners will face reduced
economic circumstances as the value of their exclusive (but
smaller) territories falls. Indeed, whenever a league moves a
team to even remote proximity of an existing owner, long
discussions ensue concerning compensation to the owner who
perceives a loss of territory.
The idea of harnessing economic competition to reduce
competitive imbalance isn’t just held by academicians in their
ivory towers, sharpening their heads to a fine point. MLB
decided to relocate the Montreal Expos to Washington, D.C.
This is the first time a team moved since 1971, when the last
incarnation of the Washington Senators (the irony drips,
doesn’t it?) left for Texas. The move was aided by the fact that
MLB had purchased the franchise from its previous owner
just a couple of seasons earlier.
During the MLB owners’ deliberations on what to do with
their Expos, John Shea of the San Francisco Chronicle
(August 24, 2003; or www.sfgate.com) plainly put it that
236

http://www.sfgate.com

http://www.sfgate.com

moving the Expos to New York (!) would have important
virtues:
It has been 46 years since three [MLB] teams played in
New York, and it might be another 46 years before it
happens again. But by moving the Expos to the biggest
city this side of Tokyo, baseball would help achieve two
of its goals. It would give the Expos a home. It would
give the Yankees a headacheMLB would not propose a
three-team New York marketEven though a Gotham
trifecta would be good for baseball, the Yankees and
Mets aren’t interested in helping the game as much as
themselvesThe 29 other teams own the Expos, and they’d
draw top dollar by selling to New York ownership.
They’d also guarantee high attendance, which wouldn’t
be a lock in Washington, D.C., Northern Virginia or
Portland.Move the Expos to New York, and the Yankees,
along with the Mets, would be forced into a regional
form of revenue-sharing, and that would go a long way
toward balancing the economic playing field.
However, in order for owners acting as a league to use
economic competition to reduce competitive imbalance, the
owners themselves would have to vote in favor of it. But it
probably will never happen. In pro sports leagues,
supermajorities are required for important league actions. In
a league of 30 teams, if a 75 percent majority is required, then
it would only take eight owners to kill a move toward
economic competition if such a movement ever arose in the
first place. And moving a team into the actual territory of an
existing owner requires unanimous consent. Along with
compensation demands by harmed owners, supermajorities
have operated for nearly 50 years to keep a third team out of
New York. If the forces of economic competition were at work,
there would already be a third team in that megalopolis
market.
237

SECTION 6
Chapter Recap
Exclusive territories generate different revenue streams that
lead to payroll imbalance and competitive imbalance. Theory
predicts this outcome, as we discovered by analyzing the two-
team league case. However, whether or not this is a problem is
an empirical issue. Revenue imbalance, payroll imbalance,
and competitive imbalance have always been part of league
play, and it isn’t clear that any of these types of imbalance is
any worse than ever, historically speaking.
The data on competitive balance within the season, analyzed
with Gini coefficients and the standard deviation ratio, reveal
that generally (but not in all sports) revenue imbalance has
improved steadily in pro sports leagues since the 1950s.
Payroll imbalance has worsened. Finally, winning percent
imbalance has declined modestly while taking a roller-coaster
ride through the decades. Championship imbalance has
always plagued pro sports leagues; although the teams that
one would identify as larger-revenue teams have changed over
time, they nonetheless dominate championship play.
The data on revenues, payrolls, and winning percent also are
correlated in just the way suggested by theory. Examining
MLB, the correlation between revenues and payrolls is
typically strong and positive. The same can be said for the
correlation between revenues and winning percent. This
suggests that arguments by the press and owners that payroll
imbalance drives whatever problems leagues face are in error.
It actually is, as theory predicts, the difference in revenue
potential across teams that explains competitive imbalance.
However, even if competitive imbalance had improved over
the decades, fans could dislike whatever level of competitive
imbalance they saw. If they responded at the gate and in their
viewing choices, leagues could face a reduced financial future.
While the work on the feelings of fans toward competitive
balance is ongoing, leagues have taken actions in the name of
reducing competitive imbalance.
Joint venture remedies for competitive imbalance claimed by
leagues include national TV revenue sharing, local revenue
sharing, luxury taxes, player drafts, and salary caps. Theory
suggests that national TV revenue sharing, local revenue
sharing, and player drafts will have no impact whatsoever on
competitive balance. However, they will redistribute money
from players to the owners of weaker teams.
Recent changes in gate sharing in the NFL show the same
amount of money collected in total. But there is a dramatic
reduction in the net payment by the largest-revenue owners to
the other owners in the league. This should reduce the
incentive for lowest-revenue teams to skimp on talent, raising
the quality of their teams. In MLB, the most recent move to
238

pooled local revenue sharing appears to redress a problem
with the previous system where team owners near the middle
of the revenue pack were net contributors to the system. In
both leagues, as the theory predicts, competitive balance
shows nearly no change across revenue-sharing regimes. The
only exception is the AL in MLB. Revenue-sharing theory and
data also illuminate a current debate in the press about the
level of imbalance in baseball.
Graphical analysis shows that luxury taxes, set high enough,
will lead richer teams to choose less talent. The price of talent
falls and smaller-revenue market owners buy more talent so
the league becomes more balanced. But this type of tax is
unlikely to be applied to very many teams. At the current tax
rates, large gains against competitive imbalance seem
unlikely.
Salary caps, theoretically, can eliminate competitive
imbalance entirely. However, the evidence is overwhelming
that even average-team payrolls exceed the cap in both the
NBA and the NFL. The NHL cap, still in its infancy, appears to
be working better. Even though caps can reduce competitive
imbalance, they simply have not done so in their current
forms. Either caps simply are designed with too many
concessions in the name of fairness, or there is significant
cheating on the caps in both leagues.
One method of reducing competitive imbalance would be to
harness the forces of competition to that end. Leagues could
locate more teams in the largest-revenue markets, reducing
the value of those territories to owners currently occupying
them and shifting marginal revenues to the left. The result is
more balanced competition on the field. However, given that
the largest-revenue owners would be opposed and that
leagues are organized with supermajority decision rules, there
is little reason for optimism about the implementation of this
method.
239

SECTION 7
Key Terms and Concepts
You should have run into each of these in pop-ups in the text
of this chapter:
• Competitive imbalance
• Winning percent equilibrium
• Payroll imbalance
• Gini coefficient
• Standard deviation ratio
• Revenue imbalance
• Championship imbalance
• National revenue sharing
• Local revenue sharing
• Equal-proportion gate revenue sharing
• Pooled revenue sharing
• Payroll luxury tax
• Reverse-order-of-finish draft
• Payroll caps
240

SECTION 8
Review Questions
1. What is competitive imbalance? What danger can
competitive imbalance pose to a pro sports league? Be
sure to use the uncertainty-of-outcome hypothesis in your
answer.
2. Why does theory predict that marginal revenues will be
equal across all teams in a winning percent equilibrium?
(Hint: Suppose they weren’t? What would happen?)
3. What is the root cause of both payroll imbalance and
competitive imbalance?
4. What is a Gini coefficient? How is a Gini coefficient used
to identify revenue imbalance or payroll imbalance?
5. What is the standard deviation of winning percents?
What advantage is there to using the standard deviation
of winning percent, divided by the hypothetical standard
deviation from a balanced league, rather than just the
standard deviation of winning percent alone?
6. What measure is used to identify competitive imbalance
in championships? Briefly summarize the evidence on
championship imbalance in pro sports leagues.
7. Briefly summarize the evidence on the following aspects
of pro sports leagues:
a. Revenue imbalance
b. Payroll imbalance
c. Within-season competitive imbalance
d. Championship imbalance
8. What does the theory presented in this chapter suggest
about the correlation between revenues and payrolls?
Revenues and winning percents? What do the data show
for MLB?
9. Define each of the following, and state what the theory
predicts about whether each mechanism can reduce
competitive imbalance. Explain fully:
a. Gate and national TV revenue sharing
b. Local revenue sharing
c. everse-order-of-finish player draft
10. What is the difference between split-pool revenue
sharing and pooled sharing as practiced in both the NFL
and MLB? Be sure to define each in your answer.
11. Define each of the following, and state what the theory
predicts about whether each mechanism can reduce
competitive imbalance. Explain fully:
241

a. Luxury tax
b. Salary cap
12. Summarize what the theory predicts will happen to
player pay under all forms of revenue sharing, the draft, a
luxury tax, and a salary cap.
13. Briefly summarize the evidence on salary cap
effectiveness.
14. Give two reasons for your answer in the preceding
question. Can we tell which of the reasons you just gave
explains more of the evidence than the other? Why?
15. How can economic competition be harnessed to reduce
competitive imbalance? (See the Learning Highlight:
Competitive Imbalance and the Power of Economic
Competition.)
242

SECTION 9
Thought Problems
1. How do today’s pro sports leagues resemble MLB as
described by Bill Veeck back in 1958? What do you think
is the primary cause of the similarities?
2. How does economics allow us to judge whether
competitive imbalance is “too large” in some sports
league? What are the limits of the economic way of
thinking on this issue?
3. Is the two-team league winning percent equilibrium
model a short-run or long-run model? Explain.
4. Graphically demonstrate that the marginal revenue from
hiring talent is equal across larger- and smaller-revenue
markets in equilibrium. In the same graph, show how
revenue imbalance leads to competitive imbalance and
payroll imbalance.
5. Starting from Figure 6.3, suppose that general
employment opportunities in the smaller-revenue market
increase dramatically so that population begins to
increase. Show the following:
a. The price of a unit of talent will increase.
b. The smaller-revenue market owner buys more talent.
c. Spending on talent by the smaller-revenue market
owner will increase.
d. Spending on talent by the larger-revenue market
owner falls. (Hint: Is demand elastic or inelastic for
the larger-revenue market owner?)
e. Competitive imbalance is less than it was prior to the
population increase.
6. What is the real-world importance of the three
characteristics of the winning percent equilibrium
demonstrated in Figure 6.3?
7. The Gini coefficient was used in the text to examine both
revenue imbalance and payroll imbalance. Why wasn’t
the Gini coefficient used to examine winning percent
imbalance as well? (You may wish to consult the paper by
Utt and Fort [2003].)
8. Show why each of the remedies for competitive imbalance
listed in the text will or will not improve balance. Where
possible, use Figure 6.3 to make your point. On local
revenue sharing, be sure to include how a comparison of
MLB and the NFL can aid in the judgment of that
mechanism’s chances to support balance.
9. Describe all of the steps involved in calculating a salary
cap. Why is the term salary cap misleading in terms of
243

what the mechanism actually does? What name does your
author prefer?
10. The Learning Highlight: Shrinking MLB refers to a
contraction in baseball that actually occurred for the 1900
season. The NL contracted from 12 to 8 teams. What
teams were dropped? How did the teams that were
contracted come to join the NL in the first place? What
happened to competitive balance after they were
dropped?
11. Does contraction necessarily mean that competitive
balance will improve? Use the definition of the mean and
standard deviation of winning percent as a guide. (You
might also think about your answer to question 10 as
well.)
12. The table shows the Years Per Title Data for all of the
NBA teams not displayed in Table 6.4:
What additional insight does this table give on
championship imbalance? For example, how many teams
really have been out of contention, relative to the total
number of teams in the league in each year? (Hint:
Remember to include all teams in this table and Table
6.4.)
13. Why is the minimum payroll in the cap system important
from the perspective of improving competitive balance?
What are the two reasons that a cap may not be very
effective? How can these be overcome?
14. What arguments can you muster against the idea that
arose when the new location of the Montreal Expos was in
doubt and some argued they should be moved to New
York? (Hint: You may need to revisit the Learning
Highlight: Competitive Imbalance and the Power of
Economic Competition.)
15. Let’s investigate the implications in the Learning
Highlight: Competitive Imbalance and the Power of
Economic Competition a bit more rigorously. Starting
from Figure 6.3, how does entry into the larger-revenue
market alter the marginal revenue functions? What
happens to the price of talent and, subsequently, to the
quality choice by the smaller-revenue market owner?
What is the predicted impact on competitive imbalance?
244

SECTION 10
Advanced Problems
1. What is the “right” level of competitive imbalance? Your
answer to Thought Problem 2 suggests that competitive
imbalance is too large relative to the level that would exist
under freer entry and exit into pro sports leagues. But
there can also be too much of a good thing. Using your
basic economic training about comparing marginal
benefits and marginal costs, portray the efficient level of
competitive balance. Can you describe a situation where a
perfectly balanced league is efficient?
2. Compare pro sports team quality to, say, the quality of
opera (your author’s favorite nonsports entertainment).
We don’t get upset when larger-revenue markets have
higher-quality opera alternatives (indeed, most would
agree that it would be a waste if the New York
Metropolitan Opera had its home base in a small town).
So why do we get so involved with the variation in team
quality across the cities in a league?
3. Imagine a league is comprised of a larger- (L) and
smaller-revenue (S) team. WL and WS are their
respective winning percents, and their marginal revenue
functions are (values in millions of dollars):

a. What is the equilibrium price of winning percent?
Show it graphically as well.
b. What will be the values of WL and WS in equilibrium?
Show them graphically as well.
c. What will be the equilibrium total talent bill for each?
Identify them graphically as well.
4. The data show that revenues are typically significantly
positively correlated with both payrolls and winning
percents in MLB. Is the same true in, say, the NBA?
(Warning: While the question is short, the time involved
in finding the answer won’t be!)
5. Examine the NFL’s newest gate revenue sharing
arrangement. What were the impacts on larger- and
smaller-revenue market owners? Which do you think
favored the change? Explain fully.
6. Examine MLB’s newest local revenue-sharing
arrangement. What were the impacts on larger- and
smaller-revenue market owners? Which do you think
favored the change? Why?
245

7. According to Table 6.1, revenue imbalance jumped
dramatically in MLB’s AL and NL, the NFL, and the NHL
in the 1990s. What can explain this? What was different
about the NBA where revenue imbalance remained
practically unchanged? What happened to winning
percent imbalance in each league over the last decade?
Does this pose a problem for the application of theory to
the data?
8. Return to our two-team league described in Advanced
Problem 3. With the same marginal revenue functions:
a. What if a luxury tax is imposed of $45,000 on each
point (0.001) over 0.500. What is the new price of
winning percent?
b. What are the new WL and WS in equilibrium?
c. What are the new equilibrium talent bills for each
team?
9. Baseball’s Blue Ribbon Panel reached the following
conclusions (reiterated in the MLB Updated Supplement,
December 2001, p. 3). Using the summary conclusions
from the text, based on the data in Table 6.1 through
Table 6.4, comment on these findings:
a. Large and growing revenue disparities exist and are
causing problems of chronic competitive imbalance.
b. These problems have become substantially worse
during the five complete seasons since the strike-
shortened season of 1994 and seem likely to remain
severe unless Major League Baseball undertakes
remedial actions proportional to the problem.
c. The limited revenue sharing and payroll tax that were
approved as part of MLB’s 1996 collective bargaining
agreement with the Major League Baseball Players
Association (MLBPA) have produced neither the
intended moderating of payroll disparities nor
improved competitive balance.
10. A recent Sports Business Journal editorial (July 12,
2004, p. 30) listed five major unresolved problems for
MLB. The fourth is fascinating:
Ensure that revenue sharing isn’t profit-taking. Owners
on the receiving end of revenue sharing ought to be
putting the money back into their teams, not into their
pockets. Spend it on payroll, on scouting or on
marketing, but spend it on making the organization more
competitive. That benefits both the individual team and
the business of baseball overall.
Suppose that revenue sharing is chosen in such a way that
the desired level of competitive balance is achieved. Using
Figure 6.3, show what happens if owners actually follow
the editorial suggestion. What else do you expect owners
will do with the shared revenues?
246

SECTION 11
References
Fleisher, Arthur A. III, William F. Shughart II, and Robert D.
Tollison. “Ownership Structure in Professional Sports,” in
Research in Law and Economics, vol. 12, ed. Richard O. Zerbe.
Greenwich, CT: JAI Press, 1989.
Fort, Rodney. “Competitive Balance in North American
Professional Sports,” in Handbook of Sports Economics
Research, ed. John Fizel. Armonk, N.Y.: M.E. Sharpe, Inc.,
2006, pp. 190–206.
Fort, Rodney, and James Quirk. “Cross-Subsidization,
Incentives, and Outcomes in Professional Team Sports
Leagues,” Journal of Economic Literature 23 (1995): 1265–
1299.
Kesenne, Stefan. “League Management in Professional Team
Sports with Win Maximizing Clubs,” European Journal for
Sport Management 2 (1996): 14–22.
Kesenne, Stefan. “Player Market Regulation and Competitive
Balance in a Win Maximizing Scenario,” in Competition Policy
in Professional Sport: Europe after the Bosman Case, eds.
Stephan Kesenne and Claude Jeanrenaude. Antwerp:
Standaard Uitgeverij, 1999.
Kesenne, Stefan. “Revenue Sharing and Competitive Balance
in Professional Team Sports,” Journal of Sports Economics 1
(2000): 56–65.
Levin, Richard C., George J. Mitchell, Paul A. Volcker, and
George F. Will. The Report of the Independent Members of
the Commissioner’s Blue Ribbon Panel on Baseball
Economics. New York: Major League Baseball, 2000.
Quirk, James, and Rodney D. Fort. Pay Dirt: The Business of
Pro Team Sports. Princeton, NJ: Princeton University Press,
1992.
Quirk, James, and Rodney D. Fort. Hardball: The Abuse of
Power in Pro Team Sports. Princeton, NJ: Princeton
University Press, 1999.
Scully, Gerald. The Business of Major League Baseball.
Chicago, IL: University of Chicago Press, 1989.
Staudohar, Paul D. “Salary Caps in Professional Team Sports,”
in Competition Policy in Professional Sport: Europe after the
Bosman Case, eds. Stephan Kesenne and Claude Jeanrenaude.
Antwerp: Standaard Uitgeverij, 1999.
Szymanski, Stefan. “The Economic Design of Sporting
Contests,” Journal of Economic Literature XLI (December
2003): 1137–1187.
247

Szymanski, Stefan.  2004.  “Professional Team Sports Are
Only a Game:  The Walrasian Fixed-Supply Conjecture Model,
Contest-Nash Equilibrium, and the Invariance Principle.” 
Journal of Sports Economics 5(Number 2 May 2004):111-126.
Utt, Joshua, and Rodney Fort. “Pitfalls to Measuring
Competitive Balance with Gini Coefficients,” Journal of Sports
Economics 3 (November 2002): 367–373.
248

SECTION 12
Suggestions for Further
Reading
Surely, I can come up with some.
249

CHAPTER 7
The Value of Sports
Talent
I don’t think anyone’s worth this type of
money, obviously. But that’s the market that
we’re in today.
—Alex Rodriguez, MLB Player
Upon signing at over $20 million per year.
CNN.com, Alex Rodriguez Profile.
CHAPTER OBJECTIVES
After reading this chapter, you should be able to:
• See that marginal revenue product theory
provides a general and insightful explanation of
pay and hiring in sports.
• Explain why the winner’s curse explanation of
pay in sports does not fit any real-world sports
situation.
• Explain why the bidding war explanation of
player pay only holds as a few teams approach
the play-offs.
• See that the winner-take-all logic might explain
individual sport outcomes but that it does not
explain player pay in team sports.
• Recognize the variety of impacts that
discrimination can have in sports labor markets
and appreciate the limits of economics in the
analysis of gender and racial discrimination in
sports pay and hiring.

SECTION 1
Introduction
The first NFL draftee, selected by the Chicago Bears, was the
1935 Heisman-winner Jay Berwanger of the University of
Chicago (then a member of the Big Ten Conference).
However, Berwanger did not sign with the Bears because the
pay they offered could not beat his starting salary with a
prestigious degree! Things have changed since then. In 2009,
the number-one draft pick, quarterback Matthew Stafford,
signed for an average of $12 million for six years, with $41.7
million guaranteed! There is also a chance for an additional
$6 million under playing incentives in the contract.
Many look at the market for athletic talent and shake their
heads—they see both outlandish levels of pay for players and
hiring discrimination in coaching and in the front office. In
this chapter, we will explain the workings of the market for
sports talent. The tried and true marginal revenue product
(MRP) explanation gains us the most yards in terms of
explaining talent market outcomes. We will also use the MRP
explanation to clarify a few misconceptions, for example, the
idea that salary increases lead to increased ticket prices. In
addition, we will analyze three alternatives to the MRP
explanation of player pay: the winner’s curse, bidding wars,
and winner-take-all logic. Although these are appealing
explanations, we will see that they actually explain very little.
Finally, we gain some insights about race and gender
discrimination using the MRP explanation. But these results
are limited to the market outcomes for which the MRP
explanation was designed.
251

SECTION 2
The Value of Athletes
Let’s start the chapter by documenting just how large player
earnings are. Summary data and averages by sport, including
salaries and total earnings, from the SI.com’s “Fortunate 50”
for 2008 are reported in Table 7.1. For most of us, these are
unimaginable figures. The highest salary in Table 7.1 is $29
million for baseball player Alex Rodriguez. Earning an
average of $100,000 per year—a respectable wage—a person
would have to work 290 years to match that single-year
figure. And that isn’t even the highest contract offered to a pro
team sports player. Michael Jordan was offered $33.14
million ($43.4 million) by Chicago Bulls owner Jerry
Reinsdorf to play the single 1998 season—and Jordan turned
that down in favor of retirement (Sports Illustrated, April 27,
1998). With total costs of annual attendance in the area of
$40,000 or more at top universities, the amount Jordan was
offered in 1998 would be enough to put over 200 students
through four-year degree programs at prestigious universities.
In addition, for many athletes, salaries or earnings are the
smaller part of their overall earnings (just check out Tiger
Woods, LeBron James, and Dale Earnhardt, Jr.). The huge
values for players in America and abroad is also indicated by
so-called “transfer payments” from one owner to another that
allow a player to leave his or her previous contract. Just
recently, the Boston Red Sox paid over $50 million to obtain
the rights to pitcher Daisuke Matsuzaka. According to various
popular sources, Real Madrid paid $45.6 million ($54.9
million) just to acquire the rights to Zinedine Zidane from
Juventus of Italy in 2001, a record that still stands in world
football.
SALARIES
Table 7.2 shows descriptive salary statistics for all four major
pro sports leagues for the past decade (and a bit longer for
MLB). As in other cases in the text, the table is cluttered
enough reporting results all adjusted to 2009 dollars. From
the data in Table 7.2, you can see that there have always been
top-end skewed salaries in all sports. First, the median is
always dramatically less than the average salary. Second, the
top salaries have been more than ten times the median since
the 1980s. For example, in 2009, the top-heavy ranking is as
follows: Alex Rodriguez at $28.0 million in MLB, just over 25
times the median; Kevin Garnett at $24.8 million in the NBA,
eight times the median; Ben Roethlisberger at $27.7 million in
the NFL, nearly 32 times the median; and Dany Heatley at
$10.0 million in the NHL, almost eight times the median.
The data in Table 7.2 also show extraordinary salary growth
over time. The average has grown steadily in real terms over
the decades. The rate of growth shown in Table 7.2 for football
is nothing short of astonishing because the average rate of
growth in the U.S. economy is typically from 2.5 percent to 3
252

percent. Baseball during the same decade did about the
national average, falling off from stratospheric growth of
about 9 percent at both the median and average over the
decade of the 1990s.
The data in Tables 7.1 and 7.2 also raise a number of
interesting questions. How can athletes be worth this much?
What happened to cause the jumps in salaries in various
sports in different years? We will also look at why players of
different sports are paid less than others later in this chapter.
Clearly, from Table 7.2, NBA superstars are paid the most in
the current decade, followed by baseball and football players.
NHL superstars are paid the least. We will try to answer these
questions in this chapter. However, let’s first look at what else
we can deduce about sports compensation.
ENDORSEMENTS
On top of salaries, many players earn substantial
endorsement income. That is the reason for the difference
between salaries/winnings and total earnings in Table 7.1.
Deduced from Table 7.1, golf pro Tiger Woods had the highest
endorsement income among active athletes at $105 million.
Basketball’s Kevin Durant dominates the rest of the sports at
$21.7 million. Perhaps the all-time greatest sports
endorsement giant is former heavyweight champion of the
world, George Foreman. He signed a deal with Salton grills for
$137.5 million over five years earlier in the decade. A Sports
Business Journal (October 9, 2000) survey shows that
endorsement earnings typically double the income of male
superstars. For female superstars, the impact is staggering.
WNBA players such as Sheryl Swoopes and Lisa Leslie added
10 times their sports earnings through endorsements. Skater
Michelle Kwan quintupled her income, and tennis players
Serena Williams and Martina Hingis added four times their
sports earnings with endorsements.
COMPARISONS WITH OTHER ENTERTAINERS
Even though these figures seem astronomical, when you look
at Table 7.3, which lists the Forbes top celebrity and sports
star earnings of 2008, with the exception of Tiger Woods,
sports superstars do not reach the lofty compensation heights
enjoyed by other entertainment superstars. Even though
active athletes make the list, only Tiger Woods makes the top
10, but even he earned only about half of what Harry Potter
magnate J.K. Rowling earned in 2008. For a final bit of
perspective, in 2008 the entire New York Yankees team, the
highest-paid team in all of sports, earned $209 million, about
70 percent of J.K. Rowling’s earnings for that year. Even
though sports superstars are paid handsomely, they are not
even in the same league with other entertainment superstars,
generally speaking.
Enough said about the impressive earnings of sports stars.
What is the answer to the first question: How can athletes be
worth so much? There are a few competing explanations for
how sports stars get paid. Let’s look at the MRP explanation of
input payment. After all, sports talent is just an input to the
sports production process.
253

SECTION 3
The Marginal Revenue
Product Explanation
In almost all spheres of economic endeavor, payments to
inputs, such as sports talent, are determined by the inputs’
marginal revenue product, or MRP for short. Remember from
Chapter 4 that talent is hired to produce winning percent in
the long run. Given that, MRP for sports talent is defined as
follows:

W is the level of team winning percent. MP(W) is marginal
product, or the player’s contribution to winning percent. For
example, from Table 4.6, a team currently with five stars and
a winning percent of 0.528 could raise its winning percent by
0.070 to 0.598 by adding another star. MR(W) is the marginal
revenue generated by the player’s contribution to winning.
Again, revenue at the margin will depend on the level of
winning percent that is being added to the level W. In sports,
where market power leads to downward sloping demand
functions, MR decreases with output. Therefore, the MR
function slopes down for the individual team at any level of W
that might be chosen in the long run. But what does MRP
represent? Essentially, MRP is the input’s contribution to the
revenues earned by the team owner. In Chapter 4, you
calculated the hypothetical cost of winning in Table 4.7. MRP
is the other side of the comparison. This is the essential
decision made by all producers: Discover the cost of winning
and compare it to those players’ MRP in order to determine
whether they are worth hiring.
One of the major contributions of sports economics is the
actual calculation of MRP. Once again, we return to one of the
true pioneers in sports economics, Gerald Scully. In Chapter
4, we discussed his contribution to the analysis of coaches and
managers. Scully (1974) also devised a way to calculate the
value of a baseball player’s MRP in his “Pay and Performance
in Major League Baseball.” In a nutshell, Scully’s approach to
MRP was to note that (1) playing statistics are used by
managers to create wins, (2) each player has a bundle of
different playing statistics that can be valued in terms of
contributions to winning, and (3) wins are sold to fans. Simple
multiplication then yields a player’s MRP.
In more detail, Scully’s first step is to estimate the
contribution of measured player statistics in the production of
wins. The relationship for a given team is as follows:

Wins are the objective and depend on H = hitter performance
and P = pitcher performance, holding W = winning percent
(team quality) constant. W must be held constant since
marginal impacts of improved team performance are different
254

for stronger teams than for weaker teams (Chapter 4).
Further, while Scully used a variety of measures of H and P,
we will stick to just one for each type of player for easier
presentation. Assuming a simple linear relationship gives:

Under this simple relationship, b1 is the definition of ?Wins/
?H for hitters and b2 is the definition of ?Wins/?P for
pitchers. For example, for hitters, better slugging average
should increase wins so that one expects b1 > 0. For pitchers, a
higher power-pitching ratio like strike outs divided by walks
should increase wins so that one expects b2 > 0 as well. Scully
used multiple regression techniques to obtain estimates of
and b1 and b2. To keep them separate, let’s label estimates as
b*1 and b*2.
Scully’s next step is to find the contribution to winning by a
player, rather than by player statistics. Scully multiplied
individual player statistics by their marginal impact on
winning. If a given hitter had hitter performance statistic H,
then the MP of that hitter (rather than the statistic itself) is b*1
× H. But this is just the player’s contribution to team wins.
Since quality is held constant, this is the definition of the
player’s MP. Thus, for hitters, b*1 × H = MP(W). Similarly, for
pitchers, b*2 × P = MP(W).
Since , and the previous steps determine MP(W), Scully’s last
step was to determine MR(W). From Chapter 2, revenues
follow from careful assessment of demand. Generally,
according to the determinants of demand, revenues are

where R = revenues, T = ticket price, I = income, S = a
measure of available substitutes, E = a proxy for expectations
and, of main interest, Wins are held constant. A simple linear
relationship is:

While the rest of the results are also of interest, focusing on
the task at hand, the important definition is ?5 = ?R/?Wins,
the MR from winning. Thus the estimate of ?5, ?*5, is the
estimate of MR(W). Multiplying the preceding results by this
estimate then gives MRP, that is ?*5 × b*1 × H for hitters and
?*5 × b*2 × P for pitchers.
Later, in The Business of Major League Baseball (1989), Scully
shows that a one-point increase in winning percent raised
1984 revenues by $31,696 ($65,039). Because a win is 6.2
winning percent points, a win in 1984 was worth $196,515
($403,242). A solid slugger would add 63 points, or about 11
wins. The MR from just those games in 1984 was worth $2.2
million ($4.4 million). A strong pitcher can add 20 net wins,
or $3.9 million ($8.1 million) in extra revenue to an owner
holding everything else about the team constant. These are
precisely the elements that determine the MRP of players.
255

In a competitive talent market, we expect that players get paid
pretty close to their MRP. In all pro sports leagues, rules
agreed upon by owners and players through collective
bargaining (covered in Chapter 9) allow players to sell their
services to the highest bidder after a fixed number of years. In
MLB, for example, it is six years. After that period, players are
free agents, and they can sell their services to the highest
bidder. The NHL has the most cumbersome formula for free
agency, where free agency is determined both by experience
and age. Of course, players become free agents only if they are
not bound by a long-term contract to their current team. In
this free-agency setting, we would expect players to play
where they will be the happiest. Most often, this is the place
where they will be paid the most.
In order to get players to move from their current team, an
owner will have to offer more than they make with that team.
If competition over the player is brisk, then the eventual
payment will be between the highest- and second-highest
offer of the two top-bidding owners. Of course, competition
works on both sides of the market. If good substitutes exist for
any given player’s services, one would expect this competition
to dampen the size of the payment that would entice the
player to change teams. Even though this amount will not be
quite the entire MRP at the highest-valued location, it will be
between that value and the second-highest value across the
entire league.
If you have caught on to the MRP idea, then you can answer
one of the questions posed in the last section. Why do NBA
players make more than, say, NFL players? If you answer that
it is so because NBA players play about 90 games (including
the preseason) whereas NFL players play about 18 games
(including the preseason), you need to think again. The MLB
season is twice as long as the NBA season, but MLB players
earn less than NBA players, as shown in Table 7.2.
According to the MRP theory of player pay, it must be either
because an NBA player has higher MP on a basketball team
than an NFL player has on a football team or because NBA
fans are willing to pay more for added winning so that MR is
higher or both. Both explanations do seem reasonable. With a
smaller number of players on NBA teams than on NFL teams,
it is possible that the contribution to team winning is larger
for each individual NBA player than for each individual NFL
player. The MR collected by NBA owners also may be larger if
the demand by NBA fans is greater. The MRP deck seems
stacked in favor of a given NBA player earning more than a
given NFL player.
CASE STUDY: THE BARRY BONDS SHOW, 2001
In 2001, Barry Bonds of the San Francisco Giants was paid
$10.3 million ($12.4 million). He was the 22nd highest-paid
player in the league, in the top 3 percent. That year, Bonds hit
73 home runs, eclipsing the single-season home run record of
70 held by Mark McGwire. Bonds’ record chase didn’t seem to
draw the same press enthusiasm as the earlier run by
McGwire in 1998, but fans around the league were interested,
256

especially in San Francisco—and to the tune of quite a few
million dollars for the owner of the Giants.
It’s pretty easy to see that Bonds’ record chase was worth
about the same amount to the Giants’ owner as winning the
division and considerably more than Bonds was actually paid.
In 1999, the Giants finished second in the National League
West Division and drew 2,078,365 at the gate. In 2000, the
team won their division, and attendance increased to
3,315,330. In 2001, Bonds’ record-breaking season, the Giants
again finished second but still drew 3,311,958. When the
Giants won their division, they enjoyed an attendance
increment of 1,236,965. Suppose that the Giants would still
have drawn 2,078,365 finishing second in 2001 without
Bonds’ record. If so, then Bonds’ record-breaking
performance added 1,233,603 fans. In their “Fan Cost Index”
resource, Team Marketing Report (www.teammarketing.com)
lists the Giants’ average ticket price in 2001 at $19.24
($23.17). The fan cost index itself is given as $163.89
($197.41) for two adults and two kids. The additional
attendance was worth conservatively $23.7 million ($28.6
million), just multiplying the average ticket price by the
attendance increment. But the value could have been upward
of $50.5 million ($60.8 million), dividing the increment by
four and multiplying by the Fan Cost Index.
Here’s the kicker: Because seasons like this do not come along
even once in a lifetime, it is unlikely that the owners of the
Giants had built the record chase into Bonds’ contract to cover
2001. This means that Bonds’ contribution to these
unexpected revenues was over and above what the owners of
the Giants thought he probably would be worth, on average,
for the 2001 season, and we haven’t included any revenues
beyond those at the stadium. This bonus to the owners makes
it clear that Bonds was worth his $10.3 million ($12.4
million), and then some in 2001.
The value of the Barry Bonds Show also spilled over to the rest
of the league. Prior to a visit by Bonds and the Giants toward
the end of the season, the San Diego Padres hosted the
Arizona Diamondbacks (August 31 through September 2) and
then the St. Louis Cardinals (September 3 through September
5) in a seven-game home stand (there was one double-
header). The total paid attendance for both series was
154,013, or about 22,002 fans per game. Just after those seven
games, in a three-game series against the Giants, 156,183 fans
attended the San Diego–San Francisco games. That is 52,061
on average, per game, or an additional 30,059 per game over
the immediately preceding series. Again, the Team Marketing
Report average ticket price for San Diego was $13.74 ($16.55),
and the Padres’ fan cost index was $127.46 ($153.53). So the
conservative estimate of the Barry Bonds Show’s value to the
owner of the Padres was $413,013 ($497,482) per game or
$1.24 million total ($1.49 million), and the top end could have
been as much as $957,830 ($1.2 million) per game and $2.87
million total ($3.46 million). Even with any other variation
that might have occurred, such as the weather or a
particularly interesting visiting team, it is clear that the
257

Team Marketing Report

Team Marketing Report

increase in revenue can be attributed mainly to the history-
making performance of Barry Bonds.
THE EXPERIENCE–EARNINGS RELATIONSHIP
MRP develops over time. In economics, this is called the
experience–earnings relationship. MRP on all jobs starts low,
increases rather quickly, tops off, and then may decline as
skill diminishes or health declines with age. In addition, a
given bundle of skills is worth more in some places than
others through time. As a result, workers move,
geographically, toward this higher value. This applies to
sports players as well.
Table 7.4 shows data on experience, performance, and
earnings for a particular MLB pitcher, Randy Johnson. The
table shows that Johnson’s lifetime record is 303 wins and
166 losses, a 0.646 winning percent. His per-game earned run
average (ERA, or on average the number of runs he would
usually give up in nine innings) shows a typical pattern—
settling around 3.00 at the beginning and improving over
time but then rising again as he reaches the end of his career.
Of particular interest to us, however, is the way that pay and
experience are related.
Figure 7.1 charts Johnson’s pay in 2009 dollars as compared
to his experience; this is called an experience–earnings
profile. It is clear that the relationship between pay and
experience is as we would expect with any job. But there is a
special sports twist. Through his first three full MLB years
(1989–1991), Johnson’s salary was very low. In the third year,
something special happens for baseball players. They become
eligible for arbitration (covered in detail in Chapter 9). If a
player and team owner cannot reach an agreement in the
third year, then the decision may go to an independent
arbitrator, who must choose either the amount the player is
asking or the amount offered by the owner. The effect, even
for players who do not go to arbitration, is a dramatic increase
in salary. Johnson’s salary increased nearly fourfold in the
fourth year, from $549,500 to $2.1 million. Apparently, his
improvement after that was sufficiently impressive that the
Mariners’ owners increased his salary steadily. Johnson’s road
to fabulous riches hit a bump in 1998. He couldn’t reach an
agreement with the Mariners’ owners and took a short stay
with the Houston Astros. But immediately he was grabbed by
the Arizona Diamondbacks at 1.6 times his previous salary.
Ultimately, he became the highest-paid pitcher in MLB. But,
as with all earnings profiles, and at the advanced baseball age
of 45 years, it appears that Johnson is in his declining
earnings period since 2007.
People are often surprised that the pattern of earnings
(although not the level) over time for pitchers is similar to
that of any other job. This pattern of earnings and experience,
except for the level of pay, looks just like the experience–
earnings profile of any other employee hired on the basis of
MRP, from starting at minimum wage on through middle-
earnings years and ultimately to peak years. The similarity of
this profile with those working in other jobs adds another
258

piece of evidence in favor of the MRP explanation of pay in
pro sports.
THE MRP OF COACHES AND MANAGERS
The MRP explanation works for other pro sport talent as well.
Red Auerbach, coach of the NBA Boston Celtics during the Bill
Russell dynasty years, earned $10,000 ($82,000) for the
1950–1951 season (Sports Illustrated, May 19, 1997, p. 57).
ESPN.go.com reported (June 16, 2005) that Phil Jackson’s
2007–2008 contract paid him about $10 million ($10.3
million). Adjusted for inflation, that’s a 126-fold increase in 56
years in real terms, a growth rate of 9 percent (three times the
typical growth rate in the economy at large). How can this
tremendous increase be explained?
In Chapter 4, the work of Porter and Scully (1982) and Scully
(1989) on managerial efficiency was described. Essentially,
managers become more efficient the closer they get to the best
that managers with similar talent have ever done in terms of
winning. One would expect, by the MRP theory of pay, that
better managers would make more. This work shows that
there is some evidence that managers who make the most of
their talent do get paid more.
However, this dramatic increase in pay for Jackson over
Auerbach was not due to the relative success of the two
coaches. Auerbach had nine NBA championships with the
Celtics, while Jackson has 10, combined, with the Bulls and
the Lakers. Part of the difference is that Jackson’s job
description included executive responsibilities that Auerbach
did not have as a coach. The rest of the difference, according
to MRP theory, simply follows from the dramatic increase in
fan willingness to pay for NBA basketball. Nowhere is there a
better example that the value of an input has risen because the
MR aspect of MRP has skyrocketed. We saw in Chapter 2 that
the NBA is now a $3 billion industry. Even though today’s
coaches may not be any better than their counterparts in the
past, what they are doing has simply become much more
valuable over time.
259

SECTION 4
Equilibrium in the Talent
Market: Graphical Analysis
Our conclusion thus far is that MRP theory provides a general
explanation of pay in pro sports. It may be hard to believe that
the MRP of coaches and players can be so high, but it is. As
long as competition for coaches and players is brisk, they will
be paid close to the value of what they produce (in Chapter 8,
we will see what happens when the players’ labor market
competition is limited). Owners are forced to pay up to nearly
the highest MRP of players across all teams in the league in
the presence of strong competition over talent.
We can determine the theoretical equilibrium of winning
percents and talent prices in a league using a graphical
analysis of a competitive talent market. We will use the same
graphical tool that we used in Chapter 6, namely, the two-
team league with an exaggerated large-revenue market and a
small-revenue market. Remember our other assumptions.
First, we have a league where the talent supply is fixed (our
closed league) so that the sum of winning percents must be
equal to half the number of teams in the league (that is, the
sum must be one for a two-team league). Second, talent is
measured so that one more unit of talent increases winning
percent by one unit, that is, by 0.001. Thus, the x-axis
measures both winning percent for the teams and the amount
of talent it takes to generate that winning percent so that the
MR functions are also MRP functions. This means that we can
find both the winning percent for teams and their talent
choice at the same time. We begin with Figure 7.2, market
equilibrium.
From Chapter 6, we know that equilibrium can only occur at
MR*L(W) = MR*S(W) so that W*L = 1 − W*S in Figure 7.2.
Because of our convention of measuring talent in units that
generate one more winning percent point, the price of
winning and talent are both equal to P. Therefore, in
equilibrium, MR*L(W) = MR*S(W) = P. P becomes the price of
one unit of talent. We also have our three important features
of a competitive talent market equilibrium from Chapter 6:
• Marginal revenues are equal across teams, that is, MR*L =
MR*S.
• Revenue imbalance means that there will be payroll
imbalance.
• Revenue imbalance means that there will be competitive
imbalance.
Given this result, it is easy to see how much each owner pays
in total for the winning percent that they choose. The bill for
the larger-market owner is the left-hand shaded rectangle in
Figure 7.2, calculated as P × W*L, that is, the price of each unit
of winning multiplied by the number of units of winning. This
amount also represents the payment to talent because we
260

have measured talent in units that produce each unit of
winning percent. Thus, the left-hand shaded rectangle in
Figure 7.2 also represents the total payment to talent hired by
the larger-revenue market owner. Similarly, the right-hand
shaded rectangle in Figure 7.2 is the smaller-market owner’s
talent bill, calculated as P × W*S. The clear conclusion is that
the larger-revenue market owner spends more for talent than
does the smaller-revenue market owner. Therefore, along with
competitive imbalance, there will also be payroll imbalance as
long as MRs are not equal.
The second important feature of the competitive talent
equilibrium is that the larger-revenue market team has a
higher winning percent than the smaller-revenue market
team, that is, in equilibrium, W*L > W*S in Figure 7.2. Of
course, this makes sense. Every single unit of talent hired
produced winning that was more valuable to the larger-
market owner. Therefore, that owner hires more talent and
wins more. It is this important feature that leads us to the
conclusion that competitive imbalance is a fact of life as long
as there is revenue imbalance.
261

SECTION 5
Real-World MRP Insights
MRP theory provides insights into a number of important and
(otherwise) confusing sports outcomes. The very first insight
is that our description of the competitive equilibrium at the
end of the last section actually does portray outcomes in real
sports leagues. Table 7.5 shows the payrolls for the 2008
seasons for all major league teams. The values in Table 7.5 are
left in their nominal levels because the table is cluttered
already, and we really only wish to compare teams and
leagues for that year rather than over time. MLB is the only
league without a cap on team payrolls (as we saw in Chapter 6
and will discuss further in Chapter 9). MLB shows the largest
imbalance comparing the top team to either the bottom team
or the median team. The Yankees’ payroll is 9.6 times larger
than the Marlins’ and 2.6 times larger than the median league
payroll (e.g., the Brewers or the Indians). In baseball, the top
payroll teams are often the best teams. Interestingly, and in
keeping with our observations about the ineffectiveness of
payroll caps, New York teams in the NBA and NHL have
payrolls nearly twice the size of their respective league
minimums. Football is not much different, but it is interesting
that in this particular year, the smaller-revenue market
Raiders top the list (with Dallas in second).
A second real-world observation is that, in every sport,
reporters usually take a shot at owners by pointing out the
“busts and bargains” of the year. Busts are highly paid players
performing below their usual level, whereas bargains are less-
expensive players having unexpectedly good years. Reporters
are especially unkind to owners burdened with busts when
other teams in the leagues have a few bargains. Is the talent
market subject to consistent owner mistakes, as reporters
would have us believe? Or are reporters just selective in their
presentation of bargains and busts, playing to the analytical
weaknesses of fans in general?
Actually, these reporters are just pointing out that, in an
uncertain world, some players will perform at levels above
their current payment and some will perform at levels below
their current payment. On average over time, one would not
expect these outcomes to be predictable. Owners who
repeatedly hire busts in the talent market will either absorb
large losses or lose their team to a more astute owner. If some
owners really do make systematic mistakes, another owner,
with the same information, would make predictions closer to
the actual performance of players.
In this light, players are paid their expected MRP and, in an
uncertain world, on average, owners have unbiased
expectations about eventual performance. Thus, if you watch
a given team over time, you would expect that team’s bargains
and busts to offset each other. Across an entire league, this
would almost certainly be the case. If you still have trouble
relating to this, try the following. At the beginning of the
262

season (not the end), make your predictions on bargains and
busts. Check your predictions at the end of the season. If you
do better than owners over a season or two, then there were
systematic errors (and you have a bright future as a general
manager!). If not, then it is reasonable that owners make
unbiased estimates and, hence, do not make systematic errors
about uncertain player performance.
THE RELATIONSHIP BETWEEN TICKET PRICES AND
SALARIES
MRP theory also helps unravel the faulty logic behind the idea
that higher player salaries drive up ticket prices. How often
have you heard this one? Greedy players demand higher
salaries, so owners, eager to retain their talent, raise ticket
prices. This type of argument clearly casts players as the
villains behind the increase in costs to fans.
However, MRP theory dictates that players can only earn
more if (1) they become more productive (i.e., MP increases)
or (2) fans increase their willingness to pay for the result. An
increase in fan willingness to pay would lead to an increase in
the MR portion of MRP, even though the MP part of MRP
stays about the same (just as we saw with coaches in the last
section).
If this is the effect, then players are more valuable than they
used to be in the eyes of fans. This means that owners will
raise prices because fans are willing to pay more. Then, player
pay will rise as long as there is competition for the players’
services between teams. If one team does not increase pay and
the revenue is there among a few teams, then others will bid
that talent away causing salaries to increase. Therefore, the
reason players make more is that a change in some demand
parameter (e.g., income, population relevant consumption
alternatives, and preferences) has increased fans’ willingness
to pay.
The data on payrolls and ticket prices bear out this theoretical
prediction. Figure 7.3, Figure 7.4, Figure 7.5, and Figure
7.6 show how ticket prices and payrolls (both adjusted for
inflation to 2009 dollars) behaved for the pro sports teams in
the Boston area since 1990 (1994 in hockey due to unavailable
payroll data). Of course, the first thing to note is that ticket
prices in real terms did not increase uniformly over the time
period studied for these four teams. Ticket prices in baseball
rose steadily to 2002 but have been nearly flat after that
(Figure 7.3). In basketball, while prices have risen generally,
there was a four-year period of decline, 1999 through 2002,
and again in 2004 and 2005 (Figure 7.4). Turning to football,
again, prices have clearly risen over time in general. However,
years of decline outnumber years of increase by three (Figure
7.5)! Finally, after an early rise, if anything hockey ticket
prices have declined since 1995 (Figure 7.6).
As for the general relationship between ticket prices and
salaries, as judged by total payroll, Figures 7.3 through 7.6
show the following. In baseball, the overall relationship is
strong. The correlation between payroll and ticket prices is
0.96. However, from 2002 on, payrolls are a roller-coaster
263

ride while ticket prices are flat; the correlation changes to –
0.24! The overall correlation for basketball is also pretty
strong at 0.75. But, again, it is easy to find stretches where the
two appear almost unrelated—the two series move in opposite
directions in six of the season in Figure 7.4. The same story
goes for football. The overall correlation is 0.71, but opposite
behavior occurs in 13 of the charted seasons. Hockey is the
best example that salaries do not “cause” ticket prices. The
overall correlation is –0.25 to begin with, negative and
probably not significant.
In summary, if salaries cause ticket prices to change, then
they should almost always be moving in the same direction. In
the figures, it is clear that salaries move in the opposite
direction of ticket prices as often as they move in the same
direction as ticket prices. In fact, for the NFL’s Patriots,
salaries and ticket prices move in the opposite direction over
two-thirds of the time.
MRP theory dictates that players foresee larger contributions
to revenues and ask for corresponding increases in salary.
Judging from the correlations, the increase in MRP in
baseball, basketball, and football is observed partly at the
gate. But other increases in MRP, especially in hockey, must
be coming from a different impact on revenue, for example,
impacts on the value of broadcast rights.
This was especially true of one particular episode for MLB
teams in general. The explosion of player salaries in the 1980s
was due to the proliferation of media providers, especially
local and regional cable. The MR part of MRP, and as a result
salaries, rose as media providers entered the scene with
increased payment of broadcast rights fees. Therefore, the
answer to our earlier question, “Why did MLB salaries rise
steadily and in a huge cumulative amount during the 1980s?”
is media provider payments. I’ll say it again: Salaries do not
drive ticket prices, but revenues are one of the determinants
of how much players get paid.
SOCIAL VALUES: BALLPLAYERS VERSUS TEACHERS
MRP theory also helps us to understand some larger social
issues related to sports. Many people abhor the obsession
with sports in the United States. They deplore spending
precious resources that could otherwise go to more “worthy”
endeavors. How is it (they ask) that sports stars can be so
highly valued when other people working at these more
worthy endeavors are so poorly paid? Demand is certainly one
side of the explanation for sports outcomes. However, we
cannot forget the supply side, for in this particular case the
supply side explains the outcome.
Figure 7.7 shows hypothetical demands for teaching and
athletic services. Each of these demand functions is an MRP
function. The figure is set up so that the MRP of teachers lies
everywhere above the MRP of athletic services. In this
hypothetical example, the MRP functions are located
consistent with a society that values every single unit of
teaching services more than it values the services of sports
stars (willingness to pay is higher). This is the same thing as
264

saying that society generally agrees that teachers are “worth
more” than ballplayers. This really does appear to be how
society places this value. For example, earlier in this chapter
we saw how Barry Bonds was paid $10.3 million in 2001. But
the value per fan is really quite low. Even if we only count the
3,311,968 fans at the gate, that’s only around $3.11 per fan. If
we add television viewers, the value per fan falls dramatically.
But for teachers, the value per student is much higher even
though their salaries are much lower. The U.S. Department of
Labor gave the average salary of teachers in 2001–2002 as
$44,327 (it has since risen in 2006–2007 to $51,009
[$53,944]). With around 30 students per class, that means an
elementary school teacher was worth at least $1,478 per
student, or 475 times as much as Barry Bonds on a per
customer basis at the time Bonds broke the single-season
home run record.
Alas for teachers, there is plenty of quality teaching supplied
at low pay relative to pro athletes. This is portrayed in Figure
7.7 by a flatter supply function of teaching services relative to
the steeper supply of pro sports star services. The intersection
of teaching supply and demand results in a wage, WT, for
teachers. That wage is much less than the wage result at the
intersection of supply and demand for sports stars, WS. Thus,
the MRP of teachers can swamp that of sports stars generally,
but sports stars can still end up being paid more. It is
especially important to understand the underlying source of
the outcome if public resources are going to be aimed at
changing the outcome. Here, efforts should go toward
increasing the MRP of teachers or decreasing their supply if
one wishes to raise their compensation.
INTERNATIONAL COMPETITION FOR TALENT
Whenever the Canadian dollar is weak against the U.S. dollar,
owners of NHL teams in Canada argue that they cannot
compete for talent. Their argument is that Canadian NHL
owners pay the same salary as U.S. teams but in weaker
Canadian dollars. However, the weak Canadian dollar cannot
be the explanation for the inability of Canadian owners to
compete for talent; why don’t Canadian owners just simply
make up the difference with higher salaries in Canadian
dollars or just write player contracts in U.S. dollars? The
weakness of the Canadian currency is only a symptom of the
actual problem from the Canadian owners’ perspective.
The reason the Canadian dollar is weak is a macroeconomics
lesson, but the upshot is that the real income of sports fans in
Canada is lower than the real income of fans in the United
States. Thus, relative to U.S. fans, the demand for hockey by
Canadian fans is lower because willingness to pay is based on
the lower Canadian real income. The lesson from Chapter 2 is
that lower Canadian demand reduces the entire revenue
function for Canadian teams. The lesson in this chapter is that
Canadian players are paid according to their MRP. With
relatively lower revenues for Canadian owners (due to lower
relative Canadian incomes), the value of talent also declines
(the MR element in MRP is lower for Canadian owners). In
265

essence, some Canadian teams are playing in smaller-revenue
markets relative to some U.S. teams.
As long as relative real income is lower in Canada, owners of
teams in Canada will lose talent (at the margin) to U.S. teams
whose fans have higher real income. Owners of NHL teams
can only pay less, even if they do pay U.S. dollars, because
their fans’ incomes are lower. Losing talent to U.S. teams is
just the large-revenue market versus small-revenue market
problem, in an international context.
Differences in MRP also explain the international migration of
talent across formal leagues in different countries. For
example, outfielder Hideki Matsui earned a Nippon
Professional Baseball League record $4.7 million ($5.6
million) in 2002 (Sports Illustrated, January 7, 2002, p. 28).
But he left the Yomiuri Giants for MLB’s New York Yankees
for $6 million ($7 million) in 2003 and then $7 million ($7.9
million) in 2004. Of course, the reason players are worth so
much more in the United States is explained completely by
MRP. On just examining the situation at the gate only, it is
clear that Yomiuri Giants’ 2002 attendance of 3,783,500 was
similar to 2002 Yankee attendance of 3,465,807. But ticket
prices were dramatically higher in New York. The average
ticket price in Yankee Stadium was $24.26, while a general
admission ticket in the Tokyo Dome was $10.48. So there is
quite a large difference in revenues generated by players in
the United States even before we get to media revenues. Given
the greater player MRP in the United States, the migration of
top sports talent to the United States is of growing concern to
pro sports leagues abroad, not just in baseball but also in
basketball and hockey.
OVERVALUED ROOKIES?
Here is one last way that MRP theory can provide information
on a typical confusion about sports markets. How can rookies
be worth their huge bonuses and starting salaries? Rookies
are untested at the professional level, and they often are paid
more than established veterans. Let’s look at an example,
LeBron James.
As a rookie in the 2003–2004 season, LeBron James was paid
a salary of $4 million. His contract also specified an additional
$14.8 million through his fourth season. So just what did the
Cleveland Cavaliers get in return? Home attendance jumped
59 percent from 471,374 to 749,790. Ticket revenue increased
80 percent from $15 million to $27 million. And the Cavaliers
went from eighth place to fifth place, more than doubling
their winning percent and missing the play-offs by one game.
While it’s a bit clouded by the unknown signing bonus, it is
pretty clear that James was worth every cent the Cavaliers
paid. In addition, as we covered in Chapter 4, Cavaliers’ owner
Dan Gilbert (Quicken Loans fortune) may enjoy significant
values that are created when LeBron leads the Cavaliers in
Cleveland.
Let’s consider what MRP theory suggests concerning the value
of rookies, generally speaking. We look at the NFL because
the most extensive data are available on those players. Total
266

compensation packages for first-round NFL rookie draft picks
are shown in Table 7.6. The level of rookie compensation
depicted in Table 7.6 stretches the imagination of sports
writers and others who do not think that anyone untried can
be worth so much. Let’s narrow our look to just the contract
mentioned in the introduction. Quarterback Matthew Stafford
was the number one pick in the entire 2009 draft. The Detroit
Lions put together a package that could pay Stafford up to $72
million over a six-year period. This is the largest contract that
has ever been offered, and special interest in the press
concerns the record $41.7 million guaranteed to a new player
before even taking a snap at quarterback.
We begin thinking through the value of this draft pick with a
note of caution: Contracts in sports, both rookie and veteran
alike, are full of ifs, and contract clauses must be read with
care. Part of Stafford’s contract involves guaranteed bonuses
pro-rated over 2009–2014. The rest of the contract is built on
the contingency that he take 35 percent of the snaps from
center in 2009. If that happens, a variety of contract options
could drive Stafford’s contract to about $72 million over six
years, 2009–2014. But if he fails to take those snaps, the
Lions can announce they are not exercising their option for
the entire six years, a specified set of payments kick in, and
the contract then ends in 2013. Table 7.7 depicts pay to
Stafford under these two contingencies.
If Stafford fails to take 35 percent of the snaps, the Lions (and
Stafford) will be disappointed, and they can take the “No
Option” path but still have to pay the guaranteed bonuses.
Without any further considerations, the total is $17.4 million
in bonuses plus $23.25 million in salaries, a total of $40.65
million. However, a basic tenet of calculating values over time
is discounting. Employing standard discounting at a 5 percent
rate of interest, the cost of bonuses falls to $13.8 million and
the salary cost falls to $19.2 million. Thinking about it in
2009, the discounted value of Stafford’s contract under “No
Option” is actually $33 million. Remember, Stafford won’t be
a total bust under this option, he simply will not have met
everybody’s highest hopes.
If he does meet the highest expectations, guaranteed bonuses
will be paid, and additional annual earnings include salaries
plus other bonuses and incentives under the “Option” column
in Table 7.7. Here is where the $72 million number comes
from—the sum of $17.4 million in bonuses plus $54.6 million
in annual earnings. However, again, discounting puts the
consideration in its proper context. The discounted value of
the option bonus remains $13.8 million. Discounted
additional annual earnings become $43.6 million for a total of
$57.4 million. Discounting already reduces the reported
possibilities for Stafford by 20.3 percent. Take what you read
about stratospheric rookie contracts with a grain of salt.
But even at the smaller amount of $33 million, isn’t this a
ridiculous amount for an untried rookie? First, be careful with
the idea that a rookie is “untried.” Elite college athletes are the
result of a relentless selection process reaching back to high
school (and sometimes before that). I can think of no other
potential employee about whom an employer might know
267

more than a team owner knows about a first-round pick.
There remains the issue of the transition to the pros, but
owners also have the history of hundreds (if not thousands) of
other incoming players to compare as well. Nobody has more
riding on the draft than owners, and they will bring all the
information they have to bear on these choices.
Consider this. In an uncertain world, there will be some draft
busts. But there will also be some who are draft bargains,
performing beyond expectations. All in all, the rest will
perform as expected. I am constantly surprised to read that
owners and general managers are somehow expected to be
correct all of the time.
The second part of the consideration of the value of a rookie
can be untangled from the MRP perspective using just
attendance revenues. The Lions drew 435,979. If they drew
the same on average over six years, total attendance would be
2,615,874. In discounted present-value terms, in the “No
Option” case, each fan in attendance has to spend only $12.62
to cover this top draft pick ($33 million divided by total
attendance over six years). As a current point of reference, the
Team Marketing Report data give an average 2008 ticket
price for the Lions of $66.39. The $12.62 represents a 19
percent ticket price increase, which might seem high.
However, the full fan cost index is $383.57 for a family of
four. The $12.62 is only 3 percent of the fan cost index. And
we haven’t even gotten to TV revenues. So, whether a rookie
actually meets expectations is an important question. But the
rest of the issue is whether rookies generate additional
revenues that actually are quite small per team in order to
cover their cost.
268

SECTION 6
The Value of Athletes:
Special Cases
The large salaries in sports seem so outlandish that one might
be tempted to abandon the MRP explanation. How can the
same theory that explains pay and hiring from minimum wage
to middle management work for sports stars? There are a few
competing explanations for how sports stars get paid: the
winner’s curse, bidding wars, and winner-take-all outcomes.
Let’s explore these special situation explanations one at a
time.
THE WINNER’S CURSE
Chapter 3 discussed the idea of a winner’s curse in broadcast
rights bidding. Some have used this same logic to argue that a
winner’s curse exists in the market for players, too. It is
easiest to see how a winner’s curse explanation works in the
talent market by thinking of the free-agent market as an
auction. Owners are the bidders and free-agent players, the
prize. Suppose that most owners are informed and
experienced bidders in the free-agent auction, but that a few
are not. In such a case, most owners would bid only the true
expected MRP of a player. However, the other inexperienced,
uninformed bidders would bid well below or well above the
true expected MRP. Under a vigorous competitive free-agent
auction, the highest of these inexperienced/uninformed
guesses will be the winning bid. But as we discussed in
Chapter 3, it will be wrong in terms of informed guesses about
the player’s MRP. In this situation, the winner is cursed by the
size of the winning bid, and player salaries would be well
above their actual MRP.
Early on during free agency in MLB, Cassing and Douglas
(1980) found some support for the winner’s curse. Lehn
(1984) also found that owners sometimes lack all of the
information they need when they make bids and may, for
example, get stuck with an injured player at a high price.
However, it does not seem reasonable that this special
situation could apply to sports stars over an extended period
of time.
It is true that the eventual performance of players is subject to
uncertainty (deviations from career averages due to slumps or
injury or the beginning of the eventual long-term decline in
ability that happens to every player). But it is also true that no
single productive input is monitored and checked more
closely or more often than sports stars. Their performance is
often tracked on a minute-by-minute basis. Player injury
histories are well known and publicized. Medical specialists
are always available to check on a player’s status. This is just
as true for incoming rookies. From the time they show any
potential as youngsters and on through college, scouts follow
players with the eye of a detective. Pro sports combines,
where all teams get together, put the incoming talent crop
269

through workouts, and observe their performance, reveal any
weakness or deviation from perfect status.
When the free-agent market first opened, nearly everybody
was inexperienced in the process. It is hard to believe that
there would be any systematic overbidding for players based
on either inexperience or misinformation as time went by. It
is reasonable to expect that inexperience in the free-agent
market on the part of owners was short lived.
BIDDING WARS
A bidding war is a special case of the MRP explanation where
a very special circumstance is in operation. Essentially, a
bidding war ensues when coming in second place is very
costly. For example, suppose the season is closing in on the
play-offs and that two teams need just one player to clinch a
play-off spot and its rewards. At such a time, there may be
very few substitutes who satisfy the play-off needs of these
two teams. Coming in second in bidding for a single player
who can swing the play-off hopes for either team would be
expensive. The situation is compounded the more valuable
that play-off appearance becomes.
In a bidding war situation, we have to be careful to include all
elements of the player’s MRP—nothing has changed about the
player’s ability, but the value during the play-offs raises the
MR part of MRP, and coming in second is very expensive.
However, although this situation may occur for some teams as
each play-off is reached, it only happens at a particular time
during the season, and most teams do not buy their talent
under such pressures.
THE WINNER-TAKE-ALL EXPLANATION
The winner-take-all explanation of Frank and Cook (1995) is
our final special circumstance version of how players get paid.
The winner-take-all explanation is easy to see for individual
sports, such as golf or tennis. Suppose that fans want to see
only the best compete against each other. Because television
allows fans to simultaneously enjoy matches between the best
players, only those best players would ever be on TV. As a
result, some players who are only marginally worse than the
best competitors will get nearly nothing, and the best players
will get nearly all of the prize money. In such a situation, fans
will see only the best compete against each other for all the
marbles. This is important: It must be the case that winners
actually do take all.
We can take a quick empirical look at the winner-take-all idea
in golf. Figure 7.8 shows the Lorenz curves for recent
tournament outcomes in men’s and women’s pro golf. In both
the men’s and women’s tournaments, the figure shows that
the top 10 percent of players (between seven and nine golfers,
typically) earn well over 50 percent of the total event purse.
The Gini coefficients index this inequality as 0.656 and 0.620
in men’s and women’s golf, respectively. For comparative
purposes, Gini coefficients in team sports can be as high as
0.500 (Chapter 6) and, in the economy at large, are typically
270

around 0.300. It certainly looks like winners take nearly all in
men’s and women’s pro golf.
However, there are other explanations for this tournament
prize structure. Suppose that pay were not structured with a
huge payoff for the top 10 in any given golf tournament. One
could envision the top players deciding it was easier to just
take turns winning the tournaments in a given year.
Everybody shows up to play, but nine of the top 10 miss a putt
here, shank a drive there, and the winner for the week takes
home the check. Golf seems like an especially likely sport for
this to happen because it is so difficult to be consistent over
time and, as a result, fans would have a difficult time
detecting this behavior. This flies in the face of what both
tournament directors and pro golf organizations are trying to
do, namely, maximize profits from attendance and TV
broadcast contracts for the year’s tournaments. If it becomes
known that players have colluded to share tournament purses,
fan confidence in the integrity of play would disappear and,
with it, the value of fan interest at the gate and on TV. Just
imagine if a “hot hand” like Tiger Woods had agreed to such a
purse-sharing scheme. The streak of majors in 2000–2001
would never have happened, and the value of selling
tournaments would be much lower. So how can directors and
associations overcome this type of behavior?
Ehrenberg and Bognanno (1990a, 1990b) show that the
concentrated structure of pay at the top protects against these
problems. When pay is highly concentrated for the top 10
finishers, moving up one spot on the leader board for players
close to the top means a high expected return. By making the
value of outdoing the other top golfers so large, the incentive
to collude is reduced. These researchers found that players do
perform better in their last round if they are close to the
leader after three rounds; why waste huge effort moving from
53rd to 52nd? However, the increased effort to move from
sixth to fifth pays off immensely. In addition, they found that
higher prize money concentrated at the top leads to better
scores. Thus, there may be a winner-take-all aspect to
tournaments in individual sports, but concentrated prize
money also maintains integrity of competition. Without that,
tournaments would not be worth very much to fans.
Tournament design structure issues were later discussed in
full by Szymanski (2003).
The winner-take-all explanation at least sounds like it can
characterize individual sports, but does it make any sense for
team sports? First, do fans want to see only the best? If they
did, then very poor teams might cease to exist. However, in all
leagues, lovable losers have always existed. At the very least, if
fans wanted to see only the very best, many home games for
most baseball teams would never be broadcast. This casts
doubt on the winner-take-all scenario for team sports. The
real deciding fact on whether the winner-take-all explanation
can be applied to team sports relates to the outcome: Do
winners indeed take all?
Let’s look at the player most likely in recent times to be such a
winner, Michael Jordan. He was worth about $33 million
($43.2 million) to the Chicago Bulls had he played his final
271

year, 1998. The Forbes magazine survey data show that total
revenues in the NBA were about $3 billion ($3.9 billion) at
that time. Historically, NBA players receive around 60 percent
of the league’s total revenues, say, $1.8 billion ($2.35 billion)
through the salary cap process. Jordan’s $33 million is only
about 1.8 percent of the total amount enjoyed by players. This
hardly sounds like a winner taking all—especially compared to
a single player taking the majority of a purse in individual
sports. However, the pursuit of the large payoffs to pro
athletes does have some characteristics in common with the
primary troublesome result of winner-take-all markets. Many
who will never get the high payoff still invest extraordinary
amounts of time and resources trying anyway. Sadly, this
means that much of it is wasted, as shown in the Learning
Highlight: Investing in an Athletic Career at the end of this
section.
272

LEARNING HIGHLIGHT: INVESTING IN AN ATHLETIC
CAREER
Frank and Cook (1995) in The Winner-Take-All Society point
out the tragic consequences of wasting youth in pursuit of
athletic achievement in order to be one of the winners. One
must exercise extreme caution in evaluating young people’s
choice to invest their time and energy into athletic careers.
First, only the individuals investing in the sport know how
much fun is involved, the subjective probability of making it,
and their own opportunity cost. Second, it does not take a
winner-take-all explanation to foster overinvestment.
Overinvestment may occur just because young people make
mistakes evaluating the benefits, chances of advancement,
and costs of pursuing a pro career.
It seems reasonable to think that at least some of the young
people investing in athletic careers are making a mistake.
Richard Lapchick, University of Central Florida sociologist
and director of its Institute for Diversity and Ethics, reports
informal sampling that reveals nearly half of all high school
senior football players think they are going to get a college
scholarship. Noted sports economist Roger Noll (1998) points
out that a 15-year-old who is a good prospect to be a football
player stands a 3 percent chance of a college scholarship and a
0.25 percent chance of a pro career. It would appear that
many high school seniors overestimate their chances.
However, those with a good chance of making it can be
behaving quite rationally in pursuing a pro career. Noll (1998)
estimates that the present value of the college shot is
$200,000 in additional lifetime earnings and that the
discounted present value of the pro shot is $2 million in
additional lifetime earnings. The expected value calculation
gives (0.033 × 200,000) + .(0.025 × 2,000,000), or about
$11,000 ($14,390). Noll also estimates that to have a chance,
the hopeful player would need to devote 2,000 hours of effort
through high school. That’s about $5.50/hour
($11,000/2,000 hours) as an expected return for three years
of hard physical work. If the alternative has a lower expected
value, it is rational for the hopeful to invest during high school
for the chance at a top-level college scholarship that would
result in a good chance at a pro career. However, note that
$5.50 an hour identifies only those with approximately a
minimum-wage potential through high school. For others who
can make more than $5.50 an hour, the effort is not worth it
unless the joy of playing makes up the difference, and joy can
be had without devoting the effort required for a potential pro
career.
Further, from a societal standpoint, having hundreds of
thousands of teenagers dedicate their lives to a career in
athletics is a waste of resources that could be better put to
developing academic and job-related skills. Suppose the
financial payoff to the best pro athletes would stay the same if
tens of thousands, rather than hundreds of thousands, took
their shot at the big payoff (with fewer trying, there is a
chance that the level of absolute competition would fall and,
with it, the value of a pro career). The waste in resources is
273

measured by the added income that the rest would have
gained, net of their investment in earning it, learning another
marketable skill. There would have to be an awful lot of
nonmonetary joy for the rest in order to make up for the
difference.
Nearly half of all high school senior football players think that
they are going to get a collee scholarshiop, but in reality
around 3 percent actually do. [Photo HS football.]
Sources: Frank and Cook (1995) and Noll (1998).
274

SECTION 7
Discrimination in Pay and
Hiring: Introduction
Before we move on, we need to examine one more important
market outcome. To this point, competitive markets for talent
have resulted in players earning close to the MRP with the
team that values their talent the most. However, in some
situations lacking this competitive vigor, it is possible for
individuals with identical productive potential in sports to end
up with very different economic outcomes because of their
gender, race, or ethnicity. These outcomes are commonly
referred to as discrimination in pay and hiring.
THE LIMITS OF ECONOMIC ANALYSIS OF
DISCRIMINATION
Economics, by the nature of its focus on market outcomes and
incentives, can address only a subset of the issues of interest
in the study of discrimination. The tools of economics will be
useful mainly in detecting discrimination in pay and hiring in
sports. Economic tools will also be useful in analyzing how
individuals respond to the discrimination that might occur.
However, discrimination can permeate all social interactions,
not just economic ones. The attitudes passed along among
friends and business contacts influence referrals and other
“network” outcomes that, in turn, influence economic
outcomes. Thus, the tools of economic analysis come up a bit
short, analytically speaking, in the direct examination of
network outcomes.
A historical episode helps make this limitation clear. In 1961,
the first version of MLB’s Washington Senators moved to
Minneapolis–St. Paul and became the Minnesota Twins.
Attendance in Washington, D.C., had been abysmal—for the
five years prior to their first season in Minnesota, the
Washington Senators had the lowest attendance in all of
baseball. About the move to Minnesota, then-owner Calvin
Griffith remarked about a racial component of the economics
of gate attendance in the two locations—he claimed that black
people do not go to ball games and there were far fewer in the
new location (Nelson, 1984, p. 95).
Many took these to be racist remarks. However, Griffith
claimed he was being misinterpreted. He said he was not a
racist, he was just explaining the demographics and buying
habits of some fans in the market for baseball in Washington,
D.C., compared to the market in the Twin Cities. The depth of
intense emotion that such issues raise immediately pushes
any economic insight into the background. Griffith had
offered an explanation of his statement as a purely economic
one. But one of his black stars, Hall of Famer Rod Carew,
responded with outrage:
I won’t be a nigger on his plantation anymore. I will not
ever sign another contract with this organization. I do
275

not care how much money or how many options Calvin
Griffith offers me. . The days of Kunte Kinte [of Roots
fame] are over. . He respects nobody and expects nobody
to respect him. (Nelson, 1984, p. 95)
Discrimination is an extremely emotional issue, and
economics is the antithesis of that way of thinking. That
makes the application of economics to discrimination
important but limited. Any statement about discrimination in
general, based on a very small subset of the outcomes from
discrimination such as pay and hiring, will certainly be
invalid. For example, even if economic analysis suggests that
there is no discrimination in pay or hiring, that does not mean
that discrimination is no longer a problem in American
society.
MRP THEORY AND DISCRIMINATION
From the economic perspective, discrimination is primarily
an issue of compensation and hiring. Therefore, the
foundation of analysis is still MRP theory. Because MRP = MP
× MR, anything causing variation in either MP or MR will
cause variation in observed payment. Let’s first examine MP.
There are three factors that may explain variation in pay
among individuals: innate ability, training, and experience.
INNATE ABILITY
Innate ability matters in the world of sports, just as in other
areas. Some individuals are just lucky at birth; they have a
greater ability or aptitude, or their body is of the proper
proportion for a sport. All else constant, those with greater
ability contribute more of what fans pay to see and are,
therefore, expected to earn more. A seven-foot, well-
coordinated basketball player should be expected to go at a
premium relative to a tall, uncoordinated player. At the
highest levels of sports, a 250-pound offensive lineman is not
very valuable at all relative to equally able 300-pound players.
Note that there is no ground for discrimination based strictly
on innate ability. The idea itself is color blind.
TRAINING AND EXPERIENCE
Innate ability is not enough to succeed in pro sports. If those
with the most innate ability did not train, then the rest of the
players would close the gap through training and even
overtake those with innate ability. Further, ability typically
increases with experience. Thus, these two factors, training
and experience, also contribute to earnings. Interestingly,
variation in willingness to train or gained experience will
contribute to variation in income. Presumably, those with
innate ability also can train and gain experience, making them
just that much more valuable. If those with less innate ability
are able to close the gap, or if those of like ability forge ahead,
then there must have been some hindrance to training and
experience for those with greater ability. Perhaps it is a
person’s willingness to work hard. Perhaps it is a rational
economic calculation about the returns to hard work.
Training and experience require investment of scarce time,
energy, and monetary resources. The willingness to make
276

these investments can vary between people for many reasons.
Some may not choose athletic training because their
opportunities in other endeavors are too great. As we saw in
the previous section’s Learning Highlight: Investing in an
Athletic Career, the expected monetary value of a shot at a
lucrative rookie bonus and contract may not meet the returns
to other endeavors. In such a case, one would expect some
people simply not to make the investment in sports, even if
they could have enjoyed a modicum of success there.
Therefore, training choices can be due to willingness to work
or calculation of the payoff to working hard. However, there
may be a discrimination-based explanation for variation in
training and experience. If minority and women players have
less access to training, they may then fall behind in the race
for athletic success. Without training, their chances of
competing are limited. Therefore, they never catch up in the
experience department. Without the training, they never had
the chance to compete. This can affect their future
opportunities.
If discrimination is pervasive, the resources needed to
compete successfully on the field as players or off the field as
coaches and front-office executives may be denied to
minorities and females. On average, because their training
opportunities and experience have been reduced, one would
expect those who were the object of discrimination to have
less to offer at the hiring table. Here is the crux of the issue:
Sports team owners and personnel directors, bigoted or not
(more on this shortly), do not know whether any particular
player has been disadvantaged in this way. The cumulative
impact of past discrimination is an unobservable
characteristic.
STATISTICAL DISCRIMINATION
In such a hiring situation, owners and personnel directors
may adopt a rule of thumb based on past performance of the
group as a whole—but that past performance builds on the
cumulative effects of past discrimination! People of color and
women may lag behind, generally, because discrimination
limited their access to training. Although not all people of
color or women necessarily lag behind, the rule of thumb
based on a group’s past performance lumps all members of
the group together. Thus, some players who are not below
average get thrown in with the rest. Their talent does not get
the attention it deserves. This is called statistical
discrimination.
The result of statistical discrimination is clear. Rules of thumb
adopted in the face of uncertainty about a player’s ability are
based on that player’s membership in a demographic group.
This leads to a perception of lower MP for members of
particular groups. Fewer members of those groups will be
hired, and they will be paid less. Until this “perception
barrier” is broken, the biased outcomes will continue.
Robert Peterson (1970) repeatedly points out that nearly all
MLB players and owners felt that Negro League players did
not really have what it took to play in the white major leagues.
277

As it has now been conclusively demonstrated that black
players are every bit as able as white players, this is a perfect
example of statistical discrimination. As the color barrier fell,
rules of thumb about race were abandoned. However, you can
also see that an alternative explanation is that these people in
MLB may simply have been bigots. The difficulty in discerning
between economic explanations of differential pay and hiring
and plain old bigotry will be a recurrent observation in this
section.
Perception barriers may also be falling in women’s sports as
well. Recent policies aimed at gender equity in the United
States are the result of changing cultural expectations and
attitudes. Full participation for women is not just a goal, it is
the law (see Chapter 13 on Title IX). Markets for female talent
have begun to take off, with the number of new women’s pro
sports teams growing. Previously, returns were low for team
sports, so few women made the investment. But the
underlying barrier—attitudes toward women in sports—
appears to be crumbling, and both the value of talent in
women’s sports and training by women is increasing.
FAN DISCRIMINATION
So goes the MP element in the explanation of discrimination.
For the MR component of MRP, the impacts will all be based
on fan demand and its determinants. Fan discrimination
occurs when fan preferences result in lower pay and reduced
hiring by race and gender only because of race and gender,
not because of ability. In Chapter 2, we saw that one group of
students was willing to pay different amounts for men’s and
women’s sports. Because that observation held across exactly
the same set of people, the difference in willingness to pay has
to be based on preferences.
Some preferences are not subjected to social scrutiny. For
example, preferences toward absolute competition can vary
between people. If the same set of fans feels there is a higher
level of absolute competition in men’s sports and higher levels
of competition are more satisfying to them, then they will be
willing to pay more as the level of absolute competition rises.
This really is no different from the observation that people
pay more for major league sports than for minor league
sports. Few would think that this type of preference is
illegitimate.
However, some preferences are the objects of society’s
scrutiny. If some fans like men’s basketball over women’s
basketball only because women play the latter, then those fans
are sexist (by definition). Sexist preferences have become so
unpopular that many manifestations of them are illegal
(under the Civil Rights Acts of the 1960s and Title IX in 1972).
A NOTE OF CAUTION
How can an outside observer tell whether it is preferences for
higher absolute levels of play or sexism? As with our earlier
example of MLB prior to integration, discrimination can be
difficult to identify. This is further confused by the fact that
men never play against women. We simply cannot observe
them in common competition and make any objective
278

comparison of ability. For example, consider the so-called
major tournament values for tennis and golf in Table 7.8. A
number of interesting questions arise. Why is the total purse
larger for men than women in two of the four tennis majors
and in all of the golf majors? Why does the winner of the
men’s singles event at Wimbledon make so much more than
the winner of the women’s singles event? Why can’t even one
winner of a women’s major golf tournament come close to
making $1 million when all of the winners of men’s majors
earn over $1 million? The possibilities discussed in this
section might explain why, but we will not ever be able to
distinguish them because men have yet to play against women
in singles competition.
When we turn to race or ethnicity and fan discrimination,
things are greatly simplified. There is no basis for any
variation in the level of absolute competition that can be
attributed to race or ethnicity. Men may play at a different
level than women, but people of different ethnic backgrounds
compete in the same leagues and demonstrate comparable
skill. Any detectable pay or hiring difference based on race
and ethnicity must reflect preferences alone. The possible
sources of this type of preference-driven discrimination are
fans, owners, and teammates.
GRAPHICAL ANALYSIS OF FAN DISCRIMINATION
To isolate fan discrimination, let’s assume that owners are not
bigots. This assumption means that if owners maximize
profits, then they will hire the players whom fans want to see,
regardless of race or ethnicity. Figure 7.9 depicts the fan
discrimination situation. Actual MRP is shown, along with
another willingness to pay construct, MRP + Race Premium.
You should think of the actual MRP as the value of this type of
talent to an owner with nonbigoted fans. In Figure 7.9, the
race premium shifts MRP to the right because the owner
values the players under analysis at a premium over their
actual MRP. This is because fans are willing to pay more to
see sports played by this type of player over others. At every
price, the owner would hire more of the type of player that
fans will pay more to see in order to collect this premium from
fans.
We can find the pay and hiring outcome from fan
discrimination from Figure 7.9. In a competitive setting, the
result is the intersection of the supply and demand for talent.
SR and WR are the amount of the preferred talent hired and its
wage in a bigoted fan market, respectively. SA and WA are the
amount of the preferred talent and its wage in a nonbigoted
fan market, respectively. In the presence of fan
discrimination, more talent of the race or gender that fans
demand the most (SR > SA) will be hired, and it will be paid
more (WR > WA) than would be the case in a market of
nonbigoted fans. Any other choice would mean less profit for
the owner than is possible in this market. Fans are willing to
pay more for every unit of talent between SR and SA than it
costs to hire those units.
THE COSTS OF DISCRIMINATION TO FANS
279

In a market with fan discrimination, the team owner either
buys talent that wins less than a more integrated lineup could
win or pays far more than the minimum amount possible for a
given level of winning percent.
It seems reasonable that the first option is more likely. The
cost of fan discrimination is that better players are forced out
of particular locations in order to pack the lineup with the
players of the preferred race or ethnicity. Therefore, bigoted
fans pay a price for their preferences in terms of a lower
winning percent than if they were not bigots. As long as the
bigoted fans are willing to pay this price, their preferences can
persist. Only circumstantial evidence on this point exists in
real sports markets. The obvious example is the integration of
MLB, beginning with Jackie Robinson joining the Brooklyn
Dodgers in 1947. It took 12 years for MLB to become fully
integrated, when Pumpsie Green joined the Boston Red Sox in
1959. Since black players demonstrated their equal talent
immediately upon entering the league in the late 1940s, this
very gradual integration indicates resistance that can be
attributed at least in part to fan discrimination. There is also
some indirect evidence of fan discrimination in the market for
memorabilia and voting for hall of fame candidates, as
covered in the Learning Highlight: Discrimination Is Where
You Find It at the end of this section. But, as we will see
shortly, teammate and owner discrimination can also play a
role.
FAN DISCRIMINATION: OWNERSHIP IMPLICATIONS
What would happen if the owner were forced to integrate the
lineup? Fewer fans would come to the games (those without
racial or ethnic preferences), and the owner would suffer
reduced revenues as demand shifted to the left. The owner
would clearly pay a price for any policy move that forced
integration. These owners are not bigots, by our earlier
assumption, even though the pursuit of profits does lead them
to cater to the preferences of bigoted fans.
GRAPHICAL ANALYSIS OF OWNER AND TEAMMATE
DISCRIMINATION
Figure 7.10 demonstrates owner discrimination and
teammate discrimination. In this case, the premium placed on
the particular group of players is due to preferences by the
owner or teammates. If these preferences are satisfied, owners
will hire more of this type of talent (SR > SA) and pay it more
(WR > WA) than it really is worth in terms of contribution to
winning percent.
However, unlike the case of fan discrimination, owner and
teammate discrimination would not be expected to stand in a
competitive league of profit-maximizing owners. This is
because the owner, rather than the fans, pays the premium on
preferential hiring. In this case, the team is less profitable
than it could be due to the owner’s choices. The shaded area in
Figure 7.10 shows the cost of owner discrimination. First,
since the wage rises to WR > WA, all of the units up to SA are
paid more than the wage that would prevail without this type
of discrimination. The rectangle (WR − WA) × SA is a loss to
280

the owner. Second, on every unit hired beyond SA, the
payment by the owner, WR, exceeds the benefits an owner
could ever get from fans in terms of the talent’s actual MRP,
that is, the rest of the shaded area in Figure 7.10 is also a loss
to the owner.
Put another way, all of this overpayment to support bigoted
preferences by owners and teammates is money that could
otherwise be spent on team quality. The owner wins less than
is possible for the level of spending that occurs. This sort of
profit wasting usually means that another potential owner can
take over the team, integrate the lineup, and increase the
value of the team as an economic asset. If competition were
vigorous enough and profits were the object, owner or
teammate discrimination would not be expected to survive.
There is some evidence that just such a situation occurred in
the NBA. In the late 1980s, the league expanded and a new
labor agreement was struck between owners and players.
Bodvarsson and Brastow (1999) argue that these changes
increased competition for players. In the face of this increased
competition, the salary premium to white players over black
players disappeared. Because nothing happened to change fan
preferences and because the composition of rosters remained
unchanged, they concluded that employer discrimination had
been in place and wilted with increased competition.
However, two factors might hinder elimination of
discrimination by competition among owners for the best
talent. First, entry into pro sports ownership is now pegged in
the hundreds of millions of dollars. At any point in time, there
may not be many takers at that price. So owner preferences
could survive quite some time. Second, owners are wealthy
individuals. If they care about their team’s bottom line but
value their own preferences highly, they may be perfectly
happy to bear the cost of their choices. Therefore, from the
economic perspective, if owner or teammate discrimination
persists, all we can say is that competition among owners isn’t
brisk.
TALENT SUPPLY ISSUES
All of these issues stem from the demand side of the sports
talent equation. However, supply issues also influence
differential pay. Two of these issues are the opportunity cost
of time spent on the job and preferences toward the job.
OPPORTUNITY COSTS
Other job and nonjob opportunities determine the
opportunity cost of time. Investments in training and
experience qualify a person for a variety of jobs. If the demand
for one job changes, raising the return at another job, then
people can switch jobs. Because the payoff may be distant in
time, the expected value of nonsports jobs can be larger than
for sports jobs. We will never know how many high-salary
career people could have been professional soccer players, but
you get the idea. McCormick and Tollison (2001) argue that
low opportunity costs are one of the reasons that black
basketball players work for less money than white basketball
players.
281

Another factor that determines the opportunity cost of time
spent on the job is family commitments. The allocation of
time to family, such as the role of primary care provider, may
be a social convention that falls to one gender over another.
Because being the primary care provider is an arduous and
demanding task, the time remaining for participation in
sports can vary by gender. Clearly, these attitudes shape the
opportunity cost of time spent in sports pursuits for women in
the United States.
WORKER JOB PREFERENCES
The other supply consideration is the preferences of those
offering up talent in sports markets. Worker job preferences
often lead athletes not to choose their highest possible
earnings. They may choose a job based on nonwage factors. In
the next chapter, there is a full discussion on how free agency
altered many players’ decisions about where to play. Suffice to
say here that star free agents in all sports have joined teams at
lower pay to avoid the hectic pace, lack of privacy, and press
scrutiny of big cities, or because teams looked like they were
building a winner. Nonwage factors may include what players
perceive to be an owner’s commitment to winning beyond
what their market might bear in terms of revenue or a coach’s
style of play.
In terms of discrimination, preferences by one group of
players may influence their choice of team. Rod Carew made
his preferences quite clear in the wake of Twins’ owner
Griffith’s statements. One wonders if other black players
shunned the Twins as well. Although the racial climate may
have changed over time in Minnesota, to this day, some cities
have a better reputation for racial tolerance than others do.
Black players may be willing to take less money in order to
choose between cities based on their own preferences.
282

LEARNING HIGHLIGHT: DISCRIMINATION IS WHERE
YOU FIND IT− BASEBALL CARDS AND HALL OF FAME
VOTING
As stated later in this chapter, the evidence suggests that
racial and ethnic discrimination no longer plague salaries in
professional team sports. But early in the chapter, it was
stressed that this is only one part of discrimination.
Discrimination can still be active in other ways. For example,
it appears that fan discrimination exists in two other places,
baseball card markets and voting for members of the Baseball
Hall of Fame.
In baseball cards, the evidence concerns the difference in card
prices for white stars and others. Baseball cards for black and
Latino players sell for about 10 percent less than cards for
white players that have comparable historical playing
statistics. It seems black sluggers are the particular target of
this type of fan discrimination. The difference is 13 percent for
the cards of pitchers of color but, interestingly, there is no
difference for Latino pitchers. Of course, one would suspect
that if all that mattered was performance there would be no
difference in price for the cards of star hitters or pitchers
based on race or ethnicity.
There is some limited evidence that both black and Latino
retired players eligible for the National Baseball Hall of Fame
also face fan discrimination. Members of the Baseball Writers
Association of America cast the deciding votes. On the first
ballot, darker-skinned Latino players receive fewer votes than
other players with comparable lifetime statistics. This is an
interesting switch from the baseball card outcome because
there is no similar finding for eligible black players. The
reason this voting evidence is limited is because this type of
fan discrimination does not deny truly star players their place
in the hall. Instead, Latino players who were eligible but
probably would not have gained entry based purely on their
statistics are the object of voting discrimination.
Fan discrimination analysis suggests that Sammy Sosa’s
baseball card will sell for less than the card of other sluggers
and that he will receive fewer Hall of Fame votes than other
sluggers. [Sosa photo.]
Sources: Fort and Gill (2000); Jewell, Brown, and Miles
(2002).
283

SECTION 8
Economic Findings on Pay
Discrimination
The primary objective in presenting the findings of economic
analysis of discrimination in sports is not to pass judgment on
the level of discrimination in pay and hiring in sports. Instead,
the findings simply serve to reinforce the idea that examining
race and gender outcomes is a difficult prospect.
Extensive data and analyses on pay discrimination are
available. For our purposes, we will follow the reviews of one
of the most respected analysts of discrimination in sports,
Lawrence Kahn (1991, 2000). This section presents a
synthesis of Kahn’s work, but interested readers are
encouraged to read Kahn’s reviews.
Kahn (1991) cites evidence of racial pay differences driven by
fan discrimination in the NBA. Up until the late 1980s,
researchers found an 11 percent to 25 percent shortfall for
blacks of equal talent. Meanwhile, more limited evidence had
been found on pay discrimination and on fan discrimination
against French-Canadian players in the NHL and favoring
white players in MLB. As we mentioned earlier, in Grand
Slam tennis events, prize money for women is lower despite
the fact that they generate at least as much revenue as men.
Fan discrimination, then, at least for team sports, appears to
be confined to the NBA and MLB. In MLB, initially after
integration, blacks increased home attendance. By the 1960s,
revenues for teams with more black players were lower than
for teams with fewer. However, these effects disappeared by
the 1970s. Kahn also cites some evidence that nonwhite
players earn more in areas with large nonwhite populations,
and vice versa for white players. Scully (1989), updating his
original work on pay discrimination in MLB, flatly stated: “We
can conclude that holding performance constant, black and
white players are paid the same on average” (p. 178).
Kahn (2000) concludes that by the mid-1990s, pay
discrimination had pretty much disappeared from the NBA.
The only remaining evidence of pay discrimination was a
significant but very small white premium at the very top of the
NBA pay scale. There was no evidence of any pay
discrimination in the NFL or MLB by the mid-1990s. In
hockey, there is still evidence that in the NHL, French-
speaking hockey players are the victims of pay discrimination
relative to other hockey players.
284

SECTION 9
Discrimination in Hiring
With the analysis of pay discrimination behind us, the
analysis of hiring discrimination is straightforward.
Somewhat ironically, given his quote at the beginning of this
chapter, Frank Robinson did become an MLB manager for
nine years (Cleveland, 1975–1977; San Francisco, 1981–1984;
Baltimore, 1988–1989), and he came back to manage the new
Washington Nationals, 2002–2006. However, he was the first
and only black manager for quite some time. The lack of
minority hiring in the head coaching ranks of the NFL is
viewed by many to be such a problem that there is a league
policy, the “Rooney Rule,” that requires candidates of color to
be considered whenever there is a head coach opening.
Detroit Lions president Matt Millen was fined $200,000
($231,740) for failing to interview any black candidates when
he hired Steve Mariuchi in 2003 (cbs.sportsline.com, July 25,
2003).
The outcome of simple hiring discrimination is
underrepresentation of those equally qualified but denied
hiring on the basis of race, ethnicity, or gender. However,
there is another subtle form of hiring discrimination called
stacking. Stacking refers to the hiring outcome on the field.
Stacking occurs when nonwhite players are put in less
conspicuous positions, such as outfielder as opposed to
pitcher in baseball or wide receiver as opposed to quarterback
in football. Although it is clear on the field, stacking is difficult
to analyze off the field. For example, suppose nonwhites were
disproportionately represented in less-visible front-office
activities. That would be similar to stacking. However, the
term stacking is reserved for on-field outcomes.
ECONOMIC EXPLANATIONS FOR HIRING
DISCRIMINATION
Economics would suggest one of three things about hiring
discrimination. First, members of the group facing
discrimination may possess different bundles of skills.
Second, workers’ preferences vary by group so that minorities
and women choose not to take these jobs. Third, hiring
discrimination continues based on current discrimination and
the legacy of past discrimination, that is, based on statistical
discrimination. The first idea leads us right back down the
path of statistical discrimination. If there are skill differences,
what is it about obtaining skills (through training and
experience) that has precluded their acquisition by one group
but not another? Given the payoff, it seems unlikely that
individuals would choose not to train or gain experience that
would open doors to employment in higher-paying jobs.
The second explanation of hiring discrimination concerns
worker preferences. For this explanation to carry any weight
in explaining hiring discrimination economically, it would
285

have to be the case that minorities and women choose not to
engage in particular sports jobs. This is pretty difficult to
swallow. For example, society might once have dictated that
coaching or top-level management was not “women’s work.”
But that time is long past. Certainly no similar societal
explanation seems valid in the case of blacks and Hispanics.
That leaves an explanation of hiring discrimination based on
current discrimination and the legacy of past discrimination
(i.e., statistical discrimination). Stacking may occur, and
discrimination may preclude the participation of women and
minorities at the top executive level because employers have
developed rules of thumb based on the unobservable
productivity characteristics of these groups. These rules
discount the human capital of current minority and women
candidates; these perception barriers will have to be breached
in order to eliminate such rules of thumb. In the rest of this
section, we will look at specific data regarding hiring
discrimination.
FINDINGS ON HIRING DISCRIMINATION
Almost all formal economic analysis of hiring discrimination
concerns players. The formal analysis concerning baseball
managers and football head coaches is sparse, but interesting,
and reserved for this section’s Learning Highlight: Why Are
There So Few Black MLB Managers and NFL Head Coaches?
(At the end of the section.) Kahn’s (1991) review of hiring
discrimination points out the obvious absence of black players
in all major league sports prior to the end of World War II.
Since then, there is no evidence of reduced hiring in the NBA,
but there is some evidence that black players exit the league
earlier and at higher rates than whites, performance held
constant (Kahn, 1991). Similar evidence has been found in
MLB (Kahn, 1991). Kahn (2000) cites evidence of hiring
discrimination in the NFL draft. Black college players are
chosen later in the NFL draft than equal quality white players
are chosen. The opposite has been shown to be true in the
NBA draft, but just barely.
THE RACIAL AND GENDER REPORT CARD
Racial and Gender Report Card (RGRC), published annually
by the Institute for Diversity and Ethics in Sports at the
University of Central Florida, is a periodic summary of the
status of minorities and women in sports. It offers little in the
way of analysis of discrimination, but it does report current
statistics. Since it is a report card, it gives grades to
professional sports and college sports organizations based on
racial and gender opportunity in the league’s central office, for
coaches and managers and also for players. The grades are
reported as “Grade for Race/Grade for Gender.” In 2009,
MLB earned A/B, up from A–/C+ the year before, the NBA
earned A+/A– the same as it did in 2008, and the NFL earned
its highest grades in the history of the RGRC, A–/C. At least
by these grades, pro sports are doing quite well on race, and
the NBA is doing well on gender. MLB and the NFL have far
to go on gender. The NHL is not included in the RGRC.
FINDINGS ON STACKING
286

Kahn (1991, 2000) found clear evidence of stacking in MLB,
the NFL, and the NHL but no significant stacking in the NBA.
Blacks were underrepresented as pitchers, catchers, and
infielders in MLB. In the NFL, blacks were underrepresented
as quarterbacks, kickers, and linebackers. In the NHL, French
Canadian players were underrepresented as defensemen and
overrepresented as goalies. Kahn points out that it is only
speculation as to whether stacking is caused by coworker
discrimination or by statistical discrimination based on lack of
access to equivalent training.
The RGRC tally on race and position in the NFL for 2009
states, “Most observers agree that the issue of stacking in the
NFL is no longer a major concern…” The quarterback
position was the primary concern since it was so central to the
game and now that African Americans have broken down that
barrier, concern about stacking has been greatly diminished.
However, from the stacking perspective, the RGRC suggests
that MLB still has a way to go:
The Racial and Gender Report Card examines the issue
of stacking for the positions of pitcher, catcher, and
infielder filled by African-Americans. These are
baseball’s primary “thinking positions.” Five percent of
pitchers, and nine percent of infielders (both up by two
percentage points) were African-American. Thirty-two
percent of outfielders, who rely on speed and reactive
ability, were African-American during the 2008 MLB
season. This percentage was more than three times the
total percentage of African-Americans in MLB.
Historically, there have been almost no African-
American catchers and that remains the same.
287

LEARNING HIGHLIGHT: WHY ARE THERE SO FEW
BLACK MLB MANAGERS AND NFL HEAD COACHES?
As in so many other areas, when we turn to racial issues we
once again begin with pioneering sports economist Gerald
Scully (1989). In addition to his earliest work on
discrimination (Scully, 1973, 1974), he analyzed just why
discriminatory hiring occurs at the MLB-manager level.
Writing in 1989, when Frank Robinson was the only black
manager in MLB and there were only a few black coaches,
Scully noted that baseball managers’ positions during their
playing days were a major determinant in becoming a
manager—nearly 90 percent of MLB managers were infielders
during their playing days.
Scully made the simple observation that black players had
different access to infield positions and, as a result, were
destined to be the object of statistical discrimination. When
he was writing, blacks predominated in the outfield and were
badly outnumbered in the infield by white players. Further,
most managers were previously coaches. Infield coaches,
formerly infielders themselves, dominated the coaching ranks.
In the chain of events that generates an MLB manager, blacks
were outnumbered every step along the way.
Scully concluded that the lack of black managers appeared to
be a supply-side problem. It is a statistical discrimination
problem stemming from the fact that black players are usually
outfielders and miss out on the background that leads to big
league manager positions. Interestingly, Frank Robinson was
an outfielder until very late in his career. As the first black
manager, and an outfielder to boot, he began the long process
of dispelling beliefs about the lack of training for blacks and
for outfielders.
Recent work by Janice Madden (2004) finds quite a different
reason for the lack of black head coaches in the NFL. Looking
at the regular season and whether their teams make it to the
play-offs, Madden finds significantly higher success rates
among black head coaches than their white counterparts. She
also finds some evidence that black head coaches are hired by
better teams in the first place. But even holding this fact
constant, teams with black head coaches outperform teams
with white head coaches. Finally, Madden rejects the Scully-
type argument of “pipeline” effects by examining the playing
positions of coaches. Overall, Madden concludes that this
evidence is consistent with black head coaches’ being held to
higher standards to get their jobs in the NFL, clearly an
earmark of discriminatory hiring.
Frank Robinson once asked, “Why don’t we have black
managers… in baseball?” Then he became the first. [Photo
Robinson.]
Sources: Scully (1989) and Madden (2004).
288

SECTION 10
Chapter Recap
Players are fabulously compensated for their efforts, but
sports stars do not make as much as entertainment stars in
general. Over time, sports salaries have increased
dramatically, and always in favor of superstars. NBA players
are the most highly compensated, followed by MLB, NFL, and
NHL players.
Marginal revenue product offers the strongest explanation of
pay in sports. Essentially, in a competitive talent market,
players will earn somewhere between their highest and
second-highest MRP in the league. That is, players will earn
pretty close to their highest revenue contribution across all
the teams in the league. Player earnings over time also follow
the usual experience–earnings profile, and the explanation
extends quite nicely to coaches and managers.
The graphical analysis of the talent market reveals three
important things. First, marginal revenues are equal across all
teams at the equilibrium in the talent market. Second, large-
revenue market teams spend more on talent than small-
revenue market teams do. Finally, in a competitive situation,
large-revenue market teams will choose to win more than
small-revenue market teams. We should expect competitive
imbalance as long as there is revenue imbalance.
Marginal revenue product theory also sheds light on some
confusing sports outcomes. The theory helps show that while
there will be busts and bargains for any particular team at any
particular time, the market sends talent to its highest valued
use over time. Owners and their personnel directors do not
make systematic mistakes in the hiring market. Marginal
revenue product theory also helps to show that rising player
salaries do not drive up ticket prices. Instead, salaries rise
because fans’ willingness to pay increases. The theory also
shows how we can value sports stars’ services less than other
types of labor and still pay them more.
The fantastic payment to players has led some economists to
produce explanations of player pay and hiring not based on
marginal revenue product. However, these special
explanations for such high pay typically only apply in special
situations, if at all. The setting is not right for the winner’s
curse; bidding wars only make sense for a few teams and only
as they approach the play-offs; and the winner-take-all logic,
although a possibility for individual sports, does not apply to
team sports.
Economics can offer limited but occasionally strong insights
into gender, race, and ethnic discrimination in sports. Again,
based primarily on the marginal revenue product concept, the
insights are in the areas of pay and hiring differentials based
on gender, race, and ethnicity. Pay differentials are mostly a
289

thing of the past, but hiring variation and stacking still
remain. Although economics can help to identify
discrimination, it can contribute little to the debate on what
should be done about it, as that is not an economic question.
290

SECTION 11
Key Terms and Concepts
You should have run into each of these in pop-ups in the text
of this chapter:
• Marginal revenue product (MRP)
• Competitive talent market
• Free agents
• Experience–earnings relationship
• Experience–earnings profile
• Arbitration
• Talent market equilibrium
• Payroll imbalance
• Competitive imbalance
• Revenue imbalance
• Expected MRP
• Unbiased expectations
• Winner’s curse
• Bidding war
• Winner-take-all explanation
• Discrimination in pay and hiring
• Statistical discrimination
• Fan discrimination
• Cost of fan discrimination
• Owner discrimination
• Teammate discrimination
• Cost of owner discrimination
• Opportunity cost of time
• Worker job preferences
• Stacking
291

SECTION 12
Review Questions
1. Why are a league’s average and median salaries both
important when comparing salaries at the top and bottom
of the league for a given year? What is the implication for
the growth of salaries over time when the average salary
dwarfs the median salary?
2. What is endorsement income? In percentage terms, who
increases their incomes more with endorsements, male or
female athletes? Why?
3. Define marginal revenue product (MRP). Carefully
explain both elements of MRP (MR and MP). What
happens to MRP as winning percent increases? Explain
the impact on both MR and MP.
4. What is the age–earnings relationship? Draw a quick
graph of what it looks like for a pro athlete. Is this shape
any different than would be expected for any other job?
5. Define the competitive equilibrium in the market for
talent. What are the two main characteristics of this
equilibrium? What is the relationship between revenue
imbalance, payroll imbalance, and competitive imbalance
on the field?
6. Explain the idea of expected MRP. What does it mean if
owners have unbiased expectations about player
performance? If owners do have unbiased expectations
about player performance, can there be bargains and
busts?
7. What is the MRP explanation of the relationship between
ticket prices and player salaries? According to MRP
theory, which causes which? Why did salaries rise so
significantly during the 1980s even though ticket prices
typically fell over the same period in real terms?
8. What are the special circumstances that must hold for a
winner’s curse explanation of player pay? Would you
expect this explanation to hold over an extended period of
time?
9. What special circumstances must hold for each of the
following to explain player pay? When would you expect
to observe these special circumstances?
a. A bidding war
b. A winner-take-all outcome
10. How might one confuse a winner-take-all outcome with
one that is simply designed to give players in a
tournament the incentive to show up and play as well as
possible?
292

11. What are the explanations based on the marginal
product, or MP, part of MRP for economic discrimination
against women and minorities? How does statistical
discrimination follow from this explanation?
12. What are the explanations based on the marginal
revenue, or MR, part of MRP for economics
discrimination against women and minorities? Name the
two types of discrimination that follow from this
explanation.
13. What happens to an owner in a market with bigoted fans
if talent is hired purely on the grounds of ability and not
race? Explain why owner discrimination will not survive
in the face of vigorous competition over players.
14. State the three reasons for an observed deficiency of
hiring one race or gender relative to another.
15. Summarize the economic findings on pay discrimination
and hiring discrimination. What are the findings on
black/white pay differences? Racial stacking?
293

SECTION 13
Thought Problems
1. Using MRP theory, explain why NBA players make more
than other sports players. Using MRP theory, explain why
J.K. Rowling makes more than any pro athlete in Table
7.3.
2. Suppose winning one more game increases revenues by
$500,000. A solid slugger adds 55 winning percent
points. Rounding all of your calculations to the nearest
whole number, show that adding a solid slugger is worth
$4.5 million to the team owner in terms of marginal
revenue.
3. Mark McGwire held the home run record prior to Barry
Bonds. During the 1998 season when he earned the
record, compared to his career average McGwire played
in 49 more games (career average 106), had 153 more at
bats (career average 356), and 60 more hits (career
average 92), and of course, hit 40 more home runs (career
average 30). Because he earned $8.9 million but had a
career year, sports writers would have called him a
bargain. How would you evaluate this claim?
4. In our example in Table 7.4 and Figure 7.1, Randy
Johnson won 20 games with a terrific earned run average
in 1997. He was paid about $7.5 million. In 2001, he had
nearly an identical season to 1997 but earned almost twice
as much. Does this violate MRP theory? Why or why not?
5. Observers of pro sports leagues have lamented over
competitive imbalance in one league or another at
different points in time. Without exception, these
observers point to payroll imbalance as the culprit.
Correct them or agree with them, and justify your
correction or agreement.
6. The examples in Figures 7.3 through 7.6 are only that.
How would you perform a more thorough investigation of
the relationship between ticket prices and salaries? Fully
explain your method and the data you would need.
7. How can society value teachers more than ballplayers but
end up paying ballplayers more? Explain using a graph
like the one in Figure 7.7.
8. If the problem with Canadian teams is that they cannot
compete in Canadian dollars, why don’t they just write
the contract in U.S. dollars and pay players that way?
9. Looking at the outcome of overinvestment in a pro career
by some youths, it would be easy to identify it as a
winner-take-all outcome. But what is the argument
against it? How do you come down on the issue?
294

10. Explain the limits of any economic analysis of player pay
and hiring discrimination.
11. In the case of fan discrimination, depicted in Figure 7.8,
show what happens if fans learn over time and the level of
discrimination falls by half in 1 year. What if it falls by
half again the following year?
12. 1How would you design a test that would allow you to tell
whether some observed pay differential between genders
was due to statistical discrimination?
13. Now that salaries in all major pro sports are equal by
race, does that mean that discrimination is no longer an
economic problem in sports? What other indicator of
discrimination suggests that it remains an economic
problem?
14. The 2003 RGRC reported that 41 percent of the head
coaches in the WNBA were women. Does this mean that
there is hiring discrimination against women at that
position? State the arguments for and against that
position. Can you draw any conclusions?
15. Scully (1989) suggests that stacking is behind the
absence of managers of color in MLB, according to the
Learning Highlight: Why Are There So Few Black MLB
Managers and NFL Head Coaches? But what lies behind
the stacking outcome in the first place? Explain your
answer fully.
295

SECTION 14
Advanced Problems
1. What would the impact on player salaries be if a
professional sports league with free-agent players decided
to relocate one of its teams? Think in terms of MRP and
explain your answer fully.
2. Adjusted for inflation, the growth rates of player pay in
Table 7.2 (except for the NBA) are truly extraordinary. Do
you think that player salaries can rise forever? Why or
why not?
3. Starting from a separate graph of the talent market
equilibrium in Figure 7.2 for each case, show what
happens to the price of talent and the amount of talent
hired by two teams if
a. Fan demand in both cities increases willingness to pay
at the margin by 20 percent in both locations.
b. Economic hard times befall the fans in the larger-
revenue market location and incomes fall.
c. Foreign demand for American sports on TV increases,
and those fans always root for the underdog.
4. Take a two-team league as in Figure 7.2 and let the larger-
and smaller-revenue team’s marginal revenue functions
be

where L denotes the large-revenue team, S denotes the
small-revenue team, and W is winning percent.
a. Graph the MR functions.
b. Show the clearing talent price.
c. Determine what the W will be for each at the clearing
price.
d. Determine what the total talent bill will be for each.
5. Find MLB revenue data and team payrolls for 1994.
Graph payroll against revenue, and calculate the
correlation between them. What did you find? What does
that do for your thoughts about MRP theory?
6. How would you determine whether owners consistently
pay more for free agents or rookies than they are worth?
Describe the following:
a. The hypothesis you would test
b. The calculations you would do to perform your test
296

c. The data you would need in order to do the
calculations and perform the test
7. Suppose an NBA rookie will generate $5 million per year
in combined TV and gate revenue for an expected eight-
year career. At 6 percent, show that the discounted
present value of the most that an owner would pay over
those eight years for the player’s services is $31 million. If
the rookie would agree to $2 million per year for an eight-
year contract, show that the largest bonus that the owner
would pay to sign him is $18.6 million.
8. Why did it take so long for MLB to fully integrate after
Jackie Robinson joined the league in 1947? Did economic
discrimination end once the league was fully integrated?
What is the evidence that economic discrimination
continued even after that?
9. How would you design a test that would allow you to tell
whether some observed pay differential between races
was due to statistical discrimination? Fan discrimination?
Owner/teammate discrimination?
10. The following are the minimum, maximum, and average
purses and first-place prizes for men’s and women’s pro
golf tournaments in the year 2000 (extracted from Sports
Business Journal, April 2, 2001, pp. 29–33). Explain the
age and gender pay outcomes from the MRP perspective.
Explain the outcome from an age and gender
discrimination perspective.
297

SECTION 15
References
Bodvarsson, Orn B., and Raymond T. Brastow. “A Test of
Employer Discrimination in the NBA,” Contemporary
Economic Policy 17 (1999): 243–255.
Cassing, James, and Richard W. Douglas. “Implications of the
Auction Mechanism in Baseball’s Free Agent Draft,” Southern
Economic Journal 47 (1980): 110–121.
Ehrenberg, Ronald G., and Michael L. Bognanno. “Do
Tournaments Have Incentive Effects?” Journal of Political
Economy 98 (1990a): 1307–1324.
Ehrenberg, Ronald G., and Michael L. Bognanno. “The
Incentive Effects of Tournaments Revisited: Evidence from
the European PGA Tour,” Industrial and Labor Relations
Review 43 (1990b): 74S–88S.
Fort, Rodney, and Andrew Gill. “Race and Ethnicity
Assessment in Baseball Card Markets,” Journal of Sports
Economics 1 (2000): 21–38.
rank, Robert H., and Philip J. Cook. The Winner-Take-All
Society. New York: The Free Press, 1995.
Jewell, R. Todd, Robert W. Brown, and Scott E. Miles.
“Measuring Discrimination in Major League Baseball:
Evidence from the Baseball Hall of Fame,” Applied Economics
34 (2002): 167–177.
Kahn, Lawrence M. “Discrimination in Professional Sports: A
Survey of the Literature,” Industrial and Labor Relations
Review 44 (1991): 395–418.
Kahn, Lawrence M. “The Sports Business as a Labor Market
Laboratory,” Journal of Economic Perspectives (2000): 75–
94.
Lehn, Kenneth. “Information Asymmetries in Baseball’s Free
Agent Market,” Economic Inquiry 22 (1984): 37–44.
Madden, Janice F. “Differences in the Success of NFL Coaches
by Race, 1990–2002: Evidence of Last Hire, First Fire,”
Journal of Sports Economics 5 (2004): 6–19.
McCormick, Robert E., and Tollison, Robert D. “Why Do
Black Basketball Players Work for Less Money?” Journal of
Economic Behavior and Organizations 44 (2001): 201–219.
Nelson, Kevin. Baseball’s Greatest Insults: A Humorous
Collection of the Game’s Most Outrageous, Abusive and
Irreverent Remarks. New York: Simon and Schuster, 1984.
>Noll, Roger G. “Economic Perspectives on the Athlete’s
Body,” Stanford Humanities Review 6.2 (1998).
www.stanford.edu/group/SHR/6–2/ html/noll.html.
298

http://www.stanford.edu/group/SHR/6

http://www.stanford.edu/group/SHR/6

Peterson, Robert. Only the Ball Was White. New York:
Gramercy Books, 1970.
Porter, Philip K., and Gerald W. Scully. “Measuring
Managerial Efficiency: The Case of Baseball,” Southern
Economic Journal 48 (1982): 642–650.
Scully, Gerald W. “Economic Discrimination in Professional
Sports,” Law and Contemporary Problems 38 (1973): 67–84.
Scully, Gerald W. “Discrimination: The Case of Baseball,” in
Government and the Sports Business, ed. Roger G. Noll.
Washington, D.C.: The Brookings Institution, 1974.
Scully, Gerald W. “Pay and Performance in Major League
Baseball,” American Economic Review 64 (1974): 915–930.
Scully, Gerald W. The Business of Major League Baseball.
Chicago, IL: University of Chicago Press, 1989.
Szymanski, Stefan. “The Economic Design of Sporting
Contests,” Journal of Economic Literature XLI (2003): 1137–
1187.
299

SECTION 16
Suggestions for Further
Reading
Surely, I can come up with some.
300

CHAPTER 8
The History of Player
Pay
No one foresaw what was coming in salaries.
Not the players. Not the management. I
always had thought I’d be an Oakland A for
my whole career, that I’d end like Al Kaline or
Brooks Robinson. That just isn’t going to
happen anymore. It’s sad, bad for baseball, but
it is the truth.
—Reggie Jackson, MLB player
The highest paid of the first crop of 1976 free agents at
$3 million over five years.
Sports Illustrated, April 16, 1990, p. 114.
CHAPTER OBJECTIVES
After reading this chapter, you should be able to:
• Graphically demonstrate in general terms the
value to owners of reducing competition over
player talent.
• Understand both the theoretical reason that
player drafts reduce competition over player
talent and the particular ways the drafts have
been implemented in pro sports leagues.
• Understand the theoretical reason that the
reserve clause reduces competition over player
talent, and know the history of the evolution and
demise of the reserve clause in pro team sports.
• Describe the data that show how much reduced
competition over player talent has been worth to
owners, especially the value of the reserve clause
while it was in effect.
• Understand the invariance principle and its
implications for competitive balance.

SECTION 1
Introduction
After a 12-year career with the St. Louis Cardinals, Curt Flood
was traded to the Philadelphia Phillies. Under the legal
relationship in force between players and owners at that time,
he was obligated by his contract to go. Flood, however, had
established his life and business in St. Louis and did not want
to move it all to Philadelphia. In his appeal to Commissioner
Bowie Kuhn to be exempted from this trade, he said, “After 12
years of being in the majors, I do not feel I am a piece of
property to be bought and sold irrespective of my wishes.”
Commissioner Kuhn replied, “I certainly agree with you that
you, as a human being, are not a piece of property to be
bought and sold. This is fundamental to our society and, I
think, obvious[but] I cannot see its applicability to the
situation at hand” (Helyar, 1994, p. 108). The contradiction
was obvious to Flood, and he left baseball to sue MLB under
the antitrust laws. He lost, but helped set the stage for players
to win freedom over where they would play in the future.
In this chapter, we will examine the history of owners’
attempts to keep as much of the value created by players as
possible. The dramatic rise in salaries after free agency
indicates just how much owner restrictions on players were
worth. Perhaps the most important insight in sports
economics is Rottenberg’s (1956) invariance principle—
regardless of whether it is owners or players who get to keep
the value created by players, competitive balance remains
unchanged. As you will see, the draft and free agency did not
affect competitive balance. Some of the same arguments about
restricting competition for players made before free agency
are surfacing again, including the opinion voiced by Reggie
Jackson in the introductory quote for this chapter, and they
are just as incorrect now as they were then.
302

SECTION 2
Restricting Competition
over Talent
Thus far, our presentation of player pay has been in the
context of competition over player services. MRP is the most
that can be paid to athletes, and they receive close to their
MRP as long as competition prevails. However, pro owners
would certainly pay less if they could and pocket the extra
revenue. This would require making the market for player
services less competitive.
In this section, we will analyze why owners would reduce
player payments below their MRP. We will also present an
“owners’ eye” view of how competition in the talent market
could be reduced at the entry level and after players have
entered a given league.
THE VALUE OF REDUCING COMPETITION OVER
TALENT: GRAPHICAL ANALYSIS
Figure 8.1 uses the graphical analysis of the labor market
from Chapter 7 to analyze the value of restricting competition
over players. The figure shows the competitive outcome for a
two-team league where the price of talent is and the level of
winning percent is W*L = 1 − W*S. The shaded rectangles in
Figure 8.1 are prizes that can be won if employers can figure
out how to stop talent from collecting its MRP. With this
analysis, we need to recognize that there is a next-best
opportunity available to talent providers. Suppose the
opportunity cost outside the sport is N. If teams acting
together as a league can reduce the price paid to talent away
from P and closer to N, they get to keep the difference. Talent
bills are reduced to the much smaller shaded areas, and the
league keeps the amount shown by the top shaded rectangle,
namely, (P − N) × (W*L + W*S). (Note that this saving is
actually P −N, because the summed winning percents are
equal to one for a two-team league.) So far, the gains to
owners are just rectangles in a graph, but shortly we will see
just how valuable these gains may be to a league.
THE TWO-STEP PLAN TO REDUCE PAY
Put yourself in the place of owners. You have an organization
that allows you to act together with other owners to make
decisions jointly, namely, your league. If the league were to
design a system that reduced payment to talent to the lowest
possible level, you would be sure to start at the point where
talent entered the league. Then the league would think about
reducing competition for talent once it was in the league. Of
course, players would have to make enough to keep them in
their sport, and many intangibles would have to be covered in
order to keep players happy. But the bottom line is that
payment to players would be reduced and owners would keep
the difference. In the following sections we will present a
simple, two-step plan that owners have used to reduce
303

competition over talent and maximize their share of players’
MRP.
304

SECTION 3
Restricting Competition
over Incoming Players:
The Draft
The first step to reducing pay involves the entry of new talent
into the pro ranks. Under a competitive structure, all of the
pro teams would scout the country at the same time and in the
same places to turn up their own talent. Scouts from each
team would outrace and outbid each other to sign that new
talent. This type of competition drives up signing bonuses and
initial contract values to players. Think how much could be
saved if all of the league members agreed not to do that.
Spending in the pursuit of talent and payments to players
could be reduced. Note that the money does not go away but is
preserved for whomever gains rights to the player entering the
pro ranks. Somebody gets to keep the money, but not the
player. Just who gets to keep it depends on the method
devised for distributing the savings.
THE ORIGINAL IMPOSITION OF THE DRAFT: THE NFL
The NFL set the ball rolling. In 1936, NFL owners instituted a
college draft. This followed on the heels of two interesting
episodes. First, the Chicago Bears signed Red Grange
immediately after the 1925 college football season ended.
School was still in session, but Grange left and joined the
Bears for their last two regular-season games. College coaches
screamed bloody murder, fearing that the pros would start
raiding their rosters even before the season ended. In 1926,
the NFL agreed not to pursue college players until their class
had graduated.
The second episode that led to the imposition of a college
draft was the signing of Stan Kostas prior to the 1935 season.
Up to that time, any team could pursue any college player,
and competition could be brutal. In the Kostas episode, the
Brooklyn (football) Dodgers and Philadelphia Eagles bid up
his services to a $5,000 starting salary. This was as much as
the premiere players in the league earned, and owners saw the
writing on the wall. At the very next league meeting, the
reverse-order-of-finish draft was established. With this
system, the weakest teams draft first, enabling them to have
first shot at signing the best talent or (in some leagues) trade
the rights to draft the best talent to other teams. The drafted
players must play for the teams that draft them or not play at
all. The NFL did have a draft in the days before the Kostas
incident, but it was geographical rather than reverse order of
finish.
The NFL draft evolved over time, especially the “hands-off”
policy following the Grange incident. The current self-
imposed restriction is that the NFL will not consider players
until after their third year beyond high-school graduation. If
players take their usual redshirt year, they can be drafted after
two years of their college eligibility has expired. These are
referred to as “sophomores,” but this is not technically
305

correct; they would have been in college for three years. In
addition, the NFL draft also has a system where incoming
contract values are roughly inverse to the order of drafting;
higher drafted players sign larger, known contracts.
DRAFTS IN THE REST OF THE PRO SPORTS LEAGUES
The NBA college draft has existed since the formation of the
league for the 1949–1950 season, and it has undergone many
changes. Today, it consists of two rounds. In the first round,
the order of the draft for the first eight picks is determined by
lottery. After that, it is by reverse order of finish. The NBA
draft is also associated with a strict rookie pay scale based on
a player’s entry-level position in the draft. Finally, after years
of behaving to the contrary, the NBA no longer drafts players
out of high school and will not consider a player until after
one year in college (or approximately players of that age).
In MLB and the NHL, the draft started much later. The NHL
draft started in 1963. In MLB, the draft started in 1965, and it
continues today with each team drafting approximately 30
new players each year. Both leagues originally developed their
talent primarily through their own minor league systems, and
their drafts are much different than either the NFL or NBA
drafts. I won’t go into the details here since the general
impacts of the draft are our present focus.
GENERAL IMPACTS OF THE DRAFT
Today, the reverse-order draft is the standard in all pro sports
leagues (although the NBA decides draft order in early rounds
using a lottery system). The general predictable impacts of
such a draft system are clear. First, part of the value of talent
that would have gone to recruiting and signing compensation
(bonuses and initial contract levels) is preserved for the
league. Essentially, all competition over incoming talent is
eliminated since incoming players have no choice in signing
their first contract. Lately, the amount preserved for the
league seems to have diminished for two reasons. First,
training has become a year-round activity for all top-level
athletes and being forced to sit out a year is not as debilitating
as it was at the onset of drafts decades ago. Second, by and
large, most of the incoming athletes are simply worth too
much to let their services slide for a year. As a result, the
threat of sitting out a year and waiting for a subsequent draft
is a real one that owners take into account.
Second, the value that is preserved for the league goes to the
owners of the weakest teams in the league. Instead of the
owners of the strongest, larger-revenue teams signing the best
talent, the draft puts the best incoming talent on the weakest
(worst) teams by design. And players must sign contracts with
the owners of these weakest teams or wait to enter the league
some other time. The owners of the weakest teams enjoy the
benefit of the draft no matter how their fans respond to
higher-quality incoming talent.
If the weaker team’s fans get excited about the new level of
talent, then revenues respond so that the owners of weaker
teams can later hold on to that talent by increasing its pay. In
this case, the weak teams keep their high draft picks, enjoy a
306

higher level of talent, and collect the economic return
generated by higher quality. This is better for the league as a
whole in terms of competitive balance since weaker teams get
stronger. In addition, league revenues rise, and all teams
share in that success through gate-revenue-sharing and TV-
revenue-sharing arrangements (covered in Chapter 6).
Especially, the owners of smaller-revenue (weak) teams
benefit.
However, if the fans of weaker teams do not respond, then the
stronger teams will place a higher value on that talent and
obtain it from the weaker teams that drafted ahead of them.
In this case, weaker teams enjoy better talent for a short time
and earn some return when players move on to stronger
teams through contract trades and sales. For this to happen,
the teams must be able to trade or sell player contracts, and
players must be bound to follow their contract wherever it
goes. This is precisely the case for all players in all leagues for
the early parts of their careers. In this case, the value of the
draft is strictly monetary but still goes to smaller-revenue
owners that now simply channel talent to larger-revenue
owners.
This leads us to the third prediction about the impact of the
draft: Typically, weak teams are weak because that is the best
that their fans will support; the fans know that high-quality,
drafted players are coming but are not willing to pay enough
to cover the owner’s costs of keeping them. This leads to the
following prediction. It is most likely that fans will not
respond to better drafted talent and smaller-revenue owners
will just channel drafted talent to larger-revenue owners. If
this ends up to be true, the draft will not change the
distribution of talent in a league.
This third prediction is literally Rottenberg’s (1956)
invariance principle: It doesn’t matter whether players or
owners get to keep the value generated by players, the
distribution of talent will be the same either way. The
invariance principle is counter to other arguments in favor of
the draft. Owners defend the draft as a financial sanity
measure and as a tool for owners of weak teams to get talent
they otherwise would be unable to sign. Owners stress that
careful drafting and good player development can turn losing
teams into winners. This view suggests that the draft does
change the distribution of talent in the direction of improved
balance for the league. We will take a look at the invariance
principle in more detail later in this chapter because it is also
relevant for other league rules that have redistributed value
from players to owners.
307

SECTION 4
Restricting Competition
over League Players:
The Reserve Clause
The draft reduces competition over players as they enter the
league. The second step to reducing pay is the reduction of
competition over talent after it has been drafted. Suppose the
pro teams all agree that players must play for the team that
obtained them, unless that team chooses to send the player to
another team, by either trading or selling their contract. The
same rule holds for the next team that obtains the rights to a
player, and so on. This would eliminate competition and
pretty much force the players to live with the amount offered
by their current team. In the context of Figure 8.1, N would be
the best that the players could hope to do outside of their
sport, and the amount that owners would keep would be as
large as possible.
Attempts to eliminate competition over player talent are
practically as old as professional sports. James Quirk and I
(1992) trace it back to pro baseball. At its inception in 1876,
the National League (NL) had no rules governing competition
over talent. Owners and players operated in a truly free
market. Of course, team costs reflected this fact, and owners
were constantly bickering about talent raiding by their fellow
league members. Owners realized that steps had to be taken to
reduce that competition and raise owner profits at the
expense of players.
The first thing the NL did was to close the market to talent
raiding during the season. No contracts could be written until
well after the season was over. But this was just one step in
reducing competition over players. In 1879, one of the Boston
(Braves) owners, Arthur Soden, suggested that all teams
recognize a list of five players that every other team would
hold off the market at the end of the season. Under this
“reservation system,” salaries were cut in 1880, and more
teams were profitable than ever before.
Player contracts specified that they would obey the league’s
constitution and bylaws that included this reservation system,
so players had to live with it. Even though this was just the
beginning of such owner actions, John Montgomery Ward,
one of the chief architects of baseball’s first labor
organization, the Brotherhood of Baseball Players, recognized
the reservation system for what it was in 1890:
The reserve rule and the provisions of the National
Agreement gave the managers unlimited power, and
they have not hesitated to use this in the most arbitrary
and mercenary way.Players have been bought and sold
as though they were sheep instead of American citizens.
”Reservation” for them became for them another name
for property right in the player. (Lewis, 2001)
Ironically, Ward ended up as a lawyer for the NL, primarily
charged with enforcing the reserve system.
308

THE ORIGINAL RESERVE CLAUSE: MLB
By 1889, owners had inserted a reserve clause into player
contracts. It read as follows:
It is further understood and agreed that the party of the
first part [the team] shall have the right to “reserve” the
said party of the second part [the player] for the season
next ensuing[subject to the condition that] the said party
of the second part shall not be reserved at a salary less
than that [paid in the present season].(Quirk and Fort,
1992, p. 182)
Owners interpreted this clause to mean that they had a
perpetual option on the player. All they had to do was invoke
the reserve clause and extend the contract at its stated price
every year. Contract prices could only change if the player
bargained during the regular season. The only bargaining tool
players had was to threaten not to play at all for any team.
This particular version of the reserve clause did not stand
court tests, but the owners kept searching for language that
would stand. In the meantime, they kept the reservation
system in place.
In 1901, the new American League (AL) was formed. It raided
talent mercilessly from the older NL. Owners in each league
could see that the bidding up of talent in this fashion meant
lower profits for all. In 1903, they reached a “peace
settlement,” rewriting the National Agreement. One of the
primary terms of peace was the observance of all teams’
reserve clauses by all owners in both leagues. However,
players challenged the reserve clause for another 20 years.
THE FEDERAL BASEBALL DECISION
Not until the monumental Federal Baseball decision of 1922
was the issue of the reserve clause settled (Federal Baseball
Club of Baltimore v. National League of Professional Baseball
Clubs, 259 U.S. 200 [1922]). In Federal Baseball, a team in
the rival Federal League sued the NL over its reserve clause
under the relatively new antitrust law, the Sherman Act of
1890. The owner of the Federal League’s Baltimore club
argued that the reserve clause was a restraint of interstate
trade. The case made its way to the Supreme Court. Justice
Oliver Wendell Holmes ruled for the majority:
The business is giving exhibitions of baseball, which are
purely state affairs [italics added]the fact that in order to
give exhibitions the Leagues must induce free persons to
cross state lines is not enough to change the character of
the business.That which in its consummation is not
commerce does not become commerce among the States
because the transportation that we have mentioned takes
place.If we are right the restrictions by contract that
prevented the plaintiff [the rival league] from getting
players to break their bargains and the other conduct
charged against the defendants were not an interference
with commerce among the States. (Quirk and Fort, 1992,
p. 185)
309

Essentially, the court majority argued that baseball was not
interstate commerce and, as a result, did not fall under the
federal antitrust laws. The plaintiff lost, and this court
decision created what has come to be known as MLB’s
antitrust exemption. The Federal Baseball decision simply
meant that it would be nearly impossible to challenge the
reserve clause because the Supreme Court had failed to rule
against it.
Immediately, the owners inserted a binding (by Supreme
Court decision) reserve clause into every player contract.
Because player contracts are personal agreements between
parties, the precise wording of the clause may vary. However,
the reserve clause, unchanged from the Federal Baseball
decision until the 1950s, reads as follows:
[I]f prior to March 1,the player and the club have not
agreed upon the terms of such contract [for the next
playing season], then on or before ten days after said
March 1, the club shall have the right to renew this
contract for the period of one year on the same terms,
except that the amount payable to the player shall be
such as the club shall fix in said notice. (Quirk and Fort,
1992, p. 185)
Thus, owners went back to interpreting the clause to mean
that contracts never expired. When a player signed a contract
with either the AL or the NL, he accepted the entire contract,
including the reserve clause. Doing so bound the player to
that club as long as the owner submitted a new contract to the
player before March 1, whether the player signed that contract
or not. The clause was also included in the new contract as
well.
Because the contract never expired, players were bound to the
team holding their contract. Further, the players could never
let their current contract expire and seek higher compensation
in any competitive bidding process. Under this regime, as long
as contracts could be traded, owners were in control of the
value of players over their entire career.
OWNER ARGUMENTS IN FAVOR OF THE RESERVE
CLAUSE
For decades, all contracts in every pro sport contained a
version of the reserve clause. The owners in the other sports
essentially lifted the idea from MLB and imposed it on their
players to keep costs down and profits up. If a team sold or
traded the contract to another team, players had no choice but
to move, break the contract and face the legal repercussions,
or quit their sport. Competition over players was reduced, and
owners maximized their share of the players’ MRP.
Owners, of course, used noneconomic arguments in favor of
the reserve clause. First, they argued that it enhanced fan
confidence in the integrity of the outcome on the field. How
could fans believe that players were doing their best against a
team whose owner might be their employer in the next
season? By binding players under the reserve clause, the fans
could be sure that the players were doing their best. Further,
310

with players changing teams at the will of competition, fans
would lose identity with teams if rosters changed annually.
Owners went so far as to argue that abolishing the reserve
clause so that free agency ruled would be the end of sports.
They argued that in such a competitive regime the rich teams
would buy all of the good talent. Eventually, the leagues would
fall apart because the large-revenue market teams would be
unbeatable. Fans would end up watching the same few teams
in the championship every year. In time, fans would lose
interest, and the smaller-revenue teams would fold. According
to this argument, echoing Rottenberg’s uncertainty of
outcome hypothesis, the reserve clause preserved teams for
fans and cities that would otherwise not have a team. The
same idea fueled the argument in the last league to end its
player reservation system, the NFL. In discussions about
eliminating the last impediments to complete player mobility
in his league, NFL commissioner Paul Tagliabue echoed
arguments that were about 100 years old: “Total free agency
would destroy the National Football League” (Sports
Illustrated, September 10, 1990, p. 42).
THE END OF THE RESERVE CLAUSE IN MLB
In MLB, the reserve clause existed until 1975. There were
court challenges galore, but the courts’ answers always went
straight back to the Federal Baseball decision. The reserve
clause in baseball was specifically exempted from such
challenges. In 1974, MLB players finally started to overcome
the reserve clause. Under labor law, arbitrator Peter Seitz
decided that Oakland Athletics pitcher Catfish Hunter was a
free agent at the end of the 1974 season. Seitz agreed that A’s
owner Charley O. Finley had violated a clause in Hunter’s
contract concerning contributions to Hunter’s retirement
fund. With the contract violated, it was null and void, and
Hunter was free to sell his services to the highest bidder. He
ended up with the New York Yankees and became baseball’s
first million-dollar player (the million was over the duration
of a multiyear contract).
Almost immediately after that, pitchers Dave McNally and
Andy Messersmith let their contracts expire, played out the
1975 season without a contract, and sought arbitration when
their team owners invoked the reserve clause. Once again,
arbitrator Peter Seitz decided they were also free agents,
reversing the reserve clause that had existed in MLB since
1901 and that had withstood all legal challenges since 1922.
The floodgates were open, and the era of “free agency” was
born in MLB.
THE RESERVE CLAUSE IN OTHER LEAGUES
Interestingly, the courts were unwilling to extend the same
exemption to other sports. Reserve clauses began dropping
from NFL, NHL, and NBA contracts beginning in the 1950s.
But full competition for player services in these leagues did
not come until much later. In the NBA, players were free to
seek their highest payment, relatively unfettered, beginning in
1981. The NFL and NHL had the longest tenure of reserve
311

clause restrictions. Free agency was not fully established until
1993 in the NHL and 1994 in the NFL.
RESERVE CLAUSE REMNANTS
Free agency actually is limited free agency in all pro sports.
Continuing with the baseball example, free agency occurs
after only six years in the league (or in the sixth year for the
very top players). The other sports have similar restrictions on
younger players. According to the 2002–2006 Basic
Agreement negotiated between MLB owners and players, the
reserve clause for players with less than six years’ experience
reads as follows:
10.(a) Unless the Player has exercised his right to become
a free agent as set forth in the Basic Agreement, the Club
may retain reservation rights over the Player by
instructing the Office of the Commissioner to tender to
the Player a contract for the term of the next year by
including the Player on the Central Tender Letter that
the Office of the Commissioner submits to the Players
Association on or before December 20 (or if a Sunday,
then on or before December 18) in the year of the last
playing season covered by this contract. (See Article
XX(A) of and Attachment 12 to the Basic Agreement.) If
prior to the March 1 next succeeding said December 20,
the Player and the Club have not agreed upon the terms
of such contract, then on or before ten (10) days after
said March 1, the Club shall have the right by written
notice to the Player at his address following his
signature hereto, or if none be given, then at his last
address of record with the Club, to renew this contract
for the period of one year on the same terms, except that
the amount payable to the Player shall be such as the
Club shall fix in said notice; provided, however, that said
amount, if fixed by a Major League Club, shall be in an
amount payable at a rate not less than as specified in
Article VI, Section D, of the Basic Agreement. Subject to
the Player’s rights as set forth in the Basic Agreement,
the Club may renew this contract from year to year.
312

SECTION 5
The Value of Reduced
Competition over Players
In this section, we will examine just how much the reserve
clause was worth to owners. Three different analyses will be
presented. One estimates how much players should have been
worth under competitive salary determination and compares
it to their actual pay. The second analysis looks at what
happened to the salaries of that first crop of free agents in
MLB. Finally, we will examine how salaries changed in the
aggregate for all four major sports with the end of their
respective reserve clauses. At the end of the section, we will
discuss an important outcome of free agency—long-term
contracts.
ESTIMATES OF THE VALUE TO OWNERS OF THE
RESERVE CLAUSE
We will begin our examination of the value of the reserve
clause by observing the final insights from Gerald Scully’s
(1974) work “Pay and Performance in Major League Baseball.”
One measure of the amount of player MRP that was kept by
owners when the reserve clause was in force is called player
exploitation. Player exploitation is simply the percentage of
money employers keep over and above the amount actually
paid to players. Recall from Chapter 7 that part of Scully’s
approach was to estimate how much players would be worth
in a competitive market, that is, their MRP. Relevant to our
discussion here, he then went on to compare that estimate of
MRP with the actual salary paid.
Because salaries are less than MRP under the reserve clause,
the following simple ratio measures player exploitation: (MRP
– Actual Salary)/MRP. For hitters, Scully’s MRP estimates,
compared to actual salaries of the time, showed that owners
kept 79 percent to 88 percent of MRP, depending on hitter
quality. For pitchers, on average, owners kept 80 percent to
90 percent of MRP. Scully himself (p. 929) summarizes that
star players keep only about 15 percent of their MRP.
Note that if stars are receiving 15 percent of MRP, then the
owner keeps 85 percent. This implies that the owner gets
roughly six times the actual salary paid (i.e., 85 divided by 15).
The multiple jumps to nine if the owner is keeping 90 percent.
Let’s keep these percentages in mind as we look at what
happened to salaries with the advent of free agency. The
salaries that might have been for one MLB legend and
member of the Hall of Fame, Willie Mays, are in the section
ending Learning Highlight: Say Hey! The Reserve Clause
and Willie Mays.
THE FIRST CROP OF MLB FREE AGENTS
Sommers and Quinton (1982) looked for evidence of the value
of the reserve clause to owners in the salary data on the first
crop of free agents, shown in Table 8.1 (since the comparison
313

is relevant to those years, we break with our convention of
showing conversion to 2009 dollars). Except for one player,
most earned two to three times their previous salary once they
became free agents in 1976. Interestingly, and quite in accord
with the MRP explanation of player pay from Chapter 7, there
was one real “bust” in this first crop of free agents. Look at
pitcher Wayne Garland in Table 8.1. His salary increased over
eight times its previous level after only four years in the major
leagues. He pitched five more uneventful seasons as a spot
starter. This first crop also had some bargains. Gary
Matthews, also in the league for only four years when signing
as a free agent with San Francisco, eventually played 16 years
and made the NL championship series twice and the World
Series once with the Phillies in the early 1980s. On average,
this first crop of free agents included solid stars; Rollie
Fingers and Reggie Jackson were inducted into the Hall of
Fame in 1992 and 1993, respectively.
Now, the players in Table 8.1 should typically be thought of as
“stars,” so most of the salary increases in Table 8.1 are less
than Scully’s estimates suggest. Two of the tabled increases
are more consistent with what Scully expected for 90 percent
exploitation (both are pitchers, Wayne Garland and Bill
Campbell). The average of the increases is exactly the multiple
of four that is more consistent with 80 percent exploitation.
So he didn’t miss the mark by much and, besides, the years
shown in Table 8.1 were adjustment years; salaries continued
their upward adjustment to free agency for a few years after
1976. We can see this best by turning to the aggregate data for
all the pro leagues over time. In addition, turning to the
aggregate data allows us to look at the impact of free agency in
the other three major sports.
SALARY GROWTH OVER TIME
Table 8.2 shows average salaries five years before and after
free agency in each of the four major pro sports leagues (the
table is crowded so only 2009 dollar values are shown).
Baseball and hockey had no other salary restrictions, whereas
basketball implemented a salary cap two years after free
agency was put in place. Free agency and a salary cap both
were implemented at the same time in football. This variation
reveals interesting differences in the impact of free agency
that have yet to be examined in any detail.
In MLB, free agency was in effect for the 1976 seasons, and
there was an immediate increase of 9.0 percent in real salaries
(1975–1976). The real growth rate in salaries from 1971 to
1975, the five years prior to free agency, averaged 1.6 percent
annually. After free agency (1976–1980), with competitively
determined salaries, the average annual real growth rate
jumped to 17.8 percent. These changes are just what we would
expect jumping from a reserve clause to free agency without
any other restrictions on pay. Clearly, the advent of free
agency cost MLB owners dearly–all of these changes are in
comparison to real growth rate in the economy of about 3
percent.
314

The closest proximate free agency result is for the NBA (free
agency effective 1981, salary cap 1983). The immediate change
was 12.4 percent. The reserve clause was stifling prior to that
with a negative growth rate of –2.4 percent. Interestingly,
even though capped, free agency reversed the negative growth
rates to 10.6 percent annually. Indeed, while further work is
required, it could well be that the cap kept the growth rate in
average salary lower after free agency in the NBA than in
MLB.
In the NFL years later (free agency and cap, 1994), we find the
only episode where the immediate change in average salary
was negative (–7.8 percent). In addition, despite earning true
free agency, the NFL cap reduced the real annual growth in
the average salary from 15.7 percent to 8.1 percent. Clearly, at
least as indicated by the behavior of the average salary, the
cap both reduced the average immediately and reduced the
growth in the average salary dramatically.
For our final assessment, free agency in the NHL was earned
at almost the same time as in the NFL (free agency 1993). The
immediate increase in the average salary was an astonishing
32.2 percent. Interestingly, however, the real annual rate of
growth did not change after free agency, remaining steady
around 15.5 percent. While the magnitudes are different in the
NHL, this pattern is the same as in the other sport without a
cap (at that time; the NHL instituted its cap in 2005), MLB.
The two capped leagues at the time of free agency, the NBA
and the NFL, exhibit quite different occurrences than the two
leagues without caps. They have by far the lowest growth rates
in average salary after free agency, so caps appear to have an
impact. The NFL cap also appears to have had the greater
influence actually reducing the rate of growth in average
salaries. This lends some support to the view that the NFL cap
was more tightly defined and enforced than the cap in the
NBA had been, following along with our discussion in Chapter
6.
315

LEARNING HIGHLIGHT: SAY HEY! THE RESERVE
CLAUSE AND WILLIE MAYS
Willie Mays’ career ended just about the time that the reserve
clause was dying, so he missed out on free agency. His salary
history is shown in Table 8.3. It is clear that Mays was a star
player. For example, in 1971, the minimum salary was $12,750
($67,119); the average was $31,543 ($166,049); and Mays
earned $180,000 ($974,556). Earning almost six times the
average isn’t bad, but what was his real worth to his teams?
In 2009 dollars, Mays earned about $13.9 million over his
career. Scully (1974) estimated that owners kept about 85
percent of a star hitter’s contribution to revenues prior to free
agency. This means that owners made about 5.7 times what
they actually paid a star hitter. For Mays, that would have
been a cool 79.2 million that owners were able to keep due
primarily to the reserve clause. Although $13.9 million is a lot,
$79.2 million is a lot more, and that does not even include any
interest that would have been earned on portions invested
over the nearly 50 intervening years.
Willie Mays earned $13.9 million over his career, but could
have made $79.2 million in the free-agency era. [Photo Mays.]
Source: U.S. News and World Report, April 19, 1985;
Spokane Spokesman-Review, April 29, 1991, p. C2; and
Business Week, August 12, 1985, p. 44.
316

SECTION 6
Special Issue: Long-Term
Contracts
In sports, risk is a double-edged sword. On the one hand,
owners would rather not undergo the expensive negotiation
process associated with annual hiring, especially for
established players who have strong bargaining power under
free agency. There is some economy to be gained by owners if
they can assure themselves a steady supply of a player’s
services and avoid annual hiring risk. On the other hand,
players face the risk of performance ups and downs and the
risk of injury. Players would like to ensure payment in the face
of these performance and injury risks. Long-term contracts
thus serve the needs of both sides of the talent market.
For quite some time now, the price of talent has risen, and it
would be rational for owners to expect the price of talent to
continue to rise. Long-term contracts allow owners to play off
the injury risk faced by players against their own expectation
that the price of talent will continue to rise. Many players
would be willing to give up some expected salary against the
assurance that they will be paid something even in the event
of poor performance or injury. Some owners will be willing to
offer such a contract if it reduces current payments to talent
by a large-enough amount. As a result, players take a lower
price per year and avoid the risk of variable earnings. Owners
reduce talent costs and take a chance on injury. The length
and payment sequence depends on expectations and the way
the players and owners feel about risk. Further, detailed in
Chapter 9, long-term contracts are guaranteed only in MLB.
In other leagues, injury contingencies and other forms of
performance clauses are contract specific.
THE DOWNSIDE TO LONG-TERM CONTRACTS:
SHIRKING AND STRATEGIC EFFORT
On the topic of effort by players, former Pirates manager, Jim
Leyland, once said that it wasn’t money that instilled effort by
players; a player that hustles for $600,000 would do the same
for $6 million. (Sporting News, September 5, 1994, p. 14).
Sports history, however, is replete with examples of shirking,
the expending of effort at a rate below a player’s potential.
Athletes are not necessarily any different from others who
would rather exert less effort for the same salary. These
episodes can be extremely expensive for team owners. At the
same time, shirking can be very difficult to detect.
THE DIFFICULTY IN DETECTING SHIRKING: AN
EXAMPLE
In 1998, Washington State University quarterback Ryan Leaf
was the second player selected in the NFL draft. The San
Diego Chargers signed Leaf to a five-year deal worth at least
$25 million that could have topped out at $31.25 million. His
signing bonus was $11.25 million. Leaf subsequently became
317

the worst quarterback in the league, and his workout schedule
was what can only be described as “light.” The press had a
field day calling into question his determination and work
ethic.
Only Leaf knows the answers, but if effort is the issue, this is
precisely the kind of problem that long-term contracts might
breed. As a 23-year-old with $11.25 million in his pocket and a
hefty payday just for showing up, the question was whether he
had decided that he had worked hard and long enough. Of
course, owners want players at top performance at all times,
and effort can be tough for an owner to monitor. So what are
owners to do?
ECONOMIC REMEDIES FOR SHIRKING
Shirking actually is an old economic issue falling under the
heading of principal–agent problems. How do the principals
of any firm protect themselves against workers who do not
have the same interests? One answer is to expend resources
on monitoring effort. In sports, entire staffs, from personnel
management to on-field coaches, are devoted to overseeing
talent and its performance.
The sports world has monitoring tools not found in the usual
firm oversight processes. First, all players are pulled through a
grueling selection process from the time they reveal any talent
as children. Talented players who put forth great effort are
selected for higher levels of play, from youth sports on into
select teams and high school, college (for all four major
sports), and the minor leagues (for baseball and hockey).
Players that fail to pass this scrutiny fall by the wayside.
Direct oversight by parents, coaches, on-field managers,
general managers, owners, teammates, sports writers,
gamblers, and fans puts players under the microscope
through their entire career.
In addition, although observers may never know the greatest
potential performance of any given player, many other players
have approximately the same bundle of talent. Comparisons
between players are always possible. For example, players’
agents and their ownership adversaries bring a long list of
comparable players to the table in order to determine the
value of any given player during arbitration and salary
negotiations. The observations of players who do work hard at
least allow effort to be compared.
STRATEGIC EFFORT
A related, stickier issue remains. Players may work reasonably
hard all of the time but only perform at their absolute top
levels at specific, economically strategic times prior to
arbitration or contract negotiation. This is called strategic
player effort. In the unpredictable world of sports, it can be
difficult to tell whether performance patterns are due to
strategic player effort or uncertainty. Players can always say
that they just had a bad year or a nagging injury.
This type of problem is not new to those studying economics.
All types of firms have to deal with the problem of obtaining
high levels of effort from salaried employees and contract
consultants. The tried and true approach is to devise
318

incentive-compatible mechanisms that tie rewards to
outcomes.
PERFORMANCE INCENTIVES AND CONTINGENCY
PAYMENTS
In sports, incentive-compatible mechanisms are either actual
performance incentives or contingencies that are paid only if
the outcome is consistent with a high level of effort. Actual
performance incentives in contracts make up only a small
share of overall compensation due to union preferences in the
collective bargaining arena, but they do exist. Typically, they
are in the form of bonuses for playing time or for making the
postseason.
Contract contingency payments are much more common. For
example, take Peyton Manning’s original contract with the
Indianapolis Colts (Manning was drafted ahead of Leaf in
1998). If Manning played a significant proportion of the
season, he earned a shot at early free agency. But if he did, the
team could take advantage of that outcome and preclude his
free agency pursuit with an agreed-upon bump in annual
salary. Either way, Manning was best off if he exerted top
effort. In addition, long-term contract payments increase over
the duration of the contract. Barry Sanders signed a $34.56
million ($45.9 million), six-year contract in 1997 with the
Detroit Lions. The signing bonus was $11.75 million ($15.6
million). His salary increased each year from 1997 to 2002,
beginning at $1.4 million ($1.9 million) and ending at $6
million ($8 million).
Signing bonuses are also contingency payments. In the case of
Ryan Leaf, the San Diego Chargers were constantly
threatening to revoke signing bonus payments without a
substantial increase in effort by Leaf. Similarly, when Barry
Sanders retired prematurely, the Lions owners went after a
prorated share of the signing bonus. These types of
contingency payments are very common effort-inducing
mechanisms as detailed in this section’s Learning Highlight:
A-Rod’s $252 Million, or Is It $69.5 Million?
EVIDENCE ON THE DOWNSIDE OF LONG-TERM
CONTRACTS
Despite all of these mechanisms to overcome the downside to
long-term contracts, work by economists suggests that
shirking and strategic effort still persist. For example, Lehn
(1982) found that time spent on the MLB disabled list
increased with the rise of long-term contracting in the league.
If players were not getting injured at a higher rate, then they
must have been choosing to go on the disabled list. Other
analysts have examined strategic performance behavior and
long-term contracts, finding mixed results. Some have found
that it occurs (Scoggins, 1993; Sommers, 1993), whereas
others do not (Krautmann, 1990; Maxcy, 1997).
319

LEARNING HIGHLIGHT: A-ROD’S $252 MILLION, OR
IS IT $69.5 MILLION?
At this writing, an important caveat given the incredible
increases in sports star salaries, the most famous contract in
sports history belongs to Alex Rodriguez of the New York
Yankees. “A-Rod” signed a contract touted at $252 million
over 10 years, leaving the Seattle Mariners for the Texas
Rangers in 2001. But before that contract ran out, he moved
on to the New York Yankees after the 2003 season. Details of
the contract are in Table 8.4 (correcting for inflation, among
other things, is the point of what follows, so nominal values
are in the table).
Contrary to the common observation that Rodriguez would
make $25 million per year with the Rangers, team owner Tom
Hicks claimed that the discounted present value of the total
contract is closer to $180 million. First, the contract
stipulated that portions of the payments were deferred to the
10 years after the contract expires (and interest on the
deferred pay is set at 3 percent), and the entire 10 years might
never come to pass. First, after seven years (end of the 2007
season), the Rangers agreed to increase A-Rod’s salary by
either $5 million or $1 million more than the league’s highest-
paid player for the final two contract years. That made the
2009 and 2010 earnings a minimum of $32 million each year.
If the Rangers decided not to meet this requirement,
Rodriguez would become a free agent.
A much more important reason why the entire 10 years would
be irrelevant was if A-Rod moved to another team. That came
to pass at the end of the 2003 season when a deal was worked
out between the Rangers and the New York Yankees. The
Rangers agreed to pay $67 million over the remaining years of
the contract to the Yankees (the details of this payment were
not released to the public). So, at present, the contract is a
$179 million contract with the Yankees (contract value from
2004 on, in Table 8.4), with the Rangers paying $67 million of
it.
In return, the Rangers received the Yankees’ more-than-
adequate second baseman Alfonso Soriano and minor
leaguers. In addition the Rangers escaped both the remaining
$112 million in salary payments and around $13 million in
interest on deferred payments. So in the end, A-Rod’s $252
million contract with the Rangers ended up to be “only” $51
million in salary, $12 million in deferred compensation plus 3
percent interest, and $6 million in signing bonuses. And that
total of $69.5 million is a lot less than $252 million!
Alex Rodriguez signed a multiyear contract with the Texas
Rangers touted at $252 million. Contingency clauses may put
the actual value closer to only $180 million and the Rangers
actually ended up paying Rodriguez around $59.5 million in
total.
Source: Street & Smith’s Sports Business Journal, December
18, 2000, p. 1.
320

SECTION 7
What About Competitive
Balance?
Players made out quite nicely after free agency, as one would
expect if their earnings during the reserve clause period were
depressed as much as 90 percent. Earlier we discussed how
owners argued that a host of problems would accompany any
interference with the reserve clause. In the rest of this section,
we will explore these competitive balance claims in detail.
ROTTENBERG’S INVARIANCE PRINCIPLE
As you recall from Chapter 7, payments to players are at their
maximum in a competitive talent market when marginal
revenues are equal. In a two-team league model, this would
mean that , where L and S denote the larger- and smaller-
market teams, respectively. This also means that the largest
possible amount that owners can keep when competition for
players is restricted also occurs when , because it is at that
equilibrium where the profits earned from player talent are as
large as possible.
Therefore, each owner wants to hire his or her profit-
maximizing level of talent and keep the largest possible
amount that can be generated by that level of hired talent.
This means that in either the competitive case, where players
earn their MRP, or the reserve clause case, where owners keep
as much of player MRP as possible, the same amount of talent
is hired by team owners. That is, talent is hired until in either
case. Remember that larger-market owners hire more talent
than smaller-market owners when this condition holds, so
there will be the same level of competitive imbalance in either
case as well.
This was one of the most important findings in sports
economics. Economist Simon Rottenberg (1956) was the first
to discover this outcome, and it has come to be called the
invariance principle. Essentially, it is defined as follows:
The distribution of talent in a league is invariant to who
gets the revenues generated by players; talent moves to
its highest valued use in the league whether players or
owners receive players’ MRPs.
The invariance principle predicts that competitive balance is
the same whether the talent market is competitive or
governed by the reserve clause. If the league is imbalanced to
start with, it will be equally imbalanced in either case. This is
directly the opposite of what owners predicted would happen.
Here is an intuitive description of the invariance principle.
Suppose the talent market is competitive; that is, there is no
reserve clause. Players, by and large, shop for the owner who
offers the highest salary and go there. The owners of larger-
revenue teams buy the most talent. Thus, the free agency
distribution of talent has larger-revenue market owners with a
higher level of talent than smaller-revenue market owners. If
321

competition over players is restricted—that is, there is a
reserve clause in place—the only difference is that players will
do short stints at teams where their value is lower. Eventually,
larger-revenue market owners will want to buy the contracts
of better players, and better players will move to the teams
with the highest-valued leaguewide use. In this case, larger-
revenue market owners just pay smaller-revenue market
owners to get the talent rather than paying the talent
themselves. The same competitive balance result occurs in the
reserve clause case as in the competitive case.
Restrictions on competition for players do not cause any
change in competitive balance. This is the crucial invariance
principle result. However, restrictions on competition do
reallocate payments away from players toward smaller-
revenue market owners.
The invariance principle also suggests another explanation for
owner enthusiasm for restrictions on competition for players.
Both the draft and the reserve clause reallocate portions of
MRP from players to owners. Even though owners might hoist
the banner of competitive balance to support their pleas for
restrictions on competition over players, they simply are
better off, financially, with the draft and reserve clause in
place.
THE INVARIANCE PRINCIPLE AND THE DRAFT
Owners argue that the draft equalizes talent in the league,
enhancing competitive balance. According to the owners’
argument, weak teams draft the better players, strengthen
themselves, and become more economically viable.
Additionally, fans get to enjoy the stability of team rosters
over time. Better players that would have been scooped up by
larger-revenue market owners in a competitive hunt for talent
now play for the owners of smaller-revenue market teams.
The invariance principle suggests other results. The
invariance principle predicts that the better players will move
to better teams, even if a draft is installed. Larger-revenue
market owners will compensate smaller-revenue market
owners for talent rather than find the talent on their own. If
there were no draft, the players would get the money because
they would simply go directly to the teams in higher-revenue
markets and collect the returns in terms of higher starting
salaries and signing bonuses. What do the data say?
PLAYERS ACTUALLY MOVE BEFORE THEY ARE
DRAFTED!
Here is the first thing the data tell us. In leagues like the NHL
and NFL, where draft picks can be traded even before the
draft occurs, the usual perception is that teams are trying to
trade for higher draft picks. Sometimes this does occur, but it
can also make perfectly good sense to trade higher draft picks
for lower ones. On this outcome, Carolina Panthers General
Manager Bill Polian once said, “The economics, and the price
you have to pay for these rookies, has certainly affected how
you draft. You begin to erode the real purpose of the draft—
which is to equalize talent” (Wall Street Journal, April 18,
322

1997, p. B9). Polian is lamenting what has come to be called
trading down draft picks. Trading down means that a team
actually trades higher draft picks for lower draft picks (in
leagues where trading draft picks are allowed). By the way,
notice that Polian’s claim about the “real purpose” of the draft
(to equalize talent) is counter to the invariance principle.
The reason that it can make perfectly good sense to trade
higher-valued draft picks for lower-valued ones is that an
owner may not be able to afford to pay the signing bonuses.
Even though there is a reverse-order-of-finish draft, owners of
weaker teams often cannot take advantage of it. Rookie
bonuses have been driven so high that smaller-revenue
owners cannot afford to sign high draft picks. Because they
never get these better players, they also never get to enjoy the
compensation from selling the high-value talent to larger-
revenue teams. Players benefit by getting more of their
entering MRP because they are drafted by owners of better
teams at higher bonuses. But trading down at least allows
smaller-revenue owners to get something for their draft pick.
In any event, regardless of whether owners trade draft picks
up or down, talent moves to higher-valued uses (presumably
to larger-revenue market owners) even before the draft
occurs. Certainly, in these cases, the draft simply transfers
wealth without impacting the eventual distribution of talent in
the league.
PLAYERS DO NOT STAY PUT ONCE THEY ARE
DRAFTED
Consistent with the invariance principle, but contrary to the
owners’ justification for the draft, better players do move to
larger-revenue market teams after they are drafted. For years,
owners of the Seattle Mariners drafted terrific pitching.
However, their revenue position did not allow them to hold
onto those pitchers. The Mariners simply became a training
team for the rest of the pitching rosters in MLB. The reward
for finishing low in the standings was a continual
replenishment, through trades and high draft picks, of a roster
full of good minor league talent. And the Mariners then
developed that talent for the rest of the league owners.
A recent example from the NBA also points out that strong
talent does not necessarily stay with the owners who drafted
it. According to Sports Illustrated (June 23, 1997), only three
of the 10 players drafted first in the 1987–1996 drafts
remained with their original team—David Robinson of the
San Antonio Spurs, Glenn Robinson of the Milwaukee Bucks,
and Allen Iverson of the Philadelphia 76ers. Only one of the
remaining seven ended up at a smaller-market team—Joe
Smith. The draft simply does not keep players on smaller-
revenue market teams.
EVIDENCE ON THE DISPERSION OF WINNING
PERCENT
According to owners’ arguments, the draft should enhance
competitive balance. The invariance principle states that
competitive balance will not change with a draft. One way to
measure competitive balance is with the ratio of the actual
323

standard deviation of winning percents to the idealized
standard deviation, introduced in Chapter 6. The closer the
ratio is to one, the more balanced the league is in terms of
talent. If the owners are correct, then the ratio should have
fallen after the draft was put in place. Which view do the data
on winning percent support?
Table 8.5 shows the standard deviation ratio in the NFL and
MLB in the years before and after their drafts were
implemented. The ratio rose in the NFL and fell in both
leagues in MLB. In this type of analysis, the question becomes
whether the changes were statistically significant. In this case
they are not statistically significant by an arbitrary (but
reasonable) choice of, say, a required 10 percent difference.
Statistically speaking, the difference before and after the draft
in both the NFL (a 6.6 percent increase in the standard
deviation ratio) and the NL in MLB (a closer call at 11.3
percent) is probably not significant. But it does appear that
competitive balance improved in the AL since the average of
standard deviation ratios fell about 21.8 percent.
However, whether it was the draft that led to improved
balance in the AL is open to question. Recall from Chapter 3
that CBS owned the New York Yankees from 1964 to 1973.
While the Yankees won 10 of 12 AL pennants from 1953 to
1964, the year that CBS took over, the team only won one
pennant in the period after the draft, 1965 to 1976. That was
in 1976, after the group headed by George Steinbrenner
bought the club. So was it the draft, or was it a failed
experiment by CBS aimed at reducing the competitive power
of the Yankees? Because Baltimore won seven of the twelve
pennants after the draft, season outcomes were closer, but the
title was still pretty concentrated, which supports the failed
experiment explanation. Essentially, Baltimore just took over
the Yankees’ spot during the CBS ownership period.
The data reject the owners’ argument that the draft equalizes
talent in the NFL and in the NL in MLB because the standard
deviations do not fall significantly. The distribution of talent
and, subsequently, team winning percents are mostly
invariant to the existence of a reverse-order-of-finish draft, as
predicted by the invariance principle.
HISTORY REPEATS ITSELF: THE BLUE RIBBON PANEL
REPORT, 2000
Many observers worry that the level of competitive imbalance
has become a problem in MLB. Competitive imbalance has
always existed in baseball, but some think it has become
worse. Perhaps in response to these worries, MLB
commissioned a special highly respected panel to analyze
baseball’s competitive balance and make any
recommendations it saw fit to make. The “Blue Ribbon Panel”
was composed of
• Richard C. Levin, president of Yale University (and professor
of economics)
• George J. Mitchell, former senator and peace negotiator in
Northern Ireland
324

• Paul A. Volcker, previous chairman of the Federal Reserve
Board
• George F. Will, noted wit, syndicated columnist, baseball
expert, and member of the board of directors of a few
baseball teams
The Blue Ribbon Panel Report, 2000 (Levin, Mitchell,
Volcker, and Will, 2000), came down squarely that MLB had
significant competitive balance problems. It suggested
(among other things) a number of alterations in MLB’s draft
system. Our guiding light in the analysis of the Blue Ribbon
Panel Report recommendations for the player draft is the
invariance principle. The principle suggests that the draft has
no impact on competitive balance, and the evidence we have
examined tends to support this conclusion. Let’s examine the
Blue Ribbon Panel Report’s suggestions to see the invariance
principle in action.
A PROPOSED COMPETITIVE-BALANCE DRAFT
The report suggested a “competitive-balance draft” that would
let teams at the bottom choose players from teams at the top
but not from the 40-man (major league) roster. The worst
eight teams would get to draft from the minor league rosters
of the eight teams that made it to the play-offs the year before.
Actually, this is a variant of a type of draft that existed in
baseball just prior to the imposition of the current draft in
1965, and Rottenberg (1956) addressed this very issue!
These players will eventually end up at their highest-valued
location. If that location is not with their current team, it may
not be with a weaker franchise either. If the weaker team that
participates in the competitive balance draft would have been
the eventual home of players chosen in a competitive balance
draft anyway, then that special draft would not alter
competitive balance. If it would not, then players who are
drafted by weak teams in the competitive balance draft would
eventually end up where they would have gone anyway. All
that happens is that the competitive-balance-drafting team
gets the reward rather than the original drafting team. Indeed,
if the talent market is sending players to their highest league
value, in all likelihood, the players going to weaker teams in
the competitive-balance draft will just turn around and head
back to where they came from. Weaker teams will be
compensated by those receiving teams. Again, nothing
happens to competitive balance.
EXTENDING THE DRAFT TO INTERNATIONAL PLAYERS
Another reform would extend the draft to include
international players. This would end the current “who finds
them first” approach, which sounds like MLB’s mode of
domestic operations prior to the draft in 1965. However, this
simple extension will have the same impacts as the original
imposition of the draft back in 1965: Owners get to keep the
money with no impact on competitive balance.
ELIMINATING COMPENSATION DRAFT PICKS
325

Currently, MLB draft picks go to team owners who lose
players to free agency. The idea is that better teams will have
more top draft picks after they lose free agents. As a result,
they can draft sets of players to their advantage rather than
just individual players. The report calls for elimination of this
practice. But at the risk of sounding like a broken record, the
players they draft would eventually end up with that team
anyway. Therefore, competitive balance should not change if
these picks are eliminated. Besides, it is typically smaller-
revenue market teams that lose the bulk of the free agents in
MLB.
ENDING THE REMAINING ELIGIBILITY OF COLLEGE
PLAYERS ENTERING THE DRAFT
Under another draft remedy for imbalance, amateurs entering
the draft would lose their remaining college eligibility, as they
currently do in basketball and football. The report notes that
if the NCAA were to adopt this position, bargaining leverage
would be altered because players would not be able to fall
back on college ball if they were unhappy with their bonus
offer. This is good for MLB owners, but it is unlikely that
incoming college players would agree with them. Again, this
does not change who gets the players; it just changes the
balance of bargaining over the amount. It would have no
impact on competitive balance.
CREATING IMBALANCE IN THE NUMBER OF DRAFT
PICKS
Another idea in the report would have weak teams draft more
players than strong teams. Play-off teams from the preceding
year would not be allowed to draft until the second round.
Here, the impacts are the same as under the current system; it
is just that more of the value goes to weaker teams than
before. Eventually, talent ends up in the same place, and there
is no impact on competitive balance.
ALLOWING TRADING OF DRAFT PICKS
Finally, like other sports, the panel suggests that MLB teams
ought to be allowed to trade draft picks. All draft pick trading
does is move players more efficiently to the larger-revenue
market teams and provide yet another way for larger-revenue
market owners to compensate smaller-market owners at the
outset rather than after drafting has occurred. This occurs
through trading up or down, discussed earlier. Talent will still
end up where it is most valuable. Again, this will have no
impact on competitive balance.
In summary, none of these draft changes will do anything
except redistribute even more money away from incoming
players and owners of larger-revenue market teams to owners
of lower-revenue teams than currently occurs. It seems that
the invariance principle, a concept that has been around for
nearly 50 years and empirically verified over and over again,
has been conveniently forgotten when MLB owners assess
their financial welfare.
326

THE INVARIANCE PRINCIPLE AND THE RESERVE
CLAUSE
Owners had long argued that the reserve clause was essential
to competitive balance: If players moved according to the
dictates of competition, then the rich teams would buy all of
the good players. The invariance principle suggests otherwise.
As you now know, larger-revenue market owners value
players more than other owners, and they will buy talent from
smaller-revenue market owners. All the reserve clause does is
transfer the value of the players to owners holding those
players’ contracts, but the players move to their highest-
valued use to the league, nonetheless. Under a reserve clause,
the owner trades or sells the contract and gets the value of the
move. If the talent market is competitive and players are free
agents, the players run out their current contracts, move to
that location, and collect the higher value for themselves. But
here is the important point: The players move to the larger-
revenue market team with or without the reserve clause.
DO PLAYERS REALLY MOVE MORE WITHOUT A
RESERVE CLAUSE?
Let’s look first at the movement of players under free agency.
Owners raised concerns about roster stability in the absence
of the reserve clause. However, there is little evidence to
suggest that free agency had much effect on player movement.
The most comprehensive analysis of player movement I have
read is by Tom Ruane (1998). According to Ruane, player
movement is actually quite consistent decade-by-decade from
1950s through the 1980s. About 23 percent of players change
teams there was a 1 percent decrease in players moving after
free agency. The 1990s saw an increase in the average
number of moves to 28 percent.
However, movement does not capture everything. Maybe
better players are moving more often. That is, one might
compare player moves versus performance moves. Controlling
for playing-time variables (at bats for hitters and innings
pitched for pitchers), Ruane finds that the story is a little
different. Again over the 1950s through 1980s, there is a 2
percent per year trend upward in better players changing
teams. And the 1990s show a marked increase of 6
percentage points over 1990 for these better players.
Therefore, although the amount of movement did not change,
the players who were moving were a little better after free
agency than before.
Essentially, the owners overstated the actual consequences of
free agency on player movement. But even to this day, owners,
fans, and sportswriters often lament the end of the golden age
when star players finished their careers with their original
teams. Whether this ever really was true is the topic of this
section’s Learning Highlight: Was There Ever a Golden Age
of Player Loyalty?
EVIDENCE ON THE RELATIONSHIP BETWEEN MARKET
SIZE AND WINNING
327

If the owners are right, the shift to free agency in 1976 should
be accompanied by an increase in the dominance of larger-
revenue markets. Fizel (1997) reviewed the literature on this
topic and found little support for increased dominance based
on market size. Large-market teams do not consistently win
more than small-market teams after free agency. By this
measure, competitive balance survived free agency
unchanged, as the invariance principle predicts.
EVIDENCE ON THE DISPERSION OF WINNING
PERCENT
Table 8.6 shows the standard deviation of winning percents
before and after the demise of the reserve clause in MLB (you
should be able to state why it is appropriate to use the actual
standard deviations rather than the ratio measure used in
Table 8.5). Owners’ claims would have it that the standard
deviation of winning percents should increase after the
demise of the reserve clause. A look at the post-free-agency
years of 1976 to 1985 shows that the standard deviation did
not change at all for the NL. The change in the AL is, if
anything, to less balance rather than more. But, again, a
reasonable “ten percent change” criterion is not really met for
the AL either (the change is 10.3 percent). This is the
invariance principle in operation. In passing, I also found that
the initial imposition of the first formal statement of the
reserve clause in 1880 did not significantly alter competitive
balance either (Fort, 2005).
For those who might take issue with my call on the AL, there
is the following additional observation. Was the decrease in
balance the result of free agency, or was it the result of the
resurgence of the Yankees as they returned to private
ownership under Steinbrenner’s group? Evidence in support
of the latter explanation was found in Chapter 6. The standard
deviation of winning percent fell on average in the 1970s and
1980s, lending more evidence against the owners’ view.
EVIDENCE ON CHAMPIONSHIPS
The distribution of league championships provides a different
view of competitive balance. The owners’ view would be that
free agency would result in a higher concentration of
championships in the hands of larger-revenue market team
owners.
Fizel (1997) analyzed the number of contenders for division
championships within seasons. If balance were harmed by
free agency, there should be fewer contenders after the demise
of the reserve clause. Fizel also analyzed the number of
different teams winning division championships across
seasons. A decline in the number of different teams winning
championships would indicate less competitive balance. The
result of free agency is that pennant races have become
furious battles with multiple contenders, and the chance to
build dynasties has declined because the number of teams
winning championships has increased. These results are
consistent with the invariance principle but not with the
owners’ claims about free agency.
328

The Gini coefficient is another way to examine the
distribution of championships. If the Gini coefficient rose
after free agency, then championships were concentrated on
fewer teams. Such a result would be consistent with owners’
views that competitive balance is harmed by free agency.
Again, the evidence presented is from MLB (Fort and Quirk,
1995). In the NL, the Gini coefficient fell by 2 percent after
free agency, from 0.400 to about 0.392. In the AL, there is a
larger fall of about 10 percent, from 0.415 to 0.372. These
decreases in the Gini coefficient on championships certainly
do not support the owners’ claims against free agency on
competitive balance grounds.
ANOTHER TWIST: PLAYER PREFERENCES UNDER
FREE AGENCY
The fall in the concentration of championships for the NL,
along with the earlier report that the standard deviation of
winning percent also fell in MLB generally in the 1970s and
1980s, does not support the invariance principle either. Under
the invariance principle, there should have been no change in
the concentration of championships.
The invariance principle requires that owners and players care
only about profit and income. But players also care about
other things. As we mentioned in Chapter 7, player
preferences may also involve winning, locations, racial
attitudes, and press pressure. Perhaps some of the
nonmonetary elements of talent supply are responsible for
these findings. For example, Ken Griffey, Jr., asked the Seattle
Mariners for a trade in 1999 despite efforts by Mariners
owners to make him the highest-paid player in the history of
MLB. He claimed that he wanted to be closer to his family and
reduce their travel. He also claimed that he would take less
money in order to satisfy these other goals.
If nonmonetary preferences are prevalent, then some players
may not move to their highest financial value to the league.
They may stay with teams that do not pay as much as the
teams that owners would have sent them to under the reserve
clause. This type of behavior might actually enhance
competitive balance; some better players will stick with some
smaller-revenue market teams.
The Yankees always get all the best talent, right? Well, think
again. It’s really the case that they do not get all the players
they seek. The reason, of course, is the same as in all job
markets. There are nonwork aspects to every job. In 1993, for
example, the Yankees offered $10.4 million ($15.3 million)
more for six free agents—Barry Bonds, Greg Maddux, Doug
Drabek, David Cone, Terry Steinbach, and Jose Guzman—
than those players eventually accepted with other teams
(Sporting News, March 8, 1993, p. 13). The duration of
contracts was the same, and the Yankees offered an average of
$400,000 ($590,000) more per player. Only Bonds actually
took a higher-paying contract with the San Francisco Giants,
but all went elsewhere. Bonds had this to say after signing his
subsequent $90 million ($132.8 million), five-year contract
that placed him only fifth among other players in 2002: “My
heart has always been here,” Bonds said. “No amount of
329

money would make me leave San Francisco, to be honest with
you. I always wanted to stay a San Francisco Giant. Unless
there was a blockbuster, out-of-the world offer, I wasn’t going
to leave. All I want now is a World Series ring” (espn.go.com,
January 14, 2002). Bonds sounds like he is singing that old
Beatles tune with the famous line “Money can’t buy me love.”
We also have more on Greg Maddux’s choice. Yankees GM
Gene Michael had this to say after losing Maddux to the
Braves: “Salaries throughout the industry have outgrown a
player’s financial need to come to New York—and Maddux is a
perfect example” (Sporting News, March 8, 1993, p. 13). The
same article quoted Maddux as saying, “I felt the Yankees
were in a rebuilding process; that Atlanta’s ability to win was
better than New York’s. The decision was not hard at all.
Atlanta was where I wanted to go and as soon as the Braves
made room for me, the decision was over.” At the margin, it
appears that winning mattered a bit more to Maddux than
money. The Yankees offered him $32.5 million ($48 million)
for five years, while the Braves got him for $28 million ($41.3
million) for five years, about a 14 percent sacrifice.
One of the all-time great examples of players taking less than
they could get is the Minnesota Twins in the 1990s (Sports
Illustrated, January 27, 1997, p. 58). Five free agents
collectively took $25.5 million ($41.6 million) less to play in
Minnesota rather than play for more elsewhere. Two more
didn’t even bother to wait for another offer before sticking
with the Twins. Indeed, one of these two, Paul Molitor, took a
$1.5 million ($2.4 million) pay cut to play for the Twins.
330

LEARNING HIGHLIGHT: WAS THERE EVER A GOLDEN
AGE OF PLAYER LOYALTY?
Some fans lament the instability of team rosters. Many wax
nostalgic for the days of flannel uniforms and high socks when
baseball was “just a game” and icon players ended their
careers with the team that brought them into the league. More
often than not, they blame it on free agency. Commissioner
Bud Selig echoes these laments (Stark, 2004), “You worry
sometimes that franchises won’t be identified anymore with
certain players. You hope, as a sport, to have as much of that
(player identification) as possible.”
Analysis in the economics literature suggests a detectable
increase in attendance associated with roster stability
(Kahane and Shmanske, 1997). But to the extent that we see
roster stability, it occurs because it makes some contribution
to higher winning percents and higher attendance. Once that
contribution is gone, owners alter lineups to regain their
targeted winning percent level. So there is a trade-off between
a stable roster that fans like and eventual declines in winning
percents that fans won’t stand for. All this really means is that
there is a profit-maximizing level of roster stability.
All well and good, one might say, but has the duration of
roster stability, especially as it pertains to beloved star
players, changed since free agency? The evidence suggests
that it has not. The first piece of evidence is already cited in
the text; players move at about the same rate with just a slight
increase in the quality of players who do move.
The second piece of evidence concerns the rate at which
players end up with the teams they started with. In 1934, 12
players on starting rosters had been with their teams for at
least 10 years. There were 16 teams in the league with roughly
400 players on major league rosters. That means that those 12
represent only 3 percent of the total. Similar extremely small
percentages from 2 percent to 5 percent have been found for
1964 (23 players), 1975 (15 players), 1994 (17 players), and
2000 (15 players). Interestingly, Hall of Fame players average
three different teams during their career. Even Ty Cobb spent
his last two seasons with the Philadelphia Athletics after 22
seasons with Detroit. These results suggest that long tenure
on a given team may be a matter of sheer chance.
It makes sense that this would be so, both before free agency
and after. Let’s think about baseball only after the institution
of the draft in 1965. Players are distributed through the
league, typically to teams that they eventually will leave. It
takes time for that to occur, easily three to six years just to
make the majors and as much as 10 years for some. As their
skills develop over time and are revealed, their value changes
over time and they move to those teams that value them the
most. According to Rottenberg’s invariance principle and the
evidence in this chapter, this was just as true before free
agency as after. The only difference was in whose choice it
was. And that brings us to the loyalty issue.
Prior to free agency, the period fans ironically refer to as the
golden age of roster stability, there was no such thing as
player loyalty because it was impossible for players to exercise
331

loyalty at all. Players simply went wherever their contract was
traded or sold. Apparently, only in very few cases did a
player’s contract remain most valuable to the team where he
started his career. After free agency, just as ironically, is the
only time that player loyalty can even enter into the picture.
Even then most free agents move strictly involuntarily.
Players with fewer than 10 years with the same club do not
typically have any control over being traded once they sign a
free-agent contract. Involuntary free agency is especially acute
among the oldest players (in the economics literature,
Horowitz and Zappe, 1998, document the impact of this
treatment on players at the end of their careers). Long-time
player agent Tom Reich puts it this way: “And even with free
agency, go through this year’s list of free agents. What you’ll
find is that the vast majority are involuntary free agents.
They’re not free by their own choice. They’re free because
their team doesn’t want them anymore. And there’s a
tremendous difference between voluntary free agency and
involuntary free agency” (Stark, 2004). Precisely so: Those
involuntary free agents who may have wished to remain loyal
actually have no chance to reveal any loyalty at all.
So it is only the very few superstar free agents who have any
chance to exhibit what fans think of as loyalty to their team.
From this perspective, it simply is misplaced anger that
blames free agency for the lack of roster stability. Blaming
player mobility on a lack of loyalty reveals a
misunderstanding about who can exercise loyalty and when
(historically). The text suggests that player preferences for
loyalty have been exercised in many cases, a recent example
being Barry Bonds’s love affair with San Francisco, detailed
shortly (prior, of course, to his alleged issues with
performance-enhancing drugs).
One last pair of historical comparisons serves to emphasize
that roster stability has nothing to do with free agency. While
Lou Gehrig and Joe DiMaggio spent their entire MLB careers
with the Yankees, Babe Ruth came from Boston and ended his
career there again with the Braves. A later Yankee, Roger
Maris, played his entire career prior to free agency. But he
played for twice as many teams (Cleveland, Kansas City, New
York, and St. Louis) as the player who eventually broke his
single-season home run record well after free agency, Mark
McGwire (Oakland and St. Louis).
Ultimately, there is a question of fairness here. Baseball is a
business. Why should there be more loyalty in the baseball
business than in any other? Gene Orza, chief operating officer
for the Major League Baseball Players Association, puts it this
way: “If you’re a Minnesota fan and I said, ‘You have to stay in
Minnesota for six years,’ most people would say, ‘No, I’ve got
an obligation to my family. If I get a better job somewhere
else, I’d have to take it.’ And that’s the American way. Loyalty
has to take second place to freedom” (Stark, 2004).
Sources: Canham-Clyne (1994); Oberman (1998); Caple
(2004); Stark (2004); Kahane and Shmanske (1997);
Horowitz and Zappe (1998).
332

SECTION 8
Yearning for the Good Old
Days:MLB Owner
Collusion, 1985-1987
By and large, free agency was the order of things in MLB by
the end of the 1970s. The astronomical increases in salaries
demonstrated earlier in this chapter were previously the
portion of player MRP kept by owners. The result, of course,
was that player costs increased and revenues did not. Owners
were not in danger of going out of business, but they were
making less on net than they had during the reserve clause
era. It appears that old habits die hard because during the
mid-1980s two different arbitrators found the owners guilty of
colluding to suppress pay to free agents. The episodes came to
be called Collusion I, II, and III by the press, denoting the
sequence of the findings over the years 1985, 1986, and 1987.
The presentation here relies heavily on Scully (1989) and
Staudohar (1996).
COLLUSION I
In Collusion I, outsiders could make the following
observations. At the end of the 1985 season, 62 players
became free agents. Of those, 57 ended up signing with their
previous clubs because they got no serious offers; that is, if
they got an offer, it was only from a smaller-market team. The
top free agents literally received no offers.
The players’ union filed a grievance claiming collusion by the
owners. The agreement between players and owners in force
at the time stated quite explicitly, “Players shall not act in
concert with other Players and Clubs shall not act in concert
with other Clubs” (Staudohar, 1996, p. 38). The players’ side
made the easy case first. They argued that no single team
would turn away from the best available players in the league
unless all teams did the same. Otherwise, they risked
becoming less competitive. Then the players’ side noted that
Commissioner Peter Ueberroth had scolded teams about
bidding up free-agent prices at meetings when there were
ongoing negotiations with some free agents. Finally, the
players’ side noted that advisors had told owners that free
agents typically signed to long-term contracts showed a
decline in performance (see the discussion of shirking earlier
in this chapter). Owners replied that there was no collusion,
just a natural slowdown to a stable market after the advent of
free agency.
Arbitrator Tom Roberts sided with the players. The linchpin
of his decision was that the free-agent market appeared to be
in full swing with the usual type of negotiations with free
agents until the meetings where Ueberroth scolded the
owners. The free agents were set loose again in an
extraordinary measure.
COLLUSION II AND III
333

This was not the end of it. Arbitrator George Nicolau presided
over the cases of both Collusion II and III concerning the 1986
and 1987 seasons, respectively. The players won both cases.
Every step along the way, Nicolau stated careful comparisons
between the language of the agreement between players and
owners and the actions of owners.
As a result, the owners’ “information bank,” used to share
free-agent offer information, was revoked as a collusive
activity. The owners were also hit in the wallet. Although 843
players made claims (most were tossed and received nothing),
some players received settlements as high as $2 million ($3.8
million). Eventually, the owners paid $280 million ($525.5
million), including interest and lost salary, to players. After
the decision, salaries resumed their upward climb.
The owners continued to deny they colluded after the
proceedings ended. Sports economist Gerald Scully (1989)
looked for economic evidence of collusion. First, Scully
compared the 1987 salaries of players who entered the free-
agent market in 1986 to those of players who did not enter
that market. He found that the salaries of hitters who entered
the free-agent market in 1986 were $260,000 ($506,000)
lower than the salaries of hitters who did not enter that
market at that time. For pitchers who entered the free-agent
market, the difference was $259,000 ($504,000). Scully
could not attribute this outcome to performance declines
across years—strike one against the owners’ claims of
innocence on the collusion charge.
Second, using the approach described in Chapter 7, Scully
calculated the MRP of free agents and nonfree agents in 1986.
He discovered that the ratio of free-agent salaries and their
estimated MRP was lower than the same measure for nonfree
agents. One would expect that the ratio should be higher—
strike two against the owners.
Finally, the ratio of player salary to league total revenue fell in
1987. It had remained constant or grown for many years. But
it would have required an additional $100 million ($188
million) to make the 1987 ratio of salaries to total revenue
equal to its 1982 level—strike three against the owners. Scully
comes down squarely that owners colluded.
334

SECTION 9
Chapter Recap
Graphical analysis shows that if ways can be found to reduce
competition over player services, owners get to keep the
difference between the higher competitive payment and the
lower, less competitive payment. The next-best option for
players is not their second-highest value to the league but
their highest-valued opportunity outside of their sport.
The draft and reserve clause transferred player MRP to
owners. The draft removed competition over entering talent,
reducing recruiting costs and signing amounts. The reserve
clause effectively ended the ability of players to capture the
benefits of competition for their services once in the league.
Instead, owners earned the value of players moving from one
team to another.
It has been estimated that 80 percent to 90 percent of player
MRP was transferred from players to owners. This is
undeniably a large amount of money. The first crop of free
agents in baseball, for example, typically enjoyed double or
triple their previous salaries, and a few as much as eight times
their previous salaries. It is clear how salaries in all sports
jumped at the time of the death of the reserve clause in each
league, except for the NFL, especially for superstar players.
With more on the line after free agency, owners facing the risk
of annual contracting and players facing both performance
and injury risks have come to rely upon long-term contracts.
There is a downside to long-term contracts, but both
monitoring of players and incentive arrangements counteract
it.
The invariance principle states that players go to the same
teams whether they are paid their MRP or owners are able to
extract portions of it. The implication of the invariance
principle is that competitive balance will be the same whether
players earn their MRP or not. The evidence pretty much
favors the principle. Competitive balance was for the most
part unchanged with the imposition of drafts, and it did not go
to ruin with free agency. This outcome suggests that
arguments in favor of the draft or against paying pro players
their MRP really were self-interested on the part of owners.
335

SECTION 10
Key Terms and Concepts
You should have run into each of these in pop-ups in the text
of this chapter:
• Reverse-order-of-finish draft
• Reserve clause
• Player exploitation
• Competitively determined salaries
• Long-term contract
• Annual hiring risk
• Risk of performance
• Risk of injury
• Shirking
• Principal–agent problems
Monitoring effort
• Strategic player effort
• Incentive-compatible mechanisms
• Invariance principle
• Trading down draft picks
• Player moves versus performance moves
• Distribution of league championships
• Collusion I, II, and III
336

SECTION 11
Review Questions
1. What is a player draft? Give the usual owner arguments in
support of the reverse-order-of-finish draft.
2. Describe the original hands-off policy of the NFL toward
college football in 1926. What was the goal of this policy?
3. Describe the original reserve system invented by Arthur
Soden. How did this system generate downward pressure
on player salaries?
4. What is the reserve clause? How was it used to end player
freedom of movement between teams?
5. What was the Federal Baseball decision? What effect did
it have on competition over players in MLB?
6. Give the usual owner argument in support of the reserve
clause.
7. How did players overcome reserve clauses in sports
beside baseball? How did baseball players eventually
overcome the reserve clause?
8. Define player exploitation. According to the percentage
estimates of Gerald Scully, how large was player
exploitation during the reserve clause era?
9. What is a long-term contract? Why did long-term
contracts evolve after free agency?
10. Describe annual hiring risk and risk of performance. How
do long-term contracts protect players and owners
against these risks?
11. What is shirking? What is strategic effort? What effect
does each have on long-term contracts?
12. State Rottenberg’s invariance principle.
13. What is trading down draft picks? Why does it occur?
14. What is the difference between player moves and
performance moves? Why does this distinction matter in
analyzing the impacts of free agency?
15. What do player preferences have to do with the impact of
free agency on the distribution of winning percents or
championships?
337

SECTION 12
Thought Problems
1. Using Figure 8.1, show graphically what happens to the
value of reducing competition over players when:
a. Fan willingness to pay increase
b. The opportunity cost outside of the sport falls
c. A rival league forms
2. Why did MLB and the NHL institute their drafts so much
later than the NFL and the NBA?
3. Why didn’t all players just let their current contract run
out and refuse to play until the owners gave in and
removed reserve clauses from all contracts?
4. Remnants of the reserve clause linger for the youngest
players in all leagues. What impact does it have on the
earnings of these young players?
5. Summarize the data on the value of reduced competition
over player talent. Be sure to include the Willie Mays
example, the first crop of free agents, and salary growth
over time.
6. Why did salaries grow at such a larger rate after free
agency in MLB and the NBA relative to the NFL and
NHL?
7. Why were long-term contracts nearly nonexistent prior to
free agency?
8. How do owners protect themselves against shirking in
long-term contractual arrangements with players? How
do owners protect themselves against strategic effort by
players with long-term contracts?
9. Despite how closely players are watched throughout their
careers, what evidence do we find in actual contracts that
shirking and strategic effort can still be problems for
owners?
10. If information about players is expensive, some players
may not find their way to their highest-valued league use.
The same might be true if the transactions costs (e.g.,
legal costs) of moving players between teams are high.
How do these factors temper the idea of the invariance
principle?
11. Using the invariance principle:
a. Evaluate owner arguments in favor of the draft
b. Evaluate owner arguments in favor of the reserve
clause
338

12. Does the income increase that happens for weaker teams
under the reverse-order-of-finish draft come from larger-
market owners? Players? Both? Explain fully.
13. How can trading down draft picks possibly make any
sense? What is the main factor in the market for player
talent that forces this outcome on weaker teams?
14. Suppose that player preferences are responsible for the
increase in competitive balance after free agency in MLB.
Does this negate the invariance principle?
15. What possible reason is there to care whether owners or
players get to keep the majority of player MRP?
339

SECTION 13
Advanced Problems
1. The WNBA once drafted players and assigned them
geographically to teams. What did they hope is true about
fan demand? Compare and contrast this to what the
men’s pro sports leagues believe about talent and fans.
2. The text of this chapter states a condition where the draft
actually will level the playing field. If this condition
occurs, is the invariance principle violated?
3. Explain fully what would happen to the following if the
draft were abolished:
a. Rookie player salaries and bonuses
b. Team profits
c. Competitive balance in the league
4. Explain fully what would happen to the following if free
agency were abolished and we went back to the reserve
clause:
a. Individual player salaries
b. Team profits
c. Competitive balance in the league
d. The length of player contracts
5. How would you be able to tell whether Ryan Leaf’s
performance problems were shirking or were just due to a
bad rookie year? Explain fully the data you would need in
order to tell. What are the chances that you will ever have
these data? What are the impacts of long-term contracts?
6. Suppose you took players who were up for a new contract
and then compared their efforts before and after their
new contract. Further, suppose their performance fell
after their new contract. Would you be able to tell from
this outcome whether these players were strategic in their
effort? If so, explain. If not, what additional data would
you need?
7. Why is it important to owners that the competitive
equilibrium level of talent on all teams be maintained
when measures to reduce competition over player talent
are implemented?
8. Summarize the evidence on the impact of the draft in
light of the invariance principle. Be sure to include what
happens to players before the draft, after the draft, and
the dispersion of winning percents.
9. Suppose that the NCAA followed the Blue Ribbon Panel,
2000, suggestion and ended eligibility for college players
entering the draft. Would teams in MLB be better off?
340

Why? Would college players be better off? Why? Would
there be any impact at all on competitive balance?
Explain fully.
10. Summarize the evidence on the reserve clause in light of
the invariance principle. Be sure to include whether
players really do move more, whether the relationship
between market size and winning changed after free
agency, winning percent evidence, and championship
participation evidence.
341

SECTION 14
References
Canham-Clyne, John. “Loyalty Griping Ignores Simple Truth,”
Baseball America, August 8, 1994, p. 10.
Caple, Jim. “Free Agency Simply Great for Baseball,”
ESPN.com, November 26, 2004.
Fizel, John L. “Free Agency and Competitive Balance,” in
Stee-Rike Four! What’s Wrong with the Business of Baseball?
D.R. Marburger, ed. Westport, CT: Praeger, 1997.
Fort, Rodney. “The Golden Anniversary of ‘The Baseball
Players’ Labor Market,’ ” Journal of Sports Economics 6
(2005):347–358.
Fort, Rodney, and James Quirk. “Cross-Subsidization,
Incentives, and Outcomes in Professional Team Sports
Leagues,” Journal of Economic Literature 23 (1995): 1265–
1299.
Helyar, John. Lords of the Realm: The Real History of
Baseball. New York: Willard Books, 1994.
Horowitz, Ira, and Christopher Zappe. “Thanks for the
Memories: Baseball Veterans’ End-of-Career Salaries,”
Managerial and Decision Economics 19 (September 1998):
377–382.
Kahane, Leo, and Stephen Shmanske. “Team Roster Turnover
and Attendance in Major League Baseball,” Applied
Economics 29 (April 1997): 425–431.
Krautmann, Anthony C. “Shirking or Stochastic Productivity
in Major League Baseball,” Southern Economic Journal 56
(1990): 961–968.
Lehn, Kenneth. “Property Rights, Risk Sharing, and Player
Disability in Major League Baseball,” Journal of Law and
Economics 25 (1982): 343–366.
Levin, Richard C. et al. The Report of the Independent
Members of the Commissioner’s Blue Ribbon Panel on
Baseball Economics. New York: Major League Baseball, July
2000.
Lewis, Ethan L. “A Structure to Last Forever: The Players’
League and the Brotherhood War of 1890,” Master’s thesis
(www.ethanlewis.org), Purdue University, 2001.
Maxcy, Joel. “Do Long-Term Contracts Influence
Performance in Major League Baseball?” In Advances in the
Economics of Sport. W. Hendricks, ed. Greenwich, CT: JAI
Press Inc., 1997.
Oberman, Keith. “Hall of Famers on the Hoof,” Sports
Illustrated, June 8, 1998, p. 74.
342

http://www.ethanlewis.org

http://www.ethanlewis.org

Quirk, James P., and Rodney D. Fort. Pay Dirt: The Business
of Professional Team Sports. Princeton, NJ: Princeton
University Press, 1992.
Rottenberg, Simon. “The Baseball Players’ Labor Market,”
Journal of Political Economy 64 (1956): 242–258.
Ruane, Tom. “Player Movement,” Baseball Think Factory,
www.baseballstuff.com/btff/scholars/ruane/articles/
player_movement.htm. (1998)
Scoggins, John F. “Shirking or Stochastic Productivity in
Major League Baseball: Comment,” Southern Economic
Journal 60 (1993): 239–240.
Scully, Gerald W. The Business of Major League Baseball.
Chicago, IL: University of Chicago Press, 1989.
Scully, Gerald W. “Pay and Performance in Major League
Baseball,” American Economic Review 64 (1974): 915–930.
Sommers, Paul M. “The Influence of Salary Arbitration on
Player Performance,” Social Science Quarterly 74 (1993):
439–443.
Sommers, Paul M., and Noel Quinton. “Pay and Performance
in Major League Baseball: The Case of the First Family of Free
Agents,” Journal of Human Resources 17 (1982): 426–436.
Stark, Jayson. “The Decision That Changed the Game, Part 3:
Player Movement a Fact of Life,” ESPN.com, November 23,
2004.
Staudohar P. D. Playing for Dollars: Labor Relations and the
Sports Business. New York: Cornell University Press, 1996.
343

http://www.baseballstuff.com/btff/scholars/ruane/articles/player_movement.htm

http://www.baseballstuff.com/btff/scholars/ruane/articles/player_movement.htm

http://www.baseballstuff.com/btff/scholars/ruane/articles/player_movement.htm

http://www.baseballstuff.com/btff/scholars/ruane/articles/player_movement.htm

SECTION 15
Suggestions for Further
Reading
Surely, I can come up with some.
344

CHAPTER 9
Labor Relations in Pro
Sports
Players say the owners are stupid, owners say
the players are greedy, and both sides are
right; they make one pine for simpler days
when the owners were greedy and the players
were stupid.
—Richard Corliss
On the inability of owners and players to avoid the
MLB strike of 1994.
Time, August 8, 1994, p. 65.
CHAPTER OBJECTIVES
After reading this chapter, you should be able to:
• Discuss the major elements of the modern labor
relations process, including the legal setting, how
owners and players are organized, and how they
interact to produce collective bargaining
outcomes.
• Explain why unions organize and what
organizational problems union organizers must
overcome.
• Understand that basic bargaining is the process
of looking forward and working back. The basic
50–50 outcome is adjusted by the alternatives
that are available to each side of collective
bargaining negotiations.
• Explain how work stoppages happen.
• Understand the history of labor relations in each
of the four major pro team sports.

SECTION 1
Introduction
In his book Instant Replay (1968), author Jerry Kramer tells
the story of star center Jim Ringo meeting with Packers coach
Vince Lombardi to discuss his 1964 contract. According to
Kramer, at the meeting Ringo told Lombardi that his agent
would be handling all of his contract negotiations. Lombardi
asked Ringo and his agent to wait outside, while he went in
his office. When Lombardi came back out, he announced that
Ringo would not need his agent to deal with the Packers—he
had just been traded to the Philadelphia Eagles. Although
Ringo has denied the story, in any case, he was traded to the
Eagles immediately after he asked Lombardi for a raise in
1964 (Quirk and Fort, 1992, p. 179).
Labor relations, the subject of this chapter, have come a long
way since 1964. Modern labor relations occur in an
environment governed by federal labor law that specifies the
rights of players to organize and the process of interaction
between unions and leagues. We can begin to understand
labor relations by recognizing unions for what they are,
namely, small democracies. As such, we should expect them to
pursue policies most valued by the subset of players who are
politically powerful within the union. We will also develop in
this chapter the rudimentary game theory behind bargaining.
Although work stoppages (strikes by players and lockouts by
owners) nearly never happen, they are significant events, and
we will examine how they may occur. Finally, we will cover
labor relations in each major sport separately.
346

SECTION 2
Modern Sports Labor
Relations
Labor relations—the interactions between players and owners
concerning employment, pay, and the negotiating
environment—were not a big deal in pro sports until the early
1970s (Koppett, 1991; Voigt, 1991; Staudohar, 1996;). But by
the 1970s, revenues were growing at an astonishing rate in all
pro sports. The trade union movement in sports grew slowly
and surely as the value of organizing rose. As sports evolved
into a large-scale entertainment industry, players began
looking more and more like entertainment stars. They
employed agents to negotiate contracts, and they got behind
their organized labor associations in the face of the
restrictions in place against their earnings. All major leagues
have unions in the era of modern labor relations.
EXCLUSION AND COUNTERACTING MANAGEMENT
POWER
Going back to the original trade guilds of Europe, labor was
organized in order to assure the progression from apprentice
to journeyman to master. Trade guilds served to train
individuals, but they also greatly limited competition. Later,
labor organized in order to counteract the power that firms
often have over worker earnings by reducing competition for
jobs and raising wages.
Early on, when no formal agreement existed among owners in
various leagues, pro sports players did well, jumping from one
team to another at the drop of a hat. However, owners did not
like this competition because it made it difficult to build
winning teams and pushed wages to competitive levels. Out of
this horrible situation (from management’s perspective) came
the controls covered in Chapter 8, most notably the draft and
reserve clause. Before the era of modern pro sports labor
relations, owners simply imposed drafts and reserve clauses
on players. Players confronted this power by organizing into
player associations and, eventually, legally recognized
collective bargaining units usually called unions.
In the mature stages of sports labor relations, players have
responded to management power by organizing into unions.
Although owners may wish otherwise, this type of cooperative
action is legal under labor law. But let’s not operate under any
illusions. The essence of modern sports labor relations is
power—organized owners against organized players. The
Learning Highlight: Champions of the “Free” Market for
Players? Not Who You Might Think, at the end of the chapter,
discusses the impacts of this power in more detail. In the rest
of this section, we will examine the legal setting and the major
actors in owner–player negotiations.
MODERN LABOR RELATIONS: THE LAW
347

The modern, mature structure of labor relations in sports is
shown in Figure 9.1. Owners acting through leagues and
players acting through unions interact with each other under
the legal structure that governs the collective bargaining
process. Collective bargaining just means that players are
allowed to act as a unit and owners are obligated by law to
recognize that right. The legal setting and major actors
depicted in Figure 9.1 are worth addressing individually.
THE NATIONAL LABOR RELATIONS ACT
State and federal labor laws comprise the legal structure that
governs collective bargaining. Most important in this regard is
the 1935 National Labor Relations Act (NLRA, referred to as
the Wagner Act after Senator Robert Wagner of New York)
and the National Labor Relations Board (NLRB) it created to
oversee the formation of unions and the process of collective
bargaining. The NLRA broadly covers workers involved in
interstate commerce and defines the criteria under which a
group of employees can act collectively through union
representation. It guarantees three essential rights for labor:
1. The right of labor to organize and form unions
2. The right of labor to bargain collectively through
representatives of their own choosing
3. The right of labor to use pressure tactics such as strikes
and picketing
Modern labor relations have evolved under the NLRA,
subsequent curbs on the tools unions are allowed to use under
the Labor–Management Relations Act (LMRA, 1947, also
called the Taft–Hartley Act after its sponsors Senator Robert
Taft of Ohio and Representative Fred A. Hartley, Jr., of New
Jersey), and under the decisions of the NLRB and the federal
courts. Hundreds of decisions have helped define the
collective bargaining relationship in all industries. A few of
the more important decisions in sports are discussed in this
chapter.
THE NATIONAL LABOR RELATIONS BOARD
The NLRB enforces the NLRA by deciding grievances brought
either by labor or management. Appeals of unfair labor
practices go to the NLRB for a ruling. If the NLRB finds a
problem, NLRB decisions are enforced through motions filed
in the federal courts. In sports, both unions and owners often
bring actions to the NLRB that involve violations of good-faith
bargaining on the part of the other side. Bargaining in good
faith means that opponents must be striving to reach an
agreement rather than striving to drive a wedge in
negotiations. For example, owners are required to provide
information about team finances to the unions so that the
unions can establish a basis for negotiation. Further, the
NLRB helps leagues and unions determine the scope of
bargaining, that is, what lies within the law and what does
not.
OWNERS, THEIR LEAGUES, AND THE COMMISSIONER
Owners, acting through leagues, are on one side of the process
in Figure 9.1. Remember from Chapter 5 that owners deal
348

with players’ associations as a league joint venture activity.
However, owners do not have to negotiate as a group with
their respective player associations (unions). For example,
Ford, GM, Chrysler, and American Motors choose not to join
forces in dealing with the United Auto Workers. Our
conclusion in Chapter 5 was that team owners must simply be
better off acting together to confront player associations.
Occasionally, league commissioners play a role in collective
bargaining. Technically, every commissioner is hired by the
owners, acting through their leagues and serving at their
discretion. As such, it would be wrong to say that the
commissioners act autonomously in the best interests of their
respective sports. Even though they portray themselves that
way, if they cross the owners, they can be fired. However,
commissioners do have some autonomy. For example, they
act as public spokespersons and buffers between collective
bargaining adversaries. They have also participated in labor
negotiations under the “best interests of the sport” banner.
PLAYERS, UNIONS, AND AGENTS
Unions represent players in the collective bargaining process.
Unions are governed by the very strict rules set forth in the
NLRA and LMRA. All unions are organized as representative
democracies of players. Typically, the equivalent of the trade
union local chapter is the individual sports team, with each
team electing a player representative. The representatives
meet to conduct the union’s business. All unions have elected
officers, including a president. In addition, hired executive
directors play an important role as professional negotiators.
Under the NLRA, once the players have ratified a union, it has
the right to represent them in all negotiations. Union
platforms are generated by the team representatives. All
actions by the representatives, especially work stoppages, are
voted on by the membership.
The role of agents in collective bargaining is often
misunderstood. Player agents represent players seeking their
highest financial reward in individual negotiations with
teams. However, as indicated in Figure 9.1, their access to the
collective bargaining process is limited. Agents are not
included in the collective bargaining process. Legally, they do
not have a place at the table. However, some unions have
taken a particular interest in player agents and have included
agent regulation in their agreements with owners. This is only
natural because players have agreed to allow unions to be
their sole bargaining authority under the NLRA.
The Major League Baseball Players Association (MLBPA),
National Football League Players Association (NFLPA), and
the National Basketball Players Association (NBPA) have
negotiated union-certification restrictions on agents with their
respective leagues. Owners have agreed that they will deal
only with agents certified by unions. The certification process
requires applicants to reveal any criminal past, detail their
sports experience, reveal any conflict of interest, and specify
their fees. Unions also typically require a standard contract
349

between agents and players that includes arbitration of any
disagreement. Only the National Hockey League Players’
Association (NHLPA) has chosen to stay out of the agent issue
completely. Agents do represent hockey players, but they are
not regulated in any way by the NHL.
COLLECTIVE BARGAINING AGREEMENTS
The essence of collective bargaining is that both owners and
players have relinquished some of their individual bargaining
rights to their representatives, leagues, and unions. Thus,
leagues and unions negotiate collective bargaining
agreements (or CBAs), which represent a clear specification of
the rights of owners and players in the process as well as the
items that result from the process. CBAs typically include the
following:
• The duration of the agreement (three years is the most
common, but negotiated extensions of a given CBA can be
for a longer period)
• So-called reopening clauses under which new negotiations
can occur before the expiration date of the CBA
• The rules and procedures governing the draft
• Pay issues common to all players, such as free agency,
minimum salaries, meal money, play-off pay, sharing
percentages under salary caps, and retirement funds
• Player location issues (such as the ability by veteran players
to have a say in where they will and will not play)
• Channels for addressing grievances, primarily arbitration
and mediation
Within the setting created through collective bargaining,
owners and players do retain some individual bargaining
rights. Owners and players each retain the right to negotiate
individual salaries. In addition, owners also retain the right to
sell and trade player contracts. However, this is usually
limited to younger players, because older players have refusal
rights specified in the CBA in each league. For example, MLB
has the “10-and-5” rule. Players with 10 years in the league
and five years for a given team have refusal rights over any
trade. In essence, this means that 10-and-5 players can dictate
the next team for which they will play.
STANDARD CONTRACT ELEMENTS
Player contracts have many elements that are subject to
collective bargaining. All contracts within a league include
standard sections, but those standard sections vary quite a bit
across leagues. Most contracts typically limit player
participation in other sports. Standard language appears in all
contracts covering breach of contract. This language is very
specific on redress to both parties in the event of a breach. All
contracts include standard language specifying grounds for
termination by the club—personal conduct, physical fitness,
declining skills, or failure to render services to the club. Injury
is not grounds for release. All standard contracts contain
language containing player rights to redress of grievances.
Disciplinary rules and filing deadlines, along with player
350

shares of group licensing and promotion, are also specified.
Most importantly, trades are also governed by standard
contracts. Typically, the teams have full control over trades;
however, no-trade clauses and items like the 10-and-5 rule in
MLB do exist.
SPECIAL LABOR RELATIONS TOPIC: SALARY
ARBITRATION
CBAs in the NHL and MLB explicitly cover the situation
where a player and owner cannot agree to a contract. In these
situations, individual players can exercise their collective
bargaining rights to have an independent third party decide
the issue. This is commonly referred to as salary arbitration,
and the independent third party is called the arbitrator. If a
dispute reaches arbitration, both sides must agree to abide by
the arbitrator’s decision.
In the arbitration process, players try to maximize the
expected value of the outcome. They choose the highest-
possible defensible salary and defend it by providing
information on other comparable players in the league, free
agents included, based on performance. The question then
confronting the arbitrator is, “How can a player be paid less
than comparable players?” Owners counter with a defensible
lower offer based on their evaluation of the player in question.
Arbitrators are the essential element in this process. They are
professionals serving at the discretion of both the players
associations and the leagues. Presumably, arbitrators wish to
keep their jobs. They must, therefore, decide between owners
and players in a way that both perceive to be fair. All things
considered, arbitrators strive to determine the value of a
player and then find in favor of the side that is closest to that
value assessment.
TECHNICAL ARBITRATION ISSUES: ELIGIBILITY AND
FINAL OFFERS
Two types of players are eligible for arbitration: free agents at
the end of their contract and non–free agents who can file for
arbitration according to the rules for that procedure in their
CBA. Free-agent players at the end of their contracts enter
into negotiations with their team owner. At any time during
the process, these players can file for arbitration. If an
impasse is reached, the player must decide whether to begin
arbitration proceedings or enter the free-agent market. Some
non–free agents are also eligible for arbitration (e.g., in MLB,
players are eligible for arbitration after three years in the
league but are not free agents until their sixth year). If non–
free agents fail to exercise arbitration, they simply accept the
owner’s pay offer.
In the NHL, arbitrators may choose any number they deem
fair between the owner’s offer and the player’s demand, but
MLB uses final offer salary arbitration (FOSA). This means
that the independent arbitrator cannot choose a fair
difference between the team’s offer and the player’s request.
Arbitrators must choose one or the other. Uncertainty about
the arbitrator’s choice actually gives both parties an incentive
351

to reach an agreement on their own before turning to
arbitration. Economically speaking, in order to let an
independent third party make the decision for them, parties
to a negotiation must be willing to live with the expected
outcome. If not, then the parties should work harder on
reaching an agreement prior to arbitration. As you might
expect, this contentious FOSA process is destined to create
hard feelings between owners and players.
FOSA OUTCOMES IN GENERAL
FOSA should lead to earnest and sincere salary negotiations.
Owners fear lavish decisions in favor of players, and players
fear that pushing owners beyond their ability to pay may
shorten their tenure with that team. As owners and players
fear the unknown, they should strive harder to reach an
agreement rather than face FOSA. Generally, this is the case.
Few players actually exercise their arbitration rights each
year, indicating that the salary determination process
generates satisfied owners and players in the vast majority of
cases.
Arbitration cases that do occur produce two interesting
outcomes. First, owners typically win more cases than the
players. However, simple tallies do not tell the whole story
about arbitration. This leads to the second outcome: Salaries
are higher in the presence of FOSA than they would be
otherwise. As we just saw, owners negotiate sincerely with
players who do not enter arbitration. The threat of losing
arbitration cases can lead owners to raise salaries to avoid
arbitration.
But FOSA, once invoked, also raises salaries whether players
win or lose their case. If the players win, their salaries rise
dramatically. If the arbitrator finds against a player, the
owner’s offer still is well above what the player made
previously because owners try to appear fair to the arbitrator.
Either way, the player earns more. For the players, it is a
matter of getting rich (if they lose) or richer (if they win).
Although players like this outcome, owners do not. They
dislike it because it allows players to move up the salary
distribution at a faster pace than would happen in the absence
of arbitration. Remember that arbitration is the result of
negotiation, and there is no reason to think that the system in
place before arbitration was either fair or efficient. Owners
prefer anticompetitive outcomes because they are able to keep
more money than they are able to keep under arbitration.
SPECIAL LABOR RELATIONS TOPIC: UNION
DECERTIFICATION
Union members forfeit their rights to individual legal action
when they vote for a union to represent them in collective
bargaining. For example, the antitrust laws protect all
individuals from the exercise of market power by firms in the
American economy. However, players organized as unions
cannot sue owners under these laws because they have voted
to deal with owners through collective bargaining. The
352

ultimate aim of players is to hold onto gains against owners
and, if possible, gain further concessions that enhance their
wealth and happiness. Almost always, these gains are made
through collective bargaining, but in very special situations,
individual action may further these goals better than
collective bargaining.
In these very special cases, players may decide on union
decertification to pursue gains against owners. Under
decertification, players simply vote to cancel collective
representation by their union. Such an approach is allowed
under the NLRA and LMRA and has been used to great effect
by football and basketball players (examples for each league
appear later in the chapter).
353

LEARNING HIGHLIGHT: CHAMPIONS OF THE “FREE”
MARKET FOR PLAYERS? NOT WHO YOU MIGHT THINK
Sports economist Allen Sanderson of the University of
Chicago has long pointed out that unions have acted just as
purposefully as owners have in their pursuit of power over
players’ earnings (Sporting News, September 9, 1995, p. 38).
Players do not want real economic competition any more than
owners do. For example, unions have never supported
completely unrestricted free agency. All restricted versions of
free agency restrict supply and raise the prices of experienced
players relative to complete free agency. Sanderson puts it
bluntly. Owners have monopolies over their franchises, and
the union monopolizes the labor market and uses that power
to get its share of these profits. In such a situation, players
make more than they would if economic competition ruled
baseball.
Professor Sanderson’s assertions were dismissed offhand by
then executive director of the Major League Baseball Players
Association (MLBPA), Donald Fehr. Mr. Fehr simply stated
that Sanderson’s perspective is wrong and based purely on
“philosophy.” He claimed to take the pragmatic high ground
by negotiating minimum standards and then trying to obtain
for players whatever freedom he could obtain.
Player union leaders recognized almost immediately that
complete free agency was not in their best interest, and so did
one owner. When free agency was on the horizon in MLB in
the mid-1970s, the then owner of the Oakland A’s, Charley O.
Finley, saw a chance for owners to salvage something from
free agency. If free agency was inevitable, then Finley’s plan
was to have as many players competing for each open spot as
possible, every year, which would bring salaries down.
Apparently, the then-executive director of the MLBPA,
Marvin Miller (1991, p. 370), heard of this and held his breath,
worrying that owners might actually heed Finley’s argument!
Miller observes that Finley’s point was logical and obvious
and should have been embraced by owners.
Miller meant that restricted free agency was a limitation on
freedom but one that raised the salaries of experienced
players more relative to truly unfettered free agency. Miller
and Finley both knew that annual free agency for all players
would increase the supply of labor and push wages down
relative to a version of free agency that would restrict supply
and raise the prices of experienced players. Finley simply
wanted to dance with his own devil—unlimited free agency—
rather than the one preferred by the players. Miller used the
remaining owner sentiment to negotiate limited free agency
that kept the labor market orderly because owners could only
bid up the price of the top players without setting so many
players free that salaries would fall.
Sources: Sporting News, September 9, 1995, p. 38; Miller
(1991).
354

SECTION 3
Union Goals,
Organizational Problems,
and Governance
Players formed unions to counter the market power that team
owners exercised over them and to limit competition over
their jobs. Thus, the goal of unions is to move players
economically forward against owners. However, obstacles
must be surmounted before unions can form. Once a union is
formed, its governance can lead to some surprising insights
into collective bargaining outcomes. Let’s first start with the
economic battle to be fought because it is those gains against
owners that fuel union organizations in the first place.
THE ROLE OF MARGINAL REVENUE PRODUCT
The economic struggle between owners and players is over
players’ marginal revenue product (our old friend MRP), the
contribution that players make to the value of team revenues.
At the time that unions formed in pro sports, all players
suffered from the exercise of owner power over player MRP, a
power derived from the reserve clause and player drafts.
These owner-imposed restrictions on players reduced
competition in the player market. In the presence of these
restrictions, players’ alternatives were limited to earnings
outside of their sport. As these outside earnings were typically
lower than player sport MRP, owners could keep the
difference between MRP and the next-best player
opportunity. In theory, owners would pay just enough to keep
players in their sport, and little more.
MRP is composed of two components: marginal product and
marginal revenue. Because team owners enjoy market power,
player contributions to team revenues are larger than they
would be in a competitive situation. The monopoly profits are
included in the evaluation of player contribution to revenues.
This makes the prize for union organization just that much
larger.
In a nutshell, the goal of unionization is to offset owner power
over MRP in order to drive payment closer to the level that
would be determined under a competitive pay structure. This
is the value to individual players of forming a union. In
percentage terms, the value that can be earned by individual
players is the level of salary exploitation discussed in Chapter
8:

The total value of forming the union will be the sum of these
values across all players. As noted in Chapter 8, estimates of
exploitation were in the 80–90 percent range. The value of
unionization is clearly in the hundreds of millions of dollars,
annually.
OBSTACLES TO ORGANIZATION
355

With that type of money at stake, it would seem a simple
matter to form a union. The players would just get together,
add up these values, recognize that hundreds of millions of
dollars are at stake, and set up a union structure. However,
there are quite a few obstacles in the way.
EDUCATION COSTS
Although some players may immediately recognize the value
of organization, not all will. Therefore, the potential union
incurs start-up education costs. From the very start of player
associations, players had to be educated about the salary
exploitation practices of owners and how they could be fought
through organized action. Because players are located across
the country, the costs of flying union leaders around the
country were large. Further, players had to meet in order to
hear the message, and their time was especially valuable
during the season. Collecting players and their leaders in
central places was costly.
The education task proved a formidable one for fledgling
unions. Players had a low level of understanding of their
actual value to owners and were distrustful of the motivations
of union enthusiasts. For example, as shown at the end of this
section in the Learning Highlight: Pro Athletes as “Company
Men” , most players in early testimony before congressional
subcommittees revealed very little opposition to the reserve
clause and the draft. In congressional testimony in 1958,
when asked if he favored legislation that would officially
exempt owners from the antitrust laws, Robin Roberts
(president of the MLBPA at the time) actually said, “Yes, sir. I
think if professional baseball owners themselves are in favor
of it, we are, because I believe—I would say this: I think
anything that is good for them in this particular line is
definitely going to be good for us.” Early on, players and their
representatives did not realize how much owner control over
MRP was costing them, and they would not find out for
almost 20 years. The players simply trusted that baseball
owners would look out for them.
This type of sentiment was difficult to overcome. Even years
later, when Curt Flood was battling MLB for free agency in the
courts, Marvin Miller (1991) noted this about the
unwillingness of stars to show support:
They may have wanted Flood to win, but they felt that
they had their careers to be concerned with, and that was
that. Even though I had explained the importance of
modifying the reserve clause, many of the players
remained in the dark about what the case might mean.
(p. 197)
FREE RIDING
Perhaps the most important problem facing all volunteer
organizations (unions, political action groups, and
philanthropic organizations alike) is free-riding behavior by
members. If you have ever felt cheated in a study group
because some members of the group did not carry their
weight, then you know all about free riding. Now, imagine you
are a players’ association organizer. You have done your job
356

educating players, and they now understand the exploitation
that they want reduced. Now is the time to generate official
membership and a budget so that you can carry forward the
association’s agenda. You appeal to the players for a dues
structure so that you can start your work.
Think about your request from the perspective of any given
player. The value of membership equals the difference
between that player’s MRP and actual salary. Union leaders
want part of that difference to go to union dues. However,
here is how a player might see the situation. If the rest of the
players pay their dues and one particular player does not, then
that player gets the benefits of the union activity (namely,
moving pay closer to MRP) without paying anything. The rest
of the players pay their dues, the union carries the battle and,
if it succeeds, generates salaries closer to MRP. Players who
do not contribute keep all of the gain without any of the pain
of union dues.
Therein lies the problem from the organizer’s perspective and,
ultimately, from the perspective of individual players. If all
players free ride, then no dues will be collected. The union
leadership will never be able to start their activities to wrestle
salaries closer to MRP. Free-riding behavior that seems such a
wise move from the individual perspective leads to collective
ruin when all practice it.
Unions overcome free-riding behavior in a number of ways.
First, once players get behind the union idea, there appears to
be a strong sense of solidarity. This greatly reduces free-riding
behavior. Second, there is the aspect of enforcement on the
field. An “errant” major league fastball, missed blocks on
massive defensive linemen, flagrant basketball fouls, and
menacing hockey sticks provide ample incentive against free
riding.
OWNER RETALIATION
Owner retaliation in the form of blacklisting union activists or
interfering with their professional development might also
hinder players’ union participation. Blacklisting has occurred
in the past. Shortly after World War II, Jorge Pasquel took a
shot at promoting his Mexican League to major league status.
He lured a few true stars—Mickey Owen, Luis Olmo, Danny
Gardella, Tommy DeLaCruz, and Sal Maglie—into jumping
their MLB reserve clauses for a chance at promised big
money. Owners feared the worst as this rival league reared its
ugly head. In 1946, Commissioner Happy Chandler
announced that players who jumped would be banished from
MLB for five years.
In the end, Pasquel’s dreams were bigger than both his fan
base and his checkbook. The league folded after just a year.
When players returned with their hats in their hands, owners
did not go after them in the usual way with lawsuits over
contract disputes. Instead, the disloyal players were punished
for jumping ship by the league. The banishment was reduced
to three years, and Owen and Olmo rejoined the Dodgers in
1949. (Ironically, Pasquel sued Owen for jumping back to
MLB!)
357

Penalties for jumping leagues are not technically retaliation
against union activity, but they do set the tone for
relationships between owners and players. Reports of
retaliation against player representatives were numerous
during the rise of the MLBPA under Marvin Miller’s firm hand
in the early 1960s. However, note that a statistical treatment
by Coffin (1999) did not find any difference between player
representatives and other players in time spent on a team, the
number of trades, or career length. Specific well-known
instances of retaliation are still worth exploring because their
impact on player willingness to participate in union activity
can be significant, even if the impact of occurrences on player
representatives, themselves, is not.
Owners also have interfered with the career development of
union activists. When the MLBPA’s executive counsel
returned from its meetings in Mexico City in 1967, union
activist and counsel member Jim Bunning found he had been
traded from the Phillies to the Pirates. Bunning’s pitching
statistics had been top-notch, leading Marvin Miller to
wonder, “I do not know how many times pitchers with a 2.29
ERA have been traded, but I’d bet it happens much more
often among pitchers who are player reps” (1991, p. 163).
Later, in 1968, pitcher Milt Pappas informed the MLBPA that
the Reds were not living up to collective bargaining
requirements on travel accommodations. Miller intervened,
suggesting that the team live up to its agreement or suffer the
consequences. Pappas was traded to Atlanta in the same
season.
Miller did not encourage marginal players to support Curt
Flood in his antitrust action against the owners in 1968. The
trial was during the season, and they had their careers to work
on. But Miller also noted the following:
Further, it was in the back of my mind that a great many
marginal players might be the targets of owner revenge
if Flood lost: A utility infielder who was active in the
union and made a public show of support for Flood might
find himself losing a job to a utility infielder who wasn’t
active in the union. Union reps had a tough time as it
was; they tended to be traded more often than players
who were less active in the union. (1991, p. 196)
Ten years later, the Twins signed relief pitcher and union
activist Mike Marshall as a free agent in 1978. Manager Gene
Mauch said to Marshall, “Of course, you’re not going to do any
of that Player Association stuff, are you? These people aren’t
very big on player reps” (Miller, 1991, p. 304). In June 1980,
after he allowed one run in five consecutive appearances, the
Twins released him for poor performance. The MLBPA
brought a grievance and the Twins settled, agreeing to pay the
remainder of his contract. Marshall remained active,
especially during the events leading up to the strike of 1981.
Marshall was hardly used by the Twins during these activities.
Again, these incidents do not prove owner retaliation. They
merely suggest that a very small group of players may have
suffered for their union involvement. Nonetheless, these
358

examples could very well sour some players on union
participation.
POLITICAL SENTIMENT AND ILLEGAL LEAGUE
BEHAVIOR
Still other factors can affect the ability of a union to form and
perform. First, hostile political environments can undo union
efforts. It was years before player trade unions even were an
imaginable concept in the United States. Second, there are
examples of illegal owner behavior. At the outset, MLB
owners controlled the operating budget of the MLBPA. This
practice was not only an obvious conflict of interest (how
would the union survive if it pushed the owners too hard?),
but also an illegal act under federal labor law. It had been
common practice for years before Marvin Miller took over the
union in 1966.
UNION DEMOCRACY AND EXPECTED UNION
OUTCOMES
What should we expect from union leadership in the collective
bargaining process? According to our basic idea of union
formation, the idea would be to push player salaries closer to
MRP. In addition, other compensation issues (like pension
plans) and workplace safety issues are also the objects of
collective bargaining. The form that this pursuit takes is
dependent on the fact that unions are democracies, warts and
all.
For example, why in the world would a union agree to a salary
cap? One possibility from Chapter 6 was that the cap would
enhance balance, fans would respond, and revenues would
increase for the league. Once that occurs, players’ MRP rises
and so should pay. However, as seen in Chapter 6, it ends up
that salary caps actually have not coincided with any increase
in competitive balance. So let’s turn to alternative
explanations regarding the collective bargaining process.
We have already seen that caps reduce payments to talent if
they are effective. A simple explanation is that a cap
represents a trade-off of lower pay for higher employment.
Indeed, in the NBA where it was first introduced, the cap
came during the merger of the NBA and ABA, when a small
number of teams in each league were failing. At that time, it
did look like players needed to take a pay cut in order to
ensure survival of teams and a number of jobs. But the cap in
the NFL and, recently, in the NHL occurred without any real
threat of job reduction. So perhaps there is another
explanation.
THE LOGIC OF COLLECTIVE ACTION
A more complicated model of union behavior lends significant
insight into this question. Political scientist Mancur Olson
(1965) took the approach that unions are small democracies.
The rank-and-file members elect representatives and a
leadership that participates in the formation of union policy.
The leadership then interacts with union executive directors,
primarily lawyers, and collective bargaining professionals in
359

carrying out the will of the membership. As in any democracy,
only certain parts of union policy ever actually go to a vote by
the rank and file. The rest of the policies are decided by union
representatives and apply to all union members. This means
that representatives, leaders, and directors have substantial
discretion in the formation of union policy.
What should we expect from such discretion? First, let’s think
about the rank-and-file players. On any given policy issue,
details are left to team representatives and leadership. As a
result, any given player knows nearly nothing about almost
everything that the union does. This is called rational
ignorance. Rational ignorance is fueled by the fact that it is
difficult for individual players to tell how any given choice will
impact them. It is also expensive to discover any particular
representative’s role in any given policy. Every hour spent on
union matters is an hour lost trying to keep up with the
players’ opponents on the field. That is expensive time,
indeed, in a business of very short average careers (especially
for younger aspiring players). As a result, it is rational for
players to be ignorant about union policy because it is
expensive to be informed. Rational ignorance reduces the
incentive to compete in the union process. The rank-and-file
players really do rely almost completely on their elected union
representatives and leaders. A perfect example is the NFL
strike of 1987, detailed later in the section on NFL labor
relations.
Rational ignorance is, by definition, a matter of degree. The
higher the value of being informed and participating, the
more players will strive to do so. The players who are not
rationally ignorant are those who stand to gain or lose the
most from union choices and have lower participation costs—
a characterization of established veteran stars. Their
economic stake in the outcome is large, and they already are
established stars, reducing their costs of participating; they do
not lose ground against the competition because they are at
the top of their game. Indeed, it is these players who dominate
union activity.
The one thing we know about union leaders is that they want
to be union leaders. This means that they constantly weigh the
impacts of their policy decisions and negotiations in terms of
the impacts on their chances to continue as union leaders.
They cannot please all of the players all of the time, and they
do not have to. All they have to do is please the members of
the group of players that actually determines their reelection.
Because rational ignorance prevails, nearly none of the rank
and file knows what goes on anyway. As long as union leaders
keep enough of the players happy enough, they gain
reelection. The players we would expect to be the most
influential are the league’s stars. The individual league
histories later in the chapter show that this is indeed what
typically happens.
Olson’s framework (1965) suggests that union leadership will
produce concentrated benefits (concentrated in the sense that
they are large relative to the small group of star players who
receive them) and dispersed costs (dispersed over the
majority of nonstar players in the league). This outcome may
360

not be anything close to the will of the majority of the union
rank and file, but the majority of the rank and file remains
rationally ignorant. They have bigger fish to fry—most are
struggling just to keep their jobs in an extremely competitive
environment. In this way, the concentrated group of stars gets
its way, and the costs are spread over the rationally ignorant
rank-and-file union membership.
This idea is portrayed in Figure 9.2. Powerful players control
the fate of union officials. Their preferences are made clear to
the leadership that chooses union policies (the path of
preferences points from powerful members to union officials
and not in the other direction). The policies then feed back to
powerful union members. If they are happy on net with these
policy choices, then the current union leadership gets to
continue in that role. If not, then others can be chosen who
will implement the preferences of the powerful members. The
rank-and-file players are left pretty much on the periphery,
with little impact on union outcomes.
COLLECTIVE ACTION AND THE SALARY CAP
With this somewhat more complex view of union governance,
let’s revisit the issue of salary caps. Caps are really just
revenue sharing between owners and players. Thus, caps only
establish a ceiling (and floor, in practice) on team spending on
talent. Therefore, players might find that they are able to
bargain a larger share of league revenues in the form of
defined gross revenues from owners during collective
bargaining than they would otherwise get competitively. If so,
then a cap makes sense because owners make the mistake of
giving a larger percent of revenues to players than
competition would generate. Alternatively, bargaining power
may put the union in a position to do the same thing, that is,
generate a larger than competitive share of the total revenue
pie.
However, the failure of democracy is that it sometimes
redistributes wealth away from the less wealthy majority
toward a wealthier minority. The particular way that a cap is
implemented with exemptions and special clauses may bring
about redistribution in the direction of the politically powerful
players. That is, even though there is a cap, the players at the
top of the pay scale may do just fine under the cap, even
relative to the alternative without the cap.
This is especially true in the world of pro sports, where career
lengths are short and the winners do not have to confront
those who bear the costs of their actions. If the current
powerful player group is comprised of veterans, by the time
the rank-and-file players move into that same position, the
previous veterans are long gone into retirement. This is
especially likely in the NBA, where just a few players run the
union under the threat of either forming another union on
their own or even forming their own league.
The logic of concentrated benefits and dispersed costs takes
an interesting twist in a league with many players, such as the
NFL. With so many more players, there is some evidence that
the NFLPA democracy may work differently than the NBPA
361

(Bishop, Finch, and Formby, 1990). With so many more
players, owners appear to have found an ally in the NFLPA in
taking money away from the top salary earners. In this case,
the union does not appear to benefit the narrowly defined top
players. The top players simply do not have the same power
over the NFLPA that they have in the NBA. In addition,
rookies may have voted for the cap because they liked its
minimum-spending requirement!
362

LEARNING HIGHLIGHT
PRO ATHLETES AS “COMPANY MEN”
It took players an extraordinary amount of time to figure out
the power that owners had over MRP, let alone how much it
was costing them. Nothing speaks more clearly than the
players’ own words. The following are some excerpts from
congressional testimony in the 1950s. Much of the testimony
concerned the reserve clause that essentially bound players to
their team for life, greatly reducing their bargaining power in
salary negotiations. Listen for the owners’ arguments that you
have studied in Chapter 8 and remember that these are
players talking. (Only player representatives are included
here, but other notables toeing the party line in these hearings
included Ty Cobb, Pee Wee Reese, Lou Boudreau, Stan
Musial, Mickey Mantle, and Ted Williams in baseball; Chuck
Bednarik and Red Grange in football; and Bob Pettit in
basketball.) The others mentioned in examples are members
of Congress or their staff.
As a player representative, Robin Roberts once said of owners,
“I think anything that is good for them in this particular line is
definitely going to be good for us.” Jackie Robinson, on the
other hand, argued forcefully for players’ economic freedom.
Baseball
Fred Hutchinson of the Detroit Tigers and chair of the
American League players’ representatives:
Mr. Hutchinson: In my opinion the players as a rule
have been treated fairly by the clubs and have generally
been paid in accordance with their value to the clubs.
Mr. Lane: As the players’ representative, in handling all
their grievances and complaints and being in constant
touch with them day in and day out, do you feel, in your
own mind, that the players wish to retain the reserve
clause, as a whole?
Mr. Hutchinson: Yes, sir.
Eddie Yost of the Washington Senators and chair of the
American League players’ representatives; Robin Roberts of
the Philadelphia Phillies and chair of the National League
players’ representatives; and Jerry Coleman of the New York
Yankees and players’ representative:
Mr. Singman: Do you think, then, the players are
generally satisfied with the reserve clause?
Mr. Yost: Yes, I do.
Mr. Roberts: I think most of them are and those that
aren’t feel there isn’t anything they can do with it.
Mr. Coleman: I have had no complaints administered to
as far as the reserve clause.
Football
363

Jack Jennings of the Chicago Cardinals and players’
representative:
Mr. Harkins: You are aware in your contract there is a
method by which you can become a free agent, and, after
that, presumably, any member club in the league is in a
position to hire you?
Mr. Jennings: That is right.
Mr. Harkins: That has never happened in the league?
Mr. Jennings: That is right.
Mr. Pierce: What you are saying is, no matter how we
slice this, in the long run the draft helps the club and
helps the player?
Mr. Jennings: That’s right.
Bill Howton of the Green Bay Packers and players’
representative:
Mr. Dixon: How do the players feel about this limited
reserve clause?
Mr. Howton: As we stated, it appears that it is necessary
in the livelihood of professional football, in the event the
reserve clause were dropped, the teams with the wealth
would naturally buy the best ballplayers. The teams with
smaller wealth would obviously fall in rank and
eventually die off. So the stronger teams would dominate
the game, and it seems as though it is a strong stabilizer
for the game although it puts the players in a bad
bargaining position.
A couple of voices sounded more like what one would have
expected from players who truly understood what was going
on. Here are two.
Kyle Rote of the NFL’s New York Giants and union organizer:
Mr. Rote: As far as the player is concerned, there is, at
present, no check over the activities of the NFL as relates
to the player. The draft and reserve clause are necessary
to the pro game. But they are not necessary when given
free rein and result in abuses of the players’ rights.
Jackie Robinson of MLB’s Brooklyn Dodgers:
Mr. Robinson: So I sometimes feel, myself, that when we
see the Ted Williamses and the Mickey Mantles, and the
Stan Musials down here testifying, that perhaps when
they say they like things as they are, I would certainly
have to agree because of the tremendous salaries that
they get. But I wonder whether or not the 8 or 9 or 10
men on the ball club would agree to what is happening
today as the right thing so far as the baseball player is
concerned.
Senator Kefauver: Let me ask you, first, do you think
there should be some limit on the length of time of a
reserve contract?
364

Mr. Robinson: Yes; I do, sir.
Senator Kefauver: What do you think about the draft
system? Do you recommend an unrestricted draft?
Mr. Robinson: If a ballplayer so desires, I think a
ballplayer should have a say as to whether or not he
should be subject to draft or not. In other words, I feel
that a ballplayer should have some say personally in his
baseball future.
Sources: Hearings before the Subcommittee on Market Power,
Committee on the Judiciary, House of Representatives, 82nd
Congress, 1st Session, Part 6, Serial No. 1, various dates in
July, August, and October 1951. Hearings before the Antitrust
Subcommittee, Committee on the Judiciary, House of
Representatives, 85th Congress, 1st Session, Parts 1, 2 and 3,
Serial No. 8, various dates in June, July, and August 1957.
Hearings before the Subcommittee on Antitrust and
Monopoly, Committee on the Judiciary, U.S. Senate, 85th
Congress, 2nd Session, July 1958.
365

SECTION 4
Basic Bargaining: Game
Theory
When we enter into a discussion of bargaining, we are in the
realm of game theory. Here, we will use the simplest of all
game explanations to gain insights into unions. More
sophisticated approaches are available, along with a complete
set of jargon, and your instructor may choose to show them to
you. However, in our discussion, the aim is to introduce you
to the basics of strategic behavior.
THE SIMPLEST BARGAINING OUTCOME
If you have ever shared a Popsicle with a friend, you have a
pretty good idea of the simplest bargaining setting. You both
know that the ultimate outcome without a sharing agreement
is a melted puddle. You both know that the longer you wait,
the less there is to enjoy. Thinking about what happens at the
very end of your bargaining period and the effects of the
passage of time, both of you reason your way to the up-front
50–50 split. Every step along the way, thinking back from the
end to the present, you both considered the best you could do,
given the outcome that would result without an agreement.
Choosing rationally each step along the way, both of you
reached the conclusion that the 50–50 split is best. This “look
forward, work back” approach is basic to analyzing
bargaining.
This simple Popsicle game is insightful, even for something as
complicated as sports bargaining. Here we present a
straightforward adaptation of a simple example found in Dixit
and Nalebuff (1993). Suppose you have a finite sports season
of 151 days (about five months). Agreement between the
players and owners is required before there can be any play.
Unions act on behalf of players, and the league acts on behalf
of owners. Further suppose the union makes its final demand
on day 150, with one day left in the season. Offers are made in
turns on daily intervals. Play on any given day is worth $13
million, for a total season revenue value of about $2 billion
(not far off from most pro sports). In this type of situation, if
both parties wish to do as well as they can, they would try to
structure their thinking about the problem just like the
Popsicle game—determine what would happen at their very
last chance to strike a bargain at the end of the season;
estimate the payoffs every step along the way, including the
choice that the other party would make; and then calculate the
most profitable choice of all. Table 9.1 shows this simple
bargaining scenario.
Both the union and the league know what will happen if no
agreement is reached until the last bargaining day (one day
left in Table 9.1). On that 150th day, it is the union’s turn to
make an offer. There is $13 million on the table because there
will be a one-day season if an agreement is reached. The
union offers the league $0 and keeps $13 million. Why is the
366

league forced into this corner? The union knows that if the
league says no, owners get zero anyway because the season is
over and no bargain has been reached. The union will earn a
100–0 split if no agreement is reached until there is only one
day left before the end of the season.
With this knowledge, consider the situation with two days left
in the season. It is the league’s turn to make an offer. The
league knows that if no agreement is reached, they will move
into the one-day-left situation where the union will get $13
million and the owners get $0. But with two days left, there is
$26 million on the table. The league knows that a $13 million
offer to the union, with the remaining $13 million going to the
league, will be an acceptable bargain. After all, $13 million is
the best the union can do by saying no and moving to the one-
day-left situation. The league sees right away that it can
obtain a 50–50 split.
The logic of working back from three days left all the way to
151 days left is summarized in Table 9.1. For example, with
three days left, there is $39 million on the table (three days at
$13 million each). If the league says no to the union offer, the
best it can do is to earn $13 million by moving into the two-
days-left situation. Therefore, the union offers the league the
$13 million and keeps $26 million for a 67–33 split. Note that
this is much closer to 50–50 than the next move that the
union can make on the last day of bargaining. A quick check of
Table 9.1 shows that the league is constantly reminded that
the best it can hope for is a 50–50 split.
Tracking the situation for the union in Table 9.1 shows that
ultimately the players come to realize that the 50–50 split is
the best they can do. It is better to split 50–50 on the very first
chance they get to make an offer than to lose even a single
day’s worth of $13 million per day. The league recognized this
almost immediately in its calculations. It is the Popsicle game
all over again. The payoff from a 50–50 split, reached before
the season even begins (i.e., with 151 days left), provides the
highest payoff to each party among all the alternatives. Just
like kids with a Popsicle, the union and league strike their 50–
50 bargains before the season even begins. Work stoppage
never enters the consideration.
ONE STEP TOWARD REALITY: DIFFERENTIAL
ALTERNATIVES
The real-world splits are seldom 50–50 between players and
owners. Estimates of the actual splits in the major sports
leagues are shown in Table 9.2. First of all, players have
managed to increase their share of total revenue in all leagues
except the NFL. On average, over the tabulated years, players
have managed to obtain 61–39 and 60–40 splits in MLB and
the NBA, respectively. The NHL is also in the same
neighborhood with a 62–38 split. However, NFL players have
managed only a 56–44 split. Thus, we do not observe the
equal sharing outcome in any given league. It looks like we
need to expand our basic bargaining logic just a bit to gain
some realism.
367

In the basic bargaining outcome, owners and players earned
nothing on any day that no agreement was reached. But in
most cases, even if there is no season, players and owners still
have profitable options. On the player side, one famous
example is the holdout by Los Angeles Dodgers pitchers Don
Drysdale and Sandy Koufax prior to the 1966 season. In 1965,
the two players combined to win 49 games while losing only
20. Through the 1960s they simply dominated the entire
league. Drysdale led the National League in wins in 1962,
Koufax in 1963 and 1965 (and Koufax won the Cy Young
Award in each of those seasons as well as the league MVP for
1963). Koufax led the league in ERA over the period 1962–
1965 and was the strikeout leader in 1961, 1963, and 1965.
Drysdale was an All-Star selection in 1959, and both were All-
Stars over the 1961–1965 period.
According to Drysdale (Drysdale with Verdi, 1990), he and
Koufax each earned $80,000 ($541,500) in 1965. The two
joined forces in February 1966 after Dodger general manager
Buzzie Bavasi tried to play them off against each other. They
began their holdout by asking for $1.05 million ($7.1 million)
for the next three years to be split evenly between them (about
$175,000 [$1.2 million] each, per year). When Bavasi balked,
they threatened retirement. Both signed movie contracts
(Drysdale would later become a regular celebrity TV show
guest), clearly indicating their intention to enjoy some returns
to their sports stardom. Bavasi handed negotiations over to
owner Walter O’Malley, who caved on March 30, 1966, giving
$130,000 ($879,900) per year to Koufax and $105,000
($710,700) per year to Drysdale for 3 years.
Actions taken by owners in the 1987 NFL strike and the 1994
MLB strike show that they also have options. During those
work stoppages, owners fielded replacement players. It was
not pretty on the field (owners did not even buy fitted caps for
the replacement baseball players!), but with labor costs way
down, profits were earned (at least for a short period). The
owners still claimed TV revenues plus what little gate revenue
there was. In addition, owners may purchase insurance
against revenues lost to a work stoppage or take out a line of
credit to indicate their staying power.
A HYPOTHETICAL EXAMPLE OF DIFFERENTIAL
ALTERNATIVES
Let’s suppose that owners have a $5 million per day
alternative in the event that no agreement is reached. (They
might be able to sell televised music concerts, for example.)
We will leave the players’ alternative at $0 for now. What
happens to the split between the players and the owners?
Table 9.3 shows how the logic of thinking about the result on
the final day of bargaining and working back toward the
present must be altered in the presence of this alternative.
With one day left, the union knows that the league can make
$5 million even if no agreement is reached. They offer just
that amount to the owners and keep the rest for a 62–38 split.
Comparing the same point in the decision process in Table
368

9.1, this is a decidedly lower split than the 100–0 split they
enjoyed when the owners had no alternative.
With two days left and $26 million on the table, the league
must cover the $8 million that the union will make if no
agreement is reached and bargaining moves to one day left.
They keep the remaining $18 million, a 69–31 split in the
league’s favor. As happened in Table 9.1, this same 69–31 split
confronts the owners whenever it is their turn. Finally, as
before, the union realizes that the same 69–31 split is in the
offing as it works its way back to the decision period just
before the season starts. Therefore, both parties realize that
they should go straight to the 69–31 split in favor of the
owners and waste nothing in the process. Referring to the
Popsicle scenario, with a candy bar to enjoy in the event that
your Popsicle negotiation breaks down, your friend would
actually be able to parlay that into a larger portion of the
Popsicle. Your friend would munch away happily on the
alternative goody while you watched the Popsicle melt.
A general observation can be made about this outcome.
Owners earned their next-best alternative, plus half of
whatever is left. Refer to the outcome in Table 9.3. With
$1.963 billion, the total for the season (151 days at $13
million), how does the split work out? Over that same period,
the total value of the owners’ next-best alternative is $755
million (151 days at $5 million). The total, $1.963 billion,
minus $755 million equals $1.208 billion. Splitting that in half
equals $604 million. Owners earn $755 million to cover their
alternatives and half of the rest, or another $604 million.
Their share is 69 percent [(755 + 604)/1963]. Players get the
remaining $604 million, for 31 percent.
Of course, players will also have alternative earning
possibilities in the event there is no season, and we could go
through the same exercise with this factor added. However,
there is no need. The same logic extends to cover this
additional factor. These differential alternatives, where one
party to the negotiation is in a better situation than the other
if no agreement is reached, must come out of the total. Then
the balance is split 50–50. Let X be the amount on the table.
U and L are the next-best alternatives for the union and
league, respectively. The outcome is that each earns its next-
best alternative, plus half of whatever is left after covering
both union and league alternatives:

Each participant earns its respective next-best alternative, U
for the union and L for the league. The terms in parentheses
equal the amount on the table minus the sum of the parties’
next-best alternatives, and each party splits that amount
equally. Of course, if U + L > X, then both the union and the
league are better off pursuing their other alternatives. The
opportunity cost of any season at all is too high in this case.
THE SIMPLEST STRATEGY
369

These expressions prove useful in thinking about strategies
that might occur to either party. First, you can see that it pays
to get the other side to believe that the value of your next-best
alternative is high. If the other side builds in a higher next-
best alternative for you, your share rises. Strategically, then,
in a world of uncertainty about alternatives, it can pay to
convince the other side that your alternatives are great even if
they are not. Second, it can be valuable to take actions that
affect the size of your opponent’s alternatives.
Suppose that the league can take an action that reduces the
union’s next-best alternative by an amount equal to A. The
union result is now as follows:

Thus, when the league is able to take such an action, it drives
the union’s share down by half the amount of the fall in the
union’s next-highest-valued alternative. Where did that half
go? You probably already guessed it, but let’s look at the
league’s result:

The league gains exactly half of the decrease in the next-best
union alternative. This type of strategic behavior alters shares,
but it also helps parties in a bargaining situation determine
just how much they are willing to spend in order to alter the
next-best alternative of their opponent. The answer is half of
the expected decline in the opponent’s next-best alternative.
Leagues and unions might try to impact each other’s
alternatives through public relations. They could try placing
blame in media statements, ads, and other testimony. The
calculation shows that neither party would spend any more
than half of the damage they hope to cause with the ads.
370

SECTION 5
Strikes and Lockouts
If both sides have bargained faithfully but still reach an
impasse, the NLRA allows players to strike, withholding their
services from owners, or owners can confront players with a
lockout, refusing to allow them to play. Fortunately for fans,
despite a rise in frequency lately, the most obvious historical
observation is that strikes and lockouts rarely happen in
sports. Their entire history in all sports, shown in Table 9.4,
includes only 16 total occurrences. All but four of these were
in baseball and football. Only seven were of any real duration
where games were actually lost to fans. To get a feel for the
infrequency of strikes and lockouts, player associations in
sports have existed since the mid- to late-1950s—roughly 45
years. With four leagues, that’s 180 total years, but CBAs have
typically been three years in length, leaving 60 total chances
for a strike or lockout. So negotiations precluded a significant
strike or lockout 53 of 60 times, an 88 percent “success” rate.
The basic bargaining model does not explain why strikes and
lockouts occur. We will have to look for additional elements in
the bargaining process to find the causes.
Strikes and lockouts happen for basically two reasons. First,
even though they sincerely wish play to continue, one side or
the other may miscalculate amounts on the table or the value
of their opponent’s next-best alternative. The second reason is
that it simply can be worth it, in terms of expected value, to
either strike or lockout.
SINCERITY, ASYMMETRIC INFORMATION, AND
MISREPRESENTATION
To see how play can stop due to miscalculation, suppose the
true amount on the table really is $13 million per day, as in
the hypothetical example we used earlier in this chapter.
However, if only the league knows this amount with certainty,
then the union can only offer and respond based on its
estimate of that amount. This is a classic case of asymmetric
information, where one side of the negotiations knows
something that the other does not. In this case of asymmetric
information on the part of the league, if the union estimates
were wrong, the league would refuse union offers, not because
the union or league is trying to pull a fast one but because the
union simply lacks information. Players might strike, not
because that outcome was the intent of either party but
because of the uncertainty inherent in the process.
This type of miscalculation is further complicated when
negotiators use strategic misrepresentation of their next-best
alternatives. Strategic misrepresentation means that each side
will try to make the other side believe that it will enjoy the
better situation in the event no agreement is reached. Suppose
371

the league’s next-best alternative is really $5 million, as
portrayed in the last section. However, the union knows that
the league has every incentive to overstate its next-best
alternative. Thus, the union discounts the league’s claim,
estimating it at $2 million instead. The union makes an offer
that covers their estimate of the league’s next-best alternative,
but the league refuses to accept the offer. The $2 million is not
enough to cover the league’s next-best alternative. However,
the union thinks the league is just posturing and sticks to its
guns. Owners move to a lockout. Again, uncertainty plays a
role with the added complexity of strategic posturing of
alternatives.
THE EXPECTED VALUE OF A STRIKE OR LOCKOUT
The second reason for a strike or lockout is that it may simply
be worth it to one side or the other. In this case, one side of
the negotiations really does want a work stoppage, not
because of information differences, miscalculation, or
strategic posturing but because it will be a profitable strike or
lockout.
For reasons that will become apparent shortly, let’s index time
during the strike or lockout as t = 0, 1, 2, … , S. After the strike
or lockout, time runs from t = S + 1, S + 2, … , T, so that T is
the end of the planning horizon. Because a time element is
involved, we will need to have rt as the interest rate at time t
during and after the strike or lockout. Finally, let P equal the
probability that the strike or lockout is successful and 1 – P
the probability that it is not.
VALUES AND COSTS OF A STRIKE OR LOCKOUT
Four elements must be considered here. First, At is the
alternative value open to the decision maker at time t. Owners
might be able to play games using replacement players or rent
out their facilities. Players may be able to capitalize for a while
on their fame through more guest appearances and
endorsement activity. Because these alternatives are earned
during the time play stops, from t = 0, … , S, we write the
following:

Second, Mt is the money cost at time t borne by the side
forcing play to stop. This could be lost income to players or
lost revenues to owners because no games are played. These
losses are also borne during the time that play stops, t = 0, … ,
S. Therefore, we would write the following:

The third element, Gt, is the possible gain from the strike or
lockout at time t. For players, it may be a reduction in the
number of years to free agency or a larger percentage of
revenue in a salary cap situation. For owners, just the
opposite would be gains. Gains come from concessions by the
other side after play resumes, t = S, … , T. Therefore, we would
write the following:
372

Last, Ft is losses at time t due to fan repercussions after the
strike or lockout. With any extended play interruption, fans
may hold it against players (for a strike) or owners (for a
lockout). It was widely hypothesized that MLB suffered
dramatically at the gate after a player strike cost fans part of
the 1994 season, all of the 1994 League Championship Series,
and the 1994 World Series. Time (August 22, 1994, p. 71)
reported a survey showing that fans blame owners and players
about equally and believe that owners’ profits are too high and
players make too much money. A Baseball America
(December 23, 1996–January 5, 1997, p. 26) survey reported
the same blame routine, as well as a drop from 9.5 to 8.1 on
the “1–10 scale of interest in baseball.” These costs are only
lately the point of economic analysis. For example, Schmidt
and Berri (2004) find that the attendance impacts of work
stoppages in pro sports are only temporary at best. But in our
formulation, we’ll allow that fans may not be happy when
their sport is withheld from them, and they may exercise their
wrath once play resumes, t = S, … , T. For these losses, we
write the following:

Now let’s look at the state of things in the event that the strike
or lockout is successful. If successful, then the winner earns G
+ [A – (F + M)], but this only happens with probability P. If
unsuccessful, the loser nets A – (F + M). This gives the
following expected value of a strike or lockout EV:

If we combine like terms and simplify, we get a simple
expression full of insight:

The expected value of a strike or lockout is determined by the
expected gain (P × G), adjusted by the balance of the next-best
alternative, net of fan costs and the lost value of play.
Figure 9.3 depicts the relationship graphically. Note that the
y-intercept for our expected value is portrayed as negative, as
fits a bit of logic and a real-world observation. If A − (F+M)
>0, then the value of the alternative exceeds fan irritation
costs plus the value of lost play. This means that EV > 0 no
matter how small P is (the chances of success). Whether it is
players considering a strike or owners a lockout, the choice is
a no-brainer in such a case—a strike or lockout will always be
the outcome. Since we do not always see a strike or lockout,
the negative y-intercept term in Figure 9.3 seems appropriate.
Note also that a particularly interesting interpretation can be
attached to the point where the EV function crosses the x-axis.
That point is labeled PBE, for the breakeven probability. At this
breakeven probability, the expected value of the strike or
lockout is zero. But this also serves to highlight that any
373

probability greater than PBE generates a positive expected
value. Put another way, if the chances of winning are less than
PBE, then the strike or lockout isn’t worth it. Let’s turn to a
numerical example.
Suppose G = $500 million, M = $200 million, A = $100
million, and F = $100 million. We can actually solve for the
breakeven probability in this case. If we just remember from
Figure 9.3 that EV = 0 when P = PBE, we can rearrange and get
the following:

Now just plug in the example values and see that PBE = 0.40.
Thus, if the chance of winning is less than 40 percent, it is not
in this decision maker’s best financial interest to force play to
stop. The possible gains, net of the up-front costs and any
future losses due to fan irritation, are just not worth it.
But what determines whether the chances of winning the
work stoppage are greater than PBE? From the economic
perspective, it all depends upon the ability to weather a
stoppage. This single factor has the greatest impact on the
eventual outcome. It is sort of like a medieval castle siege.
Each side measures the other’s ability to withstand the siege.
The victor will be the one that lasts the longest, and the other
side will lose a tremendous amount of resources.
Finally, using Figure 9.3, we can assess how changes in
expected gains, costs, and some underlying parameters will
alter EV. First, if the gain G increases, our function pivots to
the left, raising EV for all values of P. If A increases relative to
F + M, the function shifts up in a parallel fashion, raising EV.
Clearly, if F or M increases relative to A, then EV shifts down.
Turning to the other parameters in the explicitly discounted
future values listed earlier, if the duration of the stoppage, S,
rises or the interest rate, rt, rises, all else constant, current
costs become more important relative to future net gains, and
the expected value falls.
We’ll use this logic to get a handle on the NBA lockout of
1998. In addition, we’ll use it to gain insight about the work
stoppage in the NHL that began in late 2004. But let’s
approach these recent episodes from within the context of the
overall history of labor relations in pro sports, beginning with
MLB. I caution all readers that what follows is a general
overview. Your favorite aspect of either the evolution of labor
relations in a given sport or the administration of those
relations under a given sport’s CBA may or may not appear in
what follows. I do list the tenure of the most recent CBA, and
either a copy of each CBA or a link to it is at the textbook
webpage, Links and Fun Chapter 9 (https://sites.google.com/
site/rodswebpages/research).
374

https://sites.google.com/site/rodswebpages/research

https://sites.google.com/site/rodswebpages/research

https://sites.google.com/site/rodswebpages/research

https://sites.google.com/site/rodswebpages/research

SECTION 6
Labor Relations in MLB
In MLB, once the reserve clause was firmly entrenched in the
late 1800s, players could be sold, traded, and released as
management saw fit. Because players’ next-best choices were
outside their sport in the presence of the reserve clause,
owners had the upper hand. Players had to be paid enough to
keep them in their sport, but not much more than that.
John Montgomery Ward formed the Brotherhood of
Professional Baseball Players around 1885 and the Players
League in 1890. This “Brotherhood Revolt” involved about
200 of the greatest players of their time. But, as we saw in
Chapter 8, ideals do not stand much of a chance against the
concerted economic forces of owners acting together through
sports leagues. The Players League lasted only a year.
There were other early attempts at player organization, but
labor strife in MLB was pretty much confined to infrequent
confrontations between individual players and particular
team owners. One early attempt at organized player labor
action in baseball centered on the legendary Ty Cobb. For an
act of temper (Cobb entered the stands attempting to beat up
a heckler), Cobb was suspended 10 days in 1912. In baseball’s
first organized work stoppage, the Detroit players struck in
support of Cobb, and union sentiment spread among other
American League teams. However, the resulting Baseball
Players Fraternity failed five years later because it could not
produce any economic gains from owners for players.
After a long organized labor hiatus, the American Baseball
Guild formed in 1946 out of the turmoil caused by the one-
year stint of the Mexican League. Players were jumping to the
Mexican League, breaking their current MLB contracts
despite a promised five-year banishment. As players jumped,
MLB grew cautious and actually began to talk to the guild. As
a result, it was recognized as an official bargaining unit under
federal law and negotiated minimum salaries and a pension
plan. The guild was also instrumental in negotiating the
return of former MLB players after the demise of the Mexican
League.
The modern MLBPA began in 1954. It was not an official
federally recognized bargaining unit but continued the pursuit
of a better player pension plan funded out of All-Star Game
and World Series revenues. However, the organization was
impotent on the main issues confronting players, namely, the
reserve clause and player grievances. Further, the original
MLBPA was dominated so much by owners that they
appointed the director and paid the association’s bills, which
was patently illegal under labor law.
All of this changed with the hiring of Marvin Miller in 1966.
He immediately ended the illegal funding by owners and
375

quickly got to work reforming the pension system. The player
pension system is now funded by a negotiated contribution
from owners. In a second important stride forward for MLB
players, the formal grievance procedure was instituted in the
first real Basic Agreement in 1968. The 1970 agreement
allowed for a three-person review of grievances, one each
from MLB and MLBPA plus a permanent independent chair.
MLB has no salary cap, but shares revenues extensively and
employs a “luxury tax.” The most recent CBA runs 2007–
2011.
ARBITRATION
Owners hate arbitration. Owners gave it to the players in 1974
because they were afraid of all-out free agency. After the
famous court case that denied Curt Flood free agency in 1972,
owners feared the worst in the subsequent collective
bargaining process. In gaining arbitration, the players, in
turn, agreed to put off free agency to a study group. However,
the owners opened the equivalent of Pandora’s Box with
arbitration.
Few players actually exercise their arbitration rights, and
owners typically win more cases than the players. According
to James Dworkin (1997), over the 20-year period of 1974 to
1994, only 375 cases went before arbitrators (just under 19
cases per year, on average). In 1994, of 16 arbitration cases 10
ended in the owners’ favor and six in the players’ favor. These
16 cases represented only 10 percent of all players eligible for
arbitration. The largest number in any year was 35 in 1986,
but there were only nine in 1978. Through 1994, clubs won
209 times and players only 166. That is a 56 percent
advantage in the favor of owners.
Just keeping a tally misses the point about the winners in
arbitration. Players win just because arbitration is available to
them. Data on salaries show that, relative to years prior to
arbitration, just filing has resulted in a doubling of salary even
for those players who never go to arbitration. The threat of
arbitration has revealed that owners fear those 44 percent of
the outcomes that they lose. Thus, they raise salaries to avoid
arbitration. It is precisely because arbitration does what it is
designed to do (force bargaining in earnest prior to any filing
for arbitration) that owners hate it.
THE MLB DRAFT/ENTRY SYSTEM
All leagues have drafts subject to collective bargaining. MLB’s
draft began in 1965. The main difference in drafts across
leagues is in the number of rounds. In MLB, each team drafts
approximately 50 players. There are no supplemental rounds.
Baseball allows players to be drafted after they finish high
school. Unlike other sports, college players can return to their
university teams even if they enter the draft. No eligibility is
forfeited. There is no draft for international players, who are
scouted and signed subject to age restrictions, just as U.S.
players were before 1965. There are no restrictions on entry-
level pay, only a “suggested bonus” based on draft where the
player is picked in the draft. This stands in stark contrast to
the other three leagues, as we shall soon see.
376

FREE AGENCY
As we saw in Chapter 8, the arbitration procedure eventually
led to the end of the reserve clause. Free agency clearly has
been the major win for any players’ union. Interestingly, it
took quite a bit of trial and error to get to the modern version
of free agency. During the lockout of 1976, owners and players
designed the first among many versions of free agency
through collective bargaining. In the 1976 Basic Agreement,
players with six years of experience were free agents but had
to wait five years to become free agents a second time.
Further, teams gaining a free agent compensated the team
losing the player with a draft pick.
In the next go-round, disagreement over free agency and
compensation led to the strike of 1981. The compromise for
the 1981 Basic Agreement defined a Type A free agent as a
player in the top 20 percent of performance at his position.
Any team gaining a Type A free agent had to put all but 24 of
its own players into a pool from which teams losing a Type A
free agent would draw their compensation (not necessarily
from the team signing the free agent). Teams not gaining a
Type A free agent could protect 26 players. An exemption
could be issued if a team promised not to chase Type A free
agents for three years. The six-year rule and draft choice
compensation also remained.
For the 1984 Basic Agreement, all compensation was
eliminated, except for the draft pick for Type A free agents.
For Type B free agents (the next 10 percent of players by
performance at their position), an extra pick was sandwiched
between the first and second rounds of the next draft. This is
where the situation now stands. Players with fewer than four
years in the league are completely restricted, without free
agency or arbitration. Their situation is the same as under the
reserve clause of old.
GUARANTEED CONTRACTS
Player contracts in MLB are guaranteed. This means that even
if players are injured and cannot continue to play, they are
still paid the remaining obligation in their contract. Again,
this isn’t true in all pro sports leagues, as we will shortly see.
THE “LUXURY” TAX
MLB actually names its “luxury tax” the Competitive Balance
Tax. Only owners who spend above a threshold specified in
the CBA are subject to the tax. The tax is calculated on the
difference between their actual payroll and the stated
threshold. For example, owners who spent more than $162
million on payroll for the 2009 contract year, and for whom
this is a “first offense,” will pay a tax rate of 22.5 percent.
While a bit complicated, essentially second offenders pay 30
percent and chronic offenders pay 40 percent. The threshold
rises to $178 million by the end of the CBA in 2011.
REVENUE SHARING
According to the 2007–2011 CBA, 31 percent of each team’s
locally generated revenue, including gate receipts and local TV
377

revenue, goes into a central fund. The amount is net of
stadium costs, either direct or in terms of interest payments.
The money in the pool is then split equally among all 30
teams. An additional collection also is distributed by the
commissioner to teams meeting improved performance
criteria set forth in the CBA.
STRIKES AND LOCKOUTS
At the 2001 Washington University conference on the
business of baseball, current executive director of the MLBPA,
Donald Fehr, repeated a statement I had heard from him
before: “Owners have always said two things about baseball;
there is never enough pitching and no owner ever made any
money.” As we have seen in other chapters in this book,
owners always plead poor in negotiations with players,
especially those who have ultimately ended in strikes and
lockouts.
As shown in Table 9.4, early strikes and lockouts in MLB were
short and of literally no consequence to fans. In 1969, a spring
training delay strike preceded the players obtaining a
grievance panel. The 10-day season delay strike in 1972 over a
pension dispute preceded the players gaining arbitration.
Owners locked players out in 1976 in the dispute over free
agency, but it ended up that nothing could stop the demise of
the reserve clause.
Things got much uglier from the fan perspective with the 1981
strike. The issue was still free agency—owners wanted
compensation for lost free agents. Owners hunkered down for
the long haul, buying $50 million in strike insurance to add to
a $15 million strike fund from owner contributions over the
preceding two years. Players lived on solidarity. The strike
lasted 50 days. By all accounts, the players won, yielding little
on compensation for lost free agents. The result of so many
lost games was the infamous split-season play-off system. The
winners of the “first half” of the season (before the strike)
played the winners of the “second half” (after the strike) to
determine the overall postseason setup.
In 1985, the list of contentious issues included contributions
to the players’ pension fund from TV revenue (the national
contract had exploded to over $1 billion, and owners were
balking at their then existing 33 percent contribution),
compensation requirements for lost free agents, and
minimum salary increases. The result was a two-day strike
that had nearly no impact on fans. Owner contributions to the
pension fund doubled (even though percentage decreased);
players won arbitration after three years; and the minimum
salary was increased from $40,000 ($79,300) to $60,000
($118,900). Owners also gained further redefinition of free
agents that continued the practice of compensating owners
who lost free agents to other teams.
The lockout in 1990 was over the maximum share of league
revenues that could go to players (owners offered 48 percent
of ticket and national and local TV revenues), a seniority pay
scale, and a salary cap. Players responded with two years to
arbitration, double the minimum salary, continuation of the
378

owners’ fixed contribution to the players’ pension plan,
collusion protection language, and an increase to 25 players
on the roster. The players won, hands down, and got nearly
everything they wanted. The season was just pushed back to
make up for the lost start time, so nothing was lost in the eyes
of fans.
The MLBPA strike of 1994 was labeled the “mother of all
strikes.” Using a reopener clause from earlier negotiations,
owners wanted broader revenue sharing among themselves
but only if the players accepted a cap of the NBA variety. The
owners also wanted the end of arbitration, offering a
reduction to four years for free agency eligibility in return.
Players struck in August, when it was best for them. They had
nearly all of their annual salary, while the owners were
looking forward to the play-offs. The play-offs and World
Series were lost, and no end was in sight even after the
scheduled start of the 1995 season.
The owners could see they were getting nowhere and changed
their demands. The union would go along with the revenue-
sharing idea, but not the cap. Owners withdrew the cap
demand, substituting a luxury tax plan. Players countered
with a different luxury tax that reduced player salaries by less.
Owners countered that they would continue the season with
replacement players.
In the meantime, the MLBPA filed a grievance with the NLRB
claiming that the owners had blocked arbitration and free
agency. MLBPA’s executive director Donald Fehr said players
would return if the courts issued an injunction against these
owner actions. The NLRB agreed and District Court judge
Sonia Sotomayor (now a U.S. Supreme Court Justice) issued
the injunction. Fehr announced the end of the strike. The
owners could have countered with a lockout but chose not to
do so. Only 144 games were played in the 1995 season, rather
than the usual 162.
379

SECTION 7
Labor Relations in the NBA
In basketball, eventual Hall of Famer Bob Cousy was
instrumental in the 1954 formation of the NBPA. As was the
case in football, NBA owners simply ignored the NBPA at first.
Nevertheless, the NBPA gained momentum under Lawrence
Fleisher, beginning in 1962. Solidarity rose, and the players
gained some ground with a pension fund in 1964. In 1967, the
NBPA threatened to seek official federal government
recognition as a bargaining unit if the pension plan was not
bolstered. Under that threat, the league agreed to the NBPA’s
demands. The CBA signed that year was the first in all of pro
sports. In 1970, minimum salaries were established.
The NBA has no arbitration, and all player contracts are
guaranteed regardless of injury. In addition, only the national
TV contract proceeds are shared; there is no other revenue
sharing in the NBA. The most famous part of NBA labor
relations is the longest-standing payroll cap in pro sports
history. While there is no luxury tax, the cap includes an
escrow tax in the event that spending on players exceeds
amounts related to the cap. The most recent NBA CBA runs
2005–2006 to 2011–2012.
THE DRAFT/ENTRY RESTRICTIONS
The NBA has two draft rounds, usually drafting 58 players.
Since 1985, the first important picks have been determined by
a lottery, with the number of chances based on a team’s past
performance. For example, the first 14 picks in the 2008 NBA
draft were determined by the lottery, and the Chicago Bulls
won the lottery and had the first pick. Beginning with the
2006 draft, the NBA no longer considered any players
younger than 19 years of age who had to be out of high school
for at least one year. Under such a rule, for example, Kobe
Bryant and LeBron James would not have been able to enter
the NBA as they did out of high school. All players now have
to play in college or internationally for one year before
entering the draft. There also are other entry restrictions on
contract length and pay for new players. For example, there
are fixed three-year contracts, plus an option year, for first-
round draft picks based on the rank of their selection. Rookie
salaries also are controlled under a rookie wage.
FREE AGENCY
In 1970, thanks to a lawsuit by legend Oscar Robertson, NBA
players won free agency. In 1981, all compensation to teams
losing free agents ended. Unrestricted free agency began in
1988 for players who had played in the league for more than
four years. Earlier, players with more than four years were
restricted free agents. The restriction was that the team could
match an outside offer and retain the player. In addition, no
compensation was required if the team chose not to exercise
380

this right of first refusal. Currently, even the right of first
refusal is gone; players simply are free to sell their services to
the highest bidder after four years. However, the NBA has
maximum salaries for players of all experience levels, so free
agent salaries do have limits.
PAYROLL CAP
The NBA salary cap is the longest-standing cap in pro sports,
firmly entrenched since the 1982–1983 season. The NBA cap
until very recently was referred to as a soft cap. It had built-in
exceptions, such as the famous “Larry Bird rule.” These
exceptions allowed teams to keep their own free agents
regardless of their impact on team salary relative to the cap.
As you saw in Chapter 6, NBA teams rarely were anywhere
near the cap amount. The cap was toughened significantly
following an extended lockout in 1998 (detailed just
hereafter). However, as you also saw in Chapter 6, it is still
not working. The Learning Highlight: Another Kind of Cap
Cheating at the end of this section shows yet another way that
the cap poses problems for all involved.
The NBA has an interesting variation on prohibitions of
spending on players, detailed under the definition of the cap
in its CBA. Under its “escrow tax,” teams are taxed dollar for
dollar at payrolls above a threshold level that changes over the
course of the CBA. But the tax kicks in only if leaguewide
player salaries exceed a specified percentage of NBA revenue.
Some of the onus is on the players to keep salaries down.
Portions of their salary are held in an account that is returned
to owners to help them pay if the league exceeds the tax
threshold.
LOCKOUT, 1998
The NBA was once held up as the only league without a strike
or lockout. But there has been labor unrest. Basketball
decertified its NBPA in 1995; however, the NBPA leadership
was none too happy about it. Although the NFLPA had been
instrumental in decertifying itself, as we will see, the leaders
of the NBPA fought tooth and nail to hold onto their power.
Important players like Michael Jordan and Patrick Ewing
collected petitions from a majority of players, decertifying the
union in 1995. They then brought suit against the NBA
alleging that any joint action over NBA players by NBA
owners—be it a lockout or a continuation of the salary cap—
violated antitrust laws.
However, the NBPA did not step down after decertification. It
continued negotiating with the owners as if the petitions
decertifying the union had never been circulated or collected.
In these wacky proceedings, you had the majority of NBA
players who supported the suit and had decertifying the NBPA
on one side, and on the other side, you had the never-say-die
NBPA and the owners negotiating with each other! The NLRB
called for a new election. The NBA owners openly announced
that decertifying the union would lead to an immediate
lockout. Apparently this changed the minds of many players,
and the majority voted to retain the union. A new CBA was
reached in 1996.
381

The long-standing absence of any interrupted play in the NBA
ended as the 1998–1999 season was about to begin. Claiming
that players received 57 percent of total revenues in the 1997–
1998 season, the owners exercised a clause in the NBA basic
agreement that allowed renewed negotiations if the players’
share reached these levels. Negotiations were opened but
ended in disaster. The NBA owners announced there would be
no 1998–1999 season unless some new agreement was
reached that tightened the cap. They would lock the players
out. The players did not budge, and the result was the NBA
lockout of 1998.
Based on our discussion earlier in this chapter, it can be
shown that the owners may have locked out the players in a
move to increase profits. (The following example, attributable
to Roger Noll, first appeared in Quirk and Fort [1999]). Let’s
think about the revenues lost from a lockout, denoted M in
our earlier analysis. First, suppose an entire season were lost,
that is, S = one season. The NBA was a $1.7 billion ($$2.2
billion) industry just prior to the 1998 dispute. There was
$700 million ($915.6 million) at the gate, $637 million
($833.2 million) from TV contracts, $225 million ($294.3
million) in venue revenues, and $160 million ($209.3 million)
in other revenues. The 57 percent going to players would
amount to $969 million ($1.3 billion). That leaves $731
million ($956.1 million) for owners. About 30 percent, or
$510 million ($667.1 million), would typically go to other
expenses, leaving the owners 13 percent, or $221 million
($289.1 million). Thinking in terms of a one-season lockout,
M = $221 million.
What about the cost of fan irritation? This cost would have to
be figured into the arithmetic of a lockout, and we do not
know this value. Therefore, the losses of a one-year stoppage
are $221 million plus the losses that would be caused by hard
feelings on the part of fans.
What about the gains? Suppose that the owners are shooting
for a 50–50 split. A reduction in the players’ share from 57 to
50 percent is worth $119 million ($155.6 million) per year.
The subsequent CBA held for eight years, that is, T = 8. This
amounts to $952 million ($1.24 billion), but we must discount
those dollars to present value. At 5 percent, the discounted
present value is G = $769 million ($1 billion). Further, even
though the alternatives might be small, there is no reason to
suspect that owners would not take their resources to a next-
best use. The value of these alternative uses would offset the
losses. However, data on these alternative values are
unavailable.
Without knowledge of fan irritation costs or alternative
opportunity values, the expected value of a lockout boils down
to a pretty simple comparison (it would be fun to analyze this
case using Figure 9.3, but there simply are not enough data).
The owners had to consider whether it was worth it to risk
$221 million in hopes of gaining $769 million. That expected
value would be written as follows:
382

Now we reach the point where P looms large in the decision.
Although owners might have their own idea of the probability
of winning from a lockout, let’s have a quick look at the
breakeven probability of winning the lockout. Setting EV = 0,
the breakeven probability is as follows:

We can see that owners will consider the breakeven
probability of 29 percent, give or take the net of the next-best
alternative over fan costs, divided by the gain. Observe that
the breakeven probability rises by 1/769, or 0.0013, for every
million dollars that fan costs exceed the value of the owners’
next-best alternatives. This means that for PBE to increase by a
full point, or 0.01, fan costs would have to increase over the
alternative value by $7.7 million. To drive PBE = 0.5, costs of
fan irritation would have to exceed the value of the next-best
alternative by million. Note that this is effectively the entire
value of a season of play we pegged at $221 million. Owners
would have to believe that angry fans would stay away from
the NBA for an entire season in order to estimate that the
breakeven probability was a coin toss. So, if , then PBE < 0.29. If A < F, then PBE > 0.29 but smaller than 0.5.
Now, as it worked out, owners had a greater than 50–50
chance that the union would fold (because it did!). In the
limiting extreme of certainty, the expected value becomes
$548 million, plus (A – F). We are left to suspect that NBA
owners felt that at least a couple of hundred million dollars
could be won from players with an effective lockout. In any
event, the owners did lock out the players for a portion of the
1998 season.
Much was made of the idea that owners needed to harden the
salary cap in order to obtain some degree of cost
predictability. However, it is difficult to justify that the owners
needed any change in the cap at all. In 1998, there were 29
NBA team owners. Suppose that the owners shared all of the
league revenue equally. That would have been about $58.6
million ($76.6 million) per team ($1.7 billion/29). Even if
nothing were done to enforce the cap and players continued to
receive 57 percent of total revenue, that still would leave each
team with $25.2 million ($33 million) on average ($58.6
million – $33.4 million). Even after the 30 percent or so that
goes to other nonplayer costs, $7.6 million ($9.9 million) per
team remains ($25.2 million – $17.6 million). Thus, an
alternative to hardening the cap would be to share revenues.
But why should they share revenues if they can reduce costs
by reducing the players’ share of revenues? Beside, the New
York Knicks, Chicago Bulls, Los Angeles Lakers, Detroit
Pistons, Portland Trailblazers, and Boston Celtics all were
making more than $7.6 million ($9.9 million). The chance
that these dominant teams would share the wealth with their
weaker cohorts is pretty slim. It would make more sense,
economically, for these large-revenue market mainstays just
to form a separate league. How did the lockout go? Owners
383

demanded a hard cap, doing away with the Larry Bird
exemption that allowed a team to exceed the cap in order to
retain its own free agents (that is how Michael Jordan could
earn $33.1 million [$44 million] in his “last season” [he did
come out of retirement a few years later], 1997–1998, when
the cap was only $26.9 million [$35.7 million]). In essence,
the owners wanted no deviation at all from the agreed-upon
52 percent to players. Owners also wanted to lengthen rookie
contracts from three to five years, with only the first three
guaranteed unconditionally.
The NBPA wanted no change in either the soft cap or rookie
contracts. It also wanted to increase the minimum salary for
veteran players to $272,250 ($356,096). Finally, the NBPA
wanted to change the so-called team exception. The exception
enabled a team to sign one free agent for $1 million ($1.3
million) every two years. The NBPA wanted to alter the
exception to an annual signing at $2 million ($2.6 million) to
$3 million ($3.9 million).
Talks began in earnest in April 1998. By the end of June,
negotiations were still going nowhere, and the NBA
announced its lockout. Toward the end of August, arbitration
hearings began on whether owners had to pay guaranteed
contracts, totaling about $800 million ($1 billion), during the
lockout. This would have changed the alternatives available to
players and altered the balance of power during the lockout.
By October, the NBA announced that there would be no
exhibition season. On October 13, bargaining resumed but
lasted only a few hours. The NBA announced cancellation of
99 regular season games, lost to all teams in the first two
weeks of the season. Shortly thereafter, the NBPA capitulated,
and the owners pretty much got their way.
384

LEARNING HIGHLIGHT: ANOTHER KIND OF CAP
CHEATING
We saw in Chapter 6 that the cap is fairly ineffective in both
the NBA and the NFL. Here we present another way owners
circumvent the cap. Remember that the cap is determined
relative to defined gross revenue. It ends up that in the NBA—
the owners were caught understating these shared revenues
by the players.
In 1991, independent auditor Charles Bennett discovered a
problem. He was poring over trial documents from a 1990
antitrust suit brought by Chicago Bulls’ owner Jerry Reinsdorf
against the league, challenging its local TV rules. Reinsdorf
wanted his Bulls to be on TV more than league agreements
allowed. In the documents, Bennett discovered that the
owners had underreported defined gross revenues by nearly
$100 million ($163.1 million), thereby shortchanging players
in salary cap determinations.
The league contended that it was not defined gross revenue
and, therefore, did not need to be reported. However, the
league settled with the players in 1992 for a sum reported
around the missing $100 million. Bennett was immediately
hired by the NFLPA and, according to one NFLPA official,
helped push up the 1994 cap by nearly $1 million ($1.4
million) per team.
By suing his own league, Chicago Bulls owner Jerry Reinsdorf
unwittingly revealed that the league had shortchanged players
in salary cap determinations. [Photo of Reinsdorf with Bulls
backdrop.]
Source: Sports Illustrated, April 28, 1994, p. 10.
385

SECTION 8
Labor Relations in the NFL
The first football union, the NFLPA, was formed in 1956.
Owners were reluctant to even recognize the NFLPA, a typical
response by owners at the inception of player organizations.
However, the threat of antitrust suits brought football owners
around pretty quickly. Minimum salaries were established in
1957, and a pension plan funded by owners began in 1959.
Interestingly, the union did not achieve official federal
bargaining unit recognition from the NLRB until 1968. It was
a short-time affiliate of the American Federation of Labor-
Congress of Industrial Organizatoin in 1979.
The NFLPA did not really flex its muscle until it shut down
training camps in 1974. Player safety was always of primary
concern to the NFLPA, and it made the use of artificial turf an
element of the bargaining process (turf-related injuries had to
be covered in health care plans).
The NFLPA has never relinquished its rights to negotiate
individual salaries. It has never exercised those rights,
however, choosing instead to allow football players and their
agents to handle this task. The NFL has no salary arbitration,
and unlike either MLB or the NBA, player contracts are not
guaranteed. Teams can cut players at any time with no future
obligations except to cover any remaining signing bonuses.
The NFL employs the most extensive revenue sharing of all
pro sports, but there is no luxury tax. However, like the NBA
and NHL, the NFL employs a payroll cap. The current NFL
CBA runs 2006–2012.
THE DRAFT/ENTRY RESTRICTIONS
The NFL draft has been in existence since the 1930s. The
2009 NFL draft had seven rounds with 256 players in total.
After the draft, the remaining new players enter the league as
free agents. The NFL draft is restricted to players out of high
school for more than three years. Effectively, this means that
college players who have exhausted two years of playing
eligibility can enter the draft. However, these college juniors
and older must renounce their remaining college eligibility.
The league argues that drafting only upperclassmen protects
players from tragic career mistakes. However, NFL owners
really are protecting themselves from the wrath of colleges
that could sue or adopt their own rules that would be worse.
Previous NFL rules against drafting any but those who had
finished their college eligibility were struck down by the
courts.
NFL rookie salaries are also limited. Under the NFL CBA,
rookies are subject to a cap within a cap. About 3.5 percent of
defined gross revenues make up the “entering player pool.”
Unlike the share of defined gross revenue that goes to the rest
of the players, the entering player pool is allocated across
386

teams according to the level of their draft picks. Teams with
higher draft picks can allocate a larger share of the pool than
teams with lower picks. However, because the amount is
restricted, it is less than otherwise would be spent on new
players.
FREE AGENCY
The NFL had a completely hamstrung form of free agency
beginning in the late 1950s. Players were free agents at the
end of their contract, plus one option year. However, no
player ever changed teams under this arrangement. Owners
simply did not sign free agents. Later, free agents sued the
league for this behavior and won in the courts. In response,
then NFL commissioner Pete Rozelle designed a system of
compensation to teams losing free agents. This compensation,
coined the Rozelle Rule in honor of its inventor, effectively
stymied free-agent movement. The level of required
compensation was simply larger than the value of acquiring a
free agent.
NFL players eventually won free agency by suing as
individuals under the antitrust laws. Because individual
players forfeit their individual rights to sue when they choose
leagues as collective bargaining agents, the only way this
could happen was if the union no longer represented players.
The NFLPA was decertified in the battle for unrestricted free
agency so that players could pursue better legal chances under
the antitrust laws.
In the late 1980s and early 1990s, labor relations were going
nowhere for the NFLPA. They lost in 1987 when the owners
used replacement players to break a strike. The NFLPA also
lost important court cases aimed at free agency. In 1989, the
union decertified and became a trade association rather than
a legally recognized union. The union did this by petition. A
majority of NFL players agreed to withdraw the NFLPA’s
authority as a collective bargaining unit. This move allowed
the trade association to support individual players trying to
use all means possible to gain free agency. Of course, these
means could not include collective bargaining because the
union had been decertified. However, they could include
antitrust suits, and one such suit eventually prevailed for
players. The union then certified again and bargained for free
agency and the salary cap.
The lawsuit that eventually forced the NFL owners to free
agency is the now-famous 1992 case McNeil v. NFL. The bone
of contention at that time was Plan B. Under Plan B,
instituted some years earlier by the NFL owners in the
absence of collective bargaining, teams could protect up to 37
players from free agency. Only the remaining, weaker players
then entered free agency. If another team reached an
agreement with a protected player, the team could offer that
player the same salary and keep the player. If it did not match
the offer, then the receiving team paid the losing team
compensation for the lost player (two first-round draft picks).
Needless to say, players argued that Plan B had a stifling effect
on salaries, especially the salaries of the top players protected
387

under Plan B. You can probably guess how the owners
justified Plan B. That’s right. It was essential to maintaining
competitive balance in the league, and it enhanced fan
welfare.
Individual players, including New York Jets running back
Freeman McNeil, argued that even if Plan B did help
competitive balance, a less restrictive approach would
accomplish the same ends. The players successfully showed
that even run-of-the-mill players achieved restricted status,
proving the plan was too restrictive. Further, the players
showed that unrestricted players often received greater salary
increases than players who were identified as better by being
protected. Among other evidence, the players also noted that
no such restrictions on coaches and administrators appeared
necessary to preserve competitive balance.
The jury sided with the players. The jury found that Plan B
harmed competition for player services. It also found that
Plan B was too restrictive even though it aided competitive
balance. The jury suggested that a less restrictive approach
would produce the same balance results. Freeman McNeil and
the rest of the players named in the suit won compensation.
This case brought negotiations back into the realm of
collective bargaining.
The amount of time required for unrestricted free agency was
set to slide with the amount of time left on the CBA reached
the next year, 1993. Early in the agreement, unrestricted free
agency happened after four years. As the agreement has aged,
the amount of time to unrestricted free agency has increased
to five years. Players in year 4 of their contract are restricted
free agents. Teams can meet outside offers, and draft choice
compensation is required if one of these restricted free agents
is lost to another team.
THE PAYROLL CAP
The NFL cap also came out of the 1993 agreement between
players and owners. After the lawsuit over free agency, the cap
details were worked out through collective bargaining. With
this agreement, a negotiated share of defined gross revenues
goes toward player salaries. Those shared revenues are then
divided evenly among all teams, yielding a maximum amount
a team can spend on its players. Rules determine how the
amount spent on each player counts against this cap amount;
some elements of player compensation, such as signing
bonuses and deferred payments, receive special treatment in
the accounting process. In addition, the cap establishes a
minimum amount of spending. Otherwise, spending could not
be equalized across teams, and the point of the cap would be
undone.
If a team is over the cap, the commissioner steps in and
enforces it. If a team falls below the minimum, the
commissioner steps in, and players on that team are
compensated at the end of the season. If the commissioner
and the teams cannot resolve these issues, they move to
mediation and the courts.
388

The NFL cap supposedly was a hard cap, but loopholes such
as signing bonuses and deferred compensation have led to
routine spending in excess of the calculated cap amount.
According to cap rules, only salary counted. Bonuses and
deferrals did not. The incentive was for players to take their
compensation as bonuses, and teams could then spread their
cap spending farther. Furthermore, compensation could come
in so many ways that it was difficult to track.
For example, in 1995, Miami Dolphins’ quarterback Dan
Marino purchased stock in Republic Waste Industries on the
advice of Dolphins’ owner Wayne Huizenga. So did head
coach Don Shula and general manager Eddie Jones. Huizenga
then bought the company, and the stock value increased
dramatically. Marino invested $390,000 ($545,590) at the
starting price of $4.50 per share. After the purchase by
Huizenga, the price rose to around $21.75 per share. Marino’s
gain was a cool $1.5 million ($2.1 million) (News and
Observer Publication Co. Web site, www.newsobserver.com,
August 31, 1995).
This transaction was possibly a violation of the cap rules. All
payments, direct and indirect, by the team owner to players
count under the NFL’s salary cap. Commissioner Paul
Tagliabue vowed there would be a “thorough look” into any
possibility of cap violations in this transaction. If this was a
compensation payment, the risks were large. Marino’s
contract could have been voided and a $2 million ($2.8
million) fine levied against the owner of the Dolphins.
Subsequent adjustments to the cap language removed some of
these issues. Owners are now responsible for prorated
portions of bonuses under the cap. In addition, the remaining
portions of prorated bonuses still count against the original
owner’s cap amounts.
REVENUE SHARING
The NFL’s revenue-sharing plan, covered in Chapter 6, is the
most extensive in all sports. All TV revenues come from the
national-level sale of rights and are equally shared among all
owners, and 60 percent of local revenue is shared in a
straight-pool plan similar to that of MLB.
STRIKES AND LOCKOUTS
Early disagreements between NFL owners and players were of
little consequence. The complete list of stoppages is in Table
9.4. The 1968 strike was over pension fund contributions.
Players boycotted training camps, and owners locked them
out, but these events ended after an agreement in 10 days. In
1970, a lockout ensued while pensions, postseason
compensation, and grievance procedures were worked out.
This lockout occurred during training camp and did not
significantly affect fans.
Important labor–management disagreements have occurred
in the NFL, beginning with the strike of 1974. The 42-day
dispute was mainly over the Rozelle Rule, which was so strict
in its compensation requirements for teams losing free agents
389

http://www.newsobserver.com

http://www.newsobserver.com

that it essentially drove player movement to zero. However,
veteran players crossed picket lines and the owners won.
During the strike of 1982, players proposed that 55 percent of
leagues’ total revenues go to them as salaries and pensions.
They also demanded a wage scale based on seniority. Owners
did not budge and arranged for a $155 million ($342 million)
line of credit after the players threatened to strike once the
season started. Despite this significant display, the players did
strike. The owners offered a modified version of revenue
sharing, guaranteeing substantially less than 55 percent.
Players went along with this offer, but they continued to argue
over the dispensation of the funds. Players threatened to form
a new league and even staged all-star games. The strike ended
with owners getting pretty much what they wanted when it
looked like the entire season would have to be cancelled.
The strike of 1987 was the most wishy-washy player action
ever in any pro sport. Ostensibly, the strike was over free
agency, but most observers note that it was actually caused by
a management foul-up in the NFLPA. Executive Director
Gene Upshaw was rallying the players over free agency, to
which they were lukewarm, and simply neglected negotiations
until time ran out. A strike was announced, but the players
were not behind it. Owners put replacement players on the
field, and the striking players capitulated.
390

SECTION 9
Labor Relations in the
NHL
The NHLPA was formed in 1957. This was well after the NHL
owners on their own had established minimum salaries and a
pension plan for players. The only early confrontation over a
nonshared TV contract was resolved without much fuss.
During competition over talent from the World Hockey
Association (recall the discussion of rival leagues in Chapter
5), hockey felt the pressure of free agency in that rival league.
Later, the NHLPA made great strides earning free agency in
1995 with significant increases in player salaries. Player
contracts are guaranteed but with a twist—contracts can be
bought out at two-thirds of their value.
As a result of the 2004–2005 lockout, the entire season was
lost, all the way through the Stanley Cup Championship. The
resulting CBA includes a payroll cap, referred to as the Team
Payroll Range System, but no luxury tax. The NHL shares its
national TV contract revenues and also has a revenue
Redistribution System that sends money from higher-revenue
to smaller-revenue clubs. The logic of this redistribution is to
allow smaller-revenue clubs a chance to meet the payroll
spending minimum in the NHL cap system. The current NHL
CBA runs 2005–2011.
ARBITRATION
Players who are not yet full free agents (referred to as
restricted free agents) are eligible for salary arbitration. But
they must have been in the league for five years or be at least
24 years old. Unlike MLB, when players file for arbitration the
case moves forward regardless of whether the owner agrees.
This difference would suggest that players have even more of
an upper hand than in MLB.
THE DRAFT/ENTRY RESTRICTIONS
The NHL draft began in 1963. Players must be at least 18
years of age, and if they are drafted but not signed, they can
reenter after a year’s wait. If players are drafted a second time
after waiting to reenter and still go unsigned, they do not have
to sit out another year. These players are then free agents. In
the modern version of the NHL draft, the first 14 picks are
determined by lottery. Only teams that do not make the play-
offs are able to participate, and only the four clubs with the
lowest regular season points have a shot at the number one
pick. In 2009, there were seven rounds, and 211 players were
chosen.
In the NHL, rookie salaries are limited under the CBA. The
NHL has an age-based rookie wage scale. Players younger
than 21 years of age (the vast majority) must sign contracts
that last at least three years.
FREE AGENCY
391

In the NHL, free agency is subject to the most complex set of
rules of all sports. Players are divided into different groups
based on age and experience. Most are restricted free agents,
so that owners have the right of first refusal and receive draft-
pick compensation in the event that they lose a free agent.
Thus, most players in hockey are just like restricted free
agents in the NFL. At the unrestricted end are 32-year-old
players with four or more seasons in the league. For some of
the remaining players, there is arbitration. At the other end of
the spectrum, players with fewer than three years have no
free-agency rights.
PAYROLL CAP
The most important result of the 2004–2005 lockout was the
payroll cap in the resulting CBA. It is quite unlike any other
cap in pro sports. Starting at 54 percent for the first year of
the CBA (2005–2006), future percentages due to players were
set around league revenue targets ($2.2 billion, $2.4 billion,
and $2.7 billion). Players continue to get 54 percent if below
that target, and a small adjustment upward if revenues exceed
the targets. The owners of each club must then spend within
an established payroll range (the formula for the lower and
upper limits is complex). There is also an escrow account,
akin to that used in the NBA, that withholds player pay if the
league is spending over the players’ share or is taken from
owners if spending falls below that share.
In order to insure that smaller-revenue owners can spend at
the lower limit of the cap range, there is an interesting “Player
Compensation Cost Redistribution System.” It comes in a few
parts. Up to the first 25 percent of the redistribution comes
from any “excess” centrally generated league revenues (e.g.,
from national TV and licensed merchandise). Up to one-third
comes from any escrow deserved by the top 10 revenue clubs.
The rest of the amount destined to go to smaller-revenue
owners comes half from the play-off revenues of owners that
make the play-offs and half from “certain regular-season
revenues” of the top 10 largest revenue owners.
STRIKES AND LOCKOUTS
As shown in Table 9.4, strikes and lockouts have been rare in
the NHL; the two most recent were severe and the last made
history. The issues in the 1992 stoppage were a reduction in
the number of draft rounds from 12 to 6; if fewer players are
drafted, the remainder can sign as free agents, driving up
salaries. The second issue is familiar—compensation
requirements for teams obtaining free agents to pay the teams
losing them. Other issues were increases in the play-off share
for players and rights to likenesses on trading cards. The
strike lasted 10 days and had no impact on fans. On balance, it
appears the players won because the age requirement for free
agency was dropped from 31 to 30 years. However, the draft
was reduced by only one round, and the season’s length was
increased from 80 to 84 games.
The NHL lockout of 1994 lasted 103 days, and 468 games
were lost. The owners were worried that the players’ share of
league revenues had reached 61 percent. They argued for a tax
392

on payrolls, with the revenues going to smaller-market teams.
Players countered with their own tax plan. They got nowhere
until the season was about to be lost completely. It looks like
owners won this round. The salaries of players under age 25
were capped so that they increase by small increments over
five years, and eligibility for arbitration and free agency were
restricted for all players.
One view has it that the owners have stronger alternatives
than the players in the event of a work stoppage. Beamish
(1991) contends that because hockey ownership is actually
corporate ownership players have far to go before they can
muster the resources to match owners. For example, Molson
Companies, Ltd., owned the Montreal Canadiens. The team
was at the fourth level of their corporate organization chart, a
wholly owned and very subsidiary part of the parent company.
Because of their greater resources, it seems extremely unlikely
that owners will ever find a player association threat in hockey
believable. Of course, a mitigating factor is that corporations
seldom find it worthwhile to transfer funds from one division
to another to cover losses unless that division is contributing
at the margin to the overall value of the combined activity in
some way. This would limit the willingness of the Molson
company to ride out the losses incurred by their hockey team
during a work stoppage.
And we come to the most recent and most severe lockout in
sports history. The entire NHL season, including the Stanley
Cup Championship, was lost to the lockout of 2004–2005.
The issues mirror those of the NBA lockout of 1998. The
owners demanded “cost certainty” that they felt could be had
only through the imposition of a salary cap. The players
originally refused, but in one of the most colossal cave-ins in
union history (not just sports), they agreed to both the payroll
cap and the redistribution plan to support the smaller-
revenue owners described earlier. The eventual CBA also is a
very long-lived six years (2005–2011).
There are even more data to suggest that the owners locked
the players out to improve profits. However, as with the case
of the 1998 NBA lockout, we can only deduce information
about the potential gain from the lockout and the value of a
lost season of play. The primary resource is a comprehensive
report on the economic status of the NHL through the 2002–
2003 season by former Chairman of the Securities and
Exchange Commission, Arthur Levitt, Jr. (The Levitt Report,
2004). Only combined results are given for the league, and
there is no information on individual teams.
The Levitt Report shows NHL revenues at $2.094 billion
($2.48 billion). Throughout the lockout, the owners claimed
that players were receiving 75 percent of revenues. I could
find no way to verify this, so for our purposes, we will just take
this as given. Thus, close to the time of the lockout, players
received billion ($1.86 billion), while owners made offers
during negotiations in the area of $42.5 million ($50.4
million), or about 60 percent of the revenues in the Levitt
Report; eventually, the CBA includes 54 percent, which comes
to billion ($1.34 billion). So, the immediate gain would be
million ($521.4 million). The CBA lasts six years so that the
393

simple total is $2.64 billion ($3.1 billion). However, owners
need to think about the value at the time of the lockout in
order to determine whether a lockout is worth it or not. At a 5
percent interest over those six years, the discounted value is
$2.23 billion ($2.6 billion).
Turning to the lost value of an entire season, first note that the
owners are satisfied holding the players to $1.131 billion (the
54 percent of revenues in the CBA). The rest of revenues
would go to operating costs and profits. If one examines the
income and expense data reported in the Forbes annual
valuation of NHL teams, it is clear that expenses other than
player expenses come to about 30 percent of revenues, or
million ($744.2 million). The remainder is league profit:
million ($397 million). This is our best estimate of the lost
value of an entire season, M = $335 million.
Again, we have no knowledge of fan irritation costs or
alternative opportunity values for owners. NHL owners had to
consider whether the $335 million risk was worth the
potential gain of $2.23 billion. From our EV of a lockout, we
get the following:

Again, knowledge of P becomes important, and we write the
following:

Thus, owners will consider the breakeven probability of 15
percent, give or take the value of the next-best alternative net
of fan costs, divided by the gain. In the same way, we posed
for the NBA owners relative to their 1998 lockout. Let’s think
about what would have to happen to force the break-even
probability to 50–50. It would have to be the case that F
exceeds A by million. Angry fans would have to inflict costs
over and above the owners’ next-best alternative equal to 2.3
seasons’ worth of profit to make . Clearly, then if and if but
surely far less than 0.5. The NHL owners had a greater than
50–50 chance, anyway, since the players did fold in a historic
way. In the limiting case of certainty, the expected value
becomes $1.9 billion, plus (A – F). Owners stood to gain well
over $1 billion from their lockout.
In cases like the NBA and NHL lockouts, one reads and hears
many who chastise all involved because interrupted play can’t
be in anybody’s interest, especially the fans. Our examination
of these two episodes, in particular, suggests an entirely
different story. With just a bit of data and some business and
economic intuition, it is clear that interrupting play truly can
pay, even accounting for the lost value of play and fan anger.
Small wonder, then, that at the conclusion of the NHL
lockout, Commissioner Gary Bettman claimed the following
(www.nydailynews.com, July 23, 2005), “Today, our Board of
Governors gave its unanimous approval to a Collective
Bargaining Agreement that signals a new era for our League—
an era of economic stability for our franchises, an era of
heightened competitive balance for our players, an era of
394

http://www.nydailynews.com

http://www.nydailynews.com

unparalleled excitement and entertainment for our fans.” Our
assessment shows that the 2005 CBA was all that and a bag of
chips for NHL owners.
395

SECTION 10
Chapter Recap
In the early years of professional sports, competition between
leagues and owners kept player earnings at competitive levels.
Eventually, owners devised methods, such as reserve clauses,
to reduce player earnings below their MRP. To offset the
power of owners over player salaries, players organized
professional associations and unions.
With the National Labor Relations Act of 1935, players gained
the right to collective bargaining through associations of their
own choosing, and owners are bound to recognize those
associations. The law set forth by the NLRA is administered
by the NLRB and enforced by the federal courts.
In professional sports, owners acting through leagues and
players acting through unions negotiate collective bargaining
agreements that determine the economic relationship
between owners and players.
The NHL and MLB both have final offer salary arbitration
(FOSA). If an owner and player are unable to reach a salary
agreement, the player can have an independent arbitrator
choose either the player’s demand or the owner’s offer. If the
player loses, the player still gets the salary offered by the
owner. Regardless of which salary the arbitrator picks, the
player gets a salary increase. Fear of arbitration causes owners
to offer most players substantial salary increases so that
arbitration never becomes an issue in most cases.
If a union is decertified, players can sue owners under the
antitrust laws, an option unavailable under collective
bargaining. In 1992, NFL players decertified their union and
sued under antitrust laws to win unrestricted free agency.
In order for a union to be successful, unions must overcome
the high costs of educating members and limit free-riding
behavior. In addition, unions may face hostile political
environments and illegal actions by owners. Once a union is
formed, it acts like a small democracy. Elected union officials
serve the needs of those who control their reelection. Because
many rank-and-file members remain rationally ignorant of
union leader choices, many union leaders are able to establish
policies that favor a minority of players.
The first important observation about bargaining between
unions and leagues is that work stoppages rarely occur. Game
theory analysis of bargaining uses the idea that leagues and
unions consider the consequences of each decision in the
bargaining process, looking forward and working their way
back to the present, to make themselves as well off as
possible. Assuming full information and no alternatives in the
event a deal is not possible, the result is a 50–50 split. When
alternatives are available, the split covers the alternatives of
both sides and divides the remainder of the value equally.
396

This may cause one party to try to alter the opponent’s
alternatives. The value of this strategy is half the amount of
the change in the value of the alternative. Given this analysis,
there is no room for work stoppages.
Work stoppages occur (1) because of miscalculations due to
lack of information or (2) because it is worth it for one side to
bring a work stoppage. For the latter, the value of alternatives,
gains to be had after the stoppage, the economic costs, and
costs of lost fan enthusiasm must be considered. Perhaps
most important is the estimation of the probability of a
successful stoppage, and the breakeven probability is an
important piece of information in this regard.
The current status of CBAs varies across leagues. Both the
NFL and NBA have salary caps and restrictions on rookie
earnings. All leagues have free agency, but eligibility for free
agency varies across leagues. All leagues have entry drafts.
The history of strikes and lockouts offers many examples of
the theoretical issues presented in this chapter.
397

SECTION 11
Key Terms and Concepts
You should have run into each of these in pop-ups in the text
of this chapter:
• Labor relations
• Collective bargaining
• National Labor Relations Act (NLRA)
• National Labor Relations Board (NLRB)
• Bargaining in good faith
• Player representative
• Player agents
• Collective bargaining agreements (CBAs)
• Individual bargaining rights
• Final offer salary arbitration (FOSA)
• Union decertification
• Education costs
• Free-riding behavior
• Owner retaliation
• Illegal owner behavior
• Rational ignorance
• Concentrated benefits
• Dispersed costs
• 50–50 split
• Look forward, work back
• Differential alternatives
• Strike
• Lockout
• Asymmetric information
• Strategic misrepresentation
• Profitable strike or lockout
• MLBPA strike of 1994
• NBA lockout of 1998
• Rozelle Rule
• McNeil v. NFL
398

• NHL lockout of 1994
399

SECTION 12
Review Questions
1. Explain how organizing limits competition for player jobs.
How do organized player associations counteract owner
power?
2. What act created the right of collective bargaining? What
are the three key features to that act? How is the act
enforced? What federal agency has jurisdiction over
enforcement?
3. Explain how a modern players’ union is like a small
democracy.
4. What is a collective bargaining agreement, or CBA? How
is one formed? What are the key features of such an
agreement?
5. What is salary arbitration? What is meant by final offer
salary arbitration? How is eligibility for arbitration
determined?
6. What rights do players forfeit when they choose a
collective bargaining agent? Why might players in a given
league decertify their union?
7. What are the goals of a union? What role does player
marginal revenue product play?
8. Define each of the following: education costs, free riding,
owner retaliation, and political sentiment.
9. What is rational ignorance? Explain what is meant by
concentrated benefits and dispersed costs in decisions
made by union leaders. What does rational ignorance
have to do with these decisions?
10. What is meant by “look forward, work back” in
bargaining situations? Explain the basic 50–50 split
outcome in simple bargaining. What are the crucial
assumptions in such an outcome?
11. Explain the role of differential alternatives in
determining the split in a simple bargaining game.
12. What is the difference between a strike and a lockout?
13. What is asymmetric information in a bargaining setting?
How does it relate to strategic misrepresentation?
14. What is meant by a profitable work stoppage? What costs
do owners or players incur during a work stoppage? After
the stoppage is over?
15. What is the definition of the breakeven probability? Why
is this concept important in determining whether to force
a work stoppage?
400

SECTION 13
Thought Problems
1. Refer to the Learning Highlight: Champions of the “Free”
Market for Players? Not Who You Might Think. Why did
Charley O. Finley argue that if there was going to be free
agency, then all players should be free agents every
season? Why did Marvin Miller fear that type of outcome?
2. Is joint action by owners required to carry out collective
bargaining with players? Does labor law require it? If so,
what is the legal justification? If not, why does it happen
anyway?
3. How does final offer salary arbitration lead owners and
players to bargain more earnestly than they might
without it? Explain why arbitration would still be good for
players even if individual arbitration cases were lost more
often.
4. Suppose players in a given sport decertify their union.
What risks do they run in doing so? The NBA
decertification episode might be helpful in determining
your answer.
5. Carefully explain the role of each of the following in the
formation of a union: education costs, free riding, owner
retaliation, and political sentiment.
6. In the Learning Highlight: Pro Athletes as “Company
Men”, why do you think players so perfectly echoed
owners in their opinions on the reserve clause and the
draft?
7. What happens in the simplest bargaining outcome with
differential alternatives if U + L > X? Why?
8. Why is there no room in the simplest bargaining game for
work stoppages?
9. Explain how a work stoppage can happen despite the
wishes of both parties to avoid one.
10. Write the general formula for the expected value of a
work stoppage. Carefully describe each element. Suppose
that each element were to increase, one at a time. What
happens to the expected value of the stoppage in each
case? What happens to the expected value of the stoppage
if the interest rate increases? If the length of the stoppage
increases?
11. What role did signing movie contracts play in the
Koufax-Drysdale holdout of 1966? What were they trying
to demonstrate to Dodger owner Walter O’Malley by
signing them?
401

12. What did Marvin Miller mean when he said that the
growth in salaries attributable to arbitration was the
result of “the unconscionable exploitation of players in
the earlier years” (1983, p. 33)? What exploitation did he
mean, and how did arbitration help solve it?
13. What is the impact on player movement of the
requirement that compensation be paid to teams losing
free agents to other teams? Why would the commissioner
of the NFL, Pete Rozelle, impose such restrictions? (Hint:
Who hires the commissioner?)
14. What role did union decertification play in the advent of
free agency in the NFL? Why was it essential to those
pursuits? What happened when the NBA players tried the
same approach? Why?
15. Does there appear to be enough revenue to go around to
keep all teams in good financial shape in the NBA? Even
though simply sharing these revenues would bolster the
financial fortunes of smaller-revenue market teams,
explain why it is unlikely to happen.
402

SECTION 14
Advanced Problems
1. In Figure 9.2, suppose that another powerful group of
players were to appear. Diagram how this group would
enter the picture. What would have to happen for this
new group of players to have any impact on union leaders’
choices? How do you rate the chances of any such
alteration in the status quo in sports unions?
2. For years, the NBA had what was called a soft cap. The
NFL had a hard cap. What is the difference between a soft
and a hard cap? Explain how differences in the union
dynamic between the two leagues might produce such a
difference.
3. Table 9.1 has many blanks. Fill in the blanks, and prove
that each party will independently determine that the best
they can hope to do is to split the pot 50–50 without any
work stoppage.
4. Table 9.3 has many blanks. Fill in the blanks, and prove
that each party will independently determine that the best
they can hope to do is to split the pot 69–31 in favor of
owners without any work stoppage. Demonstrate that this
outcome fits the idea that, of what is on the table, both
alternatives must be covered and then the balance split
50–50.
5. Suppose total revenue for the NBA is $4 billion. Without
an agreement to work in a given year, players could make
$200 million. Replacement players would generate $400
million for owners. What share of league total revenue
will go to players? Now suppose owners know that if there
is a work stoppage, picketing reduces their alternative to
$300 million. If the threat of a picket is believable, what
share of league revenue will go to players? How much
would players spend to make picketing legal?
6. Using the general formula for the expected value of a
work stoppage, identify which elements are affected when
owners buy strike insurance or take out a line of credit to
be used in the event of a strike.
7. Suppose owners in the NFL are considering a lockout. It
will last one year. Owners usually make $300 million, net,
in a given year. If the lockout is successful, owners will
make $900 million the year after the lockout but only
$100 million during the lockout year. If the lockout is not
successful, owners will make $100 million during the
lockout plus the usual $300 million the next year. If the
chances of winning are 30 percent, what is the expected
value of the lockout? Should they do it? (Hint: Be careful
to consider the possible costs of fan ire in the year after
the lockout.)
403

8. Why do you think the MLBPA always comes out on top in
work stoppages in baseball? (That owners are inept is not
an acceptable answer.)
9. Under a salary cap system, who benefits from a lid on
rookie compensation? Given this, when would you expect
union leaders to agree to rookie compensation lids in
collective bargaining? Is your explanation consistent with
the idea that union leaders concentrate benefits on
politically powerful union members and disperse the cost
over the rest of the rank and file? Explain.
10. In the expected-value logic of the NBA lockout, how large
would the financial costs of fan ire, minus the league’s
alternative values, have to be to reduce the expected value
to zero if the probability of winning the lockout were 30
percent? As a percentage of owner profit (given in the
example), is this net cost of fan ire enough to cancel plans
for a lockout? Why or why not?
404

SECTION 15
References
Beamish, Rob B. “The Impact of Corporate Ownership on
Labor-Management Relations in Hockey,” in The Business of
Professional Sports. Paul D. Staudohar and James B. Mangan,
eds. Urbana, IL: University of Illinois Press, 1991.
Bishop, John A., J. Howard Finch, and John P. Formby. “Risk
Aversion and Rent-Seeking Redistributions: Free Agency in
the National Football League,” Southern Economic Journal 57
(1990): 114–124.
Coffin, Donald A. “ ‘These People Aren’t Real Big on Player
Reps’: Career Length, Mobility, and Union Activism in Major
League Baseball,” in Sports Economics: Current Research.
John Fizel, Elizabeth Gustafson, and Laurence Hadley, eds.
Westport, CT: Praeger Publishers, 1999.
Dixit, Avinash K., and Barry J. Nalebuff. Thinking
Strategically: The Competitive Edge in Business, Politics, and
Everyday Life. New York: W.W. Norton and Company, 1993.
Drysdale, Don with Bob Verdi. Once a Bum, Always a Dodger:
My Life in Baseball from Brooklyn to Los Angeles. New York:
St. Martin’s Press, 1990.
Dworkin, James B. “Final Offer Salary Arbitration (FOSA)–
a.k.a. Franchise Owners’ Self Annihilation,” in Stee-rike Four!
What’s Wrong with the Business of Baseball? Daniel R.
Marburger, ed. Westport, CT: Praeger, 1997.
Koppett, Leonard. The New Thinking Fan’s Guide to Baseball.
New York: Fireside, 1991.
Kramer, Jerry. Instant Replay. Cleveland: World Publishing,
1968.
Levitt, Arthur, Jr. “Independent Review of the Combined
Financial Results of the National Hockey League 2002–03
Season,” Prepared for the Commissioner of the National
Hockey League, New York, NY. February 5, 2004.
Miller, Marvin. “Arbitration of Baseball Salaries: Impartial
Adjudication in Place of Management Fiat,” Arbitration
Journal 38 (1983): 31–35.
Miller, Marvin. A Whole Different Ball Game: The Sport and
Business of Baseball. New York: Simon and Schuster, 1991.
Olson, Mancur. The Logic of Collective Action: Public Goods
and the Theory of Groups. Cambridge, MA: Harvard
University Press, 1965.
Quirk, James, and Rodney D. Fort. Pay Dirt: The Business of
Professional Team Sports. Princeton, NJ: Princeton
University Press, 1992.
405

Quirk, James, and Rodney D. Fort. Hard Ball: The Abuse of
Power in Pro Team Sports. Princeton, NJ: Princeton
University Press, 1999.
Schmidt, Martin B., and David J. Berri. “The Impact of Labor
Strikes on Consumer Demand: An Application to Professional
Sports,” American Economic Review 94 (2004): 344–357.
Staudohar, Paul D. Playing for Dollars: Labor Relations and
the Sports Business. Ithaca, NY: ILR Press, 1996.
Voigt, David Q. “Serfs versus Magnates: A Century of Labor
Strife in Major League Baseball,” in The Business of
Professional Sports. Paul D. Staudohar and James B. Mangan,
eds. Urbana, IL: University of Illinois Press, 1991.
406

SECTION 16
Suggestions for Further
Reading
Surely, I can come up with some.
407

CHAPTER 10
Subsidies and
Economic Impact
Analysis
Having a football team back in Houston will
bring thousands of visitors to our city, and it
will generate millions of dollars in our city. I’m
excited about our new stadium with a
retractable roof. And we’re also very happy
about getting a Super Bowl, and as you know
that’s very important economically to the city.
It will generate probably $300 or $400
million into our economy. But more
importantly, it focuses attention on a city that
people do not know enough about.
—Lee Brown, Houston Mayor
NFL Report, Winter 1999, p. 7.
CHAPTER OBJECTIVES
After reading this chapter, you should be able to:
• Explain how sports teams provide external
benefits that can justify a subsidy to team
owners.
• Understand that it can make economic sense to
subsidize a money-losing owner as long as
buyers’ surpluses are large enough.
• Appreciate that cost–benefit analysis provides a
useful framework for analyzing subsidies to
sports team owners.
• Understand that sports team subsidies do not
appear to generate new economic impacts or
enough development value to justify subsidies.
• See that some estimates of buyers’ surpluses and
external benefits indicate that sports owner
subsidies may be worth the cost but that other
estimates do not support that conclusion.

SECTION 1
Introduction
Prior to the 1950s, most stadiums were privately owned and
financed. The public share for such projects was just under 30
percent. However, this was soon to change. Shortly after his
appointment in 1951, MLB commissioner Ford Frick
announced that cities were going to have to start directly
supporting their teams through publicly owned and
maintained stadiums. His statements were based on the
observation that MLB teams were generating value to other
businesses near privately owned stadiums that owners were
unable to collect at the gate or through stadium advertising
(Miller, 1990, p. 71). Frick knew that state and local politicians
could collect taxes and provide stadiums for owners.
However, he could not possibly have foreseen the modern
consequences of his original announcement. The total bill for
25 pro sports facilities opening over the period 2000 to 2006
will be around $9.9 billion in 2009 dollars. The public share
will average about 63 percent, or roughly $6.2 billion.
In this chapter, we will see that sports owner subsidies can be
justified, but that this does not mean that any size subsidy is
justified. We will discuss how judging whether subsidies are
worth it or not is a question of costs and benefits and that,
although costs are difficult to ascertain, ongoing subsidies can
be calculated as another benchmark. On the benefit side, we
will see that direct and indirect economic activity is often
highly touted in subsidy debates. However, these benefits are
small, at best, and empirical analysis finds that there is nearly
no additional development value associated with pro sports
teams. We will see how this leaves buyers’ surpluses and
external benefits, commonly referred to as “quality of life,” to
cover the cost of the bulk of sports subsidies and discuss how
the jury is still out on the size of these types of benefits.
409

SECTION 2
The Logic of Sports Team
Subsidies
According to Mayor Lee Brown of Houston (see the opening
quote), it would seem that the value of a sports team to its
host city is so large that careful analysis of exactly how much
it is worth would be beside the point. A new stadium brings
thousands of visitors! Millions of dollars during the regular
season! Hundreds of millions of dollars for championship
games! However, let’s not forget that in raving about the
benefits of having a football team in Houston, the mayor was
trying to convince state and/or local taxpayers to provide
millions of dollars in subsidies in the form of a new football
stadium to the owners of the Houston Texans. Houston’s
mayor was not alone in making such a plea. Indeed, it has
been fairly commonplace, especially since the early 1970s, for
a city, county, or state to subsidize sports team owners
(Keating [1999] offers a complete list back to the 1920s).
Table 10.1 shows stadiums and arenas completed over the
last decade. The NFL leads MLB with 13 new stadiums
opening since 2000. The NBA then leads the NHL by one with
six new arenas opening since 2000. NFL stadiums and MLB
ballparks are the biggest spending items (even if we throw out
the new Yankee Stadium and Cowboys Stadium at over $1
billion each). While public shares don’t vary much across the
other sports at less than 60 percent, NBA arenas receive the
heaviest initial public subsidy at construction, 83 percent.
Subsidies are not just limited to stadium construction. Some
host states and cities subsidize streets near the stadium, water
and sewer services, and game-day safety and crowd control
services. In addition, the stadium and its operations are often
exempt from property taxes, representing another source of
public subsidy. Similarly, stadium leases typically grant
generous revenues and low rent to team owners as a subsidy.
Many question these subsidies in light of their size and the
fact that they provide benefits for very few taxpayers.
Economic arguments and other rationales have been used to
defend subsidies. The economic arguments are based on the
external benefits created by sports teams and the idea that
subsidizing a money-losing owner may be worth it to fans.
The other arguments for subsidies have to do with politics and
equity. Economists can only point out that equity pursuits are
not without costs of their own to taxpayers.
EQUITY/POLITICAL ARGUMENTS FOR SUBSIDIES
If judging whether subsidies to owners are worth it were such
a no-brainer, there would not be any debate over subsidies.
Subsidy foes argue that the subsidy money could be better
spent on some other, worthier endeavor. From an economic
perspective, the costs of any given expenditure are the other
values that could be had with the same money. However,
410

claims that education or some other social service would be a
better way to spend the money may miss the mark. If people
are willing to put more money into sports but would not put
any more into other endeavors, then this is a vacuous
argument. However, if it really is the case that money already
spent on other activities is being diverted to sports, then their
claims are correct. The only way to know is to look closely at
the funding arrangement itself.
Another argument regarding subsidies concerns the idea that
subsidies go to billionaire team owners to enhance the
incomes of millionaire players. Locally, this benefits very few
taxpayers except for those who are fans. Many local residents
find this to be unfair and believe that rich people should not
be subsidized. An economic explanation of this issue will not
satisfy anybody interested in the fairness of the subsidies.
Let’s suppose that a subsidy to a rich person generates more
wealth for society than any other possible use of the funds.
The world is still a better place with more wealth rather than
less. However, this does not mean that such subsidies are fair.
Economics has more to say about the argument that subsidies
are needed to keep a team competitive. For example, an
owner may argue that a new stadium is needed in order to
generate the revenue required to keep up with other owners
who have new stadiums. This would be true if the increased
revenues from the stadium were spent on better talent to field
winning teams. The increase in quality associated with greater
spending may cause fans to spend more on the team in terms
of tickets and merchandise. This sort of contagious quality
enjoyment might actually elevate the fortunes of the team and
the owner in the long run. On the other hand, if the owner
already knows the market and is already putting out the talent
level that fans will pay the most to see, then this plea is
without merit in terms of increasing quality. In this case, if
fans give the owner a new stadium, team quality would not
increase, and the owner will keep the money generated over a
few short seasons. On this important question, then, we
simply must know if increased quality of the stadium
experience has a positive impact on fans’ willingness to pay
for quality on the field.
A final argument against subsidies is that if having a new
stadium is so important, then team owners or their leagues
should foot the bill to build stadiums. Economics has quite a
bit more to offer on this point than the others. In some cases,
it may well be true that the value sought would be produced in
the absence of the subsidy. It could be a case of someone
trying to get somebody else to pay for what others want.
However, in other cases, the absence of the subsidy actually
harms society even if the money goes to rich people.
ECONOMICS AND SUBSIDIES
There are two economic explanations for how the absence of a
subsidy may hurt society. First, if owners cannot collect all of
the value their team generates, they will choose output levels
(quality in the long run and attendance in the short run) that
do not reflect the true net value of those commodities to
society. If owners could collect the true value to society, they
411

would choose higher quality and greater attendance levels. Of
course, this quality choice happens in equilibrium with all
other owners in the league, but adding these other values to
the outcome could alter balance among some teams in a
league. Second, it can make sense to subsidize an owner who
is losing money and who would rationally choose to move the
team. In some cases, fans may be better off subsidizing the
owner than losing the team altogether.
EXTERNAL BENEFITS FROM SPORTS TEAMS
On some important dimensions, a sports team bears a striking
resemblance to beekeeping. Both produce values that can be
captured by owners. Fans pay at the gate and purchase
advertised products, and the beekeeper’s customers buy
honey. Both also produce additional benefits that their owners
cannot capture. Bees pollinate nearby orchards, clearly of
value to the orchard owners, but there is no way for the
beekeeper to charge them for this service. In economics, this
pollination service for the orchard owner is called an external
benefit to beekeeping. Of course, the beekeeper does not pay
the orchard owner for the pollen that the bees bring back to
produce honey, either. The important point is that external
benefits are not counted in individual production decisions.
In sports, external benefits occur in two ways. First, there can
be measurable economic activity that occurs in relation to the
existence of the team. Newspapers and their writers report on
sports outcomes and make money, but they pay no fee to
sports team owners. TV sports channel news shows do the
same, and are therefore nonpaying beneficiaries. Other
independent writers and analysts (like your textbook author)
earn extra income from their proximity to sports teams and
also pay nothing. TV is another external benefit generator.
Once a game is aired, anybody with a television can enjoy it
regardless of whether he or she helps cover the costs. Teams
capture some of these values through advertising revenue, or
partly through some subscriptions, but many fans who never
pay a penny to teams or their advertisers enjoy televised
games.
A second type of externality concerns one of the scarce factors
mentioned in Chapter 2, namely, the commonality provided
to fans of a given team. Call it what you will—local unity, fan
loyalty, or civic pride—sports teams produce a broad array of
external benefits that can be categorized under the heading of
“quality of life.” Parents playing catch with their kids may
enjoy it more while discussing the performance of their
favorite sports stars. We talk about our teams around the
water cooler at work. We experience collective joy when our
team does well, and we mope collectively when it does not. All
of these benefits have two things in common: No price can be
charged, and no one is excluded from this type of enjoyment
when somebody else enjoys it at the same time.
Although external benefits are enjoyed by some fans, they
create a very real problem for society. The level of sports team
quality and the level of attendance do not reflect the true
value that fans place on these sports outputs. Fans place a
high-enough value on additional quality and attendance to
412

cover the costs of providing more. However, the owner can
only collect a lower value than fans are willing to pay, so
owners stop short of the quality level and attendance fans
actually would pay to see. This is the external benefit
inefficiency. Let’s look at a graphical analysis of this
phenomenon.
GRAPHICAL ANALYSIS OF EXTERNAL BENEFITS
Our analysis of external benefits follows Figure 10.1. The
figure shows a short-run situation where the graph of
attendance demand assumes that a particular level of quality
has been chosen by the owner. (We will examine the impact of
external benefits on the owner’s long-run quality decision
later.)
In Figure 10.1, DC is the usual portrayal of the demand
function. The owner can only collect the value of attendance
by choosing ticket prices along this function. The second
demand function, DS, shows the willingness to pay both by
fans and those enjoying external benefits. is derived as
follows. The MEB function represents marginal external
benefits that cannot be collected by the team owner. The MEB
function slopes downward, indicating diminishing utility from
external benefits (it could lie anywhere; there is nothing
special about the particular location of MEB shown in Figure
10.1). The vertical sum of MEB and gives the demand
function DS. If the owner could actually collect along this true
willingness to pay function, prices would be set along DS,
rather than DC.
This idea may be better expressed by thinking about what
happens in the presence of a positive externality at some
particular level of attendance, such as A1 in Figure 10.1. At
output level A1, the marginal external benefit enjoyed by
nonbuyers would be MEB1; the value that the owner can
collect would be V1; and the sum would be on DS. In the
presence of external benefits, the full value of attendance is
the vertical summation of MEB and DC.
THE PROBLEM WITH EXTERNAL BENEFITS
From bees to flu shots to lighthouses and national defense, a
competitive market in the presence of external benefits
produces a level of output that is inefficiently small. If the
producers could collect from those nonpayers enjoying
external benefits, they would increase output. The same is
true for attendance in the short run (and quality in the long
run). Although many observers think that we are wallowing in
sports, the presence of external benefits means that
attendance and attendance-related sales—plus the number of
televised games—are lower than they should be given the true
value that fans place on them.
This problem is demonstrated in Figure 10.2. The short-run
total revenue function TRC is derived from the demand
function DC. The short-run total revenue function TRS is
derived from the demand function DS. Remember that a
higher demand function would be associated with greater
team quality. In addition, short-run total costs will be greater
in the higher-quality situation, so that TCS is greater than TCC.
413

If the owner cannot capture external benefits, the profit-
maximizing level of attendance is A*C. If those benefits can be
captured, the profit-maximizing level of attendance is A*S. The
problem is that A*S > A*C. If the owners could collect from
those enjoying external benefits, their profit-maximizing level
of attendance would increase. However, because owners are
unable to collect, they make less, and fans get less than they
are willing to pay for.
External benefits have the same impact on the long-run level
of team quality. Because the externality is present for every
possible quality owners might choose, whenever the owners
consider profits for any level of quality, they will be smaller
than if the owner could collect along the true social value of
that quality. Therefore, owners choose a lower level of quality
relative to the true underlying social value fans place on
quality.
EXTERNAL BENEFIT REMEDIES
Either taxes or subsidies can allow the owner to achieve the
output level A*S. For example, local government officials could
inform the owners that they must increase team quality by
hiring better players, and if the owners did not respond, a tax
of sufficient size could be levied on deficient winning in such a
way that the owners would choose to increase quality rather
than pay the tax. While taxes have been used successfully to
change other types of decisions, they have not been used on
sports owners. Instead, subsidies are paid, typically for
stadium construction and operations. In the presence of
external benefits, these subsidies must be paid to owners in a
way that gets them to buy better talent and increase
attendance. If the form of the subsidy is not tied to increases
in team quality, then taxpayers run the risk of subsidizing
owners and getting no increase in quality in return.
SUBSIDIZING OWNER LOSSES
Owners may lose money on their teams despite their market
power position. We can envision two contributing factors in
this case. First, it could be that even though the owner has
adjusted quality in the long run to earn the highest possible
return, that return is negative because actual costs of
operation are not covered by revenues. If this is the deciding
factor, the best such an owner could hope to do is drop down
to minor league status; fans simply are not willing to pay for
true major league production. But let’s not forget the second
contributing factor. Remember that economic costs include
the opportunity cost confronting owners. Their next-best
option may be running their major league franchise in a
different location that offers higher profits. These opportunity
profits are imputed into the owner’s cost functions at their
current location. If this second is the deciding factor, we
might well see an owner argue that they are losing money
even though they may show positive accounting profits on an
annual basis. The best thing for this owner to do is to move to
the new location.
Of course, determining which mix of these two contributing
factors actually is in operation in any particular case is
414

complex. Our analysis in Chapter 2 casts serious doubt on the
general poverty claims of owners, but that does not mean
specific owners cannot on occasion lose money. It can be
difficult to distinguish a long-term change in an owner’s
fortunes with a short-term fluctuation in the determinants of
demand. Moreover, even if an owner’s situation worsens, that
does not mean the owner is losing money in the accounting
sense. Profits could fall but still be positive, and given the
difficulty of discerning profits by outsiders, one would be
suspicious that owners would claim they are losing money just
so that they can get a subsidy. And the final complication we
have already seen—owners acting as a league actually act to
keep viable threat locations open so that potentially profitable
other locations are almost always available. This makes it easy
for any current owner to claim they could be making more
money someplace else, whether it is actually true or not. We’ll
say more about this last complicating issue in Chapter 11.
Let’s suppose that the local sports team owner actually is
confronting long-run losses. In addition, let’s suppose that
there are no external benefit problems so that we can focus on
the issue of an owner losing money. In this case, collections at
the gate; from TV contracts; and from concessions, parking,
and memorabilia sales are not covering the owner’s cost
(which always includes the opportunity profit at another
location), and the owner is losing money in the long run. We
would expect that the owner would move the team elsewhere
or sell it to someone who would do the same. In some cases,
the buyers’ surpluses enjoyed by fans (the excess of
willingness to pay over the actual price paid) might be large
enough to cover the owner’s losses at the current location, and
then there would be no reason to move the team. If some
means of collecting the surplus from fans can be found, and
the subsidy sufficient to cover losses is actually given to the
owner, then the owner will stay put. We can gain significant
insight with a more in-depth look at this situation.
GRAPHICAL ANALYSIS OF SUBSIDIES FROM BUYERS’
SURPLUSES
We begin with a team owner who has chosen the long-run
profit-maximizing level of quality, leading to the short-run
costs and demand for attendance shown in Figure 10.3. For
this owner, the long-run prospect is losing money at the
profit-maximizing combination (Am, Pm). The loss per unit of
attendance is the deficiency of price (revenue per unit of
attendance) below average cost (per unit of attendance)
shown by line segment EC in Figure 10.3. Total losses are
calculated by multiplying average losses (per unit of
attendance) by the number of fans through the gate. At the
profit-maximizing level of output, total losses are the
rectangle DECPm. In this situation, we would expect the team
to fold in its current location and either shut down or move to
a more profitable location.
However, hope remains for fans based on the following
important observation. Because the demand function already
accounts for all of the consumption trade-offs facing fans (and
it already has external benefits included by assumption), then
415

buyers are better off with this team than without it. In
addition, if they gain enough satisfaction over and above their
other payments to the team, then a subsidy may exist that
would keep the team in its current location. Let’s examine this
possibility using Figure 10.3.
Again, the team’s losses are rectangle DECPm, and fans enjoy
buyers’ surpluses equal to the triangle APmC at the profit-
maximizing level of attendance Am. If buyers’ surpluses
exceed the losses and a portion of the surpluses sufficient to
cover the losses can be collected from fans, then a subsidy
exists that would keep the team in town. For the particular
situation depicted in Figure 10.3, part of the buyers’ surpluses
triangle APmC already covers part of the loss, namely, area
DBCPm. With that area out of the way, the comparison boils
down to the remaining buyers’ surpluses, triangle ADB, and
the remaining losses, triangle BEC. If the area of triangle ADB
exceeds the area of triangle BEC, then buyers’ surpluses
exceed the team’s losses. Because the line segment AD is
longer than line segment EC, simple geometry dictates that
the area of triangle ADB does, indeed, exceed the area of
triangle BEC.
Thus, in the case shown in Figure 10.3, a subsidy does exist
that allows two things:
1. The team would continue to operate at this location
because losses are covered by the subsidy. Remember, the
cost function already includes the value of the next-
highest use of the resources going into team production.
Therefore, the minimum subsidy just equal to rectangle
DECPm is all that is required to keep the team in town.
2. Fans would be happier with the team, even though they
would have to pay the subsidy, than they would be
without the team. After all, at the minimum subsidy, some
buyers’ surpluses remain that are equal to the area of
triangle ADB minus the area of triangle BEC.
IMPORTANT CONSIDERATIONS AND CAUTIONS
The theory is sound, but a number of other factors must be
considered. The most obvious follows from the discussion of
subsidizing a money-losing owner. Just because a particular
subsidy can be identified that will cover externalities and
possible losses does not imply that a subsidy would exist in
every case that might be analyzed. Sometimes, economically
speaking, a team just has to go. It all boils down to whether
line segment AD is longer than line segment EC in Figure
10.3. If it is not, then no subsidy exists that would keep the
team in town and satisfy fans. The subsidy required to cover
losses would exceed buyers’ surpluses. Furthermore, over
time, the subsidy issue would need to be revisited because
cost and demand functions can shift.
NOT JUST ANY OLD SUBSIDY LEVEL
For both the external benefit and money-losing owner cases,
our analysis suggests that there is a minimum level of subsidy
that will do the job. Just because a particular subsidy can be
416

identified in each case does not mean that any sized subsidy
would be justified. The minimum subsidy does two things:
1. It gives the owner the incentive to produce the efficient
level of winning in the face of external benefits. The
minimum subsidy increases attendance from to in Figure
10.2.
2. In the event of any losses, the minimum subsidy covers
the area of rectangle DECPm in Figure 10.3.
The minimum subsidy is all that is required to keep the team
in town. The owner may wish to obtain more, but no addition
to the minimum subsidy is required to entice the owner to
keep the team at its current location.
COLLECTION ISSUES
Collection issues also exist, all of which are based on fairness.
Let’s define fairness as collecting from those who enjoy the
benefits of the team. For example, in the case of external
benefits, this idea of fairness would dictate that the subsidy is
collected from those enjoying the externality. However, how
do you find these people, and how much do you charge each of
them? It would be impractical to drive around and tax people
playing catch with their children. Even if you tried, how would
you decide the level of their enjoyment in order to levy an
appropriate tax? These are classic issues in public finance that
are not easy to overcome. Typically, governments resort to
clumsy instruments that neither set taxes proportional to
benefits nor collect just from those who enjoy the benefits.
In the case of trying to collect buyers’ surpluses, the target
would be fans who buy the team’s outputs at the gate, by
subscription, or purchase goods advertised during TV
broadcasts. Ticket surcharges are one way to collect from fans
who buy tickets and enjoy buyers’ surpluses.
However, government choices are not always so precise. A tax
on hotels and rental cars is another common device aimed at
collecting buyers’ surpluses from out-of-town fans. Although
some hotel and rental car users may be fans from out of town,
most are not. In this case, many who are not fans are also
charged a portion of the subsidy. This is just as true of tax
schemes covering all local taxpayers; not all residents are
sports fans enjoying buyers’ surpluses or external benefits, but
they pay just the same. In this case, all the money does not
come from fans, and there is no reason to expect these
nonfans to be very happy about their forced contribution.
THE POLITICAL PROCESS
Finally, buyers’ surpluses exist for team owners who are not
losing money. Profitable owners may push for state and local
governments to provide subsidies. Owners actually earning
profits can use the accounting tricks discussed in Chapter 2 to
argue that they are losing money and deserve such subsidies.
Although subsidies are sometimes required for team owners
who lose money in the long run, the political process can end
up providing subsidies for a team that is doing just fine.
417

SECTION 3
Cost-Benefit Analysis
Concepts in Sports
As the preceding section points out, a particular subsidy is not
the same as any subsidy somebody might want. Cost–benefit
analysis offers a way to analyze whether a particular subsidy is
worth it. In a nutshell, the analyst lines up all of the costs and
all of the benefits of a subsidy. If the benefits exceed the costs,
then the project at least is worthy of consideration. The
analyst’s next step would be to compare the net value of the
project to other possible projects before committing public
funds to any project at all. A cost–benefit analysis identifies
which project has the larger net benefit.
Cost–benefit analysis proceeds on the basis of the project in
question. If it is a large project, costs and benefits will be
large. If it is a smaller project, costs and benefits will be
smaller. A cost–benefit analysis cannot tell a person what size
project to choose. Why build a stadium for $400 million?
Why not $250 million, instead? Cost–benefit analysis is no
help on that question. All it can do is compare costs and
benefits of projects of different size.
For now, suppose we already have a project and we want to
assess its benefits and costs in order to determine whether a
positive net benefit exists. Comparing benefits and costs is
fundamental to any economic decision. If such a comparison
is not done or is done poorly, you cannot make yourself as
well off as possible. Overvaluing benefits or undervaluing
costs leads to decisions that waste resources. We all do cost
and benefit comparisons—it is actually pretty easy to
conceptualize. However, the actual analysis presents
difficulties. For the rest of this section, let’s look at the cost–
benefit approach in general terms and the difficulties of
actually performing one.
418

SECTION 4
Sports Subsidy Costs
Let’s start off identifying the costs. State and local government
can subsidize sports by providing public infrastructure for the
owner, subsidizing team operations at the stadium, and
building or renovating stadiums. All of these are subsidy
costs.
INFRASTRUCTURE SUBSIDIES
Infrastructure is a catchall name for the public services
provided to the facility where the owner’s team plays. These
include public services, such as water, electricity, and streets,
and safety services, such as police (both for security and for
traffic management), fire, and ambulance services. State and
local government may foot part of the bill for these types of
services. Infrastructure subsidies are payments by cities and
states to owners to pay for infrastructure costs. While our
focus is on infrastructure subsidies to team owners, similar
subsidies are typically offered to any large business as location
incentives by state and local governments.
OPERATING SUBSIDIES
A second type of subsidy goes to owners through leases. Few
owners actually own their own venue. Most owners lease
facilities from the city or state. Clauses in the lease contract
between team owners and the government often include
subsidies to ongoing team operations called operating
subsidies. For example, some owners pay no rent, others pay
large rent, and rent can be a flat fee or based on the level of
attendance.
Subsidies based on attendance can overcome external benefit
problems associated with sports teams. However, rent is
typically graduated so that it increases with attendance. This
graduated increase does not give an owner incentive to
increase attendance to overcome external benefits problems.
Thus, the overall level of the subsidy, plus monitoring by
government officials, will have to generate increased quality
and attendance to take care of external benefit inefficiencies.
The rest of the overall operating subsidy includes all of the
remaining revenues and costs at the stadium. Some owners
get all of the concession and parking revenue, whereas others
get none. Some owners get all of the nonsports revenue. Some
owners are responsible for maintenance and upkeep. Similar
variations in lease outcomes exist for signage and advertising
revenues, as discussed in the Learning Highlight:
“Subsidese,” the Language of Lease Contracts at the end of
this section.
Property tax treatment can also be included in operating
subsidies. All property owners pay property taxes. The chance
419

for a subsidy here lies primarily with the case of publicly
owned facilities. Sports team owners may not pay any
property tax outright, or they may pay it through their rent
payment. Indeed, local authorities do not impute forgone
property taxes into the cost of stadiums and arenas. However,
if the property were leased or sold outright to another type of
business, property taxes would be included. This special
treatment of sports team owners is another element of the
operating subsidy.
STADIUM CONSTRUCTION SUBSIDIES
The most talked about of all subsidies are stadium subsidies
payments by taxpayers to team owners to support the
construction and operation of a stadium for the owners’ team.
Construction costs are usually publicly stated, as we saw in
Table 10.1. These amounts are so large that construction is
financed through borrowing, either directly with loans or by
issuing public bonds.
Revenue bonds must be paid off strictly from revenues
generated by the facility. However, this dramatically limits the
subsidy aspect of construction and renovation; therefore, this
type of bond is very seldom used. General obligation bonds
can be paid off through a wide variety of sources, which
broadens the ability to subsidize owners considerably. Ticket
surcharges and rent can be used to pay on these bonds, but
other methods that fall less on fans are also popular. Taxes on
goods and services like hotels and rental cars can also be used
to pay bonds. This method is politically popular because out-
of-towners typically pay the tax. Another popular method is to
increase state and local sales taxes by a small amount.
Sometimes a local government is granted sales tax forgiveness
for a specified time period by its state in order to use those
funds to make bond payments. In addition, this method of
bond financing may put other costs off on taxpayers in a
larger jurisdiction. Zimmerman (1997) points out that the use
of tax-exempt bonds puts costs onto federal taxpayers, who
must make up the difference.
WHY STADIUMS RATHER THAN A CASH PAYMENT?
Siegfried and Zimbalist (2000) pose and then answer the
question, “Why subsidize a stadium rather than just make a
cash payment to the owner?” The authors give the following
reasons why a stadium subsidy might be chosen over giving
money directly to an owner:
1. There will be political support from local contractors and
property owners at and around the site. Construction
certainly will make contractors who do construction better
off, and property owners near the site may see their
property values increase (whether this is a true increase in
value to the entire paying jurisdiction is another matter).
2. Stadium subsidies can tie the team to the city through the
lease. Some lease clauses impose dramatic penalties on
teams that try to leave the city, including repayment of the
stadium subsidy.
420

3. The enhanced revenue stream that a stadium subsidy
represents must be collected over time and through good
team management. A single lump-sum cash payment does
not provide the team owner with any incentive to manage
well in order to obtain the payment.
4. Cash payments to the owner may draw other parties to
demand cash payments as well.
5. Stadium subsidies are more politically palatable than cash
grants to wealthy people.
ANALYTICAL DIFFICULTIES IN EVALUATING
SUBSIDIES
To avoid waste, subsidies must be evaluated relative to all the
costs. In economics, opportunity cost is the guiding principle,
and careful attention to opportunity cost leads us to consider
secondary costs. Dollar costs, such as the amount of money
that must be borrowed in order to build the stadium, may be
very well known. For example, in Table 10.1, people in the
Milwaukee area were asked to pay a subsidy of $310 million
($373 million) as their share of the total for a new ballpark.
However, a full evaluation of the opportunity costs may go
beyond just the dollars cited. For example, the $310 million
may not include property taxes forgone for a publicly owned
stadium. In addition, the full implications of the costs
imposed by the subsidy-funding method must be considered.
Further, because loan or bond payments are made over time,
uncertainty is also an important cost factor. Future inflation
and interest rate behavior are also important cost-
determination factors.
We should not be naïve about the stadium-funding process.
Stadium supporters will overstate the benefits and understate
the costs. Detractors will pursue an offsetting strategy. Cost
overruns will occur. Intuition suggests that cost overruns are
systematically understated in all subsidy considerations. If
waste is to be avoided, all costs must be included, and
distortions of this type must be recognized in the analysis.
A BASIC FORMULA FOR CALCULATING SUBSIDIES
A straightforward way to calculate the annual costs of stadium
subsidies is available (Okner, 1974; Quirk and Fort, 1992;
Long, 2005). Once the actual amount of public spending is
known, then the annual costs are depreciation, the
opportunity cost of funds, and property taxes forgone. Thus,
on an annual basis, if we cast operating benefits as net
operating revenues from the facility to local government (the
net revenue specified in the lease), then the annual subsidy
formula is as follows:

As we will see, benefits can also be difficult to calculate.
However, if the subsidy is known, we can at least ask if the
benefits are worth at least that much.
421

LEARNING HIGHLIGHT: “SUBSIDESE,” THE
LANGUAGE OF LEASE CONTRACTS
Some lease outcomes might make for very interesting
questions on an accounting quiz. For example, owners of the
Colorado Rockies keep all revenues from the stadium except
parking revenues. The owners are also responsible for
maintenance and operations. However, if the partners take a
cash return in any lease year equal to 5 percent of the original
amount of capital they paid to become part owners, then they
pay 2.5 percent of the team’s net taxable income (after
reduction by the 5 percent return) to the Denver Stadium
Authority.
Some operating subsidies in leases are difficult to calculate.
For example, Tampa Bay Buccaneers owner Malcolm Glazer
also owns 70 percent of the Houlihan’s restaurant chain. In
the Buccaneers’ lease, the Houlihan’s chain pays $10 million
annually for naming rights at Raymond James Stadium.
However, under terms of the lease, all naming rights revenue
goes back to the team owner, Glazer. Because he could simply
give the money back to the restaurant chain, sponsorship
could effectively cost Houlihan’s nothing. Meanwhile,
approximately 70 percent of any increase in Houlihan’s
profits as a result of this savings goes back to Glazer
personally because he owns 70 percent of Houlihan’s.
Leases also contain very interesting “nuts and bolts” clauses.
Typically, teams receive all of the revenue from signage
advertising. Few owners get none of these revenues. Most
owners get at least a portion of signage revenue. In baseball,
the Twins get all of the signage revenue, but no advertising is
permitted on the scoreboard. The Yankees get 100 percent of
signage revenue, except for signage on the exterior of the
stadium. The Pirates get 33 percent of revenues from the
concourse only. The Mariners pay $318,000 for signage rights
and get to keep 75 percent of signage revenues, except that
they only get to keep 50 percent of revenues from the
DiamondVision scoreboard in Safeco Field.
In basketball, the Mavericks get 50 percent of signage from
static advertising and 100 percent from nonstatic advertising
(all ads that are in animated or video form). The Nuggets get
70 percent of signage up to $200,000 and 100 percent
thereafter. The Clippers get 100 percent of signage from
basketball but only 50 percent from nonbasketball events. The
Heat pays $275,000 for 100 percent of signage. The Spurs get
100 percent of arena and concourse signage but only 40
percent of any other signage. The Supersonics must pay
$750,000 over 15 years for all of the signage.
In football, the Cardinals keep 100 percent of the revenue
from message board and video display advertising. The Giants
and the Jets occupy the same stadium, and each keeps 50
percent of net signage except from the scoreboard. The Rams
get 75 percent of signage up to $6 million but only 90 percent
after that.
Finally, in hockey, the Ducks get 100 percent of hockey-
related advertising and 50 percent of the rest. The Sharks get
422

100 percent of hockey-related advertising and 50 percent of
fixed signage.
Meanwhile, just figuring out occupancy charges (the
aggregate of rent and any other charges) can be quite a chore.
Because everything else (sponsorship, signage, other ads,
concessions, and parking) are shared benefits, we will stick
with any ticket or seat-based payments that owners must
make and any amount of team profits that go to the stadium
authority. The following are just a few examples. The only way
to know how much Safeco Field is worth to the Seattle
Mariners is to understand the details of their lease.
In baseball, the Mariners pay 7 percent of revenue on the first
million tickets sold or $160,000 plus 5 percent of revenue on
the first million tickets sold, whichever is greater; 60 percent
of net suite rental revenue on 48 of 77 suites; $9,750 for 1996
game-day expenses; and 5 percent of reported net operating
profits.
It does not get any simpler in basketball. The Mavericks pay
$10,000 or 7 percent, whichever is less if per game receipts
are less than $324,000, and $10,000 plus 5 percent of excess
of per game receipts over $324,000.
From football, the Jaguars pay $250,000 per year through
2000; $500,000 through 2005; $1 million through 2015;
$1.25 million through 2025; $2.50 per ticket surcharge; and
$2 per car surcharge.
Finally, as you might suspect from the complications for their
cotenants, the NBA’s Mavericks, the hockey Stars pay the
lesser of 7 percent or $10,000 on gate receipts up to
$324,000; $10,000 plus 5 percent on gate receipts over
$324,000.
Sources: Sports Leases on CD-ROM, Major League Facilities,
vols. 1 & 2 (Chicago, IL: Team Marketing Reports, 1998);
Roger Noll and Andrew Zimbalist, “Build the Stadium—Create
the Jobs!” in Sports, Jobs and Taxes: The Economic Impact of
Sports Teams and Stadiums, Roger Noll and Andrew
Zimbalist, eds. (Washington, D.C.: Brookings Institution,
1997a).
423

SECTION 5
Estimates of Sports
Subsidy Costs
We have laid out the conceptual approach to costs. The sizes
of the obligations cities and states assume are shown in Table
10.1. Monthly payments are easy to determine once the bond
issue and loans are signed. But then cost measurements
become difficult. The sources used to make the monthly
payments, detailed in earlier sections, carry a host of
additional considerations. When all of these are considered,
the basic formula gives a close assessment of the size of the
subsidy. The subsidy is the excess of depreciation, opportunity
cost of funds, and forgone taxes over net operating revenue
from the stadium that actually goes to state or local
government. In this section, we will examine subsidy costs in
detail.
EARLY SUBSIDY ESTIMATES
Benjamin Okner (1974) was the first to consider stadium
subsidies under the basic subsidy formula. He compiled a
survey that produced data on gains and losses of 30 publicly
owned stadiums. Table 10.2 contains a summary of his data.
Only five of the 30 stadiums showed any net gain. The rest
lost money. Converting Okner’s 1970–1971 findings in Table
10.2 to 2009 dollars, fully 20 percent of the stadiums had
losses in excess of $5.3 million. Okner found that the total
subsidy for stadiums showing losses was about $23 million
($121.1 million) in 1970–1971. The average loss was about
$4.8 million (2009 dollars) per year for each of the 25
stadiums in the red.
MORE RECENT SUBSIDY ESTIMATES
James Quirk and I (1992) report the results of a similar
analysis on a more modern set of stadiums. The subsidies to
25 publicly owned stadiums and arenas were around $187.4 in
1989 ($322.2 million), for an average of $12.9 million per
facility in 2009 dollars. This represents just under a threefold
increase in the average subsidy per facility per year over the
earlier findings by Okner.
Table 10.3 shows A few of the highlights of our calculations.
For the discussion here, all values are converted to 2009
dollars. The minimum subsidy was for Green Bay’s Lambeau
Field at about $325,000 (the data here are before Lambeau’s
renovation in 2003, see Table 10.1). The facility closest to the
average subsidy was Atlanta-Fulton County Stadium (the
previous home of the Atlanta Braves and Atlanta Falcons), at
about $12.9 million. The real white elephant is the Superdome
in New Orleans with an annual subsidy of about $72.5 million
(still the home of the New Orleans Saints).
In the most recent (and exhaustive) assessment of public
subsidies, Long (2005) develops an extensive data set on all
424

99 facilities in use in the four major leagues as of 2001. Her
data allow an in-depth examination of the basic formula for
calculating subsidies. The data on the opportunity cost of
public funds include not just building costs but public land
and infrastructure costs as well. She also has extensive data
on lease arrangements that allow a more precise assessment
of the net operations portion of the subsidy calculation. In
2001 dollars, the total subsidy is about $17.3 billion ($20.8
billion) across all 99 major-league facilities for the years in
Long’s sample, 1990–2001. On average, that would be $1.6
billion per year ($1.9 billion) and $16.2 million ($19.2 million)
per facility annually. Table 10.4 shows the breakdown by
facility type. It’s interesting that dual-purpose stadiums and
arenas have the lowest subsidy levels. Long concludes that a
much larger share of the total cost has been borne by the
public than previously realized:
Overall, the findings refute the much-touted claim that
during the 1990s, team owners and other private entities
were “partners” in sharing the burden of facility
financing with taxpayers. Instead, the analysis shows
that upfront private contributions are often substantially
recouped through lease-based subsidies and exemptions
from property taxes. Although industry sources estimate
that the average public share of costs for a new sports
facility is 56%, my findings show that after adjusting for
omitted subsidies, the average public share is 79%—an
increase of 23 percentage points. (p. 139; Judith Grant
Long, Journal of Sports Economics (Vol. 6, Issue 2), pp.
25, copyright © 2005 by SAGE Publications. Reprinted
by Permission of SAGE Publications.)
So we have the following comparison spanning Okner’s
original work for 1970–1971 to my work with Quirk for 1989
and on to Long’s findings for 2001. All measured in 2009
dollars, the average subsidy grew from $4.8 million in 1970–
1971 to $12.9 million in 1989 and on to $19.2 million in 2001.
That’s a real annual rate of growth of 4.7 percent, or about 57
percent greater than the 3 percent typical real growth rate in
the economy at large. This is compelling evidence that the
process generating stadium and arena subsidies deserves
scrutiny, and we will move on to that in the next chapter.
Subsidies may also include a federal subsidy component.
When stadiums are subsidized with tax-exempt bonds, taxes
that otherwise would have been collected for the federal
treasury must be made up by all federal taxpayers. On the
same sample of stadiums used in Quirk and Fort (1992),
Zimmerman (1997) calculates the maximum federal subsidy
total to be another $24.3 million ($32.3 million), over the life
of those 25 stadiums.
WHAT ABOUT EXTERNAL COSTS?
Proponents of new facilities always tout their external
benefits. And they tend to come in second in terms of the
amount of public discussion and press coverage, immediately
behind economic activity estimates. However, little attention
is directed to the other side of the externality coin. Just as
425

some people earn benefits that owners cannot capture, some
people bear costs that those same owners do not consider
either. Neighborhood groups typically voice opposition,
fearing damage to the character of their neighborhoods, and
their respective property values, due to rezoning for
businesses around the new facilities. To date, I have seen no
estimates of these very real costs due to public subsidization
of sports facilities. These opponents receive occasional, light
press treatment, but their main avenue into the process, and
then only in a very few cases, is through the courts.
426

SECTION 6
Sports Subsidy Benefits
Having dealt with cost concepts, let’s turn to the benefit side
of the evaluation. Analysts have broken the types of subsidy
benefits into three categories: economic activity benefits
(direct and indirect), development value, and quality of life
values measured by buyers’ surpluses and external benefits.
In the following discussion, all of these benefits are
considered in the context of either having a team or losing
that team altogether in the absence of a subsidy.
ECONOMIC ACTIVITY BENEFITS
Economic activity concerns the measurement of economic
welfare created by a team and its stadium both during the
construction phase and after the stadium is completed and
occupied by the team. The most important issue here is
making sure the welfare is new, rather than simply reallocated
from other spending. In addition, extreme care must be
exercised to properly ascribe only those values that occur due
to the presence of the team.
Direct economic activity is economic activity that occurs at the
stadium. During the construction phase, all site planning,
preparation, and construction are direct economic activities.
After construction is completed, all activity that is generated
by stadium and team operations and the support businesses
that serve the team is direct economic activity. An easy
measure of direct economic activity after construction would
be team total revenues.
Indirect economic activity describes the benefits accruing to
other businesses that do not pay anything to the team for
those benefits. As such, they are part of the external benefits
produced by the team but only represent the measurable
impact on surrounding businesses. Proximate businesses are
the most obvious recipients of indirect economic activity.
They do not directly deal with the team, but they do rely on
fan traffic for part of their income. Nearby restaurants and
retail outlets are good examples of proximate businesses. The
idea behind indirect economic activity is captured by
spending on food, lodging, and transportation by those
enjoying the team.
It is important to distinguish indirect economic activity from
other externalities produced by the team. For example, the
impact of the team on the local newspaper is not included in
indirect economic activity but is an externality produced by
the team. Suppose a city failed to provide a subsidy and lost
its NFL team. How would a newspaper fare without a home
team on the sports pages? Circulation and ad revenues would
probably decline. The same goes for local TV broadcasts. One
would expect that the advertising slot value of the sports
segment of the local news would decline in value if the city
lost its team. But only additional sales to people who come to
427

watch the team’s games would enter the economic activity
calculation. All of the rest of the values are discussed under
the section “Quality of Life”.
Table 10.5 lists some economic activity values that were
calculated by analysts in Atlanta. The sum total is impressive.
But is the host city’s economic activity value really this large?
And where does the value go? Attempts to assess this value
fall under the general heading of economic impact analysis,
which we will shorten to EIA. A number of guiding principles
determine correct EIA.
NEW ECONOMIC ACTIVITY
First and foremost, only new economic activity matters in
EIA. New economic activity is any economic activity created
by the subsidy that did not exist before the subsidy. This
definition is nothing more than the marginal distinction
common to all economic decisions. In a given jurisdiction, the
opportunity cost of the subsidy (including all secondary
effects on other programs) is the additional, or marginal, cost
to taxpayers. The proper comparison is to the additional, or
marginal, benefits that will flow from the subsidy. New
economic activity is commonly referred to as economic
growth. Therefore, only if new economic growth attributable
to the subsidy exceeds the value of the subsidy is the subsidy
worth considering.
This seems straightforward, but it is the most common cause
of confusion in EIA. Let’s remember how growth happens. A
new firm might enter a jurisdiction. It raises wages by
competing for labor. Rising wages represent increased income
for the existing labor pool. As income rises, new economic
impacts occur that can be measured through spending. Thus,
in this case, observed increases in spending are due to growth.
Other factors that result in growth are an increase in demand
for things produced by people in the jurisdiction or enhanced
education and training that raises productivity. Note that all
of these result in greater income than previously was observed
in the jurisdiction.
However, government subsidies that alter the flow of
spending in a given jurisdiction need not create any growth at
all. If the subsidy comes from rearranging taxes, then taxes
that would have been spent somewhere else are now spent on
the subsidy. Unless growth happens as a result of this
rearrangement, all subsequent measurable spending will just
be rearrangements that have nothing to do with growth. If
government borrows to provide the subsidy, the same logic
holds but the opportunity costs are borne over time. If there is
no new growth, then the spending is just a transfer (over
time), and observed changes in spending patterns are just
rearrangements of economic activity that would have occurred
anyway. No new income is created.
The question boils down to the likelihood that stadium
construction and operations subsidies will create economic
growth. Compare these types of subsidies to spending on
education. The increased productive capacity provided by
education leads to higher incomes, higher spending, and
economic growth. However, what enhancement to
428

productivity follows the construction of a ballpark or a more
favorable lease? There may be a flurry of activity, but unless
that flurry of activity generates new income, it is all
rearranged spending that would have happened anyway. If
there is no new income, no new economic impact is generated.
Consider fans and a direct economic activity—attending a
game. At the ballpark, fans buy tickets, pay for their parking,
and buy souvenirs and food. If there were no team and no
ballpark, the same fans might go to the movies or a play. Once
again, tickets would be purchased, parking would be paid, and
food would be bought. Admittedly, more movies might be
attended, but spending on the part of fans is just rearranged.
No new income is created.
Let’s look at an example concerning indirect economic
activity. Suppose a restaurant moves from a neighboring city
to enjoy the crowds created by a new ballpark. This is a new
economic impact only if growth occurs. That is, the move
represents a new economic impact if the restaurant bids up
the price of labor in the city so that incomes rise. Otherwise,
employment just moves from other restaurants or elsewhere
in the service sector, and the net impact of the restaurant is
zero. Even increased revenue enjoyed by the relocated
restaurant owner will not be new spending if it is simply
rearranged from another activity in the city. This would be
even more likely for a restaurant that moved from one part of
the same city to be closer to the ballpark. Again, there is no
growth. In this case, it would be incorrect to count the entire
value of a restaurant that moved close to the stadium, even if
from another city entirely. If all of this activity were counted
as a value flowing from the subsidy, the analyst would be
guilty of double counting. Rearranged spending simply is not
new activity.
By the same analysis, subsidies will not generate growth if all
they do is keep an existing team from moving. In this case,
substantial alterations in business patterns will not occur
because business patterns were already established around
the existing stadium. Therefore, any new economic impact is
unlikely if the subsidy is to keep an existing team rather than
to entice a new one. The only chance for growth if a team
already is in the area is any increased value that fans put on
the stadium and team. The resulting increased spending must
come from new income rather than from other spending.
CAREFUL BOUNDARY DEFINITIONS FOR CORRECT EIA
EIA must assess the benefits of a subsidy using the same
political boundaries of the party that provides the subsidy. For
example, suppose a city government is considering a stadium
subsidy for its NFL team. Benefits that would occur outside
the city limits would be irrelevant to the city decision makers.
On the other hand, if the state government is providing the
subsidy in question, statewide estimates of the benefits would
be the focus of EIA. It is easy to see why proponents of a
subsidy would like the jurisdiction of EIA to be larger than the
spending jurisdiction. If benefits outside of the jurisdiction of
interest are included but not identified as lying beyond the
spending jurisdiction, it certainly would make a subsidy look
429

more valuable relative to the costs. EIA practitioners must be
ever vigilant on this issue, or they may dramatically overstate
the benefits of a subsidy.
THE USE AND ABUSE OF MULTIPLIERS IN EIA
Whenever a dollar of new income is created, it changes hands
many times, creating income along the way. The spending
multiplier shows how many times a new dollar changes
hands, generating multiple impacts on incomes and spending.
The idea of the multiplier is an important one in EIA. The
dollars enter the economy directly (e.g., at the ballpark) or
indirectly (e.g., at a hotel or restaurant). The bank balances of
firms receiving these dollars rise. However, the dollars begin
to move out of these firms’ accounts. Some dollars go to local,
state, and federal governments. Some go to business
purchases. Some go to wages. Some “leak,” or move outside of
the jurisdiction. Then, another round occurs. Governments
send the dollars to other businesses and households through
investments and transfer payments. Households send the
payments they receive into savings and local and nonlocal
purchases. The cycle then repeats.
A multiplier is an important element in the determination of
the value of a team to a city, county, state, or region, but
extreme care must be exercised in applying a multiplier. First,
a multiplier must only be applied to new value added. Our
most important guiding principle cannot be avoided. If there
is no new income, then there is no new spending, and the
multiplier cannot be applied.
Second, care must be exercised to evaluate the multiplier from
the original injection of a dollar to the end of the line. Dollars
might be observed at different points along the spending trail,
and different multipliers are appropriate at different points.
Typically, there are two types of multipliers measured in
terms of dollars: sales multipliers and household income
multipliers. The first applies to an added dollar of economic
activity, such as the dollar that comes into a hotel or ballpark.
The second applies only once the dollars have reached
household bank accounts. Because a dollar has much farther
to travel for the former, sales multipliers are larger than
household income multipliers.
Which of these multipliers is most relevant depends on your
purpose. If you want to know how sales dollars multiply, then
use the sales multiplier. If you want to know the impact on
household incomes, then use the income multiplier. It is
tempting to say that it is the latter that matters in EIA because
households will fund the subsidy and enjoy the benefits. But
that idea shows just how difficult a concept EIA really is. Most
income that would be calculated in the case of stadium
subsidies actually comes in the form of wages. But wages are a
cost of providing the stadium, not a benefit. To use the income
multiplier at the level of wages to stadium workers would be
inappropriate in the case of stadium subsidies and would
overstate the costs. (Crompton [1995] is a good source for
more on the use of multipliers in EIA.) In the best of all
possible worlds, one would have data on all sales and use the
sales multiplier only on added sales.
430

Third, multipliers depend on the complex interactions of
businesses, consumers, and governments. A multiplier that
works in one geographic location will not necessarily hold in
another, and a multiplier that works for one sector of a
defined economic area may not work for another. For
example, a retail clothing sector multiplier will not be the
same as an entertainment sector multiplier, and
entertainment industry multipliers in, for example, Seattle
and Baltimore need not be the same either.
Finally, it can take quite some time before all of the rounds
are complete in the calculation of a multiplier. This brings
time into consideration and, along with it, discounting
considerations. As future dollars are worth less than current
dollars, later dollars must be discounted. As a result, the
longer it takes a dollar to work its way through the spending
jurisdiction, the smaller the multiplier is.
The multiplier concept can be abused. Proponents of a
stadium can just use the highest multiplier that can be found
and apply it to all of the dollars in the stream. Opponents can
do the opposite. Each strategy is unsupportable and incorrect
from an accounting perspective. A correct EIA will use the
appropriate multiplier and only apply it to new economic
impacts.
WHAT ABOUT JOBS AND TAXES?
Proponents of construction subsidies stress that they also
create jobs and additional tax revenue. But this misses the
point from the perspective of evaluating the benefits and costs
of the project, itself. Just as materials must be purchased, and
equipment rented, labor must be paid. Certainly, politicians
tout the creation of jobs and those people who get the jobs are
glad they have them. But from the perspective of the project
itself, wages are a cost, not a benefit! Further, taxes do not
accrue as a project benefit but, rather, as an external benefit.
Besides, new taxes can only be generated if there is growth in
the first place. So special care must be exercised in calculating
the externality of tax revenues.
DEVELOPMENT VALUE
Some believe that cities with sports teams have higher
economic growth rates than cities that do not and that part of
that difference comes from the presence of the teams. This
development value is a highly touted form of benefit. The idea
behind development value is that cities with a pro sports team
will have higher growth rates than cities without them. In a
sense, sports teams are a “big-league city” viability indicator
for the purposes of attracting and keeping business. As former
Kansas City mayor Emanuel Cleaver once said, “Without the
Chiefs and the Royals, Kansas City would be nothing but
another Wichitaor Des Moinesor Omaha” (Sporting News,
August 11, 1997, p. 38). Note that this type of big-league city
growth would be over and above any that would occur due to
actual construction and recurring economic activity
associated directly with the team or indirectly with business
relying on fan traffic or game outcomes.
431

Consider the following example of development value.
Suppose a major firm is relocating its corporate headquarters.
Further, suppose that both Portland, Oregon, and Seattle,
Washington, have put together the same set of location
incentives for the firm in question. That is, just based on the
incentives, the firm is indifferent between the two Pacific
Northwest cities. But Seattle has the MLB Mariners and NFL
Seahawks, whereas Portland has only the NBA Trailblazers. If
there is any development value inherent in the presence of
sports teams, or the number of them, the firm should choose
Seattle over Portland because it has more major league teams.
In addressing development value, let’s be sure to remember
one of the most important general principles of benefit
analysis: Only new value matters. New development value
would be relative to the next-best development that could be
obtained. If not a sports team, then some other type of
development could be subsidized. In addition, if the subsidy
just keeps existing teams in town rather than drawing new
teams, the chance for new value is slim. Further, as noted in
an earlier section, uncertainty plays a role in subsidizing
stadiums. At the end of this section, the Learning Highlight:
Build It and They Will Come? shows examples where
subsidized facilities were built to try to entice teams but failed
to do so.
QUALITY OF LIFE: BUYERS’ SURPLUSES AND THE
REST OF THE EXTERNAL BENEFITS
The last type of benefit from a subsidy concerns the question,
How much would you pay to keep your sports team around,
over and above costs of attendance and spending on
advertised products, and over and above any income you earn
due to the presence of the team? Any answer greater than zero
means that you enjoy either buyers’ surpluses or external
benefits or both. We have already addressed buyers’ surpluses
and sports teams in the case of subsidizing a money-losing
owner. But it is important to point out that buyers’ surpluses
comprise part of the benefits received by those subsidizing pro
team owners.
We also discussed external benefits in earlier sections of this
chapter. However, we should be cautious in adding them to
our list of benefits. Indirect economic activity is a benefit
created by the presence of a sports team that owners cannot
capture. However, for the most part, an outside analyst can
find indirect economic activity just by identifying those
sectors of the economy that should enjoy it. Then, if there is
any increase in the revenues in that sector that is greater than
reductions somewhere else in the jurisdiction, those benefits
count as indirect benefits. Therefore, indirect activity can be
measured fairly simply. Again, only new benefits should be
included. Without a team, newspapers would still have sports
sections, but they would be limited to reporting the scores of
other teams. The same goes for local TV news broadcasts. The
measure of indirect economic activity is how much income
these beneficiaries would lose in the absence of a team.
432

This still leaves external benefits that cannot be added up
because no market creates any price for them. Remember that
the crucial distinguishing feature of external benefits was that
no price could be charged, and enjoyment by one did not
preclude enjoyment by any other person at the same time.
Although for many these external benefits are small, for
others they can be quite large. Summed across an entire
jurisdiction that might be considering a subsidy for the local
pro sports team owner, the total can be huge. Returning to the
case of Kansas City, columnist Jason Whitlock put it this way:
“The only self-esteem this city gets is from its sports
franchises” (Sporting News, August 11, 1997, p. 39). As a
sports writer, Whitlock has some self-interest tied up in that
claim, but I think you get the idea. We’ll explore economists’
estimates of these values later in the chapter but first let’s turn
to estimates of subsidy costs.
433

LEARNING HIGHLIGHT: BUILD IT AND THEY WILL
COME?
There is a famous line from the movie Field of Dreams: “Build
it and they will come.” Although this proved to be true in the
movie—Kevin Costner’s character built a ballpark, and players
from the past came to play in it—it does not always happen
this way in real life. A prime example is the Suncoast Dome,
completed in 1990, in St. Petersburg/Tampa Bay, Florida.
The Suncoast Dome was built for $138 million ($210 million)
with public money in order to draw an MLB team or gain an
expansion team in 1992. However, neither happened (recall
their successive disappointment with the White Sox,
Mariners, and Giants detailed in Chapter 5). The Suncoast
Dome stood empty until the NHL expansion Tampa Bay
Lightning appeared for their 1992–1993 season. The
Lightning used the stadium, renamed the Thunderdome, as a
temporary home until the Ice Palace was completed in 1996.
Eventually, in 1998, Tampa Bay did get the MLB expansion
team the Devil Rays. The Thunderdome was refurbished for
baseball-only use for another $70 million ($87 million), $62
million ($77 million) of which was publicly funded. Renamed
Tropicana Field, it opened in time for the 2000 baseball
season.
Let’s think about the costs incurred in this episode. Including
the two years to build the original Suncoast Dome, it was 10
years before MLB came to Tampa Bay. In the meantime, the
facility stood empty from 1988 to 1991, was partially occupied
from 1992 to 1995 by the NHL Lightning, and was empty
again from 1996 to 1997. Had the $138 million original
construction cost been put in the bank at 6 percent interest, it
would have grown to about $247.1 million ($306 million) by
the year 2000, when a team actually occupied the stadium.
And the additional $62 million would have grown to $69.7
million by 2000 ($86 million). Therefore, just the funding
cost of building the facility in anticipation of a team would be
$316.8 million ($392 million) in public money, minus
miscellaneous revenues (concerts and the like) and revenue
from the Tampa Bay Lightning lease (if any). The moral of this
episode appears to be that it would be better to have a team,
then promise to build it a stadium, than to not have a team
and incur empty stadium costs.
The Suncoast Dome was built in 1990 to attract major league
baseball to Tampa Bay, but it wasn’t until 1998, reborn as
Tropicana Field, that the facility hosted the MLB Tampa Bay
Devil Rays. [Photo of original SunCoast with sign.]
Source: Devil Rays Web page, www.devilrays.com.
434

http://www.devilrays.com

http://www.devilrays.com

SECTION 7
Estimates of Sports
Subsidy Benefits
IIn this section, we will review the findings on the benefits of
subsidies to pro team owners. EIAs play an important role in
determining these benefits, but as we will see, they typically
do not follow the conceptual guidelines outlined in previous
sections, and they do not cover development value, buyers’
surpluses, or external benefits. We will see this as we look at
EIAs done for subsidies in Baltimore and Seattle.
MEASURING THE BENEFIT OF SUBSIDIES IN
BALTIMORE
Table 10.6 and Table 10.7 show an EIA done in 1987 on
subsidies proposed for the Camden Yards baseball and
football complex in Baltimore. EIAs are often commissioned
by local governments; sometimes they are commissioned by
team owners. Let’s see if any insights can be gained from this
Baltimore example based on the guiding principles presented
thus far.
Note that the jurisdiction issue was unresolved at the time of
the EIA. Therefore, the analysts offered an array of possible
jurisdictions, including the city, a five-county area, and the
entire state. The issue was eventually put to a vote at the state
level, so we will pay attention to the state specifications in
Tables 10.6 and Table 10.7 for the remainder of this section.
Further, the EIA reports both impacts for the construction
phase and those that would recur annually after construction
was completed. Also, keep in mind that no other types of
values are included in either of the tables. We will examine
development value, buyers’ surpluses, and external benefits
later in this section.
The EIA is careful to derive new economic impact. Indeed,
only new value, the category “new value added,” is reported.
New value added equals new value minus wages, a cost to the
project under consideration. Unfortunately, the method of
estimating new impacts is not reported. Again, for this
discussion, all values are adjusted to 2009 dollars. From
Table 10.6 and Table 10.7, the state-level estimates of new
economic impacts from baseball and football construction
were reported at $258.2 million and $227.3 million,
respectively. Recurring new state-level economic impact, net
of wages, was estimated at $82.6 million for baseball and
$64.9 million for football.
An extremely important issue was whether the proposed
subsidy would keep current teams in town or draw new teams.
At the time, the answer was mixed. The MLB Orioles already
called Baltimore home, but the football spending would have
been used to draw a new NFL team. Baltimore lost its NFL
Colts in 1984 when owner Robert Irsay moved them to
Indianapolis (the Cleveland Browns did eventually move to
435

become the Baltimore Ravens in 1996, and a new football
stadium was completed in 1998). Thus, it would seem much
more likely that any new economic impact would be attached
to the football stadium spending rather than the baseball
stadium spending. However, the reported totals for both
construction and recurring impacts are larger for the baseball
portion of the proposed stadium subsidies.
The basic issue remains to be considered, even for the football
spending. How much actual growth can stadium spending
create? Would people in Maryland work harder, creating new
income growth, in order to enjoy these amenities? Or is it
more likely that these estimates include rearranged spending
that would have happened anyway? Fortunately, other
investigators have supplied some answers to these questions.
THE HAMILTON AND KAHN REVIEW
Hamilton and Kahn (1997) review the actual Camden Yards
outcome and suggest that it fell far short of the projected net
economic impact. The Orioles began play in their new
Ballpark at Camden Yards in 1992. Hamilton and Kahn
estimate that the state of Maryland loses about $9 million
($12 million) per year on the ballpark. The ballpark generates
enough revenue to cover capital and operations, but the team
owner keeps that revenue under the lease agreement.
Therefore, the park is a loser in dollar terms to Maryland
taxpayers. In addition, according to the authors of this
critique, the national economy loses about $12 million ($15.9
million) per year because the ballpark was financed with tax-
exempt bonds. It appears that the people of the state of
Maryland paid around $200 million (original construction
costs, 1992) ($303.9 million) in order to lose $9 million ($12
million) per year.
Hamilton and Kahn (1997) also show that a similar outcome
can be expected for the NFL Ravens and their stadium at
Camden Yards. They estimate losses of around $13 million
($17.3 million) per year. Because the Ravens were new to
Baltimore and the stadium was nearly brand new at the time
of the authors’ critique, Hamilton and Kahn conceded that
potential offsets might remain to be seen. For example, any
new non-NFL activity occurring in the stadium would offset
the losses. Losses to the national economy generated by the
subsidy for football are about the same amount as for the
Orioles’ ballpark. The expected cost of the Ravens stadium
also was around $200 million. At least in its early stages, the
Ravens stadium was no bargain, financially speaking.
Hamilton and Kahn (1997) do note that no other values are in
the original EIA such as external benefits or buyers’ surpluses.
But they also note that, at least in the case of the Orioles, there
is no reason to suspect that this type of benefit will increase
because the team gets a new stadium as opposed to playing in
its old one.
MEASURING THE BENEFIT OF SUBSIDIES IN SEATTLE
The EIA in our previous example of Camden Yards did not
separate direct and indirect economic activity. In our next
436

example, direct economic activity and indirect economic
activity are broken out for separate treatment. However, as in
the Baltimore example, neither buyers’ surpluses nor external
benefits are covered in the Seattle EIAs. Further, there is no
published review of these EIAs, so we are on our own when it
comes to a review.
Table 10.8, Table 10.9, and Table 10.10 show summaries
of the EIAs produced during subsidy debates over new
stadiums for all of Seattle’s pro teams, the MLB Mariners, the
NFL Seahawks, and the NBA Sonics. The Mariners and
Seahawks analysts, Conway and Byers (1994, 1996), provided
only recurring impacts. Further, they stress that indirect
impacts are only for businesses relying on fan traffic. This
means that the tables do not include other types of indirect
monetary inputs. The Sonics analysts, Berk and Associates
(2007), provided construction impacts as well, but they are
not shown here in order to focus on comparison across the
reports. Finally, since the point here is to discuss the
techniques used, all tabled values are left in their original
dollar terms. Conversions to 2009 dollars happen only in the
discussion and evaluation.
Unlike the Baltimore case, the jurisdiction of interest, the
state of Washington, was known at the time the EIAs were
provided for the Mariners and Seahawks. In both cases, the
EIA appeared during the debate over statewide subsidy
referenda put before Washington voters. The Mariners
subsidy failed to garner a majority of votes (the stadium was
subsidized anyway through alternative approaches), but the
Seahawks subsidy was approved by the voters. No subsidy
agreement was reached for the Sonics, and the team moved to
Oklahoma City for the 2008–2009 season. For comparison
purposes, let’s think like voters and pay attention only to the
state columns in Table 10.8, Table 10.9, and Table 10.10.
EVALUATION
Once again, our guiding principles provide insights into the
Seattle EIAs. All EIAs identify new economic activity. We can
also bring the value-added notion into play, but unlike the
Baltimore EIA, the Seattle EIAs do not calculate it for us. New
value added is found by subtracting wages from spending.
New value added at the state level for the Mariners was $42.9
million minus $25.4 million, or about $17.5 million ($23.2
million) annually. About $9 million ($12 million) of this new
value added comes directly from team and stadium operations
and supporting businesses, whereas $8.5 million ($11.3
million) comes indirectly from other businesses that rely on
fan traffic. A similar line of reasoning yields about $28.3
million ($37.6 million) in new value added at the state level by
the Seahawks. About $16 million ($21.2 million) of this new
value added comes directly from team and stadium operations
and supporting businesses, whereas $12.3 million ($16.3
million) comes from indirect business activity. The
breakdown for the Sonics is $30.1 million in new value added
with $4.6 million direct and $25.5 million indirect. In
summary, all in 2009 dollars, the Seahawks have the largest
annual new value added ($37.6 million), followed by the
437

Sonics ($30.1 million) and Mariners ($23.2 million), a total of
$90.9 million annually.
By way of comparison, in 2009 dollars, the recurring new
value added by the Camden Yards stadiums was about $56.7
million for baseball and $50.1 million for football, a total of
$116.8 million in today’s dollars. The Seattle estimates for
baseball and football are $60.8 million, or about 52 percent of
the Baltimore estimates. It does not seem unreasonable that
Baltimore teams would be more valuable to their city than
Seattle teams would be to Seattle. After all, the Orioles were
an extremely successful franchise at the time, and Baltimore
has a much longer football and baseball history than Seattle.
In addition, other demand factors probably were higher in
Baltimore.
The ultimate question always boils down to this: Regardless of
the relative value to cities, will the subsidies under
consideration add any income growth? If not, there can be no
new economic impact of any sort. Would people in the state of
Washington work harder in order to enjoy the amenities
offered by new stadiums or just rearrange spending? It seems
extremely unlikely that any of the results in Table 10.8, Table
10.9, and Table 10.10 actually pertain to new impact that
would not have occurred anyway, despite the estimates in the
EIAs.
438

SECTION 8
Are the Costs of Subsidies
Worth It?
A review of the verdicts on the various claimed values leads
naturally enough to a final conclusion on whether or not the
cost of the subsidies that occur for stadiums and arenas are
worth it or not.
THE VERDICT ON ECONOMIC ACTIVITY VALUE
The general verdict on EIAs is that they overstate the value of
economic activity. Baade and Dye (1988), the pioneers in
reviewing EIAs, devised an interesting scenario to illustrate
this point. First, they suppose that Cook County, Illinois
(where the city of Chicago is located), might contribute $100
million to a stadium subsidy. For a three-year period of
construction, the $100 million would represent about 3.11
percent of the entire Cook County capital project expenditure
budget. Even if all of commercial sports in the county were
produced in this hypothetical stadium, the result would be
about $50.4 million in personal income in the county. That
ends up to be less than 0.5 percent of the county income.
Strictly based on economic impacts, Baade and Dye
determined that it would make little sense to commit so much
of the county capital budget for so little in return.
The collection of work in Noll and Zimbalist (1997b), as well
as the survey by Siegfried and Zimbalist (2000), offers nearly
nothing in support of any positive economic activity from
such subsidies. The upshot of all the works in the former
volume is most likely that there are no new economic impacts
from sports stadium subsidies and that this is especially true
for subsidies aimed at keeping an existing team as opposed to
drawing a new team to an area. Typically, EIA claims of new
impacts are just rearrangements of activity that would have
occurred anyway. The implication is that EIAs have been
produced primarily by proponents to overstate the benefits of
subsidies. In fact, Siegfried and Zimbalist (2000) put it this
way:
Few fields of empirical economic research offer virtual
unanimity of findings. Yet, independent work on the
economic impact of stadiums and arenas has uniformly
found that there is no statistically significant positive
correlation between sports facility construction and
economic development. (p. 103)
THE VERDICT ON DEVELOPMENT VALUE
Economic activity arguments offer nearly no support for
subsidies. However, other values not examined by EIAs
should be considered when evaluating a subsidy. The first,
development value, is always touted as an important but
difficult-to-measure benefit. One way to assess development
value is to find a sample of cities that are similar except for
sports teams and see if the presence of sports teams coincides
439

with higher economic growth rates. This is exactly the
approach taken by Baade and Dye (1988). Their study found
no difference in manufacturing activity attributable to the
presence or number of major league sports teams in a city, but
there remains some debate on the issue.
Baade (1996a) offers an updated evaluation that arrives at the
same conclusion—that teams do not provide development
value. The primary beneficiaries of stadiums are the owners
and players, not the taxpayers. Chema (1996) says that
stadiums are part of an integrated growth approach for cities.
He argues that cities must create chances for interaction and
socialization or they are doomed. In his opinion, stadiums
generate these chances. Chema says we just have to look
harder to find the benefits that everybody knows sway the
decision in favor of subsidies. However, he offers no analysis
to support his thoughts.
Baade (1996b) replies that Chema’s argument must be
discounted by the fact that no difference in growth rates can
be found based on the presence of pro sports teams. Baade
does agree that additional analysis of external benefits is
necessary for a complete picture of the benefits of subsidies,
but he suggests that additional analysis probably will not
change the outcome. But as with Chema’s original point,
Baade offers no evidence. This interaction sets up a discussion
of the other values involved in subsidy decisions, namely,
buyers’ surpluses and external benefits. We’ll turn to those
values after this.
The most recent academic examination of the growth issue is
Coates and Humphreys (1999). They investigate whether
cities with pro sports teams and facilities have higher real per
capita income. Their data cover 1969–1994 for cities with
MLB, NBA, and NFL teams. Coates and Humphreys find no
evidence of higher income, and in fact, there is some evidence
that some cities with pro sports franchises actually have lower
per capita personal income. In addition, they find no effect on
the growth of income at all. Their conclusion is clear—they
find no evidence that new teams and facilities spur any
economic growth.
QUALITY OF LIFE: BUYERS’ SURPLUSES AND THE
REST OF THE EXTERNAL BENEFITS
Researchers have begun to estimate the size of buyers’
surpluses in sports. Irani (1997) analyzed buyers’ surpluses
enjoyed by fans at eight MLB stadiums. Using estimates of
demand, he calculated the net present value of buyers’
surpluses over a reasonable stadium life. The total of buyers’
surpluses over a reasonable subsidy period averaged about $9
million ($12 million), with a high of $61 million ($81 million)
at Dodger Stadium in Los Angeles. Because these are totals
over fairly long periods, the buyers’ surpluses estimated
actually are quite small.
Alexander, Kern, and Neill (2000) estimated buyers’
surpluses for teams and compared them with stadium
payments. They calculated buyers’ surpluses from published
reports of team total revenues. This makes their estimates
440

dependent upon where buyers are on the demand curve, that
is, the elasticity of demand. Their results are presented in
Table 10.11. Remember from Chapter 2 that most estimates
of the elasticity of demand are at or just below 1.0. For an
elasticity of 0.75, buyers’ surpluses are in the range of $15
million ($18.6 million) to $17 million ($21 million). These
estimates dwarf those by Irani (1997) because they are annual
estimates whereas Irani’s covered the entire length of the
project.
Turning to external benefits, Rosentraub (1996) stressed that
the study of the “psychic values” of teams and stadiums has
been neglected and that these values can be large. For
example, after development that included a new baseball
park, downtown Cleveland is now an exciting “hot spot.”
Similarly, the values behind the sentiment expressed in the
earlier quote by Mayor Cleaver of Kansas City have not been
analyzed. It could be that sport as a cultural icon is valuable
enough to sway the margin in subsidy assessments. Finally,
Rosentraub echoes Chema on one of the other benefits. Sport
has value as coalition glue; without sports in the mix, it can be
difficult to get any agreement on any public spending plan.
Of course, Rosentraub’s “psychic benefits” are actually the
external benefits discussed and analyzed early in this chapter.
As with buyers’ surpluses, economists have also begun to
estimate the external benefits associated with sports output.
Johnson, Groothuis, and Whitehead (2001) estimated
external benefits generated by the Pittsburgh Penguins using
consumer valuation surveys. Their upper-bound estimate of
external benefits for just one NHL team equals around $3.9
million per year ($4.7 million).
Carlino and Coulson (2004) approach quality of life values by
examining the difference in rental prices and wages in cities
with NFL teams compared to similar cities that do not have
teams, circa 1999. On the one hand, if quality of life is higher,
then people should be willing to pay higher rents to enjoy it.
On the other hand, people should be willing to accept
relatively lower wages to live in a place with higher quality of
life. Carlino and Coulson’s strongest evidence concerns
central cities where rental prices were 8 percent higher in
NFL cities (they find only very weak evidence of a much
smaller impact on wages). For an average rent of around $500
per month ($640), an additional 8 percent added about $480
($614) in additional value per year per renter. The level of this
rental difference applied to the average size of the renting
population in central cities yielded a possible amount near
$139 million ($178 million) annually. Carlino and Coulson
compared this amount to the subsidies paid on stadiums that
Quirk and I found earlier, and they conclude that stadiums
are easily worth it to city residents. Indeed, comparing their
finding to Long’s more extensive treatment would produce the
same conclusion—$178 million in added rental value swamps
the average annual subsidy.
On a related note regarding wages, Coates and Humphreys
(1999, cited earlier) did find that per capita income was
actually lower in some pro sports cities. Carlino and Coulson
did not find this result, but Coates and Humphreys suggest
441

that there are alternative explanations for lower wages. One is
the Carlino and Coulson argument that quality of life is
higher, so people take lower wages to enjoy it. But another
explanation is that sports subsidies divert taxes from other
spending so that tax bills in pro sports cities must be higher in
order to cover the usual public offerings. In this case, the
anticipation of lower net wages would reduce the supply of
labor in pro sports cities.
All in all, the analysis of external benefits is in its infancy, and
many questions remain. For example, while denizens of the
city’s central core may benefit, governments typically collect
from a much broader population base (the collections issue
raised earlier in this chapter). Because Coulson and Carlino
also find that the rental value does not extend to these other
payers, their conclusion is tied up with transfers from other
residents to central city residents. In addition, work in the
area has yet to test the competing explanations for lower
wages and income in pro sports cities. Finally, none of the
work in the area addresses the basic issue of just why public
funds should be spent to raise private rental values in the first
place. Clearly, more work on external benefits is needed, but
the evidence is mounting that this type of value is real and
should be counted in the assessment of public subsidies to
sports team owners.
A NOTE ON EVENTS
Our same guidelines remain in operation for individual
events. Just hearing how politicians describe pro sports
events should clue us in to that idea. In reference to Super
Bowl XXXVIII, Texas comptroller Carole Keeton Strayhorn
said, “Texas will be a winner, no matter who wins the big
game” (Window on State Government, news release from the
Texas comptroller, Friday, October 31, 2003). The values to
the state of Texas of Super Bowl XXXVIII played at Houston’s
Reliant Stadium in 2004 were estimated as follows by the
comptroller:
• 87,700 visitors from outside the state (85 percent of total
visitors)
• At least $8.7 million ($9.8 million) increased state tax
revenue
• $165.5 million ($186.8) from all visitors
• 5,637, full-time jobs for the year
• $336 million ($379 million) total economic impact
This litany of benefits is familiar, and the same careful EIA
approach must be exercised in assessing the value of events.
Economist Philip Porter (1999), in his in-depth study of the
value of Super Bowls, provides some shocking revelations
about the absence of Super Bowl value to host cities, as you
can see at the end of this chapter in the Learning Highlight:
Super Bowl? Super Dud!
ARE THE COSTS OF SUBSIDIES WORTH IT?
442

Returning to Table 10.1, subsidized projects are just under
$300 million range for arenas and right around $500 million
for stadiums and ballparks. The most recent and
comprehensive treatment on subsidies by Long (2005)
suggests annual subsidies around $19 million in 2009 dollars.
Are the benefits created by these projects—economic activity
value, development value, and buyers’ surpluses and external
benefits—worth these costs?
In the previous sections, we found that direct and indirect
economic activity and development value alone do not provide
a solid argument for subsidizing stadiums. Almost all
measured value is simply value rearranged from other
sources, and analysts can find no development value. The
most recent estimates of buyers’ surpluses range from $18.6
million to $21 million per year in 2009 dollars. Finally,
external benefits for a hockey team have been estimated at
$4.7 million 2009 dollars per year, whereas NFL teams could
raise some rental rates in the central core of cities by as much
as $178 million 2009 dollars annually. Even at the lower
levels of the estimates, buyers’ surpluses and external benefits
appear to be about equal to the average subsidy. However,
external costs should also be subtracted in this assessment.
Unfortunately, estimates of external costs are not available in
the literature to date.
However, we must be extremely careful in our evaluation.
First, only a few studies have even examined the benefits or
the costs of subsidies and external costs have received no
formal treatment at all. Second, subsidy amounts vary
considerably and across locations, so comparisons at the
average will be misleading in in any particular case analyzed.
Clearly, more analysis is needed before any definitive answer
to the subsidy question can be obtained. Third, just because
we can identify subsidy benefits does not mean that collection
of funds for subsidies will be easy or fair. And, finally, there
will always be vigorous argument over whether government
should be undertaking actions only because they generate
benefits for some people in excess of the costs to others.
443

LEARNING HIGHLIGHT: SUPER BOWL? SUPER DUD!
Economist Philip Porter has shown that Super Bowls are
actually super duds when it comes to direct and indirect
economic activity (1999). Income and taxes simply do not
materialize as promised, even though they are used repeatedly
as justifications for city–county subsidies in the pursuit of
megaevents such as the Super Bowl, NCAA play-offs, or even
the Olympic Games. An NFL study found that the value of the
1995 Super Bowl to Miami was $365 million. Porter found
conceptual errors in that analysis and notes that the projected
massive activity from such events simply does not occur.
When there are perfect complements to events, such as hotel
rooms, that have capacity constraints, there is nearly no net
increase in local incomes if local suppliers raise prices in the
face of dramatic demand increases. If hotels already are at
capacity, prices rise but the number of sales cannot. Prices
rise for local purchasers of goods and services, as well. The net
activity is certainly much smaller than touted by the NFL. The
gains to hotel owners are lost through higher prices paid by
other local residents. Apparently, the NFL missed Porter’s
analysis; their commissioned study estimated the expected
benefits for the immediately following Miami Super Bowl in
1999 at $396 million ($507 million) in total economic impact,
an additional $31 million ($40 million) in just four years
rather than the downward adjustment suggested by Porter.
Another unexpected reduction in hotel-related spending may
also occur with events. When Tampa, Florida, got the 2002
Super Bowl, 35,000 rooms were immediately reserved.
Proprietors required a minimum seven-day reservation for
Super Bowl weekend. This was great for hotel owners because
they enjoyed greater paid-occupancy rates. But their guests
actually only stayed for the weekend, crowding out a few days’
worth of other tourists who would have occupied the rooms
and spent over the entire week. Occupancy at hotels was high,
but spending was lower than it otherwise would have been in
the absence of the Super Bowl. But true to form, the NFL had
already estimated that Tampa Bay would enjoy $250 million
($296 million) in total economic impact hosting that Super
Bowl. More recently, the NFL clearly has continued to turn a
deaf ear. A study jointly funded with the San Diego Super
Bowl Host Committee claims that the 2003 Super Bowl
generated $367 million ($425 million) in total economic
impact, up from the city’s last Super Bowl in 1998 by $72
million ($94 million).
Recently, the National Association of Sports Commissions
provided guidelines for uniform study of these benefits.
Essentially, they suggest ascertaining the number of visitors
multiplied by their spending, plus administrative and
operations spending. Then, a reasonable multiplier can be
used for total impact. But this clearly misses the point: It is
“new” spending that matters, not “total” spending. As Porter
so aptly puts it, what matters is how much more an event
brings to a location over and above what already would have
occurred anyway. In the case of sports megaevents, other
444

activity-generating alternatives would have occurred in their
absence.
Source: Philip Porter, “Mega-Sports Events as Municipal
Investments: A Critique of Impact Analysis,” in Sports
Economics: Current Research, John Fizel, Elizabeth
Gustafson, and Laurence Hadley, eds. (Westport, CT: Praeger,
1999); Sports Business Journal, January 15, 2001; NFL.com
(NFL News), May 13, 2003.
445

SECTION 9
Chapter Recap
It is possible to justify subsidies that benefit team owners and
fans. The presence of external benefits is one justification for
sports team subsidies. When teams produce benefits to
nonpayers, team outputs are lower than society would prefer.
Subsidies to sports team owners can lead owners to produce
greater attendance and TV output until fans receive the level
of quality and attendance for which they are willing to pay. A
second justification occurs when a sports team owner has a
money-losing team. In such a case, if buyers’ surpluses are
large enough, a subsidy that keeps the team in business is
better for fans than losing the team.
Just because subsidies provide the benefits does not mean
that any subsidy size is justified. Cost–benefit analysis is
crucial in assessing the size of a subsidy. However, assessment
of benefits and costs is a complex problem. Costs are in the
form of infrastructure subsidies, operating subsidies, and
stadium construction subsidies. Some costs can be estimated,
whereas others can only be identified. For example, external
costs have not been estimated to date. Components affected
by time require discounting. Although dollar costs can be
obvious, opportunity costs require deeper inspection. Finally,
disinformation for political purposes complicates the
determination of subsidy costs.
Care must also be taken in evaluating subsidy benefits.
Economic activity benefits (direct and indirect) occur during
construction and after the team has moved into a facility. Only
new economic activity counts in the analysis of benefits.
Boundaries for the analysis of benefits must coincide with the
spending jurisdiction. The correct spending multiplier should
also be used. Development values, buyers’ surpluses, and
external benefits should also be considered when assessing
subsidy benefits.
The cost of a subsidy is approximately the sum of
depreciation, opportunity cost of funds, and forgone taxes,
minus the net operating revenue from the stadium that
actually goes to the state or local government. Subsidies at the
state and local level are around $19 million 2009 dollars
annually on publicly owned facilities, plus a small amount
more in federal subsidy if the facility is funded through tax-
exempt bonds.
The Baltimore Camden Yards and Seattle Mariners and
Seahawks subsidy examples make it clear that the economic
impacts of subsidies are close to zero. Nearly all of the
economic activity from the sport is just rearranged from other
spending that would have occurred anyway. The bulk of
economic analysis supports this conclusion. In addition, so-
called development values do not appear to exist, despite their
intuitive appeal. This leaves buyers’ surpluses and external
446

benefits to make up the bulk of the subsidy. Few studies have
examined these last two types of benefits, but buyers’
surpluses may be as high as $21 million per year in 2009
dollars, and external benefits have been estimated for hockey
at about $4.7 million per year, while annual rental rates in
cities with NFL teams, representing another take on quality-
of-life values, may be as high as $178 million annually. The
small number of studies, their limitations regarding external
costs, and the variation in their outcomes across teams and
locations make it difficult to pass judgment on whether sports
subsidies are worth it.
447

SECTION 10
Key Terms and Concepts
You should have run into each of these in pop-ups in the text
of this chapter:
• External benefit
• External benefit inefficiency
• Minimum subsidy
• Cost–benefit analysis
• Infrastructure subsidies
• Operating subsidies
• Stadium subsidies
• Annual subsidy formula
• Direct economic activity
• Indirect economic activity
• Economic impact analysis
• New economic activity
• Spending multiplier
• Development value
• Federal subsidy component
• New value added
448

SECTION 11
Review Questions
1. Define external benefit. What distinguishes the two types
of external benefits that sports teams produce? Give
examples of each.
2. Describe external benefit inefficiency. How can it be
remedied?
3. Define the minimum subsidy. Remember to include both
external benefits and money losses to owners in your
definition.
4. What is cost–benefit analysis? Why is it a valuable tool?
5. Can cost–benefit analysis tell us how much to spend in
subsidies to sports teams? Explain.
6. What are the three main types of sports subsidies? Give
an example of each.
7. Define the following terms: stadium rent, concessions,
parking, signage, and nonsport revenue?
8. What is the difference between a revenue bond and a
general obligation bond? Which is most commonly used
to finance sports stadium subsidies? Why?
9. State the annual subsidy formula. Carefully describe each
element of the formula.
10. Define economic activity benefits. What is the distinction
between direct and indirect economic activity benefits?
Give examples of each.
11. What is economic impact analysis (EIA)?
12. What is the difference between total economic activity
and new economic activity? Which is more important in a
cost–benefit analysis? Why?
13. What is a spending multiplier? What is the difference
between a sales multiplier and an income multiplier?
Should they be applied to total or new economic activity?
Why or why not?
14. What is meant by development value? Is there a
distinction between “total” and “new” for this type of
subsidy benefit? Explain.
15. What is new value added? How would you calculate it?
449

SECTION 12
Thought Problems
1. Is it efficient to subsidize team owners for the benefit of a
small proportion of taxpayers who are sports fans? Is it
fair?
2. Explain how a new stadium might enhance the
competitiveness of a particular team. Be very careful to
state the particulars of the situation where this can
happen.
3. Graphically demonstrate the inefficiency that occurs
when sports teams produce external benefits. How will a
subsidy remedy this problem?
4. In Chapter 2, we downplayed the idea that owners lose
money. Can owners actually lose money? If so, can it
make sense to subsidize a money-losing owner?
Graphically demonstrate this situation.
5. If a subsidy is justified on external benefit grounds, is any
size subsidy justified? Discuss the size of a subsidy in
terms of the definition of a minimum subsidy. Would you
expect owners to try to obtain larger than minimum
subsidies? Why? Why might they be successful?
6. List the problems that confront state and local
governments if they are to collect portions of buyers’
surpluses and external benefits to provide subsidies to
team owners.
7. Under general obligation bonds, hotel and rental car taxes
are sometimes used to generate revenue for the bond
payments. With this method, who pays for the subsidy? Is
this fair? Why are these types of taxes politically
attractive?
8. Why doesn’t an estimate of the original construction costs
give a full accounting of the costs of a stadium subsidy?
What other subsidy elements does this original
construction cost miss?
9. Why do we resort to the annual subsidy formula in
describing the costs of stadium subsidies? What is gained
by using this formula? What is lost?
10. Do you think the economic activity values given in Table
10.5 for Atlanta refer to total economic activity or new
economic activity? Why?
11. Under what circumstances would the movement of a
restaurant from a more distant location to a location close
to a new stadium be new economic activity? (Hint: There
are two explanations.)
450

12. Why is it less likely that new economic activity will
accompany a subsidy to keep a team that already exists
compared with a subsidy used to lure a brand new team?
13. Why aren’t wages a subsidy benefit?
14. Compare and contrast the two EIA examples for
Baltimore and Seattle, for baseball and football (e.g.,
which values are included in each, how were they
determined). Which is more believable? Why? Does either
prove conclusively that there are large economic activity
benefits? The critique by Hamilton and Kahn (1997) and
the evaluation for Seattle suggest these benefits are very
low. Does that mean the subsidies were not worth it?
15. Summarize the general verdict on economic activity
values from sports owner subsidies. Do the same for
development value. Why is the final evaluation of the net
value of sports owner subsidies given in the text a
qualified one? What additional work needs to be done to
reach a more definitive conclusion on the value of
subsidies?
451

SECTION 13
Advanced Problems
1. How must rent and shared revenue be structured in a
sports team lease in order to take care of external benefit
problems? Are rents that increase with attendance
consistent with this setup? How else should subsidies be
structured to overcome external benefit problems?
2. Refer to the Learning Highlight: Build It and They Will
Come? What was the cost of the stadium construction
subsidy to the citizens of the Tampa Bay–St. Petersburg
area? Show how the cost was calculated. Will the Devil
Rays provide enough benefits to make up for this
subsidy?
3. Why do you think that the average subsidy has grown
threefold over the period from early estimates (the early
1970s) to the more recent estimates (late 1980s)
discussed in the text? (Remember, the threefold
statement already includes inflation.)
4. The Sports Business Journal (November 6, 2000, p. 15)
reports that the New York Times stepped up its usual
World Series coverage by 12 to 15 percent for the subway
series. The Daily News had its largest Sunday edition ever
for game one of the series and printed 1 million copies
daily (from Saturday to the end of the series), when its
regular subscriptions are only 730,000. What type of
benefits are these? What were they worth to the Daily
News? Relate your answer to arguments for sports
subsidies.
5. Read the buyers’ surpluses analyses of Irani (1997) and of
Alexander, Kern, and Neill (2000). What were their
assumptions? What were their statistical techniques? If
they used surveys, how did they treat nonrespondents?
Does this treatment of nonrespondents influence their
results? Why do you think the average discovered by
Alexander, Kern, and Neill (2000) is so much larger than
the average discovered by Irani (1997)?
6. Read the external benefit analysis of Johnson, Groothuis,
and Whitehead (2001). What were their assumptions?
What were their statistical techniques? If they used
surveys, how did they treat nonrespondents? Does this
treatment of nonrespondents influence their results?
What are the limitations of an analysis of one team in one
league?
7. Write a brief response to the remarks by Texas
comptroller Carole Keeton Strayhorn in the section on
events. Be sure to use the ideas presented in the Learning
Highlight: Super Bowl? Super Dud!
8. According to the Sports Business Journal (January 15,
2001, pp. 21–22), the National Association of Sports
452

Commissions (NASC) estimated that the total economic
activity of the Indianapolis 500 in the year 2000 was
$336.6 million. Total economic impact was calculated as
follows:
V = Number of out-of-town visitors
S = Average spending per day
D = Length of visit in days
TVS = Total visitor spending = V × S × D
TAS = Total administrative spending to make an event
possible
TDS = Total direct spending = TVS + TAS
M = Regional spending multiplier
TEA = Total economic activity = TDS × M
Suppose you are trying to figure out how much to
subsidize this historic race. Describe all of the possible
problems with using the NASC estimate to arrive at the
value of a subsidy.
9. Which pro sport would produce the most positive
economic impact for a community? Why?
10. The city of New Orleans recently renegotiated the
Superdome lease of the New Orleans Saints (Sports
Business Journal, July 16, 2001, p. 8). Prior to the
renegotiation, the Saints were 25th in local revenue in
1999 ($35.2 million from tickets, special seat boxes, and
concessions) and were expected to be last by 2003. Under
the 2000 renegotiation, an additional $12.5 million goes
to the team owner over the 2001 and 2002 seasons. The
owner kept all concession revenue (previously, 40
percent), and occupancy was rent free (previously, rent
was 5 percent of gross ticket revenue). If naming rights to
the Superdome were sold, the Saints received all the
revenue. In return, the owner agreed to stay in New
Orleans for two more years, while negotiations for a new
football stadium were underway. If those negotiations
failed, the team owner could buy out of the lease early at a
greatly reduced amount. Does this renegotiation help
overcome external benefit problems associated with the
Saints? Does it fit the Siegfried and Zimbalist (2000)
itemized list of reasons to give a subsidy rather than a
cash payment to an owner? How?
453

SECTION 14
References
Alexander, Donald L., William Kern, and Jon Neill. “Valuing
the Consumption Benefits from Sports Franchises and
Facilities,” Journal of Urban Economics 48 (2000): 321–337.
Baade, Robert. “Professional Sports as Catalysts for
Metropolitan Economic Development,” Journal of Urban
Affairs 18 (1996a): 1–17.
Baade, Robert. “Stadium Subsidies Make Little Economic
Sense for Cities, a Rejoinder,” Journal of Urban Affairs 18
(1996b): 33–37.
Baade, Robert, and Richard Dye. “An Analysis of the
Economic Rationale for Public Subsidization of Sports
Stadiums,” Annals of Regional Science 22 (1988): 37–47.
Berk and Associates. “An Economic & Fiscal Benefits
Assessment of the Proposed King County Events Center,”
Seattle, WA. April 4, 2007.
Carlino, Gerald, and N. Edward Coulson. “Compensating
Differentials and the Social Benefits of the NFL,” Journal of
Urban Economics 56 (2004): 25–50.
Chema, Thomas. “When Professional Sports Justify the
Subsidy, a Reply to Robert A. Baade,” Journal of Urban
Affairs 18 (1996): 19–22.
Coates, Dennis, and Brad R. Humphreys. “The Growth Effects
of Sports Franchises, Stadia, and Arenas,” Journal of Policy
Analysis and Management 18 (1999): 601–624.
Conway, Richard, Jr., and William B. Byers. Seattle Mariners
Baseball Club Economic Impact. Dick Conway and Associates
and Department of Geography, University of Washington,
Seattle, WA, August 1994.
Conway, Richard, Jr., and William B. Byers. Seattle Seahawks
Economic Impact. Dick Conway and Associates and
Department of Geography, University of Washington, Seattle,
WA, March 1996.
Crompton, John L. “Economic Impact Analysis of Sports
Facilities and Events: Eleven Sources of Misapplication,”
Journal of Sports Management 9 (January 1995): 14–35.
Hamilton, Bruce, and Peter Kahn. “Baltimore’s Camden Yards
Ballparks,” in Sports, Jobs, and Taxes: The Economic Impact
of Sports Teams and Stadiums. Roger Noll and Andrew
Zimbalist, eds. Washington, D.C.: Brookings Institution, 1997.
Irani, Daraius. “Public Subsidies to Stadiums: Do the Costs
Outweigh the Benefits?” Public Finance Review 25 (1997):
238–253.
454

Johnson, Bruce K., Peter A. Groothuis, and John C.
Whitehead. “The Value of Public Goods Generated by a Major
League Sports Team: The CVM Approach,” Journal of Sports
Economics 2 (2001): 6–21.
Keating, Raymond. “Sports Park: The Costly Relationship
Between Major League Sports and Government,” Policy
Analysis No. 339. Washington, D.C.: The Cato Institute, April
5, 1999.
Long, Judith G. “Full Count: The Real Cost of Public Funding
for Major League Sports Facilities,” Journal of Sports
Economics 6 (2005): 119–143.
Miller, James E. The Baseball Business: Pursuing Pennants
and Profits in Baltimore. Chapel Hill, NC: University of North
Carolina Press, 1990.
National Sports Law Institute. “Sports Facility Reports,
Volume 9,” Marquette University, 2009. URL:
law.marquette.edu.
Noll, Roger, and Andrew Zimbalist. “Build the Stadium—
Create the Jobs!” in Sports, Jobs, and Taxes: The Economic
Impact of Sports Teams and Stadiums. Roger Noll and
Andrew Zimbalist, eds. Washington, D.C.: Brookings
Institution, 1997a.
Noll, Roger, and Andrew Zimbalist, eds. Sports, Jobs, and
Taxes: The Economic Impact of Sports Teams and Stadiums.
Washington, D.C.: Brookings Institution, 1997b.
Okner, Benjamin. “Subsidies of Stadiums and Arenas,” in
Government and the Sports Business. Roger Noll, ed.
Washington, D.C.: Brookings Institution, 1974.
Porter, Philip. “Mega-Sports Events as Municipal
Investments: A Critique of Impact Analysis,” in Sports
Economics: Current Research. John Fizel, Elizabeth
Gustafson, and Laurence Hadley, eds. Westport, CT: Praeger,
1999.
Quirk, James, and Rodney D. Fort. Pay Dirt: The Business of
Professional Team Sports. Princeton, NJ: Princeton
University Press, 1992.
Regional Science Research Institute. Camden Yards Studies.
Philadelphia, PA, 1987.
Rosentraub, Mark. “Does the Emperor Have New Clothes? A
Reply to Robert J. Baade,” Journal of Urban Affairs 18 (1996):
23–31.
Siegfried, John, and Andrew Zimbalist. “The Economics of
Sports Facilities and Their Communities,” Journal of
Economic Perspectives 14 (2000): 95–114.
Sports Leases on CD-ROM, Major League Facilities. Vols. 1 &
2. Chicago, IL: Team Marketing Reports, Inc., 1998.
Zimmerman, Dennis. “Subsidizing Stadiums: Who Benefits,
Who Pays?” in Sports, Jobs, and Taxes: The Economic Impact
of Sports Teams and Stadiums. Roger Noll and Andrew
Zimbalist, eds. Washington, D.C.: Brookings Institution, 1997.
455

SECTION 15
Suggestions for Further
Reading
Surely, I can come up with some.
456

CHAPTER 11
Government and Sports
Facility Outcomes
USA Today: The excitement from that playoff
run (in 1997), and the excitement you provided
as a young superstar, saved baseball in
Seattle. Do you feel in some way that Safeco
Field is the House That Junior Built?
Ken Griffey, Jr.: No. The people of Seattle built
it. They’re the ones who went out and said
“yes” to keep this ballclub here. The 25 guys on
the field helped, but it was eventually the
people of Seattle who said they wanted the
ballpark and wanted the team to stay. They
could have said “no” and we would have ended
up somewhere else.
www.usatoday.com, July 15, 1999.
CHAPTER OBJECTIVES
After reading this chapter, you should be able to:
• Understand the basic concentrated benefits and
dispersed costs prediction of the rational actor
explanation of representative democracy
outcomes.
• Explain how league control of team location gives
team owners the upper hand in facility
negotiations.
• Argue that the concentrated benefits and
dispersed costs explanation depicts the typical
facility subsidy outcome.
• Explain that direct democracy through referenda
and initiatives has not curbed the excesses of
facility outcomes because it has both theoretical
and practical limitations.
• Understand why facility outcomes can be
realistically reversed only if consumer interest
groups fight the current winners in the facility
subsidy process.

http://www.usatoday.com

http://www.usatoday.com

SECTION 1
Introduction
Sometimes it is simply amazing how history is rewritten by
sentimentality and selective memory. The opening quote is a
great example of this. At the time Ken Griffey, Jr., made his
remarks, he was moving from the Seattle Mariners to the
Cincinnati Reds. Apparently, his memory was short and a
little bit fuzzy. It is true that there was a stadium vote in King
County (not just in the city of Seattle) for a new ballpark for
the Mariners, but the people did not say yes. The referendum
failed by the narrowest of margins but fail it did. The people
actually said no.
However, despite this defeat, the Mariners did get a new
ballpark. The Washington state legislature developed a
funding package for the ballpark that raised the price to
taxpayers and spread the costs out over many more of them
than was specified in the original referendum. If anything, the
will of the people was not obeyed. Voters voted against the
ballpark, but they got one anyway. Astonishingly, it came at a
greater cost than they refused in the referendum. Let’s hope
Griffey remembered his manager’s signals better than he
remembered the ballpark voting episode in Washington state.
In this chapter, we develop an explanation for why so many
sports facilities are built and at such extravagant public cost.
The important role of owners and leagues in sports facility
outcomes receives particular attention. In addition, the role of
direct democracy (referendum and initiative votes) is
presented. As we will see, theoretically, weaknesses in direct
democracy reduce its ability to the excesses of sports facility
outcomes. Examples of MLB and NFL stadium financing in
Seattle are used to illustrate the points made throughout the
chapter. Finally, because the outcome is political, we will see
how an alteration in the power of interest groups is the only
realistic hope of altering these outcomes.
458

SECTION 2
Rational Actor
Explanations in Politics:
Insights
As we saw in Chapter 10, sports team subsidies are large and
can be justified in only a few cases. The highly touted
increased economic activity and development value are not
large enough to justify the large pervasive subsidies we
observe. The most important subsidy to sports teams is the
provision of public stadiums. In this section, we provide the
general framework used by economists and other analysts to
explain how stadium subsidies come about.
Social scientists have been examining the logic of collective
action since the original works of Nobel Prize winner James
Buchanan, his colleague Gordon Tullock, and Mancur Olson
in the mid-1950s and early 1960s. We used a version of the
logic of collective action when we analyzed union governance
in Chapter 9. The rational actor explanation is their
description of political outcomes.
The rational actor explanation assumes that people pursue
their own self-interest in politics just as they do in private
markets. Voters look to obtain benefits that they either cannot
get from markets or can get more of from the political process.
Politicians, for whatever reason, enjoy being politicians and
pursue reelection in the political choices and legislation that
they propose and support. The outcome of an election is
determined by the relative political potency of groups. Those
groups that provide the most effective political support for
politicians tend to gain from the political process. The rational
actor explanation makes a very special prediction: In the
political process, for any particular special interest under
consideration, benefits will tend to go to the politically
powerful groups, while costs are spread over those without
power. In essence, if some group is not fighting for you on a
particular issue, then you will tend to pay rather than receive.
SOURCES OF DISSATISFACTION WITH MARKET
OUTCOMES: TURNING TO GOVERNMENT FOR
REMEDIES
It is important to understand that government and markets
are two different methods of satisfying people’s desires.
Before we turn to an explanation of government outcomes,
let’s recognize why groups turn from markets to governments
in the first place. One reason is to remedy market failures. In
sports, the failures are market power and external benefits.
More generally, hospitals, electric utilities, and even tree
surgeons are regulated for quality and market power in the
name of consumer protection. Interestingly, despite their
obvious positions of market power, sports leagues remain
unregulated (see Chapter 12 for why this is the case).
459

The external benefit problem and how it relates to team
production was presented in Chapter 10. The argument is that
sports teams create value to fans that team owners cannot
capture. As a result, attendance (in the short run) and quality
(in the long run) are lower than they would be if owners could
capture more of the value they create. These benefits are
small, but that does not prevent owners from trying to make
the most of them in arguments for stadium subsidies.
Finally, government has the power to redistribute wealth from
some people to others. Sometimes such wealth redistribution
occurs out of a sense of fairness. For example, all taxpayers
help pay for programs that transfer purchasing power to those
defined as poor. However, sometimes redistribution occurs
just because a group has the power to get more of what it
seeks through political approaches than through market
approaches. As we will see, this last type of demand for
redistribution can help explain the behavior of team owners
as they pursue stadium subsidies.
RATIONAL ACTOR POLITICS: AN OVERVIEW
Rational actor politics predicts that powerful groups win in
the political process. We can see how they get the upper hand
by thinking about political participation, in general, using the
following identity (note that everything cancels so that the left
and right sides of the equation are identical):

Vm is the number of people casting votes on a ballot measure
identified by the subscript m. The measure could be a ballot
for candidates to elective office, or it could be a direct
democracy ballot (referendum or initiative) for a stadium
subsidy. Note that there are E members in the eligible voting
population. Clearly, then, the only time that the ratio on the
left side of the identity equals one is when every member of
the eligible voting population actually casts a vote on the
particular measure. However, universal voting is far from the
norm. If enough people do not vote on a particular issue or for
a particular elected official, then the outcomes may favor
special interests. We will examine how this occurs.
COSTS OF REGISTRATION
On the right-hand side of our identity, R voters register out of
the eligible population, E. This first ratio on the right side is
less than one for at least a couple of reasons. Some are
“conscientious objectors” who intend to never use their vote.
These protesters would find no reason to ever register. In
addition, voter registration is not free. It requires an eligible
person to find a place to register, incur costs in getting there,
and fill out paperwork. Given the costs, some people do not
feel strongly enough about participation to register. Those
interested in raising the registration rate often attack the
problem by trying to lower the costs of voter registration. For
example, so-called motor voter registration allows eligible
individuals to register when they pay their vehicle license fees
or renew their driver’s license.
460

COSTS OF TURNING OUT TO VOTE
The second ratio on the right side of the identity indicates that
B of the R registered voters actually make it to the booth. This
is typically called turnout. Again, this ratio is also less than
one. Even after a person has registered, the actual act of
getting to the booth has a cost. The cost may include time
away from work and fees for transportation and childcare.
These costs may cause participation to wane. In some states,
to overcome this, all ballots (not just absentee ballots) are
actually mailed with return postage, and mail-in voting across
entire counties is becoming common.
FALL OFF, RATIONAL IGNORANCE, AND FREE RIDING
Finally, only Vm votes are cast out of the B voters that actually
turn out. This phenomenon is called fall off. One of the most
compelling reasons for fall off in the booth is the idea of
rational ignorance discussed in Chapter 9. Because
information is expensive, it is rational for individuals to be
ignorant about most things. Typically, individual voters
overcome these costs by being informed only concerning
those few issues about which they most care. The cost of
becoming broadly informed is just too high. This is one of the
reasons that voters typically receive a state-sponsored voter’s
pamphlet with balanced descriptions of the issues on the
ballot.
Rational ignorance is the most compelling explanation of the
fall off ratio on the far-right side of our voting identity. Why
else would people fail to cast a vote on some particular issue
once they have overcome all of the other costs? The only
answer is that they have not taken the time to become
informed on all of the issues and are unwilling to make an
uninformed vote. Rational ignorance also plays a role in the
other ratios on the right side. If voters remain rationally
ignorant, they may never see the point in registering or
turning out in the first place.
Another factor affecting the right side of the identity was also
discussed in Chapter 9, namely, free riding. For example, if
other voters are informed, and many of them are like me, then
I might believe that the result will be the same regardless of
whether or not I vote. The idea that follows this reasoning is
that if there are others just like me and I think enough of them
will vote, then why should I vote? This free riding contributes
to fall off.
THE RATIONAL LEVEL OF VOTING AND THE
BEHAVIOR OF ELECTED OFFICIALS
The rational level of voting is the total number of votes on any
given issue or politician after people fail to register, do not
turn out, and fall off on the given vote. Thus, the rational level
of voting is lower than the total number of eligible voters.
Here is the kicker: The more specialized the voting measure
becomes, the lower the expected turnout and the number of
votes cast on the particular measure. Indeed, it may well be
that the rational level of voting has a distinct minority of the
eligible voting population deciding many of the issues on a
given ballot.
461

Perhaps the most important observation by rational actor
theorists is that politicians understand the rational voting
behavior of their constituents and, consequently, make
choices based on this knowledge. Politicians seek to satisfy
groups of constituents that play a prominent role in
reelection. If Vm is a reelection vote for them, politicians
know that the majority of their constituents never cast votes
in the first place. Thus, whenever possible, in the bundle of
choices politicians must make, many of the decisions will
provide benefits to politically potent constituent groups and
spread the costs out in small amounts onto others. This is the
concentrated benefit and dispersed costs outcome we saw in
Chapter 9 at work in unions, only here it occurs in the larger
context of state and local government special interest issues.
As those bearing the costs are, for the most part, rationally
ignorant, they will never figure out what hit them, and even if
they do, the per capita costs will be small. Those bearing the
costs are unlikely to raise a politically potent response to
special interest outcomes. Special interests, on the other
hand, bring block votes from potent groups and campaign
contributions from lobbyists that politicians require in order
to gain reelection. Sports leagues have learned the lobbying
lesson as well as any participant in the political process, as
shown in the Learning Highlight: MLB Lobbying Efforts, at
the end of this section.
A DIAGRAMMATIC MODEL OF RATIONAL ACTOR
POLITICAL OUTCOMES
A useful diagram of rational actor political outcomes is shown
in Figure 11.1, which presents a generalized version of the
triangle describing union governance we saw in Chapter 9.
The reelection constituency is at the top of the triangle.
Members of this group are politically potent potential
suppliers of votes and campaign contributions to elected
officials. Elected officials make policy choices in a given policy
area to keep their reelection constituency happy. That
constituency then evaluates the outcome and decides whether
to reelect the responsible elected officials. If the constituency
is pleased, elected officials retain their seats. If they are not, a
pool of potential elected officials always is waiting in the
wings to give politically potent groups what they want. The
general constituency sits by the sidelines, bearing the
dispersed costs of policy choices because the members are
rationally ignorant and do not participate in this particular
policy area. Figure 11.1 suggests that we look closely at
politically potent participants if we want to explain stadium
subsidy outcomes.
462

LEARNING HIGHLIGHT: MLB LOBBYING EFFORTS
Lobbying is one way by which politically potent groups
influence the political process. That sports leagues lobby as
heavily as many larger industry interests may come as a
surprise. Indeed, some sports leagues, including MLB, have
full-time offices staffed in Washington, D.C., devoted to
lobbying efforts. In 2001, MLB was the first pro sports league
to form its own political action committee. In 2008, MLB
spent $840,000 lobbying Congress on issues including
alcohol and drug abuse, gambling, immigration, taxes,
copyright, and radio and TV (Center for Representative
Politics, www.opensecrets.org). The MLBPA spent $205,000.
The NFL spent $1 billion, and the NFLPA spent another
$100,000. The list of issues is similar but also includes real
estate. Other leagues and unions spent less, but spend they
did. The NBA spent $227,000, and the NHL spent $50,000.
As Joe Browne, executive vice president for communications
and public affairs of the NFL, puts it, “The first amendment
gives every individual and organization in this country the
right to petition the government. We’re there to let them know
where we stand on issues, just like every other entity and
industry in the country” (Sports Business Journal, January 12,
2004, p. 18).
Compared to the real giants of lobbying, MLB’s efforts are
trivial; the U.S. Chamber of Commerce led the 2008 “top 20
list” at $91.7 million, General Electric was 10th at $14.3
million, and rounding out at 20th was General Motors at
$13.8 million. But MLB’s lobbying efforts are only topped by
the NFL in their entire government-defined sector,
recreation/live entertainment. In fact, in 2008, MLB spent as
much on lobbying as did many major universities lobbying for
research support—USC ($880,000), Johns Hopkins
($850,000), and Cal ($810,000).
MLB’s main lobbying efforts in the recent past were pushing
legislation that fortified the league’s antitrust exemption and
fighting legislation aimed at reducing that exemption. For
example, the Stadium Financing and Relocation Act, which
would have forced baseball franchises to fund stadiums
privately in order to keep their antitrust exemption, was
introduced by Arlen Specter (R-PA). He made no bones about
the fact that it was intended to curb the type of spending that
had happened in Pennsylvania at the time. Pennsylvania state
legislators had approved stadium funding for MLB’s Pirates
and Phillies and the NFL’s Steelers and Eagles.
Other nuts-and-bolts issues also received MLB lobbying
attention:
• Trade with Cuba (mainly clearance for the Baltimore Orioles
to play exhibition games in that country)
• Communications (prohibiting Internet gambling and
increasing satellite broadcast regulation)
• Copyright law (especially important to MLB broadcasts)
• Tax law (governing deductions when entertaining at sports
events)
463

http://www.opensecrets.org

http://www.opensecrets.org

Alan Sobba, MLB director of government relations with
offices in Washington, D.C., made it clear that MLB
participates in the lobbying process with the same goals of any
potent political participant, “We’re up there [Capitol Hill] on a
day-to-day basis to make sure we get a fair shake on these
pieces of legislation.” In congressional hearings on steroid
abuse in early 2005, oversight committee members had
received $8,000 from MLB’s PAC and faced the man in
charge of those political contributions, Commissioner Bud
Selig.
Sources: Sports Business Journal, October 11, 1999, p. 6, and
January 12, 2004, p. 18; BostonHerald.com, March 17, 2005;
DailyNews.com, March 20, 2005.
464

SECTION 3
Owners as Beneficiaries of
Rational Actor Subsidy
Politics
Team owners are an important reelection constituency for
state and local politicians. Let’s take a look at the value of the
benefits owners seek from the political process. The size of
those benefits should clue us in on how much owners will flex
their political muscle in order to get stadium subsidies.
Two things are expected from a stadium subsidy. First,
subsidies raise owner profits. Subsidies are expected to
increase revenues at the gate and, in the event of a better
lease, owners will receive a greater share of concessions and
parking. In addition, under a better lease, costs such as rent
and stadium maintenance could fall. Second, in the eyes of
fans, the team should improve. However, the relationship
between subsidies and team quality can be complicated. As we
learned in earlier chapters, revenues can rise, but the team
may not get any better. The owner will be better off in any
case, but the profit-maximizing use of the subsidy may not be
to increase the team’s level of quality.
From Chapter 4, we know that owners are constantly striving
to find that level of team quality that maximizes profits in the
long run. From that perspective, here is how an injection of
new revenue would not alter team quality. If the subsidy
revenue comes in, but the owner does not think that fans will
support increased quality in the long run, then no more
spending on talent will occur out of subsidy revenues. We
should expect the owner to pocket the subsidy. However, if
the owner perceives that fans will support a higher level of
talent in the long run, then subsidy revenues will go to hire
more talent, and the team should improve.
It is difficult to know which will occur beforehand, and
looking at revenues and team quality after the fact is not
always enlightening. Suppose that both revenues and quality
increase after a particular team owner obtains a stadium
subsidy. Did the subsidy lead to the increase in quality? The
answer can only be maybe. The team may have been
improving prior to the subsidy and simply have continued a
long-run plan that would have occurred without it. In this
case, the team improved because the owner had made the
decision prior to the subsidy that profits would improve with a
higher-quality team. Again, because the subsidy had nothing
to do with the team’s quality improvement, we would expect
that the subsidy would just go in the owner’s pocket and that
no spending on talent beyond that already planned prior to
the subsidy would occur.
Another explanation exists for an observation of improved
quality prior to the subsidy. The owners could have been so
sure that the subsidy was forthcoming that they went out on a
limb and began hiring talent before receiving the subsidy. The
owners may have forecast that their fans would pay more for
465

higher quality and borrowed against the future subsidy in
order to increase quality. As you can see, there are two
explanations for the same outcome. One does not depend on
the subsidy, the other does.
Let’s keep all of these ideas in mind as we look at the evidence
on revenues, spending on talent, and team quality after
stadium subsidies. In addition, remember that owners will
always be better off with a subsidy than without one. They will
either spend the subsidy on higher quality and earn greater
profits in the future or they will just pocket the subsidy. In our
empirical analysis of subsidies and revenues, to which we now
turn, we will focus mostly on MLB.
AN EARLY ANALYSIS OF SUBSIDIES, REVENUES, AND
WINNING
In our 1992 book Pay Dirt: The Business of Professional Team
Sports, James Quirk and I detail the impact of new stadiums
on attendance and performance in MLB over the period
1960–1982. Focusing on the five-year average attendance
before and after a new stadium, we found that MLB teams
with new stadiums enjoyed an attendance increase of about
624,000 per year. A similar calculation for the remaining
teams in the league (those without new stadiums) revealed
only a 96,000 increase per year. Thus, teams with new
stadiums enjoyed about 6.5 times the attendance increase of
the rest of the league. New stadiums draw more people, just as
we would expect from our analysis in Chapter 4, because the
quality of the stadium experience has increased. The long-run
stadium decisions of teams should lead to improved fan
enjoyment.
Quirk and I also show that the on-field performance of teams
with new stadiums improved over that same period (1992).
The same five-year average comparison before and after a new
stadium showed that teams with new stadiums enjoyed a 35-
point winning percent increase. Each game won represents
about 6.2 points (1,000 total possible points, divided by 162
possible wins, yield 6.2 points per win). So 35 points would be
worth about 5.6 more games per year. From our analysis of
winning percent increments and final league standings in
Chapter 4, that is enough to move a team out of the cellar and
is a goodly portion of the amount needed to move a team
through the rest of the standings. At least on one sample, new
stadiums improve attendance and revenues, and those
improvements are parlayed into better teams on the field.
Quirk and I (1992) also noted that part of the overall increase
for teams with new stadiums would be bound to occur
because four of the 12 teams in the analysis were expansion
teams. Expansion teams slowly improve over time. We offered
the same explanation for our findings that you would offer
from your understanding of owners’ long-run profit pursuits:
A new stadium will tend to improve the drawing
potential of a team, for a given roster of players. As
makes sense intuitively (and from economic theory) the
stronger the drawing potential of a team, the more a
profit-maximizing team finds it worthwhile to spend in
466

improving the caliber of the team. In effect, a new
stadium converts a small-town market, such as Kansas
City, into a market that is not quite so small as before,
and the profit incentives this creates lead predictably to
the team’s acquiring higher quality players, producing a
better performance on the playing field. (Quirk and Fort
1992, p. 139; QUIRK, JAMES; PAY DIRT. © 1992
Princeton University Press. Reprinted by permission of
Princeton University Press.)
This explanation is entirely consistent with our findings.
However, team quality does not always increase following a
move into a subsidized stadium.
Hamilton and Kahn (1997) note that attendance rose
significantly for the Orioles after they moved to their new
Camden Yards Ballpark for the 1992 season. In the five years
before the Orioles began play in the new park, attendance
averaged 29,459 per game per year. In the five years after, the
average increased by 53 percent to 45,034. Further, they note
that stadium revenues rose by $25.5 million ($30 million)
after the Orioles paid about $1.8 million ($2.4 million) to the
stadium authority. Thus, the owners acquired about 93
percent of the stadium revenue increase. This clearly indicates
that the Orioles’ owners benefited from a new park. How did
the new park affect the quality of the Orioles? In the five years
prior to the 1992 season, the Orioles’ average winning percent
was 0.434 (with a range from 0.335 to 0.537). For the five
years from 1992 through 1996, their average winning percent
was 0.534, a full 23 percent increase in quality. It does not
appear that this dramatic improvement was by prior design,
as the Orioles’ winning percent had fallen steadily from 1989
to 1991.
Some evidence shows that the new stadium revenues went to
talent, as the Orioles’ team payroll jumped from $14 million
($21.9 million) in 1991 to $24 million ($37.6 million) in 1992
and continued to increase by about $8 million per year
thereafter, whereas leaguewide the average team payroll rose
about $500,000 per year over the same period. Remember,
we cannot know whether the subsidy drove the higher
winning percent outcomes observed here, but in the Orioles’
case, it appears that the stadium subsidy, in addition to
making more money for the owner, did coincide with a better
team.
Of course, we cannot know whether this increase in quality
was planned regardless of whether a subsidy was obtained,
and moreover, the possibility that the subsidy was just
pocketed by the owner cannot be ruled out either.
Interestingly, the team did not make it to the play-offs in the
five years prior to moving into Camden Yards, and they did
not improve enough to make it to the play-offs after that even
with the revenue injection from their new stadium.
A MORE RECENT ANALYSIS OF STADIUMS, REVENUES,
AND WINNING
In my 1999 article “Stadium Votes, Market Power, and
Politics,” I took an updated look at team owner benefits from
467

the most recent stadium subsidies where data were available.
Looking at MLB, data availability narrowed the field to the
White Sox (New Comiskey Park), Orioles (Camden Yards),
Indians (Jacobs Field), and Rangers (Ballpark at Arlington).
The data in Table 11.1 show the annual averages for the five-
year periods before and after the new stadiums (since they are
averages over periods of time, we leave them in nominal
terms). The impacts from the recent stadiums are huge. The
smallest annual average attendance increase was 941,000 for
the Sox, and the remaining teams enjoyed average increases
of over 1 million fans at the gate per year.
Because we have already discussed the winning percent
results for the Orioles, let’s focus on the remaining teams.
Winning percent increases were enough to move the White
Sox and the Indians up an entire place in the standings, and
this improvement was enough to push the Indians into
pennant contention once they began play in their new Jacobs
Field. Team salaries also followed a familiar pattern. For the
White Sox, team payroll increased $7 million ($11 million) in
1991, $14 million ($21.3 million) in 1992, and another $12
million ($17.7 million) in 1993. In 1994, the Indians’ team
payroll went up $15 million ($21.6 million), and then another
$9 million ($12.6 million) in 1995 and $7.5 million in 1996
($10.2 million). Success followed the revenue injection from a
new stadium for these two teams. Again, whether these
quality increases were planned with or without the subsidies
from new stadiums cannot be known. However, it is clear that
winning percent increased coincident with owners’ receiving
subsidies.
The Rangers illustrate the cautions discussed at the beginning
of the section. First, to be sure, Table 11.1 shows that gate and
venue revenues took a decided jump up. However, no increase
in quality followed their movement into their new park in
1994. This should come as no surprise because the Rangers’
owners spent hardly anything more on talent after the move.
Team payroll fell $3.6 million ($5.2 million) in 1994 and
increased just $3.5 million ($4.9 million) and $4.8 million
($6.5 million) in the next two years. It appears that the
Rangers’ owners did not put much, if any, of the increased
gate and venue revenues toward talent. The theory in Chapter
4 would suggest that this was because the owners saw no
gains in terms of fan willingness to pay for increased quality
in the long run. We can reach no other conclusion except that
the owners pocketed the subsidy.
Quinn, Bursik, Borick, and Raethz (2003) extend the
examination of the relationship between new stadiums and
winning to all four major pro sports leagues. They take both
seven-year and three-year before-and-after looks at all teams
in all four leagues. They find that there is no improvement in
team quality measured by winning percent in any league
except MLB. They also find that the magnitude of the impact
of a new stadium on MLB on-field success is about what Quirk
and Fort (1992) found, on the average nearly six more games
per season.
468

WHAT TO EXPECT FROM OWNERS
Although revenue injections from stadium subsidies appear to
lead to higher team quality only in MLB, new stadiums do
help owners by increasing revenues and possibly lowering
costs; profits should, therefore, increase. Stadium revenues
have always been important to all teams in all pro sports
leagues. In Table 11.2, one part of stadium revenues, ticket
revenues at the gate, are shown both in dollar values and as a
percentage of total revenues for all four major pro sports
leagues. Unfortunately, data limitations do not allow any
other insights into the rest of stadium revenues (attendance-
related revenues like parking and concessions, or signage
advertising, for examples). However, at least attendance-
related revenues are highly correlated with ticket revenues at
the gate.
What we should expect from team owners is indicated by the
large gap between the gate revenue haves and have nots. By
simple ratio comparisons of actual gate revenue values, the
gap is largest in MLB. The Yankees dominate the median gate
revenue teams like the Brewers and Mariners by a factor of
3.6 and the lowest gate team, the Marlins, by a factor of nearly
10! The ratio comparisons are closer in the remaining leagues,
but substantial gaps exist there as well. The percentages add
more insight. Again, since the differences are most apparent
there, look at MLB. Again, teams like the Brewers and
Mariners are typical in terms of the importance of gate to
their overall revenues. But those revenues are less than half as
important in percentage terms compared to the Angels. And
pity the poor Marlins (again)—gate revenues are less than half
as important as the median teams! Things are not nearly quite
so bad in the other leagues, measured by revenue percentage,
but substantial differences do occur.
Although most of this difference is determined by whether a
team is located in a large-revenue market, that will not deter
owners who suffer a “stadium revenue disadvantage” from
pushing their hosts for a larger subsidy. Owners will use the
argument that such subsidies will help them stay competitive
with teams with better stadiums and lease arrangements. For
years, this is precisely the plea of the Florida Marlin in MLB
and the San Francisco 49ers in the NFL, and Table 11.2 shows
at least that the facts of the comparison are correct—they are
at the bottom of the stadium revenue heap. In the NBA,
events transpired to simply move the Seattle Sonics to
Oklahoma City for the 2008–2009 season.
We have just seen that teams usually do not follow through on
their claims to put better teams on the field once a subsidy
increase occurs. But, we also know from Chapter 5 that team
owners can point to other cities that may offer a large stadium
subsidy as a potential place to move if their current hosts do
not help them to keep up with the Joneses. Therefore,
although all revenues are important in determining a team’s
competitiveness, getting a better stadium deal is important to
any given team if for no other reason than that the money is
nice to have.
STADIUM SUBSIDIES AND THE NFL SALARY CAP
469

NFL owners would like to pursue stadium subsidies like any
other owners. But any argument they make that subsidies will
increase quality faces a direct challenge. The NFL shares
nearly all revenues and has a salary cap, ostensibly designed
to equalize spending and level the league, competitively
speaking. So where can there be any competitive advantage in
a stadium subsidy? The answer lies in the fact that not all
stadium revenues are shared with other teams, and these
unshared revenues increase the salary cap. The cap goes up,
but only some teams actually have the increased revenue to
increase the quality of their team.
For example, stadium sponsorship and so-called luxury-box
revenues are not shared with other teams in the NFL. Suppose
the configuration of a new stadium includes more of these
unshared revenues than the owner enjoyed in the old stadium.
This NFL owner might be able to obtain a stadium-specific
revenue advantage over the rest of the owners. The revenue
from that owner’s stadium advantage goes into defined gross
revenue, raising the leaguewide cap on salaries. Because the
cap has increased, all teams are now supposed to spend more
on talent, and the minimum-spending requirement has also
increased. However, only the owner who gained the stadium
advantage actually has the increased revenue to spend up to
the new cap. The rest of the teams are left behind because
they do not have any increased revenue to put toward talent.
Now, as we saw in Chapter 5, NFL owners typically spend
more than the cap amount on players. However, this extra
spending varies considerably. In general, the sort of one-
upping behavior driven by stadium-specific revenue
advantages will always leave some teams behind,
competitively speaking, at any point in time. The result is that
every time one owner gets a better stadium arrangement or a
new stadium, demands by the rest of the owners without these
revenues increase.
470

SECTION 4
The Role of Leagues in
Rational Actor Subsidy
Politics
Leagues play an important behind-the-scenes role in the
modern determination of venue subsidies. In fact, they have
always played a role in stadium subsidy issues. As detailed in
Chapter 5, it is clear that MLB chose not to expand westward
until well into the 1950s, despite congressional scrutiny and
pressure from a potential rival, the Pacific Coast League. The
result of keeping the West open was quite advantageous for
the owner of the Brooklyn Dodgers, Walter O’Malley.
O’Malley, increasingly vocal about needing a new stadium,
could get nowhere with powerful New York City’s political
boss, Robert Moses, in his pursuit of a Brooklyn location for a
new stadium. In the end, O’Malley moved the Dodgers to Los
Angeles after the end of the 1956 season. The Dodgers went
from being the most financially successful team in baseball, in
Brooklyn, to being one of the most successful sports
franchises in any league in history in their Los Angeles home.
O’Malley had this opportunity because MLB had kept the
West open.
LOCATION MANAGEMENT BY LEAGUES
Leagues facilitate the market power position of team owners
by restricting the number of teams in the league. In the
stadium subsidy context, this artificial restriction stacks the
deck in favor of teams in their stadium negotiations with host
cities. As long as leagues carefully manage team location,
existing teams will always have a believable threat location in
their negotiations with their current hosts. The relationship
between subsidies and believable threat locations is clear.
There has not been an MLB team move since the Senators left
for Texas after the 1971 season. This is because nearly every
team in baseball has gotten either a new stadium or a
significantly renovated stadium in the meantime.
The San Francisco Giants are a perfect case in point. After
searching in vain for a new location willing to build a new
stadium in the Bay Area and a bit inland through the early
1990s, the owner sold the team in 1992. Immediately, new
owner Peter McGowan said, “I think the people now know if
we don’t get a [new] stadium, the team will be forced to move”
(Sporting News, November 23, 1992, p. 5). Shortly thereafter,
the owner managed to get some infrastructure subsidy for the
Giants’ current home, SBC Park, completed in San Francisco
in 2000.
The most recent NFL expansion, when the league was
deciding between Houston and Los Angeles, is also instructive
on this point. In general terms, the decision between the
locations hinged on two considerations regarding the Houston
and Los Angeles markets. First, the league considered the
financial contribution that either location would make to the
471

league. Second, the league considered the value of an open
location and the negotiating advantages it provided to current
league members. Keeping the best believable threat location
helps owners in negotiations with their current host cities.
SPECIAL PROBLEMS FOR THE NFL
In the NFL, through a combination of lawsuits and owner
preferences, teams are able to move between cities much
easier than in other leagues. This means that a given
community might perceive that they actually have a good
chance to replace their team if stadium subsidies lead their
current owner to leave. Some evidence supports the view that
alternative threat locations are less important to NFL owners
than they are to, say, MLB owners. In every case, a city that
lost its team got a new one within a few years (except for Los
Angeles).
In the most famous case, the Oakland Raiders moved to Los
Angeles and back to Oakland again within a few years.
Baltimore lost its Colts to Indianapolis but landed the new
Ravens (previously, the Cleveland Browns) within a few years.
As detailed in the section-end Learning Highlight: The
Stadium That Didn’t Get Built and Then Did, when St. Louis
did not meet the stadium demands of Cardinals owner Bill
Bidwell, he took his team to Arizona in 1988. St. Louis was
without pro football until they built a new stadium, luring the
Los Angeles Rams to St. Louis in 1995.
Thus, in the NFL, some cities have been able to cut
demanding NFL owners loose and find another team. Los
Angeles remains the only exception, but both the former
Seahawks owner and the owners of the Buccaneers have
shown an interest in filling the hole in Los Angeles. However,
the league refused those moves. Perhaps the NFL has
something greater in store for that city, or perhaps they are
just waiting for the right offer.
In contrast to the NFL, in MLB, with owners unwilling to
replace an owner who threatens to move, cities have paid
dearly and more often to keep their teams. Former
Washington, D.C., mayor Sharon Pratt Kelly put it this way:
If no mayor succumbs to the demands of a franchise
shopping for a new home then the teams will stay where
they are. This, however, is unlikely to happen because if
Mayor A is not willing to pay the price, Mayor B may
think it is advantageous to open up the city’s wallet. Then
to protect his or her interest, Mayor A often ends up
paying the demanded price. (Shropshire, 1995, p. xi;
Reprinted with permission from the University of
Pennsylvania Press.)
Given that leagues manage locations to keep teams the only
game in town, a local government must deal with its current
tenant or lose a team and hope that it can woo an unhappy
owner or gain a team in the next expansion round. Arguably,
this single factor contributes more to the modern venue
subsidy picture than any other element in that process. Now
that we understand the role of politics, owners, and leagues,
472

we are ready to bring the rational actor explanation to bear on
the stadium subsidy process.
473

LEARNING HIGHLIGHT: THE STADIUM THAT DIDN’T
GET BUILT AND THEN DID
The Cardinals are the oldest team in the NFL, their roots
going back to Chicago in the 1890s. Owned by the Bidwell
family since the 1930s, they moved from Chicago to St. Louis
to play the 1960 season. The team sank into mediocrity in St.
Louis. In 1985, owner Bill Bidwell announced that he was
exploring alternatives for his team. The football and baseball
Cardinals shared Busch Stadium, and average attendance had
been the lowest in the NFL, at about 48,000 per game.
Bidwell had two alternatives in mind: St. Louis could build a
new stadium for the Cardinals and Bidwell would stay in St.
Louis, or St. Louis could refuse and he would move the team.
He argued that Busch stadium was too small (even though the
football Cardinals had never sold it out at its 56,000 football
capacity) and objected to the sale of beer in the stadium
(unlikely to change because the Busch family, of Anheuser-
Busch fame, owned the stadium and the concession rights).
Two groups formed in response to Bidwell’s demands. One
group wanted a privately owned stadium well out of town. The
idea that it would be privately owned was a stretch, because
the proposal called for city and county bonds to finance $120
million ($238 million) out of the $150 million ($297 million)
construction cost. The second group wanted a publicly owned
stadium downtown. Both proposals included a domed
stadium that would seat around 70,000.
Engineering estimates suggested that the out-of-town version
would leave a $2 million ($4 million) annual operating
surplus, whereas the downtown version would lose $5 million
($10 million) annually. After an economic analysis of the
situation by noted economist James Quirk, both of these
estimates were revised downward: $5 million ($10 million) in
losses out of town and $10 million ($19.8 million) in losses
downtown.
In the end, the mayor rejected both proposals in favor of a
renovation project for Busch Stadium with luxury suites and
added seating. Bidwell rejected it, and in 1988 moved the
team to Phoenix, where mediocrity reigned for the NFL
Cardinals until their first 0.500 season in 1994 (they appeared
in the playoffs only once before their first-ever Superbowl
appearance in 2008). The Bidwells finally received the new
stadium they desired. The people of the state of Arizona voted
for a new stadium in November 2000. The location was up in
the air for years before settling on Glendale, Arizona. The
$450 million facility ($557 million), currently known as
University of Phoenix Stadium, opened in 2006.
St. Louis also ended up building a new stadium to attract an
NFL team. In the early 1990s, the people of St. Louis
approved a multimillion dollar bond issue for a downtown
stadium. In 1995, when the TransWorld Dome (capacity
66,000) was nearing completion at a cost of about $290
million ($405.7 million), a team finally committed to move to
St. Louis. Georgia Frontiere, owner of the Rams, agreed to sell
a portion of the team to local St. Louis interests, and the new
474

owners of the St. Louis Rams moved the team in 1995 under
the following agreement:
• A $30 million ($42 million) relocation fee was paid to the
NFL stadium fund.
• The NFL would keep the Rams’ $13.5 million ($18.9 million)
share from the next expansion.
• A $12.5 million ($17.5 million) payment was made to cover
the expected reduction in the FOX broadcast contract with
the team in St. Louis rather than Los Angeles.
• A $17 million ($23.8 million) payment was made to a St.
Louis group that raised enough money through PSLs to get
the NFL’s consent to move the team.
All things considered, that is about $73 million ($102 million)
and a reduction in ownership rights as the cost of moving the
team. The move by the Rams eventually led to a Super Bowl
title in 2000.
Previous St. Louis Cardinals owner Bill Bidwell could not get a
new stadium and left St. Louis. The Los Angeles Rams moved
into the new TransWorld Dome in 1995 and were Super Bowl
champs in short order. [Photo TransWorld Dome field.]
Sources: News & Observer online (www.newsobserver.com),
1995; James Quirk and Rodney D. Fort, Pay Dirt: The
Business of Professional Team Sports (Princeton, NJ:
Princeton University Press, 1992) (QUIRK, JAMES; PAY
DIRT. © 1992 Princeton University Press Reprinted by
permission of Princeton University Press.).
475

http://www.newsobserver.com

http://www.newsobserver.com

SECTION 5
Sports Facility Outcomes
Explained
Elected officials make stadium decisions according to the
rational actor logic just developed. That logic leaves us every
reason to believe that subsidies will be too large and paid by
many who couldn’t care less about pro sports. Let’s walk
through the logic applied to the determination of venue
subsidies, then turn to a real-world case in Seattle.
RATIONAL ACTOR POLITICS AND VENUE SUBISIES
Imagine that a stadium subsidy issue is brewing somewhere.
Subsidy critics will deplore both the size and purpose of such
spending. On the size dimension, opponents will push any
and all estimates to the upper limit of belief (and possibly
beyond), causing rationally ignorant voters to get an uneven
portrayal of the costs. The voters will also be made to feel
guilty about supporting the subsidy. Who can reasonably
argue that stadiums and pro sports teams are more important
than schools? Surely, say the critics, voters will not be able to
look themselves in the mirror if they support such spending.
Stadium supporters will be quick to defend the subsidy.
Supporters typically chalk up the entire amount of economic
activity on the benefits side, dramatically overstating the
benefits of the subsidy. Reports from consultants are
presented that tout the massive economic benefits of the
team’s presence; payback periods are often shown to be less
than 10 years. Once again, voters get a skewed perspective,
but this time on the benefit side. Few form a reasonable view
of the new net economic activity from a proposed subsidy.
Further, both sides argue about the “intangible” benefits of
the stadium we referred to as quality of life values in Chapter
10. Supporters overstate them, and opponents minimize
them. At the very least, there are comparable situations that
could be examined but never are. The reasons for this are
clear. Team owners employ a “short-fuse” ultimatum strategy;
the pot may simmer for a year or two, but then state and local
government officials are given a very short time to generate
the subsidy or lose the team. Another reason solid analysis
never comes into play could be that owners and their
supporters already know that, in all likelihood, the net values
will not justify the size of the subsidy they seek. Again, voters
are left with nothing but rhetoric.
There is more. Remember from Chapter 10 that a minimum
subsidy charges users the extra costs of operation, and
government collects a tax that makes the stadium worth it.
However, the actual type of tax mechanism chosen almost
never matches this subsidy-minimum prescription. Tax
collection options range from hotel taxes, which are paid by
both fans and nonfans, to sales taxes, to bond issues
guaranteed out of the general fund. A favorite is property tax
relief for the stadium and/or sales tax relief for the local
476

government. In each case, the amount available for other
purposes, statewide, is reduced, and nonfans pay the vast
majority of the price of keeping a team in town.
Nearly all of this is driven by the maintenance of believable
threat locations by owners acting as leagues. Fans are
confronted with an all-or-nothing alternative. If the state or
local government fails to meet the owner’s subsidy demands,
then the team will move to one of the locations kept open by
the league.
This sets the political stage. Detractors emphasize only the
costs and supporters overstate the benefits. Little in the way
of independent analysis is forthcoming. The poor taxpayers
are left with only extreme information with which to weigh
the situation, and that is if they are even paying attention in
the first place. In such a setting, one would expect benefits to
be concentrated in the hands of politically powerful stadium
proponents and costs to be dispersed among the rest of the
taxpayers. You have already seen this logic in action in
Chapter 9 when we covered union outcomes.
A DIAGRAMMATIC MODEL OF VENUE SUBSIDIES
The venue subsidy process can also be portrayed as a rational
actor politics triangle, as shown in Figure 11.2. Powerful
local and state political interests make their demands known
to local and state elected officials. These officials produce
financing arrangements for a stadium subsidy (or take it
directly to the people with a referendum). The demands of
powerful groups are met, and these groups reward politicians
with votes and election resources. The general taxpayer is left
on the sidelines. New York City councilwoman Letitia James,
commenting on the move to use public money to build a
Westside stadium, said, “When I run for office, I knock on
hundreds and hundreds of doors, and nobody ever says we
need arenas or stadiums” (Sports Business Journal, December
27, 2004, p. 18). But she is just describing how the general
taxpayer is left on the sidelines in this particular policy arena.
As predicted, benefits are concentrated and costs dispersed,
and there is always the possibility that costs exceed the
benefits.
The economic beneficiaries of a stadium subsidy are usually
well organized and easy for politicians to recognize. The team,
ardent fans, the press and TV media, members of the
construction and concession industries, and businesses near
the stadium stand to gain because the subsidy keeps the team
in town. These interests have a very large stake in getting a
stadium improved or approved. It is reasonable to expect that
they will reward politicians who give them what they want
with the resources that politicians need to campaign
successfully.
On one occasion, the public did exact political punishment for
a massive stadium subsidy. At first, Wisconsin state senator
George Petak (from Racine) opposed a legislative financing
scheme for a new MLB Brewers stadium in 1995. However, in
short time and under heavy lobbying by the governor, he
changed his vote. His constituency back in Racine was so
enraged that he was recalled by popular vote—but the rarity of
477

this event makes it the exception that proves the rule.
Politicians do not want to lose their team and the election
support that would follow a team move. Even if the team
really wouldn’t move, special interest politics can still yield
the same outcome.
WHAT ARE TAXPAYERS TO DO?
There are two ways that taxpayers can affect the subsidy
approval process, neither of which can be expected to slow the
stadium juggernaut. One way is through direct democracy. We
will look at this topic in detail in the next section. The other
method is through the review of the choices made by elected
officials the next time they are up for election. In this case,
rational ignorance takes over with a vengeance. Even voters
vehemently opposed to stadium construction will have a
difficult time trying to hold local and state representatives
accountable. Rationally ignorant voters will never really know
the benefits that were provided and to whom, and the costs on
a per voter basis will be quite small. As long as these elected
officials provide benefits from other decisions that offset the
very small costs borne by the general voter, they will be safe in
their reelection pursuits. At least that is the outcome
predicted by the rational actor explanation. Some hold out the
hope that direct democracy (referenda or initiatives) is a way
to improve the venue subsidy process.
478

SECTION 6
The Role of Direct
Democracy
One avenue of public participation in decisions about sports
stadiums is direct democracy through referenda and
initiatives. But can direct democracy help alleviate the what
most see as overly high subsidies? Although uncommon, some
subsidies have been voted on in referenda and initiatives. In
this section, we will examine the relationship between
subsidies and direct democracy.
Typically, when a stadium issue appears on the ballot, it is the
culmination of a complex political process. In some cases,
when taxes are changed or bonds are issued, a referendum
may be required by the state constitution. In other cases,
public officials can place a stadium measure on the ballot even
though it is not required. In still other cases, citizens use the
initiative process to force a public vote.
HIGH HOPES FOR DIRECT DEMOCRACY
It is easy to see why subsidy opponents might think that direct
democracy can save the public from stadium subsidies. The
whole point of direct democracy is to allow voters a direct
means of expression. In theory, a more “hands-on” decision
process will produce outcomes that are (1) more legitimate in
the eyes of participants and (2) a better reflection of the “will
of the people.” Direct democracy takes the decision away from
elected representatives.
Direct democracy has other appealing characteristics. On
stadium subsidy ballot issues, the choices confronting voters
are quite simple and cover a single element, spending. The
alternatives seem quite clear: the ballot choice versus the
current situation. In addition, information presented by
subsidy supporters and foes may generate well-reasoned
debates prior to the vote so that voters are no longer rationally
ignorant.
WEAKNESSES OF DIRECT DEMOCRACY
Direct democracy, however, has some important weaknesses
relative to representative democracy. The outcome of a
democratic process depends on which voters turn out to vote.
This, in turn, depends heavily on the amount and quality of
information that is available to citizens. As we have seen in
previous sections, both turnout and information can be
biased, and the same holds true for public votes on stadium
issues. In addition, elected officials may control the
alternatives presented to voters, and these alternatives may
actually force higher spending than would occur through a
representative decision process.
BIASED TURNOUT
Turnout can be biased due to the registration and voting costs
discussed earlier. Further, rational ignorance and free riding
479

drive both turnout and fall off. The incentive to turn out and
vote is dramatically less for the rationally ignorant voter who
will eventually bear the dispersed costs of stadium subsidies.
Those standing to gain typically have a much larger stake in
the outcome and are more likely to vote. This is the biased
turnout result for direct democracy.
BIASED INFORMATION: THE SEATTLE SEAHAWKS
VOTE
Let’s turn to the biased information question. In most
elections, the information that is offered to fuel “well-
reasoned debate” is polarized. Proponents present
information that overstates the benefits. Conversely,
opponents present information that overstates the costs.
Independent analysis is rarely offered. A recent example
makes this very clear.
Proposition 48, the referendum on a new stadium for the
Seattle Seahawks, reached a vote in 1997. The information
distortion surrounding Proposition 48 was incredible.
Supporters of the new stadium funded a huge information
blitz. Paul Allen alone spent around $3 million in advertising,
whereas opponents spent only $160,000. A full-page
newspaper advertisement that ran in one of the state’s major
newspapers (Spokane Spokesman-Review, May 12, 1997, p.
C4) touted the benefits of Proposition 48. The article was paid
for and presumably formulated by “Our Team
Works” (henceforth, OTW), the publicity organization in
support of the stadium. In this ad, OTW attacked the then
home of the Seahawks, Seattle’s Kingdome:
Remember the day the Kingdome roof fell in?
The disaster left property taxpayers with a $70 million
hangover. This adds up to $5 million per year in
property tax debt for which our taxpayers receive no
services. In fact, the Kingdome has cost taxpayers more
in the last 2.5 years for repairs than the original
construction costs and now $42 million more is needed
for basic repairs.
The good news? The funding package for the new
stadium and exhibition center [Proposition 48] will retire
all Kingdome debt, and free the property taxes for better
purposes. (Spokane Spokesman-Review, May 12, 1997, p.
C4)
Picking apart this distortion is like shooting fish in a barrel.
First, OTW forgot about inflation. It is invalid to compare the
1996 repair bill of $70 million to the $67 million original
construction cost in 1976. That $67 million, adjusted for
inflation, would have been worth about $187 million at the
time of the vote in 1996. This changes the comparison
considerably. Far from being over twice the original cost, the
cost of repairing the Kingdome after 20 years of faithful
service actually was less than half the original cost.
The second distortion also follows from the observation that
the Kingdome was 20 years old by 1996. All capital equipment
480

wears out eventually and requires upkeep in order to remain
productive. The simple fact that repairs were needed is not
damning criticism of the structure.
A third observation about the OTW’s claims concerns the idea
that no services were gained from the repair costs. Without
the repairs, the Kingdome would have had to close quite some
time before the election. In that case, the NFL and MLB teams
would have been without a stadium, and nonsports-related
functions would have been without a venue. Therefore,
services certainly were obtained through maintenance and
upkeep. The question is whether it is worth spending on
maintenance relative to the alternatives, such as a new
football stadium.
The question of whether repairs are worth it is just a version
of the classic used car problem. Suppose you have $1,000 in
cash plus an existing used car worth $1,000 to put toward a
transportation upgrade. But then the car breaks down,
requiring repairs of $1,000. Should you put $1,000 into a car
that will have a market value of $1,000 after the repair? If
your next-best alternative is to buy a car costing $1,700, but
you only have $1,000 plus a broken-down trade-in, then
putting the $1,000 into the broken-down car clearly is
rational. Just because the Kingdome was breaking down a bit
does not mean that it was irrational to fix it.
As a fourth observation, the remaining debt on the Kingdome
does not magically disappear just because it is “retired” under
the new funding plan. Indeed, the balance was simply rolled
over into the debt represented by Proposition 48. To return to
the used car analogy, suppose you trade in your current used
car for a price less than you owe on it. The remainder just rolls
over into the new contract. The debt does not go away; it still
has to be paid.
One final observation is critical. Tearing down the Kingdome
makes the new stadium/exhibition center the only place in
town that large non-NFL activities can take place. Because the
lease arrangement with the Seahawks turns all nonfootball
revenue over to the team’s owner, the owner has a monopoly
on the only venue in town for professional and amateur
soccer, trade shows, community festivals, and other
entertainment events. If the Kingdome were not torn down
and remained as a viable alternative for these activities, price
competition would increase the number of these types of
events and lower the rental price charged. Therefore, the team
owner benefits from removing a competing facility.
AN IMPORTANT BUT TECHNICAL ISSUE: THE
ALTERNATIVES CONFRONTING VOTERS
In cases where the alternatives that actually confront voters
are controlled by a few people, such as the mayor or city
council, problems can arise if elected officials have a
preference for high levels of stadium spending. If the
alternative to the high-spending proposal is complete loss of
the team or if the current stadium really is decrepit, those in
control of ballot alternatives on which the voters vote can
actually force high spending levels even though the voters may
481

favor a lower amount of spending. Those in control of the
ballot alternatives have come to be called “the setters,” and
the following uses have come to be called “the setter model” of
referendum outcomes.
This possibility is shown in Figure 11.3, an example from my
1997 chapter “Direct Democracy and the Stadium Mess” (Noll
and Zimbalist, 1997). In Figure 11.3, a one-dimensional line
represents the level of possible stadium spending and four
possible outcomes. The leftmost outcome is no team and no
spending. This is the preferred position of citizen A. To the
right of this position is the current spending level. Citizen B
prefers this outcome, with a team playing in a dilapidated,
out-of-date facility. Next to the right is a new facility that
would be an improvement over the existing obsolete stadium
but without any frills. Both citizens C and D prefer this option.
Finally, far to the right is an elaborate facility with luxury
boxes, restaurants, lavish offices, and an internal shopping
mall. Citizen E prefers this outcome.
If our five citizens could debate and consider all of the issues,
the most likely winner is the no-frills stadium. It beats the no-
team option hands down (B, C, D, and E defeat A). Similar
thinking shows that the no-frills stadium beats the obsolete
stadium option by a vote of three to two. The no-frills stadium
also beats the elaborate stadium four to one. But suppose that
the alternatives put before the voters are restricted so that this
pair-wise comparison never occurs in the voting booth. This
might occur because citizen E is the most powerful in terms of
election resources, and the ballot issue is framed in citizen E’s
favor. In this case, “the setter” confronts voters with
alternatives that move the outcome closer to citizen E’s
preferred stadium.
When “the setter” does this, the most elaborate option would
be offered on the ballot, and both elected officials and citizen
E would try to convince voters that the no-team option would
occur if the election failed. Leagues reinforce the fear by
keeping believable threat locations open for the owner of the
team. The outcome in our case is as follows. Voters C and D,
and maybe even voter B, might prefer the elaborate stadium
to no team at all. So the vote is going to be at least 3–2 in
favor of the elaborate stadium (and quite possibly 4–1), even
though all voters but E would prefer the no-frills facility to the
elaborate option. Even if voters believe that the team will stay
and play in the obsolete stadium, the elaborate option is likely
to win if voters C and D marginally prefer keeping the team
and having a new stadium to keeping the team in a decrepit
stadium.
Again, the role of leagues in maintaining believable threat
locations open plays a crucial role. In a real-world example,
Bob Lurie (chairman of the San Francisco Giants at the time)
wrote a very forceful, thinly veiled all-or-nothing threat in the
San Francisco ballot statement of 1987:
The issue is not whether or not we should build a
ballpark for Bob Lurie or for the Giants. This franchise
will be around a lot longer than I, and they will always
482

have a place to play. I just think it would be tragic if it is
not in San Francisco.
The believable threat location in this case was Tampa Bay/St.
Petersburg (detailed in Chapter 5 and a learning highlight in
Chapter 10). If fans believe they will lose their team to another
location, the spending level that may ultimately win support is
higher.
EVIDENCE FROM REFERENDA
Although there is ample theoretical reason to doubt that
direct democracy will improve the stadium subsidy picture,
what do actual voting outcomes reveal? Table 11.3 shows the
outcome of all facility referenda since 1995. Prior to the 1990s,
facility votes typically did not pass (Fort, 1997), but what a
turnaround for the NFL from 1995 on! NFL owners enjoyed 12
successes against only two failures. MLB and the NBA just
break even, but they still enjoyed five and four successes,
respectively. The NHL had only two successes, but that’s all
they tried. Interestingly, these direct democracy votes occur
predominantly in cities trying to keep their current teams.
Only the 1996 election that eventually built a stadium for the
NFL Houston Texans was aimed at drawing a team to that
city.
DO REFERENDA RESULT IN LOWER SPENDING?
The effects, if any, of direct democracy can be seen in two
comparisons (Fort, 1999). First, we can compare the spending
levels of all issues that passed with those that were funded
without a vote. If direct democracy offers an improved venue
subsidy result, we would expect spending to be lower for
facilities voted for by the public. The actual results are mixed,
as shown in Table 11.4. Spending under direct democracy is
about the same for MLB and much higher for the NFL.
Spending is only lower for the single issue that occurred for an
NHL/NBA arena.
Second, we can compare those votes that failed but were
eventually funded anyway with issues that never came to a
vote. Again, if direct democracy has an ameliorating effect,
spending should be lower for those votes that failed but found
some other funding path. Interestingly, of those that failed,
half eventually were funded outside of the direct democracy
process. This outcome echoes a sentiment of Chandler,
Arizona, Mayor Jay Tibshraeny, who was in favor of
upgrading a spring training facility, “I believe the citizens
should have a say in this issue. If the voters pass this, we will
move forward. If the voters don’t pass this, we will still move
forward” (Phoenix Gazette, October 13, 1995, p. A1).
Table 11.4 shows that for both MLB stadiums and NHL/NBA
arenas, funding levels were dramatically higher for issues that
failed but found eventual funding than for those that never
went to a vote. Maybe proponents knew that voters wanted
them to spend more. The alternative explanation is that direct
democracy did not have any dampening effect on spending
even when the vote failed. Voters said no, and some issues
moved forward anyway at larger spending levels than would
have occurred even if there had not been any referendum.
483

THE PUBLIC–PRIVATE SPENDING MIX
More evidence of the impacts of direct democracy can be seen
in patterns of public–private spending on stadiums and
arenas shown in Table 11.5. In the 1970s, the public’s share
for such projects rose to a high of about 90 percent and fell
back to around 60 percent through the 1980s and 1990s. The
available data on stadiums and arenas opening in the early
2000s (Table 10.1 in the previous chapter) show that the
public’s share is still around 60 percent. If direct democracy
reins in stadium spending, then the public share should be
lower for those projects put to a vote. The data show that the
lowest public shares occur when funding is determined
outside of the direct democracy process (Fort, 1999). Further,
if you look at the level of spending, public shares under direct
democracy have more at the high end of the distribution than
occurs for other types of issues. So public shares determined
by election are higher, and higher on average, than occur for
nonelection stadium subsidies.
EVIDENCE ON THE CONTROL OF ALTERNATIVES
CONFRONTING VOTERS
There is some evidence that “the setters” controlling the
alternatives confronting voters are nudging spending upward
(Fort, 1997). The implication from our simple theoretical
depiction in Figure 11.3 is as follows. Suppose that the
stadium that will exist if the vote fails is far below the
preference of most citizens. If those in control of the ballot
alternative present voters with the highest possible spending
level that will pass, then the voting outcome should be barely
enough to win. If there really is control over the alternatives
and voters believe the worst in the event of failure, then votes
should pass by the barest majority.
Now, among votes that passed, 70 percent of the stadium
votes that received at least a majority were clustered around
the bare majority for passage. This is not due to random
chance; bare majorities are common in stadium elections.
Therefore, when a referendum passes, cities are more likely to
spend more on stadiums than the middle-ground preferences
of most citizens. Not much can be said about referenda that
failed. Perhaps those in charge of ballot alternatives either
guessed wrong, or the alternative in event of failure may not
have been so bad in those cases.
484

SECTION 7
First Down and $779
Million to Go in Seattle
In this section, we will look at the recent Mariners and
Seahawks stadium outcomes in detail (the discussion here is
from Fort [2000]). Many of the issues surrounding these
stadiums were very similar to those encountered 25 years ago
when the Kingdome was built. Speaking of the Kingdome, let’s
remember that it represents the alternative if the teams were
to stay in Seattle. In 1995, at the time the discussion came to a
boil over the fate of the Kingdome, the ongoing subsidy from
original construction, covered in the next sections of this
chapter, was around $18.6 million ($27 million) annually. In
addition, there were $70 million ($97.3 million) in repairs in
1995, and another $42 million ($58.8 million) in repairs
needed in 1997. That’s a total of $183.1 million in 2009 dollars
at about the time the Kingdome turned 20 years old. With
that type of renovation, let’s suppose the Kingdome would
have lasted another 20 years. At 6 percent for 20 years, that
raises the payment by about $11 million annually, for a total
(including the ongoing subsidy) of about $38 million per year.
We’ll use this comparison value shortly.
As discussed earlier, Safeco Field for the MLB Mariners was
denied public funding by referendum. However, a creative
funding package was moved through the representative
process. In contrast, spending on a stadium/exhibition center
for the NFL Seahawks was approved by referendum.
Therefore, the Seattle story offers an interesting example that
includes nearly everything that can occur in stadium financing
and construction: the intrigue of threats to move teams,
public revelation of spending preferences through referenda,
and the response by state and local politicians to rejection by
voters.
THE MARINERS AND SAFECO FIELD
Previous Mariners owner Jeff Smulyan paid about $77 million
($132.4 million) for the team in 1989. The current owners
bought him out for about $106 million ($161.1 million) in
1992. Although Smulyan cried poor, that was a 22 percent
increase in real value in just four years. It did not take long
before the Mariners’ “stadium problem” surfaced. In the usual
way, the new owners made it clear that they were losing
millions of dollars on the team annually. Our look at paper
profit reports in pro sports in Chapter 4, along with the return
earned by Smulyan, would lead us to doubt these claims.
Nonetheless, under this pressure, a stadium referendum issue
went before state voters.
THE PEOPLE SAY NO BUT SPEND $417 MILLION
($583 MILLION)
On September 19, 1995, in a special election, the stadium vote
in King County failed by a narrow margin (50.1percent to 49.9
485

percent). Voters were unwilling to raise the sales tax by half of
a percent toward the projected $240 million ($335.8 million)
cost of the new stadium plus repairs to the Kingdome. As
would be expected when there are believable threat locations,
team executive John Ellis announced that the Mariners would
be sold. The Washington, D.C./Northern Virginia area seemed
primed to receive the team.
In the end, the team did not move. A special funding package
was passed in the state legislature. To see why this would
happen, imagine the situation facing state and local
politicians after such a close vote. If politicians let the vote
stand, the politically potent groups in favor of the stadium in
the more densely populated part of the state would feel
abandoned. Remember, if state politicians irritate the
politically less potent and help the politically powerful,
election chances can be enhanced. As one would expect, the
will of the people revealed through the referendum process
was overridden by the representative government.
Senator Slade Gorton and Governor Gary Locke found state
funds for the project and devised a way to ease a spending
limitation in effect for King County at the time. Easing the
limitation required the approval of the King County council,
which was granted in early October. In 1997, $414 million
($550 million) in spending was authorized, to be funded
through state bonds. The city also provided land worth $33
million ($44 million). The team agreed to pay $45 million
($60 million), plus any overruns beyond a specified buffer as
their share of the spending. Thus, the intended total was $459
million ($610 million).
However, as is usually the case with stadium issues and public
funds, the total amount spent missed the planned budget of
$459 million ($610 million) by a long shot. By the time it was
occupied in 1999, Safeco Field cost $517 million ($662
million) in total (Seattle Post-Intelligencer, Web site
[www.seattle-pi.com], April 13, 1999). Team owners increased
their share to $100 million ($128 million) and kept the public
share close to the original plan. The final tab was so much
larger than the initial authorization due to overruns beyond
the buffer specified in the original plan. Most of these
overruns were due to the relentless push by the Mariners’
owners to meet the 1999 opening date.
All-in-all then, the 1997 amount of $414 million ($550
million) planned public dollars for Safeco Field ended up at
about $419 million ($536 million) by 1999. Suppose Safeco
lasts as long as our assumption about the renovated
Kingdome, 20 years. Let’s keep in mind that, at 6 percent for
20 years, that’s about a $46 million annual payment in 2009
dollars just on the borrowing. Even though it appears that
most of the added spending was paid by owners, a look at
their lease shows otherwise.
THE MARINERS’ LEASE
The Mariners’ Safeco Field lease is very generous to the
Mariners owners. Of the owners’ promised share, $30 million
($40 million) came from naming rights (hence, Safeco Field)
486

http://www.seattle-pi.com

http://www.seattle-pi.com

granted to the team in the lease. These are just stadium
revenues collected for the stadium authority by the owners.
The authority could certainly have just collected the revenues
and kept the money. The Mariners also get all revenues
generated by use of the stadium, including parking and
concessions. The team gets all of this for $700,000
($895,800) in rent, down from the $1.39 million ($1.78
million) it paid at the Kingdome. The lease also includes some
profit-sharing arrangements that can only come to pass under
extremely unlikely circumstances. At most, James Quirk and I
(1999) calculated that the team might pay $2.6 million ($3.3
million) per year. The team has also agreed (so far) to pick up
the $10 million ($12.8 million) in annual maintenance. The
total amount the Mariners will pay each year to play in Safeco
Field is between 12 and 14 percent of their 1998 total revenues
of about $89.7 million ($117.3 million).
The lease also obligates the Mariners to stay in Seattle until
2020. But who would not agree to that stipulation?
Essentially, the state built the owners a premiere facility and
turned the keys over to them for less than 10 percent of their
upgraded total revenues. The stipulation that they must stay
and enjoy this for 20 years actually is a subsidy guarantee.
The result of the building of Safeco Field is a substantial
injection of revenue into the Mariners at significant public
expense. Fans kept their team, all taxpayers are footing the
bill, and as detailed earlier, it was not clear at the time of the
funding agreement whether the team would improve in the
long run. As it turns out, the Mariners did end up a better
team during the regular season. Their winning percent rose
from 0.508 in the five years prior to moving into Safeco to
0.583 after they moved in 1999. However, the team was less
successful in the postseason, appearing only once after
moving into Safeco as compared to twice in the five years
prior to the move.
THE SEAHAWKS’ NEW STADIUM/EXHIBITION CENTER
New stadium demands by the previous owner of the NFL
Seahawks, Ken Behring, began in 1995. Behring claimed that
his Kingdome lease had been invalidated using some trumped
up geographical data showing the Kingdome’s vulnerability to
earthquakes. Behind the scenes, Behring had been negotiating
a new stadium in the Los Angeles area. When Seattle officials
refused to build a new stadium, he proceeded to load his
Seahawks into moving vans, turned south on I-5, and headed
for Hollywood Park, California. The action was stopped after
the NFL stepped in to remind Behring that he could not make
such a move without league approval.
During the subsequent “time out,” billionaire Paul Allen
purchased an option to buy the team. The option contract put
conditions on his willingness to pay Behring’s $250 million
($350 million) asking price for the team. The conditions were
pretty predictable. Allen stated that the $250 million ($350
million) would be forthcoming only if the people of the state
of Washington would cover the lion’s share of a new stadium/
exhibition center. If not, then Allen would just let the option
expire. If Allen did not purchase the team, it was almost
487

certain that the team would leave Seattle. Allen even offered
to personally finance the cost of running a referendum
election on the issue. The exhibition center component of the
package that included demolition of the Kingdome was
virtually ignored, but as noted in the earlier discussion on
misinformation, it was an essential element of the deal.
VOTING ON A “COMPLEX” SIMPLE REFERENDUM
On June 17, 1997, for the second time in two years, state
voters were asked to vote on a stadium subsidy. The total cost
of the facility was set at $425 million ($565 million). The
roughly $325 million ($472 million) public share was clearly
stated, and the vote was a simple yes or no. But underlying
this simple decision was a confusing hodgepodge of funding
mechanisms:
• A sales tax credit of $101 million ($134 million), an amount
that usually would go to the state fund now stayed in King
County to cover stadium costs
• New lottery games expected to add $127 million ($169
million)
• A ticket tax of $52 million ($69 million)
• A room tax extension in King County of $40 million ($53
million)
• A stadium parking tax of $4 million ($5.3 million)
• A $27 million construction tax break ($36 million)
• $14 million ($19 million) in interest on a $50 million ($66
million) escrow account set up by Paul Allen (in essence,
Allen just added the interest to his contribution)
The total of $365 million ($485 million) in the plan,
mentioned earlier, was more than the $325 million ($472
million) public share that appeared on the ballot because of a
contingency reserve for cost overruns and money for ball
fields around the state. The ball fields were viewed as a return
to the eastern part of the state for supporting a stadium that
pretty much benefited westside residents.
In addition to the funding mechanisms, the Kingdome debt
would be retired and that facility torn down. Remember that
retiring the Kingdome debt and demolishing the structure did
two things. First, all remaining Kingdome debt was rolled into
the bill for the new stadium. Second, with the Kingdome gone,
the new exhibition center was pretty much guaranteed
monopoly status for a wide variety of local events.
Allen promised to pay the balance of $100 million ($133
million) on the $425 million ($565 million) facility and cover
any cost overruns beyond a specified buffer. It is important to
note that the anticipated lease (at the time) was as generous
as the Safeco Field lease. For example, Allen would keep all
sponsorship revenues. Again, because the stadium authority
could just as easily collect and keep these revenues
themselves, having Allen do it for them and pay the proceeds
as part of his $100 million ($133 million) promise actually
raised the public share.
488

Here is how the election played out. The issue passed 51.1
percent to 48.9 percent. Unlike the Mariners issue, which
failed by about the same margin as this one passed, elected
representatives took no further action behind the scenes. The
Seahawks moved to their new stadium in 2002. The final tally
on the stadium/convention center hit about $430 million and
didn’t miss the original $425 million or so (76 percent public
money) originally approved. So we are safe calling the public
share $365 million ($485 million). The annual payment under
the same 6 percent for 20 years would be about $32 million
annually just on the borrowing. Once again, it was unclear
whether the Seahawks would improve with this revenue
injection or not. And the result was marginally more success
than in the Mariners case; the team averaged one more win
per year in their new stadium compared to the three years
before 2002 and raised their number of Wild Card
appearances in the play-offs from one to two.
COST COMPARISON WITH THE KINGDOME
This summary of events brings us to a point of cost
comparison between keeping the Kingdome and building both
a new baseball park and a new football stadium. As noted at
the outset, the annual cost in 2009 dollars of keeping the
Kingdome in action would have been around $38 million per
year for 20 years. The annual cost in 2009 dollars of just the
borrowing on the two new facilities totals $78 million
annually ($46 million for the Mariners and $32 million for
the Seahawks). In addition, the previous borrowing cost
remaining on the Kingdome would be rolled into the new
funding amount. At least some of the same subsidy also would
be present since public property and forgone taxes also went
along with the new spending arrangement. It’s entirely
possible that the new facilities have an annual cost of over
$100 million in 2009 dollars, easily three times the annual
cost of keeping the Kingdome in operation. Just because the
new facilities cost dramatically more does not mean they
should not have been built. From Chapter 10, we know that
this amount would have to be covered by additional buyers’
surpluses and positive externality values because economic
activity and development values typically are quite small.
Because surpluses and externalities are most likely in the tens
of millions of dollars (again, from Chapter 10), it appears that
the logic of overly high venue subsidies struck again in Seattle.
THEY SHOULD HAVE KNOWN BETTER: THE
KINGDOME
The Seattle example is clear testimony to the power of pro
sports teams in the political process. After all is said and done,
the people of Washington may end up kicking themselves, but
it will be because they ignored their own history. The history
of the Kingdome certainly was available to Seattle voters at
the time they were voting on the roughly $779 million in
stadium spending for baseball and football. It is a history of
delays, frustrations, and cost overruns. Let’s look at that
history and then revisit the estimates of the subsidy to the
quickly forgotten Kingdome.
THE HISTORY OF THE KINGDOME
489

In 1966, an election was held in King County for a $25 million
($165 million) domed multipurpose stadium to draw NFL and
MLB expansion franchises to Seattle. That election failed, but
just two years later, in 1968, a $40 million ($245 million )
referendum passed. No doubt, this change of heart followed
the granting of the expansion MLB Pilots franchise to Seattle
in 1968.
During the debate surrounding the 1968 election, the
projected opening of the Kingdome was set at 1970, so the
Pilots played the 1969 season in a dilapidated minor league
park, old Sick’s Seattle Stadium. After the election results
were in, a revised opening date of 1972 was given. A delay of
two years in just a few days! The Pilots went bankrupt after
one season and left Seattle for Milwaukee to become the
Brewers, beginning play there in 1970. A lawsuit by the city of
Seattle forced MLB to promise the city an expansion
franchise. However, a team was not yet committed, so the
voters were facing the prospect of a wonderful new domed
stadium with no pro team occupants.
And perhaps the timing of the arrival of their teams helps
explain the ensuing delays in building the Kingdome. After
rejecting one stadium site, the final site on the edge of
downtown was approved in 1972. Ground had not even been
broken by the revised date that the facility was supposed to
open. This was not the end of the delays. By 1973, the revised
opening date was pushed back to 1975. Eventually, the facility
opened in 1976, in time for the first Seahawk season (the
Mariners began play in the Kingdome the following year). If
one had believed the original debate, this was a six-year delay.
In addition, by 1976, the final cost of the stadium had
mounted to $67 million ($251 million), about 3 percent more
than the original referendum statement in 1968.
THE ONGOING KINGDOME SUBSIDY
The Kingdome is one of the stadiums included in the Quirk
and Fort (1992) subsidy analysis cited in the preceding
chapter. Table 11.6 shows that Kingdome subsidy
calculation, following the basic subsidy formula of Chapter 10.
Circa 1989, the Kingdome was losing about $1.5 million ($2.6
million) just from operations. Depreciation, the opportunity
cost of capital, and forgone property taxes totaled about $12.8
million ($22.0 million). This all added up to an ongoing
annual subsidy of about $14.3 million in 1989 ($24.6 million).
Measured in 2009 dollars, the Kingdome cost about $251
million. For that price, those paying the bill bought into a
$24.6 million annual subsidy. Although not technically
correct because the subsidy probably was larger early on,
when the Kingdome was demolished at the ripe old age of 20
years, the subsidy total was something on the order of $510
million in 2009 dollars. Let’s not forget the federal
component. Zimmerman (1997) calculates the total Kingdome
federal subsidy at an additional couple of million dollars. It
seems a lesson should have been learned from the Kingdome,
but it appears that it has to be learned anew every generation
or so.
490

491

SECTION 8
Altering the Politics of
Facility Outcomes
Venue subsidy outcomes are political outcomes. Hopes that
either review of elected officials by their constituents at
reelection junctures or appeals to direct democracy will
ameliorate subsidy results through the rational actor
approach are unrealistic. But that same rational actor
approach does give us some insight into how venue subsidy
outcomes might be improved in the future.
Because they are political outcomes, only a political response
can change them. This will require altering the political status
quo that now favors lavish stadium and arena construction.
According to the theory presented in this chapter, we see
massive and repeated stadium subsidies because special
interest politics dictates that outcome; it is in the best interest
of elected officials to have it that way. Two things would have
to change to alter the politicians’ current perceptions.
First, a new group must enter the fray, become politically
potent, and wrestle its way into the triangle portrayed in
Figure 11.2. Fan and taxpayer advocates would have to
displace the current team and press, construction, and
business beneficiaries. Forming a coalition that could displace
a politically powerful group is an expensive proposition, much
like forming a players’ union, as described in Chapter 9. A
broadly distributed group of current losers would have to be
educated about their losses. In addition, free-riding behavior
would have to be overcome, and finally, the organization
leaders would have to carry the fight and win a place in the
stadium policy arena.
These types of changes do not happen often in politics, but
they do happen. The environmental movement is one example
of a group’s gaining political power. Once a loose-knit group
of outdoor enthusiasts, the environmental movement is now
potent enough to have its own set of laws and policies under
the National Environmental Policy Act (1969) and the
Endangered Species Act (1973). Lately, environmentalists
have had impacts on global economic planning processes in
the international arena.
The rise of fan organizations on the Web is a sign that such a
fan/taxpayer movement might be underway. In recent
testimony before Congress, Frank Stadilus, CEO of United
Sports Fans of America (USFans), a Web-based fan interest
group, was there with MLB commissioner Bud Selig, former
senator and Blue Ribbon Panel 2000 member George
Mitchell, commentator Bob Costas, syndicated columnist and
Blue Ribbon Panel 2000 member George F. Will, and (yours
truly) the author of your textbook. That Stadilus was invited
to testify reveals that some members of Congress recognize
fan/taxpayer groups. However, it will take more than just
testimony; political potency requires funding and action.
492

The second thing that would have to happen to alter venue
subsidy outcomes follows from the first. The second step
would be reducing or eliminating the ability of owners, acting
through leagues, to maintain believable threat locations.
However, reducing the leagues’ power over team locations
would require changing the status quo in sports politics at the
federal level, which is the topic of our next chapter.
493

SECTION 9
Chapter Recap
Stadium subsidy outcomes are politically determined. The
rational actor explanation of political outcomes explains why
this is so. Because voters are rationally ignorant and tend to
free ride on the voting behavior of others, only those with the
most to gain typically participate in politics. Elected officials
know this, and the rational actor explanation predicts that
politicians will concentrate benefits to politically potent
groups and spread the costs out as thinly as possible over the
rest of the taxpaying public.
Owners are especially important elements of this political
process. Owners gain from new stadiums. Attendance, gate
revenues, and venue revenues all increase with a new
stadium. In addition, if the fans in a given team’s market are
willing to pay, then a new stadium can also raise a team’s
winning percent. It appears that increased revenues go hand
in hand with increased team quality, but this does not happen
for all teams that receive a stadium subsidy. Given that
owners always benefit from subsidies, we should expect team
owners to pursue them vigorously, whether they end up
improving the quality of their team or not.
An especially important part of the stadium subsidy process is
the careful control of believable threat locations by the
leagues. If a given locality fails to meet the subsidy demands
of a particular owner, other locations hungry for a team just
might. Thus, team owners have the upper hand in the
negotiations over stadium subsidies.
In this setting—political determination of subsidies and team
owners with the upper hand in bargaining—the rational actor
explanation makes a clear prediction. Powerful pro sports
team supporters will obtain stadium subsidies for team
owners that are paid for by nonsports fans. The bulk of the
total cost is paid in small per capita amounts. This is precisely
what we observe.
There is little hope that direct democracy can curb venue
subsidies. Participation and information bias will plague
direct democracy approaches, just as they do other political
decisions. Those in control of the ballot alternatives can pose
the issue to voters in such a way that preferences for a
medium-level outcome can be overcome in favor of large-scale
spending on elaborate facilities. The data show direct
democracy neither reduces amounts spent nor the public
share of spending. In addition, it appears that those in control
of ballot alternatives have also nudged spending toward
higher levels.
Seattle offers an example of nearly all of the ideas presented
in this chapter. The MLB ballpark issue was denied public
funding by referendum, but state legislators approved a
494

funding package anyway. The NFL stadium/exhibition center
was approved by referendum. The result is about $779 million
(2009 dollars) in facility subsidies to extremely rich team
owners with very little in return to those footing the majority
of the bill. They should have known better because similar
problems plagued the Kingdome 20 years earlier.
Because these outcomes are determined by special-interest
politics, the only realistic hope for improving subsidy
outcomes is a change in the balance of power toward
taxpayers. Antisubsidy groups will have to become politically
potent before changes will occur. This is unlikely to happen,
but reversals of this type have occurred. An integral element
in the reversal of fortunes for taxpayers would be in reducing
the power of leagues over team location.
495

SECTION 10
Key Terms and Concepts
You should have run into each of these in pop-ups in the text
of this chapter:
• Rational actor explanation
• Market failures
• Wealth redistribution
• Rational level of voting
• Subsidies raise owner profits
• Subsidies and team quality
• Subsidies and believable threat locations
• Subsidies and direct democracy
• Biased turnout
• Biased information
• Control of ballot alternatives
496

SECTION 11
Review Questions
1. What is the rational actor explanation of political
outcomes? What do voters seek? What do politicians
pursue? What is the very special prediction from this
explanation?
2. What types of market failures characterize sports
markets?
3. Beside market failure remedies, what other economic
remedy do people seek from government?
4. Why doesn’t everybody register to vote?
5. Why doesn’t everybody turn out to vote?
6. Why doesn’t everybody who turns out necessarily cast a
vote on every issue on the ballot?
7. Why is the rational level of voting less than the total
number of eligible voters?
8. How does rational voting contribute to the concentrated
benefits and dispersed cost outcome that characterize
rational actor politics?
9. How do subsidies increase owner revenues? Give
examples.
10. How do subsidies decrease owner costs? Give examples.
11. Are owners always better off with subsidies than without
them? Explain.
12. Insofar as leagues do not negotiate directly with host
cities, how do they contribute to stadium subsidy
outcomes?
13. Why don’t those paying for venue subsidies vote the
responsible politicians out of office at the next election?
14. What is direct democracy? What are the reasons behind
the hopes that direct democracy will improve venue
subsidy outcomes?
15. What is the source of turnout bias? What is the source of
information bias? What does it mean to have control of
the ballot alternatives that actually go before voters in
direct democracy? How do turnout bias and information
bias reduce the chances that direct democracy will curb
venue subsidies?
497

SECTION 12
Thought Problems
1. Using Figure 11.1, explain why the general constituency,
as opposed to special interests, is not connected to the
triangle that determines policy outcomes.
2. How do subsidies increase owner profits? Remember,
profits consist of both revenue and cost considerations.
3. Owners will always be better off with subsidies, but the
quality of their team may not increase. Why?
4. Why will new stadiums increase attendance? In analyzing
the impact of subsidies on team attendance, why is it
important to compare attendance with those teams that
do not have a new stadium?
5. In analyzing subsidies and winning percent outcomes,
why is it not important to compare the winning percents
of teams with a new stadium with those without one?
6. Why is it so hard to figure out whether increased revenues
from subsidies cause team quality to improve? How
would you be able to determine if subsidies improved a
team’s quality?
7. Walk through the evidence from the study by Hamilton
and Kahn (1997) that proves the move into Camden Yards
raised the quality of the Baltimore Orioles. Was it
necessarily the move into the new stadium that caused
this outcome? Why or why not?
8. Walk through the evidence from my study (1999) that
shows the move into the Ballpark at Arlington did not
raise the quality of the Texas Rangers. Why would the
owner of the Rangers reduce spending on talent despite a
fairly large revenue injection from moving into the new
ballpark?
9. Why might an MLB team’s local TV contract value be the
leading indicator of whether team quality will improve
after moving into a new stadium?
10. Explain the stadium revenue disadvantage logic behind
small-revenue market owners’ pleas for new stadiums. Is
that argument believable? If so, under what
circumstances?
11. How does the greater legal mobility of teams in the NFL
dampen the propensity of local government hosts to
provide subsidies to “their” NFL teams?
12. There are two reasons why tearing down the Kingdome
might have been a bad idea, economically speaking.
Provide these two reasons. Even if the value of using the
Kingdome into the future was less than the cost of repairs,
what is one reason it should have remained standing?
498

13. Why might those in control of the ballot alternatives that
confront voters in direct democracy prefer to influence
voters toward higher rather than lower spending on
stadiums? Explain using Figure 11.2.
14. Explain how the data on spending on stadiums and
arenas refute the idea that direct democracy curbs venue
subsidies.
15. Explain how the data on public spending on stadiums and
arenas support the idea that there is some control of the
alternatives confronting voters under direct democracy by
those who prefer higher rather than lower spending levels
on stadiums and arenas.
499

SECTION 13
Advanced Problems
1. Control of believable threat locations is an essential
activity for pro sports leagues. If laws aimed at curtailing
MLB’s power over team location were passed, the losses
to MLB from these laws could be dramatic. Suppose that
the interest rate is 5 percent, the value of power over team
locations is $30 million annually, and the planning
horizon is 10 years. How can it make sense for MLB to
spend around $1.8 million lobbying against these laws for
the year 1999?
2. What is the one piece of evidence suggesting that the
improvement in the Baltimore Orioles’ winning percent
after moving into Camden Yards was not due to the extra
revenue generated by the new stadium?
3. Why have stadium revenues, as a percent of all revenues,
grown in importance in MLB but not in the NFL?
4. Explain how a new stadium can increase an NFL team’s
ability to compete on the field despite the presence of a
salary cap and nearly complete revenue sharing in the
NFL. Does the observation that teams all violate the cap
anyway have any bearing on this explanation?
5. Using the rational actor politics model in Figure 11.2,
explain failed attempts at obtaining stadium subsidies.
6. In the Learning Highlight: The Stadium That Didn’t Get
Built and Then Did, voters and politicians in the St. Louis
area denied Bill Bidwell a new stadium and then turned
right around and built the TransWorld Dome, luring the
Rams to town. What special problem that plagues a
league that shares revenues such as the NFL might have
led to this interesting result?
7. Suppose there is going to be a stadium vote and the voter
preferences are as shown in Figure 11.3. Further, suppose
that voters believe that “no team” will occur if the election
fails to pass. If the elaborate stadium option is put before
the set of voters shown, why will the vote be 3–2 in favor
of the subsidy? What will be the voting result if voters
believe the team will stay and play in the “obsolete
stadium”? Redraw the figure to show where voters C and
D would have to be located along the line for the issue to
fail. Given your answers, is it always the case that control
of the ballot alternatives will lead to an elaborate new
stadium? Explain.
8. Using Figure 11.2, explain the Safeco Field outcome in
Seattle. Remember, in that case, the voters said no in a
referendum, but the stadium was funded and built
anyway at a higher level of spending than was originally
refused in the referendum.
500

9. Even though the ballot issue was a simple yes or no to a
new stadium/ exhibition center for the Seahawks, explain
how the issue actually was much more complex than
portrayed to the voters. Did this impact the perceptions of
people voting on the issue? Did they even know how to
figure out the costs? Why is this consistent with the idea
that politicians know that taxpayers are rationally
ignorant?
10. Suppose you were going to start a fan protection action
group. Describe the difficulties you would confront in
becoming a politically viable interest group using the
diagram in Figure 11.2.
501

SECTION 14
References
Coates, Dennis, and Humphreys, Brad R. “Proximity Benefits
and Voting on Stadium and Arena Subsidies,” Journal of
Urban Economics 59 (2006): 285–299.
Fort, Rodney. “Direct Democracy and the Stadium Mess,” in
Sports, Jobs, and Taxes: The Economic Impact of Sports
Teams and Stadiums. Roger Noll and Andrew Zimbalist, eds.
Washington, D.C.: The Brookings Institution, 1997.
Fort, Rodney. “Stadium Votes, Market Power, and Politics,”
University of Toledo Law Review 30 (1999): 419–442.
Fort, Rodney. “Stadiums and Public and Private Interests in
Seattle.” Marquette Sports Law Journal 10 (2000): 311–334.
Hamilton, Bruce, and Peter Kahn. “Baltimore’s Camden Yards
Ballparks,” in Sports, Jobs, and Taxes: The Economic Impact
of Sports Teams and Stadiums. Roger Noll and Andrew
Zimbalist, eds. Washington, D.C.: The Brookings Institution,
1997.
Keating, Raymond J. “Sports Pork: The Costly Relationship
Between Major League Sports and Government,” Policy
Analysis No. 339. Washington, D.C.: Cato Institute, April 5,
1999.
Noll, Roger, and Andrew Zimbalist, eds. Sports, Jobs, and
Taxes: The Economic Impacts of Sports Teams and Stadiums.
Washington, D.C.: The Brookings Institution, 1997.
Quinn, Kevin G., Paul B. Bursik, Christopher P. Borick, and
Lisa Raethz. “Do New Digs Mean More Wins? The
Relationship Between a New Venue and a Professional Sports
Team’s Competitive Success,” Journal of Sports Economics 4
(2003): 167–182.
Quirk, James, and Rodney D. Fort. Pay Dirt: The Business of
Professional Team Sports. Princeton, NJ: Princeton
University Press, 1992.
Quirk, James, and Rodney D. Fort. Hard Ball: The Abuse of
Power in Pro Team Sports. Princeton, NJ: Princeton
University Press, 1999.
Shropshire, Kenneth L. The Sports Franchise Game: Cities in
Pursuit of Sports Franchises, Events, Stadiums, and Arenas.
Philadelphia, PA: University of Pennsylvania Press, 1995.
Swindell, David. “Public Financing of Sports Stadiums: How
Cincinnati Compares,” Policy Solutions. Dayton, OH: The
Buckeye Institute for Public Policy Solutions, 1996.
Zimmerman, Dennis. “Subsidizing Stadiums: Who Benefits,
Who Pays?” in Sports, Jobs, and Taxes: The Economic Impact
of Sports Teams and Stadiums. Roger Noll and Andrew
502

Zimbalist, eds. Washington, D.C.: The Brookings Institution,
1997.
503

SECTION 15
Suggestions for Further
Reading
Surely, I can come up with some.
504

CHAPTER 12
Taxes, Antitrust, and
Competition Policy
Anyone who quotes profits of a baseball club is
missing the point. Under generally accepted
accounting principles, I can turn a $4 million
profit into a $2 million loss, and I can get
every national accounting firm to agree with
me.
—Paul Beeston, Toronto Blue Jays Vice President
Zimbalist, 1994, p. 62.
CHAPTER OBJECTIVES
After reading this chapter, you should be able to:
• Know the special tax and antitrust advantages
afforded to owners of sports teams.
• See the value of the special status afforded to
owners of teams and their players and the costs
of these benefits to fans and other taxpayers.
• Appreciate the rational actor model of politics
explanation of why pro sports owners enjoy
special tax and antitrust status.
• Understand the somewhat pessimistic view that
extremely unlikely changes will have to occur
before politicians see fit to alter these special
advantages.
• Discuss the role that competition policy can play
in fixing the ills attributed to sports market
power.

SECTION 1
Introduction
The United States Football League (USFL) disbanded after the
jury decision in its lawsuit against the NFL in 1986. The USFL
alleged that the NFL had violated the antitrust laws by
tampering with the USFL’s network contract negotiations.
The USFL sought $567 million ($1.1 billion) in damages that
would be trebled under the law to $1.7 billion ($3.3 billion).
The jury found that the NFL had tampered with the
negotiations, but then something interesting happened. The
total value of the damages to the USFL was set at $. How
could these antitrust violations be worth only $1? The jury
cited USFL mismanagement as the reason for their paltry
award. However, another answer lies in the ability of
dominant leagues to drive out rivals. It could be the case that
after the USFL failed to compete against the existing
dominant NFL that it was practically worthless for all intents
and purposes.
In this final chapter on pro team sports (we turn to college
sports in Chapter 13), we will extend our look at the
relationship between local, state, and federal governments
and the sports business. The chapter offers a three-step
approach to understanding the special tax and antitrust status
of pro sports. First, we will identify and describe these two
situations. Second, we will see how they are valuable to team
owners and players. Third, we will use the rational actor
description of democracy we developed in the previous
chapter to gain some insight into why leagues are able to
maintain these special benefits. We wrap up the chapter, and
our look at pro sports, with a look at competition policy. As we
will see, all of the evils commonly attributed to sports can be
remedied with a stiff dose of competition, but our rational
actor model suggests that dramatic reversals of sports policy
in favor of fans and other taxpayers should not be expected
without some very important changes to the system.
506

SECTION 2
Taxes
Bill Veeck, the famous baseball business entrepreneur, was no
fan of business income taxes. He felt that MLB deserved a tax
break. In The Hustler’s Handbook (1962), he says, “Look, we
play the ‘Star Spangled Banner’ before every game—you want
us to pay income taxes too?” In his efforts to avoid personal
income taxes, he created a lucrative special tax situation for
pro sports team owners. This raises the issue of sports
accounting and tax implications (also touched on in Chapter
4). In this section, we will examine the special tax situation of
pro sports. Later in the chapter, we will look into why pro
sports enjoy special tax status in the first place.
PLAYER ROSTER DEPRECIATION REVISITED
When a team is purchased, the new owner reorganizes it as a
new business enterprise. This accomplishes two things. First,
the player roster and other assets like the value of broadcast
contracts are depreciable under tax law. In Chapter 4, we
discussed the validity of such an approach, noting that players
do not depreciate and, even if they did, they already are on the
expense side of the ledger for tax purposes. Player roster
depreciation can easily erase any team taxable income for
owners.
The second thing that reorganization accomplishes follows
from the form of reorganization. As we saw in Chapter 4,
because roster depreciation typically generates huge losses for
the first few years of ownership, millions can be saved on
owners’ other Form 1040 earnings. However, the owners can
only do this if they are able to pass losses straight through to
their personal income tax forms. The form of reorganization
must facilitate this pass-through, typically as a partnership or
Subchapter S corporation. In the Seattle Supersonics example
in Chapter 4, an old version of roster depreciation was in
force. For the purposes of understanding the relationship
between government and the sports business, it is informative
to trace the historical development of this type of special tax
treatment to the present.
In 1959, Bill Veeck bought the Chicago White Sox for the first
time (he would own them again in the 1970s). Always an
innovator on and off the field (recall the Chapter 2 Learning
Highlight: Bill and Mike Veeck), he argued that players
“waste away” like livestock, and because livestock could be
depreciated, why not players? As covered in Chapter 4,
players bear the loss due to their depreciation, not owners. In
addition, tax rules also allow the annual cost of players, their
salary, to be expensed instead. While the justification for this
roster depreciation is suspect, the IRS agreed and roster
depreciation was born (Quirk and Fort, 1992; Coulson and
Fort, 2009).
507

Bud Selig, who owned the Milwaukee Brewers prior to his
current reign as MLB Commissioner, played a role in the
development of roster depreciation back in 1970. A group led
by Selig bought the bankrupt Seattle Pilots in April of 1970 for
$10.8 million ($59.3 million) and moved them to Milwaukee,
the second incarnation of the Brewers. Selig attributed $10.2
million ($56 million) of the purchase price to the player
roster, set up his depreciation schedule accordingly, and
submitted his tax forms. The IRS challenged Selig’s claim that
94 percent of the purchase prices could be attributed to the
player roster and demanded a large tax adjustment. Selig sued
(Selig v. U.S., 565 F. Supp. 524) and eventually won the case.
After a few more IRS challenges of subsequent large roster
depreciation claims by other owners, the Tax Reform Act of
1976 set the level of roster depreciation at 50 percent of the
purchase price and a five-year depreciation schedule. Any
claims over these limits had to be justified by the taxpayer.
This version of roster depreciation was still in force and
applicable to the 2000-2001 Seattle Supersonics example in
Chapter 4 and held until the Tax Act Amendments of 2004.
Under these recent amendments, owners can depreciate 100
percent of the team sale price over no more than 15 years.
The IRS justified this change on two grounds (Wilson, 2004).
First, the IRS would spend less taxpayer money litigating
depreciation claims by owners. In addition, the IRS argued
that owners would actually pay more taxes under the new
rules. The first of these justifications is obvious, but let’s
investigate the second in some detail. In what follows, we’ll
look only at the tax shelter value of the pass-through because
any reasonable amount of roster depreciation should push
team taxable net revenue to zero.
As always, we need to make a few assumptions because this is
a complex issue. We’ll examine an MLB team without any
regional television ownership or stadium. Further suppose net
operating revenue, nonroster depreciation, and amortization
remain constant over time. In addition, the team is organized
as a pass-through for tax purposes (Subchapter S corporation
or partnership); profit or loss is claimed on the owner’s
personal Form 1040 at the personal income tax rate. For now,
also suppose that net revenue after subtracting nonroster
depreciation and amortization are positive but less than the
value of roster depreciation. Finally, IRS rules require
straight-line depreciation for “intangible” assets like this.
A little notation and some algebra become extremely useful at
this point. Under the old plan, roster depreciation is half the
purchase price P over five years, that is, Dold = (0.5 × P)/5.
Under the new plan, roster depreciation is the entire price
over 15 years, or Dnew = P/15. Let G be the gain (or loss) under
the new version of roster depreciation, compared to the old
version:

where ? is the owner’s personal income tax rate (also assumed
constant over time), E is net revenue after nonroster
508

depreciation and amortization, and r is the real rate of interest
(because the gain covers discounted present values over
different lengths of time). Clearly, the gain hinges upon the
different paper loss pass-through amounts, Dold − E and Dnew
− E, and the difference in the periods of time. Our earlier
assumption guarantees that both of the pass-through values
are greater than zero.
Because the numerator of each term on the right-hand side is
constant with respect to time by virtue of our previous
assumptions, we write G as:

Now, if G < 0, the IRS is correct—the tax shelter value of the new version is less than the old version, and owners will pay more taxes. Otherwise, owners will have greater shelters under the new version. At this point, let’s make the algebra really easy by assigning reasonable numerical values to some of the variables in G. Our point of reference is the year of the tax change, 2004. First, let P be the average team sale price over the previous decade or so, $355 million 2004 dollars (Fort, 2006). Further, let the real rate of interest be r = 0.03, the typical real rate of growth in the economy at large. Finally, any owner almost certainly faces ? = 0.35, the highest marginal income tax rate. These choices yield the following: Dold = (0.5 × 355)/5 = 35.5; Dnew = 355/15 = 23.7; and with r = 3 percent the summations are 11.9 and 4.6 for the fifteen year and five year periods, respectively. Substituting these values into G, along with ? = 0.35, yields the following: The graph of G = 42.7 – 2.6E in Figure 12.1 shows that the IRS could possibly be right. If net operating revenue after nonroster depreciation and amortization but before roster depreciation is subtracted exceeds $16.4 million, then the owner is actually worse off under the new version compared to the old because G < 0. Apparently this is what the IRS thinks will be true. But Figure 12.1 also shows that as E decreases, the gain under the new version of roster depreciation increases. The gain turns positive once E < $16.4 million and grows larger from there. Once E = 0, the gain is $42.7 million. The gain grows ever larger as E continues to become increasingly negative. So, whether or not the IRS ends up being correct depends on the level of E, the net revenue from team operations. Some additional data, however, cast doubt on the IRS claims. Forbes listed 2004 operating revenues of MLB teams before taxes, depreciation, and amortization, and the average of those values is $4.4 million. Reports from the same source for the other leagues are available only up to 2003; calculating the averages yields $9.6 million for the NBA, $26.6 million for the NFL, and –$3.2 million for the NHL. Even before 509 nonroster depreciation and amortization, only the NFL is a candidate for G < 0 and higher tax payments under the new version of roster depreciation. At the average, NFL teams must have nonroster depreciation and amortization of less than $10.2 million as well in order to pay more taxes under the new version. All of this suggests that existing teams with some remaining depreciation will see some benefit, and it appears pretty likely that all future buyers are better off under the new version of roster depreciation. This last observation means that current owners who have already exhausted their depreciation will nonetheless be able to sell their teams for more because future owners will enjoy a larger tax shelter value. Estimates by financial analysts at Lehman Brothers back this up. They have publicly stated that the new version of depreciation will add about 5 percent to sports team values across all leagues (Wilson, 2004). Forbes’ total franchise values for all four major leagues in 2004 were $45.9 billion, so owners might expect a windfall gain of $2.3 billion, or an average of about $19 million for all 121 team owners. A number of notables have downplayed that owners are typically better off under the new version of depreciation (the quotes in this paragraph are from Rovell, 2004). NBA commissioner David Stern says: “It’s definitely not a significant motivating factor in purchasing a team. No matter what the tax treatment is, if the team isn’t profitable, it can’t provide a substantial benefit.” According to Jeff Smulyan, previous owner of the MLB Mariners: “When owners are losing millions of dollars in real money, it’s crazy to think that they’re fine with it because they are not paying as much tax.” As per David Samson, president of the Marlins: “The benefit of depreciation is far outweighed by the reality of cash losses when you are losing money operationally.” Now, especially at today’s high team purchase prices, the depreciation value may be relatively small, but none of these notables suggests it is zero. The analysis aforementioned, plus the assessment by Lehman Brothers, suggests millions of dollars gained, certainly for owners in all leagues except the NFL and probably for many individual owners in that league as well. While public statements downplay the value of depreciation, it is also true that MLB actively lobbied for the revisions and that the NFL supported the new version of depreciation (Wilson, 2004). Further, Smulyan is a principal in one of the groups vying to buy the new Washington Nationals. Don’t forget, while it should be the case that disputes over allowable write-offs would decline and so would their costs to the IRS, this is just sauce for the goose; legal fees for team owners should fall as well, increasing net operating revenues. Bill Veeck cracked his joke about taxes and the National Anthem relative to just the original version of roster depreciation. One can only expect that he’s smiling in that great grandstand in the sky over the new and improved version! IRS claims of higher tax payments appear unjustified, so owners will do better under the new rules, and their legal costs should fall. Later in the chapter, we’ll explore just how it 510 is that team owners end up with this special type of tax treatment. Roster depreciation is far from the end of the tax advantages enjoyed by owners. Another special tax decision allows owners both to postpone tax payments and to pay at a lower rate than the prevailing personal income tax rate. Things have changed in the past 30 years, but the first to observe this type of tax advantage was Okner (1974) who pointed out, “Nevertheless, the apparent irrationality of even the worst sports investment is largely explained by the effects of full utilization of the available tax benefits” (1974, p. 164). Ambrose (1985) also detailed the tax law changes of 1976 that determined tax status at the time of team sale. Let’s look at these tax advantages in more detail. CAPITAL GAINS TAX ADVANTAGES FOR SPORTS OWNERS Suppose an owner has run out the player roster depreciation advantage after 15 years. That owner faces two options: continue on or sell the team. Either choice confronts the owner with tax problems. REORGANIZING AFTER ROSTER DEPRECIATION RUNS OUT If the owner continues on after 15 years organized as a pass- through firm, the outcome can be very costly. At this point, the team is organized in order to pass all profit or loss directly through to the owner’s personal 1040 Form. If the pass- through is positive, then the owner pays taxes at the higher personal income tax rate. At the end of the 15 years, the owner starts paying back what was made in the first few years from the depreciation advantage in terms of the higher personal income tax rate. However, if the team were organized as a corporation, taxes would be at the lower corporate tax rate. It is in the owner’s best interest to change the structure of the business. Under IRS rules, any firm organized to pass business income through to a personal 1040 Form can reorganize for tax purposes just once. Once reorganized into a regular corporation or limited partnership rather than a pass- through corporation or partnership, the tax obligation is at the lower corporate tax rate. For owners who intend to keep the team, the special tax status of pro sports allows them to take advantage of the depreciation tax shelters for the first few years. Then, the owners can reorganize the team for tax purposes once the depreciation shelter runs out. After reorganization, the owners never pay back any of the shelter and suffer no other disadvantages into the future relative to any other corporation. IRS RULES AND THE AVOIDANCE OF EXCESS DEPRECIATION AND CAPITAL GAINS But what happens if the owner decides to sell, rather than continue to run the team reorganized for tax purposes? In this case, it is important to note that player rosters did not really depreciate during the tenure of the previous owner despite the 511 fact that the owner took out millions in roster depreciation. Therefore, the previous owner will sell the team for the discounted future profitability it represents. In a sense, this type of owner was in it for the tax shelter and sells the asset at market value after the tax advantages have run out. The market value of the team will include the tax shelter value to the next owner. There is one inconsistency with this practice. If the previous owner claims the roster is worth something when selling it to the new owner, it is an admission that the depreciation never really took place even though millions were claimed. In such a case, the IRS would typically rule that there was excess depreciation by the previous owner, and that owner would be liable for taxes on the amount of excess depreciation. When the IRS pursues excess depreciation, it does so at the higher personal income tax rate, and two things happen to the team owner. First, all of the income sheltered by the pass- through depreciation losses is taxable at the personal income tax rate. Second, any actual net operating revenue net of real depreciation is also taxable. Because the firm was organized at the time to pass that income through to the owner’s personal 1040 Form, it would be taxed at the higher personal income tax rate as well. However, no excess depreciation cases have ever been brought by the IRS against sports owners selling teams. For years, the IRS just agreed that the “new” team (for tax purposes) was a collection of new players ready to be depreciated all over again. By this IRS ruling, the previous owner did not have to reduce the sale value of the team because the roster was already completely depreciated. Thus, there never was any revelation of any excess depreciation. The new owner could start all over again doing precisely the same thing as the previous owner. This roster depreciation rollover suggests a strategy: Use the depreciation shelter for 15 years and then continue to run the team or sell it as an entirely depreciable new team. Of course, the previous owner still must pay capital gains tax on any increase in the sale value of the team. However, even here the depreciation shelter is valuable. Capital gains taxes are paid on the difference between the sale value of the team and the owner’s report of its profitability based on the book value of the firm over time. This reported profitability based on book value is called the basis at sale. Capital gains taxes are not meant to double tax the owner who would already have paid taxes on the profits during ownership. Because of the depreciation shelter, the book value of the team is very low during the early years as the owner reports losses or profits of zero. This means that the sale value minus the basis is larger during the early years of ownership than it would be without the depreciation shelter. However, the owner still pays less in taxes as long as the capital gains tax is lower than the personal income tax. Rather than paying at the personal income tax rate for what profits could have been shown, the owner only pays at the lower capital gains tax rate 512 at the time of sale. Owners enjoy the tax shelter from team losses on their 1040 Forms. Playing off the personal and capital gains tax gap, the depreciation shelter raises the amount subject to capital gains, but some of the amount would have been taxed at the higher personal tax rate without the shelter. From the perspective of the current owner, this is a very attractive special tax advantage to sports teams that cannot be had anywhere else. TYPICAL SPORTS TEAM ORGANIZATION FOR TAX PURPOSES Here is how a team is typically organized for tax purposes. For 15 years, organized as a pass-through firm, the team shows losses due to player depreciation. The owner earns the entire net operating revenue after depreciation but pays no tax on the team. Any paper losses go straight through to the owner’s personal Form 1040 as a tax shelter. After the depreciation shelter runs out, if the owners continue on, they reorganize to take advantage of a better tax situation and reduce future tax payments. If the owners sell, they have postponed payments on actual income taken out of the team and, at the time of sale, pay a possibly lower tax bill on their capital gain. Any new owner buys into exactly the same proposition starting from scratch. The sum of these tax advantages has been worth millions of dollars to sports team owners. James Quirk and I (1992) reported that these advantages could triple an owner’s rate of return on team ownership. An interesting question is why the sale price faced by new owners would not have already incorporated the tax advantage value in the first place. If all owners understand the tax advantages, then the price of the team should include imputed tax advantage values. This just means that the value of any tax advantages represents a value of owning a team. As such, it will be counted in the price that potential owners are willing to pay for the team. In such a case, no owner should earn more than a normal return including the tax shelter. As we will see in a later section of this chapter, sale prices appear to be greater than the normal rate of return would dictate. The personal income tax rate and the capital gains tax rate were set equal to each other in the 1986 tax reforms, and a large part of the tax advantage to team ownership disappeared. The player roster depreciation advantages remained but not the value of the difference between the capital gains and personal tax rates. However, the capital gains rate was changed to 20 percent in 1997. We are now back to the situation of the days before 1986. The capital gains rate is lower than the personal income tax rate, resulting in millions of dollars in tax value for owners. 513 SECTION 3 Antitrust The antitrust laws are designed to protect consumers from market power. In a nutshell, the laws dictate that firms cannot exercise their market power unhindered by government. They cannot attempt to extend their market power into other economic endeavors either on the input or output side, and they are subject to scrutiny by the Federal Trade Commission (FTC). The Department of Justice (DOJ) carries out the legal battle when the FTC finds an antitrust violation. Individuals also can bring antitrust actions if they feel a firm has violated these important market power laws. From our studies in Chapters 5 and 8, it would seem that sports leagues are prime candidates for constant vigilance under the antitrust laws. Leagues are organized to guarantee that teams enjoy market power through territorial exclusivity. Leagues also offer superior outcomes for member owners in negotiations with media providers, players, and host cities. Finally, leagues by and large put teams where they please, either through expansion or decisions over team relocation. All of these exercises of market power are possible violations of the antitrust laws. MLB once enjoyed antitrust exemption that resulted from the Federal Baseball decision (discussed in Chapter 8), but that only pertained to players and was undone by the Flood Act in 1998 (discussed in the next section of this chapter). And no similar restriction on antitrust intervention exists for other pro sports. Indeed, the only formal exemption enjoyed by sports leagues pertains to joint venture TV negotiations allowed by the Sports Broadcasting Act of 1961 (also discussed in detail in a later section of this chapter). A reasonable question is why pro sports leagues appear to enjoy all of the rest of their special treatment under the antitrust laws. Let’s revisit some of the issues presented in Chapter 8 and raise some new ones that lay bare the antitrust status of sports leagues. PLAYERS AND ANTITRUST The antitrust issue in player markets concerns how owners have turned their league market power over the production of sports into market power over players. If players have only one major league option, teams will recognize their superior position and try to take advantage of it. The history of league power over players, primarily through the reserve clause and player drafts, does not need to be repeated here, and players have gained significant ground in this regard. Long-standing league impositions such as the reserve clause were removed by court decision and the rise of unions. EASING OF RESTRICTIONS ON PLAYERS’ ANTITRUST RIGHTS 514 Players’ rights to take individual antitrust action against sports team owners remain quite restricted. Until 1998 the National Labor Relations Act required players to yield their individual antitrust rights when they entered into collective bargaining through certified unions. Recent changes allow players in all pro sports leagues to sue individually if behavior prohibited under the antitrust laws leads to a breakdown in labor negotiations. MLB players were the last group of players to gain this protection. Following the MLBPA strike in 1994–1995, Congress sought a way to reduce the chances of future work stoppages. The result was the Curt Flood Act of 1998. The Flood Act puts MLB players on the same antitrust footing as other professional athletes. However, this act only grants rights in the special case where antitrust violations lead to a breakdown in labor negotiations. LIMITATIONS OF THE FLOOD ACT Congress made it clear that the Flood Act was very limited, providing an itemized list of exemptions. These crucial limitations guarantee that market power over output will continue in its current form in every important respect. Specifically exempted are the following: • Minor league players (especially in draft matters) • The entire relationship between the minor and major leagues • Expansion and team location issues • Team ownership and ownership transfer • The relationship between owners and the MLB commissioner • All marketing of the entertainment product • The joint marketing of broadcast rights, under the Sports Broadcasting Act of 1961 Therefore, almost all MLB business remains exempt, from franchise exclusivity, team movement, and expansion to output management activities such as season length, play-off structure, and broadcasting sales. THE ROLE OF DECERTIFICATION In other leagues, players decertified their unions in order to sue under antitrust laws. As discussed in Chapter 9, NFL players voted to decertify the NFLPA as their collective bargaining agent. Then, individual players brought suit under the antitrust laws, earning free agency victories in the courts (McNeil v. NFL). After their victory, the union was recertified. However, players may face decertification dangers using this approach. For example, a videotape distributed by Michael Jordan to other NBA stars practically guaranteed that any reformation of the NBPA would have been contentious. Jordan urged the stars of the league not to allow decertification and argued that if decertification did occur, then the union leadership should be replaced. Had the players decertified the NBPA, the reformed union would look much 515 different from its current makeup. It would have happened due to the actions of a few stars rather than through majority rule of the union rank and file. FRANCHISE MOVES AND ANTITRUST Under the operating rules in all leagues, a three-fourths majority is required before a team can move (and unanimous consent is required in any move that puts a team in an existing franchise area). If that rule is not arbitrarily applied, leagues have nearly complete control over franchise moves and team location. This, as we have discussed in Chapters 2 and 5, is the main source of the market power enjoyed by individual team owners. THE RAIDERS CASE AND RECENT NFL TEAM MOVEMENTS One application of the three-quarters rule was found unreasonable in the famous NFL Raiders case (L. A. Memorial Coliseum Commission v. NFL). In 1981, Al Davis and the operators of the L.A. Coliseum sued the NFL under the antitrust laws when the league blocked the move of the Raiders from Oakland to Los Angeles. Essentially, the court said that the three-quarters rule had been applied in the interest of the other Los Angeles–area team, the Rams. They had been the Los Angeles Rams since 1946 and played in the same L.A. Coliseum the Raiders occupied until moving just up the freeway to Anaheim for the 1980 season. No doubt, Ram ownership feared lost revenues. The courts ruled in favor of Davis and the Coliseum operators. At the time, this ruling was expected to reduce the power of leagues over the movement of individual teams. However, over time, it barely made a dent in that power. The NFL cited the Raiders case as justification for not blocking the move of the Cleveland Browns to Baltimore in the late 1990s. However, it ignored the Raiders case entirely when dealing with Ken Behring’s proposed move of the Seahawks to the Los Angeles area. The NFL also hindered the move of the Rams from Los Angeles to St. Louis, discussed in Chapter 11, until a mutually agreeable payment was made to the league stadium fund. As you can see, the Raiders case has not effectively constrained the NFL in its ability to decide where to put teams. It is easy to see why the league behaved this way regarding these team moves. The NFL’s behavior had nothing to do with fan happiness and everything to do with owner and league wealth maximization. The move of the Browns to Baltimore saved one of the league’s owner-icons, Art Modell, a considerable amount of money at no cost to the league. A team was lost from one longtime NFL city, Cleveland, but gained by another longtime NFL city that had gone without a team for a few years, Baltimore. On net, Modell was better off and the league was not hurt, especially because it expanded back into Cleveland shortly thereafter. In the Seahawks case, a move from Seattle to Los Angeles would have erased a prospective large expansion fee from the future owner of the 516 extremely valuable and, at that time, vacant Los Angeles market. League members were better off with that decision, looking forward to expansion fees that are rapidly approaching the $1 billion mark. Finally, the NFL earned $43 million ($61.9 million) or so from the Rams when they moved to St. Louis after the 1994 season ended. CONGRESSIONAL HEARINGS ON TEAM LOCATION Because it is clear that leagues will continue to exercise power over team location and movement to the detriment of fans, government has the opportunity to step in under the antitrust laws. However, as history shows, government has chosen not to protect consumer (fan) interests in the case of franchise movement and location (a history of early congressional activity in sports, up to 1978, can be found in Johnson [1979]). As early as 1951, Congress investigated the power of leagues over team location and movement. Congressman Emanuel Celler began what became known as “The Celler Hearings” with some opening remarks: Organized baseball affords this subcommittee with almost a classroom example of what may happen to an industry which is governed by rules and regulations among its members rather than by the free play of competitive forces. Without knowing at this time whether such regulation is in the best interest of baseball because of its many unique characteristics, we may at least learn something of importance about how an industry operates itself instead of being forced to comply with the antitrust laws. However, at the end of the 1951 hearings, Congress chose not to revisit the antitrust status of MLB on the output side. Congress did, however, make it clear to MLB that its reluctance to expand westward was puzzling and encouraged the league to consider such movement. Shortly thereafter, the Dodgers and Giants moved to California. As discussed in Chapter 5, this instigation on the part of Congress probably doomed the potential for a reduction in MLB’s market power. When MLB moved west with its ready-made Dodger–Giant rivalry, the very successful PCL was reduced to minor league status despite its demands to join in as another major league. Congress could certainly have facilitated competition between MLB and the PCL but chose not to. Similarly, after “The Kefauver Hearings” in 1961 led by Senator Estes Kefauver, Congress again chose to let stand the market power of MLB. Almost instantly, in 1962, the New York Metropolitans (the Mets) and Houston Colt-45s (now the Astros) were added to the National League (NL). The American League (AL) quickly followed with the addition of the Los Angeles Angels and the last incarnation of the Washington Senators. However, as we saw in Chapter 5, these MLB actions, coupled with the move of the previous version of the Senators to Minnesota, doomed the Continental League proposed at the time by Branch Rickey and William Shea. 517 Another franchise move with antitrust implications occurred between congressional hearings. Noll and Zimbalist (1997) reported on congressional inquiry into antitrust after the last version of the Washington Senators left for Texas to become the Rangers after the 1971 season. That inquiry led to a recommendation that MLB’s antitrust exemption be removed. However, Congress only recommended further study and no further action by the entire House. In 1976, Congressman B. F. Sisk led “The Sisk Hearings” on the issue of market power in sports in general. Despite the clear testimony of every leading analyst on the harm done to consumers by the market power of the leagues, Congress again made no moves against the power of pro leagues over team location. In the early 1990s, another wave of antitrust interest followed the possible move of the San Francisco Giants to Florida, but no action was ever taken. Interestingly, the most recent congressional inquiry into franchise movement and location, concerning the movement of the NFL’s Cleveland Browns to Baltimore, actually attempted to give the NFL an exemption similar to the one enjoyed by MLB. The hope was that location stability would return to the NFL. No further action beyond the inquiry occurred. BROADCASTING AND ANTITRUST Now let’s turn to the antitrust status of pro sports in broadcasting. In the late 1950s, the DOJ held that league negotiations on the part of all member teams for a leaguewide broadcasting contract were an antitrust violation. This move could have been heralded as a solid move forward for consumers of broadcast games, but Congress is nothing if not consistent. It basically reversed the DOJ’s ruling. Under the Sports Broadcasting Act of 1961, leaguewide TV contracts in all sports are exempted from antitrust. This exemption continues to the present day, despite the fact that the Supreme Court struck down precisely the same type of behavior by the NCAA (NCAA v. Board of Regents, 1984, discussed in Chapter 13). Congress also took one further step that harms broadcast viewers and benefits leagues. If a team does not sell out a game, it can kill a contracted broadcast in its own home- viewing area. The effect of these so-called blackout laws is to reduce the number of people who can enjoy a given game if the game is not sold out. As you can see, acts of Congress reinforce market power in broadcasting rather than suppress it. MERGERS AND ANTITRUST Congress has been an active facilitator of pro sports league mergers (Quirk and Fort, 1992, 1997, and 1999). In MLB, the AL was a successful rival to the NL in 1902–1903. Players were jumping between the two leagues and pushing their payment closer to their marginal revenue product. For the 1904 season, an agreement was reached between the two leagues to observe each other’s reserve clauses and crown a 518 champion. That agreement established the modern version of MLB, the league we know today. Despite the fact that the antitrust laws had been in force and used in other areas for over 10 years, Congress made no effort to even investigate this agreement. The rest is market power history. In other cases, Congress actually took formal merger facilitation steps in pro sports rather than carefully applying antitrust laws to the benefit of sports consumers. The fourth incarnation of the American Football League (AFL, with Al Davis at the helm) established itself as a successful rival to the NFL in the mid-1960s. However, Congress acted to formally exempt the AFL–NFL merger from antitrust in 1966, producing the modern NFL with the American and National Football Conferences (AFC and NFC). In other sports, Congress did not act to formally exempt a merger, but it brought pressure to bear on both the NBA–ABA and the NHL–WHA mergers. As James Quirk and I (1997) argue, this merger activity has played a primary role in the elimination of effective economic competition in pro sports. Thus, despite repeated investigation, congressional action either has not been forthcoming or has been directed toward enhancing market power in pro sports. Congress has failed to intervene in league practices concerning team movement and location, reversed antitrust decisions by the DOJ, passed blackout laws in broadcasting, and acted to exempt or facilitate league mergers in every sport. Why has Congress behaved this way? Before we apply the rational actor explanation to this question, let’s have a look at what this behavior is worth to team owners. 519 SECTION 4 The Impacts of Special Tax and Antitrust Status The impacts of the special tax and antitrust status of pro sports leagues are felt by all participants—fans, taxpayers, media providers, advertisers, owners, and players. In this section, we will focus mainly on the impacts on owners and players, but we will briefly examine the impact on the other participants. IMPACTS ON FANS AND TAXPAYERS From the fans’ perspective, market power has the general tendency to reduce output and increase price. In the case of sports, consider the control that leagues and conferences have over team location and movement. Clearly, because so many cities line up when a team announces that it wishes to move or when a league announces expansion, the number of teams is restricted below the level that fans prefer. The reason leagues maintain control over location and expansion is to accumulate all of the value of increases in fan willingness to pay for the sport output for themselves. This value comes either from increases in demand for a given number of teams or from expansion fees. Fans pay more for less in the presence of market power. From the taxpayers’ perspective, tax breaks for some taxpayers raise the tax bill for the rest of the taxpayers for a given level of spending by governments. All of the types of tax breaks that have been discussed in this text have the effect of reducing the taxes paid by team owners. This is perhaps most clear for property tax forgiveness and the use of tax-exempt bonds that goes along with occupation of a publicly owned facility. Without that publicly owned facility, another tax- paying activity would occupy that space. Because property taxes typically go into both local and state treasuries, other revenue sources must make up the rest for some chosen level of public spending. The same goes at the federal level for the depreciation-driven shelter enjoyed by owners. IMPACTS ON MEDIA PROVIDERS The antitrust status of pro sports has clear negative impacts on media providers and advertisers. The special antitrust status of pro sports means that media providers must negotiate with the league rather than individual teams. This reduces competition and raises the price of programming. As we saw in Chapter 3, leagues collect the value of this market power. Subject to the relationship between the supply and demand for ad slots, media providers pass a part of this higher price of programming on to advertisers. The higher price is shared between the media providers and advertisers. IMPACTS ON OWNERS AND PLAYERS 520 Now we turn to the main beneficiaries of special tax and antitrust status—owners and players. For owners, we will see that special tax and antitrust status increases franchise values. Interestingly, the existing group of players also benefits. As long as they are free to move between teams, players earn more when teams have market power than when they do not. TEAM VALUES AND GROWTH IN TEAM SALE PRICES Table 12.1, Table 12.2, Table 12.3, and Table 12.4 present the most obvious starting place for an analysis of the value of these types of advantages for pro sports team owners. I argue that estimates of sports team values like those in these tables end up a bit low relative to actual observed sale values (Fort, 2006) because they ignore some of the values of ownership discussed in Chapter 4. However, teams simply don’t sell that often so, while the values in these four tables concern only part of the value of ownership, they do give a general feel for that value. The majority of teams in hockey and basketball are valued at more than $200 million and $372 million, respectively. The majority of MLB teams are valued at more than $472 million. However, the NFL is at the top of the heap, with the majority of teams valued at more than $1 billion. Indeed, the NFL values are so lopsided that only one team in any other pro sports league is valued at more than $1 billion, namely, MLB’s New York Yankees. Leagues protect territories, and the market power position in these territories is valuable. The special tax and antitrust status plays an important role in creating that value. The special status of teams makes all of the following more valuable than they would be without that status: • Revenues at the gate • Venue revenues • Media rights contracts • Claims on future expansion fees The result is that team values average over $1 billion in the NFL and hundreds of millions of dollars in all other leagues. Table 12.5 shows that the rate of increase in the actual sale price of teams has been, at times, simply staggering. The rate of return on the Standard & Poor’s Index (a weighted average of common stock prices) in Table 12.5 is meant to give a feel for the opportunity cost faced by team owners. Owners could certainly invest the amounts paid for their teams at that rate of increase rather than continue on with ownership at any point in time. The lowest rate of increase in sale value is in MLB, at times nip and tuck with the Standard & Poor’s Index. Teams in the other leagues are no-brainers. If you have the money, get an NFL or NBA franchise! As the value of teams has increased, the character of ownership has changed as well. Few owners now depend exclusively (or even to any large extent) on the earnings of their team. The team is just one element in the portfolio that helps them gain wealth. But the teams do help them gain wealth, and special treatment by government makes for part 521 of this value, relative to other investments. If teams in sports leagues did not have special tax or antitrust status, they would not be worth as much. Interestingly, this all relates back to our discussion of the values of team ownership in Chapter 4. One type of value does follow from annual operating revenues. But all of the other values, including imputed values from special tax advantages and antitrust status, also are part of the sale price. Remember, a well-respected financial group like Lehman Brothers has already gone on the record that the most recent change in pro sports team special tax status will raise franchise prices in all sports by around 5 percent. In a recent article (Fort, 2006), I was able to use the franchise valuations published annually in Forbes, compared to actual team sale prices, to try to untangle these values. The results suggest that values of MLB ownership other than annual operating profits (if any) can be in the tens of millions of dollars. TEAM SALE VALUE PUZZLES In a world where everything was certain, the most anyone would pay for a team would be the present value of the profits that the team would generate. This is one of the most basic notions of finance. If the price were lower than the present value of profits, a prospective owner could just borrow the lower amount, buy the team, and later sell it for a profit over and above the interest paid on the loan. If the price were higher than the value of discounted profits, then anyone who owns the team would just sell, put the money in the bank, and earn more than the team would have generated over the same period of time. Thus, in a world of certainty, the price must be equal to the present value of profits, and the team value could only grow at the rate of interest. However, the observation from Table 12.5 is that franchise prices have consistently (but not always) beaten a reasonable opportunity rate of return, the Standard & Poor’s Index. How can this result be explained? BENCHMARK CASE: NO UNCERTAINTY Let’s stick with a world of certainty. Without any uncertainty, values can beat the normal rate of return, but the flow of profits must be negative. If the franchise price were expected to increase at greater than the interest rate, then a prospective owner would do the following: borrow the money, buy the team, and make more than enough at the end of the period to pay off the loan. Any positive profit on top of that is gravy. However, all prospective owners would know this (it being a certain world, after all), so they would bid up the price to an amount equal to the excess over borrowing plus the leftover profits. The only thing that can stop the price from rising to meet the rate of interest is negative expected profits. However, profits cannot be negative for all years because a team with negative profits for all years cannot have a positive price. Thus, the explanation for team prices growing in excess of the rate of interest in a world of certainty seems completely unlikely. 522 UNCERTAINTY CASE 1: TEAM OWNERS AS BAYESIAN UPDATERS Let’s look at an explanation in the more realistic situation of uncertainty. One kind of decision maker in an uncertain world is a Bayesian updater (after the famous English theologian- statistician, Thomas Bayes, 1702–1761). Bayesian updaters are people who dampen their own expectations about the future based on prior experience. All this really means is that they must be shown something (repeatedly) before they really believe it. Thus, a lag occurs before Bayesian updaters update their beliefs about a change that has already occurred. Let’s suppose that the returns to owning a team have been steadily chugging along, generating a 3 percent real rate of return for a number of years, shown in Figure 12.2 by the dotted line. Then suppose a new media source (like the advent of cable TV) makes its appearance at time T0. This new media source raises media revenue so that a team generates a 5 percent real return, shown in Figure 12.2 as the top dashed line. Thus, at time T0, the return stream jumps discontinuously from the dotted line to the dashed line. However, Bayesian updaters will not believe the discrete jump at time T0. Instead, based on their prior belief about steady returns at 3 percent, they will only come to realize that returns have increased after some time has passed, say, at time T1. In this case, instead of having franchise prices increase immediately to reflect the new 5 percent return, they grow over time from T0 to T1. The slope of the adjustment path represents the rate of growth in the actual present value as it approaches the new higher 5 percent return. You can now see that the average rate of growth over the adjustment period can be dramatically larger than some typical opportunity rate such as the Standard & Poor’s Index. Indeed, during the adjustment period, growth rates double the opportunity interest rate may occur. At time T0, the slope of the adjustment path is easily twice as steep as the eventual 5 percent return. Therefore, if owners are Bayesian updaters, then growth rates in franchise prices can beat the rate of interest for as long as it takes these decision makers to believe in the new, higher franchise price growth rate. However, barring any other increase in returns, eventually, growth rates must reflect the present value of team profits. So there would have to be continual unpredicted increases in order for slow Bayesian updaters to generate prolonged periods of high rates of return. UNCERTAINTY CASE 2: TEAM PRICES AS SPECULATIVE BUBBLES Owners do not have to be Bayesian updaters for franchise values to grow at greater than the opportunity rate of interest. Instead, some episodes of high rates of growth in franchise prices may be examples of what is called a speculative bubble. It is easy to see what a speculative bubble is from an old story. A stockbroker touts a given stock to a client. The client then buys 1,000 shares at $10 each. The broker proudly reports 523 that the price has risen to $12 in only a week. The client is impressed by that stock performance and buys another 1,000 shares. This goes on for several weeks. The price rises, and the client buys more. When the price reaches $25 per share, the client decides to do a little profit taking and tells the broker to sell 3,000 shares. The broker responds, “To whom?” Clearly, the price rise was fueled by the expectations of this single investor. When those expectations collapse, those buying just before the end of the bubble are left holding the bag and suffer losses equal to their holdings. The important requirement for a bubble in prices is that somebody continues to believe the price will rise. If somebody believes the price will rise, it will. Data on sports team prices do not support arguments that teams are overvalued. James Quirk and I, in Hard Ball (1999), noted that through the 1970s, 1980s and 1990s, NBA team prices have increased around 26 percent per year, MLB teams at around 14 percent per year, and NFL teams at around 22 percent per year. Thirty years seems too long a period of increases to be explained by a bubble. It is possible that it is a bubble but not very likely. If any teams are artificially overvalued, the bubbles will eventually burst. When they do, those unlucky few holding teams that are overpriced due to inflated expectations will have to eat their losses. THE ROLE OF GOVERNMENT IN RISING FRANCHISE PRICES Government affects the expectations of team owners. Some choices by state and local governments, for example, can lead to extended periods of increases in the returns to those holding teams. For example, as we saw in Chapter 11, NFL teams have been the beneficiaries of 13 votes in a row to fund new NFL stadiums. These new stadiums increase team franchise values. If owners, leaguewide, are Bayesian updaters, this series of increases in franchise values could take a long time to update. The result would be extended periods where franchise prices increase at greater than the standard rate of return. These same types of impacts may occur at the federal level. As we saw in the preceding section of this chapter, periodically the federal government maintains league market power and often facilitates its growth. These actions by the federal government increase the value of franchises, league wide. If owners are Bayesian updaters and slow to adjust their beliefs after these federal tax and antitrust decisions, then franchise values might grow at greater than the rate of interest for a prolonged period. THE VALUE TO PLAYERS The value of special taxes and antitrust status to players can be seen in an analysis of MRP in the presence of these circumstances. Figure 12.3 facilitates our discussion. In the absence of market power, MRP in a competitive output situation determines the value of athletic services offered by players. The result in a competitive player market would be S0 units of athletic service hired at a wage equal to W0. However, owner profits are greater with market power. Because players 524 contribute to a higher value in this situation, the owner demand for athletic services rises, as shown by the MRP with market power in Figure 12.3. The result would be an increase in the quantity of talent hired to SMP and an increase in the wage to WMP. Players contribute to an enterprise that generates more than it would without special tax and antitrust status. This is no different from the idea that if the demand for cars increases, so does the demand for autoworker skills. Both the quantity of autoworker skills hired and autoworker pay would increase. So it is for players. Imagine if the market power of sports leagues disappeared. Profitability would fall and so would the value that players create. One would expect that salaries would fall as well. Thus, harkening back to Chapter 9 on labor relations, it is no wonder that players are staunch supporters of the market power positions of their respective leagues. 525 SECTION 5 Rational Actor Politics, Taxes, and Antitrust Let’s examine the rational actor explanation of the relationship between government and the sports business. As with labor union governance and venue subsidies, we will use a rational actor model of politics to investigate the special tax and antitrust status of pro sports team owners. RATIONAL ACTOR POLITICS AND TAXES The rational actor model explanation of why there is special tax status for sports teams is the same as the explanation for any federal tax loophole. In a given tax setting, a particular group presents its elected official with a reason why it should be afforded special tax treatment. If the loophole favors a particular group, then the cost of that loophole is spread thinly across the rest of the taxpayers in terms of higher taxes to maintain the same level of government spending. As long as the specific, politically potent reelection constituency in favor of the loophole delivers in terms of votes and vote- generating resources, they get what they want. Groups can also make their case at the agency level. The IRS hears cases from special interests about just why the interpretation of a particular part of the tax laws should go a given way. The IRS, sensitive to its congressional overseers, makes a ruling. If Congress likes the IRS decision, the IRS receives more resources and approval, which matters to the agency leadership in its career pursuits. Through either the direct congressional route or the indirect agency route, the important interest group gets its way. Figure 12.4 is our familiar triangle model applied to tax policy for pro sports teams. Pro sports team owners and their various supporters are an important special interest. Elected officials are interested in pleasing the strong sports reelection constituency. The costs of serving them are borne in small ways by the rest of the taxpayers. The IRS, in this case, is just the agency that carries out the desires of Congress to do this service. In this way, owners managed to “convince” the IRS that players were depreciable assets. How is it that sports constituencies are powerful enough to gain legislative approval of their demands? To the extent that pro owners are located in the top 30 cities, nationwide, they and their supporters are important constituents to about 50 very influential members of the Senate (a few states have more than one team) and at least that many members of the House. This means that the logic we have developed is binding on many of the most important members of Congress. Because so many elected officials share a similar interest, it should come as no surprise that owners and their supporters get their way. Opposition to owner benefits may be vocal and may make the news, but apparently it is not strong enough to 526 pose much of a challenge to the status quo powers in sports policy. In short, if elected officials wanted to alter the tax status of pro sports teams, they certainly could. However, they have not done so, either directly through legislation or indirectly through dictates to the IRS. We are left to conclude that pro sports enjoy their special tax status because special interest groups and elected officials benefit from it. RATIONAL ACTOR POLITICS AND ANTITRUST Figure 12.5 is the triangle diagram for the antitrust outcome. Suppose you are a member of Congress with a team in your geographic jurisdiction. The team is enjoyed by some of your constituents, but it also contributes to the wealth of your reelection constituency. A vote to limit sports league market power will change that. As long as reelection drives your decisions, you will not vote against sports or encourage the DOJ to do so. You have just walked yourself through the predicament facing members of Congress in roughly the top 30 population areas in the United States. All have at least one sports team in their jurisdiction. This means that the pro sports league support group in Congress will include many influential politicians. Asking them to support antitrust intervention in pro sports leagues is asking them to act against their own reelection interest. However, it cannot be denied that the political margins do not always weigh in favor of market power. The Flood Act of 1998 is one example of how political margins sometimes change, even in the world of sports. The MLB strike of 1994–1995 was a killer, politically speaking, because nobody was happy except for a few hundred players. The MLB strike hurt politically powerful sports interests and the fans. Members of Congress had little to lose in making small changes to MLB’s antitrust status in the interest of preventing future strikes. Interestingly, Congress did not respond in a similar fashion to any owner-instigated lockout (NBA, 1998, or NHL, 2004– 2005). Apparently, and consistent with the rational actor politics explanation, now there are politically powerful interests in favor of maintaining the ability to stop play. REALITY AND THEORY MAY NOT MATCH Before we move on, let’s remember that the real world can be more complicated than our simple rational actor model. Sometimes, outcomes are more of a two-way street. For example, it appears that MLB is cognizant of the market power that is granted to it by Congress and responds, occasionally, to the needs of elected officials. Shortly after the Celler hearings in 1951, MLB teams moved to the Pacific Coast. Partly in response to congressional interest (and partly due to the threat of the Continental League), MLB added a team in New York in 1962. Although the Giants did not move to Florida, the expansion Florida Marlins did come into existence shortly after that threat in 1993 (along with the 527 Colorado Rockies). Interestingly, this happened after hearings instigated by Florida senator Connie Mack and Colorado senator Tim Wirth. As recently as 2000, hearings led by Senator Mike DeWine of Ohio saw MLB promising to mend its competitive balance problems, as detailed in this section’s Learning Highlight: Professor Fort Goes to Washington. It appears that Congress can get concessions out of sports leagues and owners every once in a while. 528 LEARNING HIGHLIGHT: PROFESSOR FORT GOES TO WASHINGTON In November of 2000, I was honored to have been asked to testify before the Senate Committee of the Judiciary, Subcommittee on Antitrust, Business Rights, and Competition. The title of the hearings was “Baseball’s Revenue Gap: Pennant for Sale?” I saw and heard many interesting things in this up-close look at the rational actor explanation of sports outcomes. The focus of the hearings was competitive balance, and the deck was pretty well stacked in favor of MLB’s position. Commissioner Selig went first. The gist of his presentation was that baseball has real problems but that baseball could solve them without government intervention. Former senator and peace negotiator George Mitchell followed. He had served on the Blue Ribbon Panel 2000, appointed by MLB to make recommendations about how to solve its competitive imbalance problems. Senator Mitchell argued that at this triage stage, there was no more time to waste in saving a clearly sick patient, MLB. The final testimony came from a panel consisting of Bob Costas (noted commentator and true baseball enthusiast), George F. Will (another true enthusiast, member of the boards of MLB teams, and nationally syndicated political columnist), Frank Stadilus (CEO of USFans, a fan interest group with a strong Internet presence), and yours truly. I was the token analyst and, I must admit, I felt more like window dressing than anything else once the panel was over. Now, except for me, think of the composition of the testimony with reference to Figure 12.5. On baseball’s side, we have Selig, Mitchell, Costas, and Will. In addition, Cincinnati Reds owner Carl Lindner was in the audience and received special recognition from the chair, Senator Mike DeWine of Ohio. Indeed, Senator DeWine made it clear there was a constituency interest here: “As a lifelong fan of baseball, I am concerned about the future ability of Ohio’s teams—the Cleveland Indians and the Cincinnati Reds—to compete with big-money teams, like the Yankees, the Braves, and the Diamondbacks.” The fans received recognition through the presence of Mr. Stadilus. The players had no representative on the panel. What transpired in this real-life example of the policy triangle in action? My testimony showed that (at that time) there is competitive imbalance in baseball but probably no more than had existed historically. Senate staffers tried to refute my testimony with a few vague charts. Costas eloquently pleaded for fiscal sanity and a return of baseball’s soul. Will pointed out that there is nothing like a sports league, implying MLB knew what was best for MLB, and finished with the idea that baseball is not Bangladesh. It could be healed, but both players and owners would have to do it. Stadilus reminded everybody that the fans are the whole point, and without them the whole thing comes tumbling down. 529 It is my opinion that the point of the hearing was to publicize MLB’s problems as extraordinary, the line from all the baseball people and clearly the belief of Senator DeWine. Then the senator, representing the antitrust investigating arm of the Senate, suggested that the commissioner and MLB could take care of its problems. Indeed, Commissioner Selig had emphasized that MLB had found solidarity and willingness by owners to do whatever it takes to create the economic blueprint for the future. Senator DeWine concluded his remarks as follows: While there is no consensus on the correct approach to remedy the decline in competition in the business of baseball, one thing is very clear: The status quo is simply unacceptable. Unless something is done to correct the payroll and revenue disparities among the teams—unless we untie the stranglehold around the small- and medium-market teams by increasing competition—baseball cannot survive. We look this morning to Bud Selig to outline his blueprint as Commissioner of Baseball on what he plans to do to save baseball. We know the problems, we will hear about the recommendations, but what the game really needs now is leadership. The immediate challenge is clear. How will the commissioner convince the owners, particularly the larger market owners, as well as the players, that their future is tied to the health and survival of their sport? It is now up to them —owners and players—to step up to the plate. And, as always, we, the fans will be watching. To put it simply, I look forward to hearing how the commissioner will lead the owners and players toward real solutions to the problems that plague baseball. I hope that for the sake of the game, they do the right thing. Interestingly, nobody was there to speak for the players. Perhaps that was because collective bargaining began again at the end of the 2001 season. You can download all of the testimony at www.gpoaccess.gov/chearings/ (you will need to use the site’s search routine, checking the 106th Congress (1999-2000), Senate Hearings, keywords: “Revenue Gap”). It is pretty good bedtime reading. Hearings on MLB’s imbalance problems led this author to believe that the special antitrust status enjoyed by that league really is a rational actor political outcome. [Photo Selig and Costas.] Source: “Revenue Gap: Baseball’s Pennant for Sale?” www.gpoaccess.gov/chearings/. 530 http://www.gpoaccess.gov/chearings/ http://www.gpoaccess.gov/chearings/ http://www.gpoaccess.gov/chearings/ http://www.gpoaccess.gov/chearings/ SECTION 6 Competition Policy Throughout this text, we have exploded myths, enhanced your ability to explain sports business outcomes, answered questions, and generated new ones. By way of a basic understanding of the economics of pro sports, there is only one more thing to cover, namely, competition policy. Competition policy concerns market power problems and what can be done about them. THE MARKET POWER PROBLEM The problem addressed by sports league competition policy is market power (the discussion in this subsection follows Quirk and Fort [1999]). Most sports fans can easily identify the results of market power. High ticket prices have made it practically impossible for the average fan to enjoy many home games in person. Growing revenue imbalance has made the final league standings a forgone conclusion. Fans have lost touch with players as player salaries skyrocket. Don’t even get a fan started about work stoppages! Fans could point their fingers at any one of a number of culprits responsible for these problems—the media, owners, leagues, players, unions, and politicians. A fitting summary of what you have learned in this text is a review of the role each of these potential culprits plays in the problems confronting pro sports. We will then get back to the real problem, market power, and discuss the types of competition policy that can be used against it. THE MEDIA In Chapter 3, we learned that media providers do nothing more than collect revenue from advertisers and pass most of it on to pro sports leagues and teams. To media providers, sports are just another type of programming that reaches demographic groups of interest to advertisers. Because teams are allowed to act jointly through their leagues and college conferences in the sale of TV rights, it should come as no surprise that it is the leagues and teams that earn the bulk of the return from sports broadcasts. OWNERS The lesson of Chapter 4 was that owners simply act to maximize their wealth. Their team is just a very valuable asset that contributes to that pursuit. Owners will temper their pursuit of winning because it costs more to field a winning team. Thus, love them or hate them, it is not the personalities of owners that result in many of the problems facing sports; it is the discipline of the bottom line. Moreover, as future profits are imputed into the original price of the team, we should expect owners to exploit their market power. If they do not, then they will fail to recoup their initial purchase costs. 531 PRO SPORTS LEAGUES Chapter 5 explained how owners act jointly through leagues to further their individual economic interests. Leagues exercise market power over both inputs and outputs. On the input side, Chapter 8 discusses how owners acting through leagues enjoyed the lion’s share of player marginal revenue product under the reserve clause and player drafts. The result for fans, on the output side, has been clear since the time of Adam Smith—restricted output, prices greater than marginal cost, and profits greater than the normal rate of return. However, we should exercise caution before we pass judgment on leagues. Remember, market power and the single dominant league outcome do not necessarily have to happen. The market power that is exercised is the culprit, not those who exercise it. After all, if the existing owners did not exercise this power through leagues, there are plenty of potential buyers who would. If the antitrust laws were enforced, market power could be regulated as it is in other industries. PLAYERS Many find players’ seeming insatiable demands at least partly to blame for rising ticket prices. However, in Chapter 7 we covered two ideas that dispel this notion. First, why do fans think that players are any greedier than anybody else, including themselves? How many fans would be willing to take less than they could possibly make? Second, sports salaries simply reflect the value that fans place on player talent. Changes in players salaries do not cause changes in ticket prices. Instead, it is the willingness of fans to pay ever more for sports that raises player salaries or encourages owners to seek new ways to collect revenue. Players are no more to blame for the high price of sports than fast-food workers are responsible for the rising price of hamburgers. Fan demand drives the high-return activity that players engage in. Beside, if players did not take their share, would anybody reasonably suggest that owners would rebate the balance to fans? Admittedly, some players may not be very likable, and some run afoul of the law, but the economics of player pay makes them no different than any other star entertainer. They are paid much the same as any person is paid, according to their contribution to the value of their employer. PLAYERS UNIONS There can be no doubt that unions have fundamentally altered the face of professional sports. In Chapter 9, we studied how unions have displaced nearly all of the mechanisms that owners previously used to restrict the free movement of players between teams. The result is that players now receive salaries much closer to the value of their contribution to team revenues. But are unions responsible for the ills that many claim plague sports? Some fans might begrudge players their huge salaries, but the money that fans spend on sports will not go away if, 532 somehow, salaries were restricted. As we just mentioned, if players do not get the money, then owners will keep it. Salaries are large because leagues earn more than the normal rate of return; it is the profit earned by leagues that is up for grabs in player–owner negotiations. Without these profits, unions would not have much to do except negotiate minimum salaries and benefit packages. Unions have not created the fabulous wealth available to athletes. They have just been proficient at moving that wealth from owners to players. POLITICIANS Typically, when market power runs wild, we hope for government intervention to protect consumers. However, this appears to be frightfully optimistic in the case of pro sports. If anything, the behavior of politicians has facilitated market power all the way back to their lack of any legislative response to Federal Baseball. Witness what are viewed by many as overly high subsidies plaguing so many current pro sports team hosts. But are politicians, per se, to blame for these outcomes? The lesson in this chapter is that the reelection imperative confronting politicians drives their decisions. Politically potent groups control reelection by providing money and votes to politicians who serve their interests. In this sense, it is difficult to blame politicians as opposed to the system in which they operate. If politicians work against the powerful group, they risk losing in the next election. It seems unreasonable to expect politicians to commit political suicide on a regular basis. AN INDICTMENT OF THE PROCESS It appears that the market power problem is ultimately caused by failures with the political process. Special interest politics at the state and local level are responsible for venue subsidies. Special interests also influence politics at the federal level, maintaining the special tax and antitrust status of pro sports leagues. The results of this special status are exclusive franchise rights for teams, league control over market power, and a complete stifling of competing leagues. It is a testament to the political power of owners, fans, and other sports supporters that leagues have enjoyed this special status for nearly 100 years. There you have it. Ticket prices are high because there are no competing teams in the same geographic market to push prices to marginal cost. Competitive balance is lacking because leagues restrict the number of teams in large-revenue markets. Salaries are higher than they would be under competition because some of the profits from market power accrue to players in a labor market that is carefully managed by unions. Strikes and lockouts occur as owners and players lock horns over the division of league profits. As leagues carefully manage the number of teams, output is restricted and prices rise, and some cities are held open as relocation threats against current host cities. Host cities are confronted with all-or-nothing propositions because of the lack of 533 substitutes for professional sports teams. All of this is the result of the pursuit of self-interest through the political process. RATIONAL ACTOR POLITICS AND PESSIMISM ABOUT REFORM If market power is the problem, then its regulation or removal is the answer. There are a variety of approaches to enhancing economic competition or introducing new economic competition into pro sports. However, we would be naïve to think that just listing the good things about economic competition is enough to make it happen. After all, current political margins dictate the ongoing state of affairs in pro sports in the first place. Calls for alterations in the level of economic competition in sports must first come to grips with this simple fact of life: Unless new powerful interest groups replace the current ones, politicians will not change the economic outcomes in pro sports. This is the same observation made in Chapter 11 on altering venue subsidy outcomes, only generalized to the special tax and antitrust status of pro sports. New powerful interest groups must arise to do battle with the status quo. In our various triangle diagrams, those on the outside looking in at the current triangle outcome must become insiders. Until this happens, our rational actor approach suggests that the chances for strident competition policy to be applied in sports are pretty low. In their 1997 book Sports, Jobs, and Taxes, Noll and Zimbalist put it this way: Unfortunately, the same forces that have impeded effective legislation also stand in the way of antitrust action that would lead to divestiture. Like Congress, the Antitrust Division of the Department of Justice is susceptible to political pressure not to upset sports, and so a large, influential monopoly remains unregulated and de facto immune from antitrust prosecution by the government. (p. 505; Reprinted with permission from the Brookings Institution) However, those with an interest in reforming pro sports do have some hope. When a problem becomes important enough to enough voters, a new political mobilization can occur that alters the status quo. This is an expensive process. Those bent on such change must successfully accomplish an overwhelming educational mission and overcome the high costs and free-riding behavior associated with organizing a politically potent group. Then, once mobilized, the dog-eat- dog world of advocate politics begins. Small wonder that market power has ruled in pro sports given the obstacles to bringing it down. REMEDIES THROUGH COMPETITION POLICY Pessimism over the chances for reform does not stop us from examining the type of competition policy that might work if changes in rational actor politics ever occur. Two kinds of 534 interventions come to mind: direct regulation and enforcement of antitrust laws. REGULATORY APPROACHES Regulatory approaches to sports problems were suggested as early as 1972. Senator Marlow Cook introduced federal legislation creating a separate federal agency to regulate sports. It never gained broad legislative support, but in 1993 columnist Charles Rhoden, writing in the Sporting News (April 5, 1993, p. 8), tried to resurrect the idea. Rivkin (1974) suggested that a national sports council was the best way to facilitate cooperation between leagues, Congress, and the public. Zimbalist (1992) argued that sports are a commodity supported through public expenditure and called for public- utility style regulation. Others suggest fan ownership of teams. All of these ideas have good and bad aspects. At the end of this section, the Learning Highlight: Would Public Owners Do Any Different? provides examples of fan ownership and municipal ownership. ANTITRUST APPROACHES Noll (1976), Horowitz (1976), Ross (1989, 1991), Fort and Quirk (1997), Quirk and Fort (1999), and Fort (2000) detail the usual antitrust approach of breaking up the leagues. Suppose the AFC and NFC in the NFL were turned into two separate competing pro football leagues. The idea is that competition between the leagues would remedy many of the ills currently attributable to NFL market power. However, two things must be kept in mind when considering this option. First, there have been competitive leagues in the past, but the tendency has always been back toward a single dominant league. Economic competition has not been self-sustaining in pro sports historically. Owners in rival leagues ultimately see the value in reforming into a single entity. Second, owners in the current dominant leagues have established their reputations and created a strong sense of fan identification at some expense. These owners suffered low or negative profits during the early years of their leagues and have paid public relations expenditures since that time. The returns to such an investment are the profits earned by sports teams under their current market power structure. A rival league planning to compete with existing leagues must make the same kind of investment, in addition to competing for talent, in order to demonstrate that it really is in it for the long haul at the big-league level. However, vigorous economic competition would make it difficult to recover these costs. A new league just would not survive. With these two factors in mind, let’s think about breaking up the pro sports leagues. THE HISTORICAL PRECEDENT FOR SEPARATE LEAGUES History suggests that breaking up the pro sports leagues might work. Prior to 1903, the AL and NL in MLB were economic competitors. The same is basically true of the AFC and the NFC (by and large) in the NFL. With only a few crossover teams, the conferences are about the same as the 535 most recent version of the AFL and the NFL prior to the merger in 1969. The two leagues were so economically competitive that then-AFL commissioner Al Davis (the current Raiders owner) urged the AFL owners not to merge with the NFL. In his opinion, the AFL would have taken over the NFL. As we saw in Chapter 5, the current structure of the NBA and NHL is built on mergers of previously economically competitive rival leagues. Although some would argue that the merger was necessary to salvage any sort of pro basketball league, the strength of the remaining teams in the NBA belies this. The evidence does not suggest that a merger was required for economic stability in the NFL or NHL. Therefore, a breakup of existing leagues would recreate a state of economic competition that existed in the past. One of our concerns about intervention is covered by this approach. Remember that large investments are required to generate fan loyalty and media ties. The breakup of existing leagues would allow individual owners in the resulting competing leagues to retain the fan loyalty and media ties they already have built over so many years. The leagues are already major in every sense of the word—and they wouldn’t lose the fan identification that they have cultivated over the years. However, remember that rival leagues have a tendency to fail or merge with another league. Enforcement of league breakups under existing antitrust laws would be required for competition to flourish. If leagues tried to regroup and merge, antitrust enforcement would preclude the new merged league result. BREAKING UP THE LEAGUES: ECONOMIC IMPACTS Of course, any change away from the current sports outcome will impact the welfare of leagues, players, fans, and taxpayers. Owners and players could be characterized as the losers because their welfare will fall. Overall, fans will be winners because there will be more sports to enjoy at lower prices. Taxpayers should win as subsidies would be reduced. However, some fans at locations with marginal teams could be losers; competition would drive marginal teams out of business or to other locations. But remember that our analysis of the value of owning teams in Chapter 4 shows that there are few marginal teams to worry about. Competition pushes prices to marginal cost and increases output relative to market power situations. Believe it or not, competition would create more and cheaper sports for fans to enjoy! It would drive down the price of television contracts as more teams entered into the offer process. Media providers would have substitutes if one league got out of line in their rights fee demands. There would be more games at a lower price for fans watching games on TV. Competitive balance problems would be resolved as well. Currently, larger-revenue-market teams buy more talent than smaller-revenue-market teams. Economic competition would drive profits to zero, economically speaking, because entry by new teams would equalize drawing potential, driving down 536 revenue differences between teams. The result would be more equal revenues, more equal talent, and more balanced outcomes. But it must be remembered that willingness to pay by fans for quality would still vary so competition would still be imbalanced, just less so. With true competition, bloated player salaries could become a thing of the past. Talent hiring and competitive payments to labor in a market power setting occur where the marginal revenues for teams are equal to each other. However, the evaluation of marginal revenue comes from a market power situation. Player salaries would fall in the presence of competition because monopoly profit would be driven to zero. Players simply would not be as valuable to owners as they are now. An offsetting feature would be if fan demand continues to increase over time. Salaries can still rise in a competitive economic environment. However, those same players would have earned even more during periods of increasing demand if leagues and teams had market power. What about labor–management hegemony? Again, in a competitive environment, profits are driven to zero, and if there is nothing to fight over, there should be no fights. Unions would still have issues to negotiate, such as an inherent trade-off between decreased employment overall and higher pay for players who do get hired. But mostly, unions would be left to deal with pensions and health plans and, possibly, the inequality of salary distributions in sports. Economic competition would also have impacts on team location and expansion. Currently, the decision rests with existing leagues that base team location on expansion fees, TV impacts, owner group stability, lost threat value, and preclusion of rival leagues. Under competition, there would be no such luxury. Every location that is economically viable would simply have a team. The only way a locality wouldn’t have a team is if no team could make a go of it there. The tables would be turned on teams that threatened to move. Cities could just deny team demands and rest safe in the knowledge that they would get another team by virtue of their economic viability. Because all viable locations would have a team, there wouldn’t be any believable threats. Power would be transferred away from leagues, owners, and players toward fans and taxpayers. If one league threatened to leave a viable location, another competing league would bid for that location by lowering the demand put upon the city. Indeed, this could have the potential to turn subsidies to teams into payments by teams. BREAKING UP THE LEAGUES: TEAM QUALITY IMPACTS One of the usual questions posed at this point is this: “What will the quality of competition on the field be under increased economic competition?” One of these worries concerns spreading talent too thin. However, this can only be a short- term problem, like the talent spreading that occurs with any expansion. Eventually, among all teams, talent reaches the 537 level that fans are willing to pay for in various cities. If fans in an expansion city are willing to pay for it, the level of talent they enjoy will rise. The level of talent in pro sports leagues always rebounds from short-term dips due to expansion. The clear offset to even any short-term talent diminution is that many more fans get to enjoy pro sports in more locations. And this enjoyment would be due to their own self-interested spending rather than due to profit-maximization choices by owners in a league. Another worry about the quality of talent in a more economically competitive structure concerns its distribution. With increased competition, profits from exclusive territories will fall. Because talent will be less valuable in the highest return cities, the value of talent will fall there as well. This worry is real. The distribution of quality will change. Some fans will not be as happy as they were, but their previous level of happiness was dictated by a market power situation that left many fans without any team at all. Under competitive leagues, many more fans would enjoy pro sports entertainment than ever did before. Of course, it’s never the case that everybody agrees on anything in sports, and antitrust is no exception. Famed sports lawyer Gary Roberts (1991, 2003) argues that the special antitrust status enjoyed by pro sports team owners and leagues is, on net, a good thing. According to Roberts, MLB’s antitrust exemption has a host of characteristics that enhance fan enjoyment from the sport. First, if MLB were subject to antitrust, it would be exposed to “irrational, ad hoc regulation by judges and juries.” The problem with leagues is their exercise of market power as a single entity (recall this definition from Chapter 5), not that they are illegal combinations of trade. Applying laws aimed at the second problem to try to fix the first would be irrational and costly (Shugart [1997] makes this identical argument as well). Ross (2003) shows that antitrust intervention in sports leagues actually has served the public interest in the past, but here is an example consistent with Roberts’s fears. Suppose antitrust intervention was successful and resulted in freedom of movement for all teams in MLB. If the league still controlled the number of teams, then some cities would gain from the moves, but the fans in cities losing teams would be clearly worse off. Johnson (1997, 2000) points out that the teams would move to larger-revenue markets, reducing the economic advantages of the previous single-occupant team owners. But it is the uncertainty of the results of antitrust applications that worries Roberts; the result could be a net loss. Second, according to Roberts, applying antitrust laws to the major leagues could dramatically reduce the number of minor league teams to the detriment of minor league fans. This could happen because of the intricate financial relationships between major league teams and minor league teams (recall the discussion of the Professional Baseball Agreement in Chapter 2). If champions of antitrust laws see these relationships as anticompetitive, antitrust intervention could break the relationships, reducing payments from major league 538 teams to minor league teams. If the payments from major to minor league owners were severe enough, some minor league teams wouldn’t survive. Once again, however, we should all remember that the current situation concerning the minor leagues also is an artifact of their past contracts with major league teams that have enjoyed market power positions. 539 LEARNING HIGHLIGHT: WOULD PUBLIC OWNERS DO ANY DIFFERENT? Joseph Bast, president of the educational organization the Heartland Institute, argues that fan ownership will curb pro sports market power abuses. He cites the Green Bay Packers as a case in point. A private nonprofit corporation with over 400,000 shares owns the Packers. Only the CEO is paid, and a 45-member board rotates 15 positions annually. Shares cannot be sold for more than the purchase price, and no dividends are paid. If the team is ever sold, nearly all of the proceeds will go to a local Green Bay VFW post. Because profits cannot be taken from operations or from the sale of the team, all net operating revenues go back into the team with the goal of keeping the team in Green Bay. The success of the Packers on the field is obvious from Green Bay’s nickname, Titletown. We saw in Chapter 11 that Lambeau Field in Green Bay has been the least subsidized venue in pro sports until just recently being renovated. Interestingly, NFL rules enacted after the Packers joined the league now forbid corporate ownership, nonprofit or for profit, for all teams. (In other leagues, for-profit corporate ownership is allowed, but nonprofit ownership is not.) Bast argues that municipal ownership is no substitute for fan ownership, citing “the inefficiencies of public management of enterprises of any sort.” However, the city of Harrisburg, Pennsylvania, bought its minor league Senators baseball team when the previous private owner appeared to be moving the team to Springfield, Massachusetts. The club is now run by the nonprofit Harrisburg Senators Baseball Club, Inc. The team draws about 250,000 fans and earns $2 million annually. The Senators were Eastern League champs from 1996 to 1999 and currently have a new stadium under design. In this case, municipal ownership has worked out just fine. Properly designed, it appears that public ownership of teams, like the Green Bay Packers, does help reduce the excesses against taxpayers seen in other forms of team organization. [Photo crazy GB Packers Cheesehead.] Sources: Joseph L. Bast, “Stadium Madness: Why It Started, How To Stop It,” Policy Study No. 85. Chicago, IL: Heartland Institute, February 23, 1998; Associated Press, “Minor League Teams Create Dreams for Grimy Towns,” 1998. 540 SECTION 7 Chapter Recap In this chapter, we examined the relationship between pro sports and government in detail. The history of the special tax status of pro sports dates back to when Bill Veeck convinced the IRS that the player roster was a depreciable asset. The modern version of this provides a write-off of the entire team purchase price for the first 15 years of team ownership. No real value is lost, and the owner gets to reduce both taxable team income and, if organized carefully, taxable personal income as well. In addition, when the capital gains tax rate is lower than the personal income tax rate, owners can postpone any tax payments on the team and pay at the lower capital gains rate. All this requires is a reorganization of the firm that is allowable under the tax laws. The IRS has also ruled that the next owner can start the depreciation all over again, even though the previous owner may already have depreciated the roster asset completely. These special tax advantages have been estimated to triple the owners’ rates of return on team ownership relative to a case where the advantages did not exist. In any other sphere of economic endeavor, the behavior of sports leagues would draw almost immediate antitrust scrutiny. In their relationship with players, leagues in the past sought to extend their market power on the output side over to the input side. The reserve clause was a prime example. Now, all players in all leagues can sue under the antitrust laws in the event that leagues exercise their market power in a way that leads to a negotiation stalemate. In any other case, players must decertify their union in order to sue under the antitrust laws. This is risky because the union can be quite different if and when it reforms. On the output side, leagues remain free to exercise market power over territory, franchise location and movement, broadcast rights, and merger activity with rival leagues. Indeed, Congress historically seems willing to preserve and extend this power. Both owners and players benefit from these special tax and antitrust advantages. On the owner side, the value of franchises indicates how financially advantageous this special tax and antitrust status really has been. Franchises sell for hundreds of millions of dollars, and expansion fees have been rising at astronomical rates. These franchise value increases often exceed reasonable estimates of the opportunity cost confronting owners. This raises two interesting possibilities. First, if there is little uncertainty in the world of sports ownership, teams must generate significant negative cash flows. Second, in the much more likely case of an uncertain world, franchise price increases in excess of opportunity costs indicate that either owners are slow to update their beliefs about franchise value increases or there are speculative bubbles in ownership. Government enters into the structure 541 of franchise price increases by being the source of many of the increases in rates of return. By protecting market power and raising the level of subsidies going to owners, government fuels increased franchise values. Players benefit because the value of their services and, hence, their pay is greater as teams reap the harvest of tax and antitrust special status. Why does government facilitate the special tax status and market power position of pro sports leagues? The answer offered here centers on a rational actor model of the political process. In the tax case, the explanation for sports is the same as for any tax loophole. Directly, special interests seek favorable tax treatment from Congress. Indirectly, politically powerful groups can influence decisions made by the IRS. Owners and their supporters represent a strong political constituency for representatives in roughly the top 30 population centers in the United States. Their strength in Congress cannot be denied. The explanation in the case of antitrust also is amenable to the rational actor model of politics. Powerful sports special interest groups demand antitrust protection from Congress. In the usual course of the rational actor explanation, the DOJ fails to enforce the antitrust laws because of pressure from special interests and members of Congress. When necessary, Congress itself has acted to make sure that market power is preserved by passing laws making joint venture TV negotiations legal, upholding blackout laws, and facilitating mergers. It is clear that there are problems in pro sports; just ask any fan. Because these problems stem from market power, competition policy that encourages economic competition would fix these problems. However, congressional encouragement of competition should not be expected unless fan and taxpayer interest groups become politically potent relative to the current sports support groups. A number of arrangements could facilitate economic competition in sports leagues. These include a federal sports regulatory agency, a sports council, a public utility type commission, and a breakup of the existing leagues. The latter has an important factor to recommend it; namely, it preserves the current values of investments made in the past that need to earn a return in order for existing owners not to lose their teams. However, all of the options would require ongoing vigilance because the tendency has been to reform any form of competitive league structure into a less competitive one. 542 SECTION 8 Key Terms and Concepts You should have run into each of these in pop-ups in the text of this chapter: • Reorganize for tax purposes • Excess depreciation • Roster depreciation rollover • Basis • Personal and capital gains tax gap • Imputed tax advantage values • Curt Flood Act of 1998 • Decertification dangers • Raiders case • Sports Broadcasting Act of 1961 • Blackout laws • Merger facilitation • Negative expected profits • Bayesian updater • Speculative bubble • Competition policy • Regulatory approaches • Antitrust approach 543 SECTION 9 Review Questions 1. What is the corporate income tax? What is the personal income tax? What is the capital gains tax? 2. What is a pass-through firm? How does a pass-through firm benefit pro sports team owners? 3. What is roster depreciation? How does it benefit team owners? 4. What is excess depreciation? Logically, is there any excess depreciation when a team owner sells a team? 5. What is roster depreciation rollover? Logically, should such a thing exist? 6. What is the basis? What is meant by the personal and capital gains tax gap? 7. What are imputed tax advantage values in pro sports? 8. What is the general purpose of the antitrust laws? List the practices of pro sports leagues that seem to run afoul of the antitrust laws. 9. What is the Raiders case, and why is it important to antitrust considerations in pro sports leagues? 10. What is the Sports Broadcasting Act of 1961? Why is it important in considerations of antitrust laws in pro sports? 11. What are blackout laws? 12. Explain merger facilitation. What is its importance in pro sports? 13. Explain what is meant by the term Bayesian updater. 14. What is a speculative bubble? How can a bubble persist over time? 15. Define competition policy. What are the two competition policy approaches listed in the text? Give examples of each. 544 SECTION 10 Thought Problems 1. Is being organized as a pass-through firm a bad thing for sports team owners once roster depreciation has run out? Explain. What step can an owner take after roster depreciation has run out? 2. If you bought a team, how would you structure it for tax purposes? Explain your actions. If you intended to keep the team, what would you do once the roster depreciation had run out? Again, explain your actions. 3. What are the possible costs to a pro sports team owner of being hit with an IRS excess depreciation penalty? Carefully explain the way that the penalty would be assessed and the tax rates that would apply. Why isn’t there any excess depreciation when a team owner sells the team to a new owner after fully depreciating the roster, even though the sale price includes roster value? 4. How do team owners use the personal and capital gains tax gap to reduce tax payments? How did the tax law changes of 1986 change this outcome? 5. What does it mean to say that once tax advantages are imputed into the value of a team the owner will only earn a normal return on the purchase of the team? Do owners earn just the normal rate of return? Explain. 6. Why would it seem on the face of it that sports leagues are prime candidates for antitrust lawsuits? Give examples. Explain why MLB has special status under the antitrust laws. 7. Explain the relationship between collective bargaining and the ability of players to take individual antitrust action against owners. How was that changed for MLB players by the passage of the Curt Flood Act of 1998? 8. List the limits of the extension of antitrust rights under the Curt Flood Act of 1998. What did players gain from the act? 9. What are the dangers of union decertification to players interested in having union representation? Explain. 10. Does the Sports Broadcasting Act of 1961 reduce market power or promote it? Explain. Who benefits from the act? Who loses? 11. Have congressional decisions on mergers facilitated market power or reduced it? Explain. 12. If there is no uncertainty over team values, how can the growth in team sale prices exceed the normal rate of 545 return? Explain. Can your answer persist for a long period of time? Why or why not? 13. Use the speculative bubble explanation to explain the growth rate in franchise values in excess of the normal rate of return. Can speculative bubbles last over a long period of time? Why or why not? 14. What problem is addressed by competition policy? List all of the usual suspects, and discuss their role in that problem. Name the ultimate culprit. 15. What are reasons for pessimism for solving market power problems in pro sports? Be sure to include the lessons from rational actor politics in sports in your answer. 546 SECTION 11 Advanced Problems 1. Explain the progression of organizational forms that a team owner will choose over the time of ownership. Be sure to include the tax implications of each choice. 2. Did the Raiders case have any impact on NFL control over the location of teams? Explain the outcome of the Raiders case using the logic behind Figure 12.5. 3. Characterize the behavior of Congress toward sports league market power. In particular, summarize the outcomes of the Celler, Kefauver, and Sisk hearings. In terms of the explanation in Figure 12.5, what was Congress up to during these hearings? 4. Summarize congressional behavior toward tax issues in sports (roster depreciation, firm reorganization, and excess depreciation). What explains this behavior? Use an explanation based on the discussion of Figure 12.4. 5. Summarize congressional behavior toward antitrust issues in sports (team location, broadcasting, and league mergers). What explains this behavior? Use an explanation based on the discussion of Figure 12.5. 6. If team values are uncertain, how does Bayesian updating lead to growth rates in team prices that exceed the normal rate of return? Explain fully using Figure 12.2. Can your answer persist for a long period of time? Why or why not? 7. Of the three explanations for why the growth rate in franchise prices has been higher than the normal rate of return, which is most plausible? Justify your answer. What is the role of government at both the state and local levels in your answer? 8. Demonstrate graphically how players are better off if leagues are able to keep their market power. Figure 12.3 should prove useful in forming your answer. Given your answer, are unions necessarily against the restrictions on players’ rights to sue individually under the antitrust laws? Explain. 9. Refer to the Learning Highlight: Professor Fort Goes to Washington. Draw a figure similar to Figure 12.5 based on my discussion of the competitive balance hearings in November 1999. Of the three main participants—owners and their supporters, elected officials, and antitrust authorities—who appears to have the most power in the competitive balance hearings? 10. Explain the benefits of breaking up pro sports leagues. Would you favor a breakup? Justify your answer (remember to discuss winners and losers). 547 SECTION 12 References Ambrose, James F. “The Impact of Tax Policy on Sports,” in Government and Sport. Arthur Johnson and James Frey, eds. Totowa, NJ: Rowmand and Allenheld, 1985. Associated Press. “Minor League Teams Create Dreams for Grimy Towns,” 1998. Bast, Joseph L. “Sports Stadium Madness: Why It Started, How To Stop It,” Policy Study No. 85. Chicago, IL: Heartland Institute, February 23, 1998. Coulson, N. Edward, and Rodney Fort. “Tax Policy and the Value of Pro Sports Team Ownership,” Contemporary Economic Policy (2009): forthcoming. Fort, Rodney. “Market Power in Pro Sports: Problems and Solutions,” in The Economics of Sports. William S. Kern, ed. Kalamazoo, MI: W. E. Upjohn Institute for Employment Research, 2000. Fort, Rodney. “The Value of Major League Baseball Ownership,” International Journal of Sport Finance 1 (2006): 3–8. Fort, Rodney, and James Quirk. “Introducing a Competitive Economic Environment into Professional Sports,” in Advances in the Economics of Sports, Vol. 2. Wallace Hendricks, ed. Greenwich, CT: JAI Press, 1997. Horowitz, Ira. U.S. House Select Committee on Professional Sports. Inquiry into Professional Sports, 94th Cong., 2d sess., part 2, September 1976, 131–136. Johnson, Arthur T. “Congress and Professional Sports, 1951– 1978,” Annals AAPSS 445 (1979): 102–115. Johnson, Bruce K. “An Overlooked Implication of Baseball’s Antitrust Exemption,” in Diamond Mines: Baseball and Labor. Paul D. Staudohar, ed. Syracuse, NY: Syracuse University Press, 2000. Johnson, Bruce K. “Why Baseball’s Antitrust Exemption Must Go,” in Stee-Rike Four! What’s Wrong with the Business of Baseball? Daniel R. Marburger, ed. Westport, CT: Praeger Publishers, 1997. L. A. Memorial Coliseum Commission v. NFL, 486 F. Supp. 154 (C.D. Cal. 1979), 726 F. 2d 1381 (9th Circ. 1984), 791 F. 2d 1356 (9th Circ. 1986). Noll, Roger. U.S. House Select Committee on Professional Sports. Inquiry into Professional Sports, 94th Cong., 2d sess., part 2, September 1976, 131–136. Noll, Roger G., and Andrew Zimbalist. “Sports, Jobs, and Taxes: The Real Connection,” in Sports, Jobs, and Taxes: The 548 Economic Impact of Sports Teams and Stadiums. Roger G. Noll and Andrew Zimbalist, eds. Washington, D.C.: Brookings Institution, 1997. Okner, Benjamin A. “Taxation and Sports Enterprises,” in Government and the Sports Business. Roger G. Noll, ed. Washington, D.C.: Brookings Institution, 1974. Quirk, James, and Rodney D. Fort. Pay Dirt: The Business of Professional Team Sports. Princeton, NJ: Princeton University Press, 1992. Quirk, James, and Rodney D. Fort. Hard Ball: The Abuse of Power in Pro Team Sports. Princeton, NJ: Princeton University Press, 1999. “Revenue Gap: Baseball’s Pennant for Sale?” www.gpoaccess.gov/chearings/106scat2.html. Rivkin, Steven R. “Sports Leagues and the Federal Antitrust Laws,” in Government and the Sports Business. Roger G. Noll, ed. Washington, D.C.: Brookings Institution, 1974. Roberts, Gary R. “Professional Sports and the Antitrust Laws,” in The Business of Professional Sports. Paul D. Staudohar and James A. Mangan, eds. Urbana, IL: University of Illinois Press, 1991. Roberts, Gary R. “The Case for Baseball’s Special Antitrust Immunity,” Journal of Sports Economics 4 (2003): 302–317. Ross, Stephen F. “Monopoly Sports Leagues,” Minnesota Law Review 73 (1989): 643–761. Ross, Stephen F. “Break Up the Sports League Monopolies,” in The Business of Professional Sports. Paul D. Staudohar and James A. Mangan, eds. Urbana, IL: University of Illinois Press, 1991. Ross, Stephen S. “Antitrust, Professional Sports, and the Public Interest,” Journal of Sports Economics 4 (2003): 318– 331. Rovell, Darren. “Tax Break Losing Value,” ESPN.com, April 15, 2004. Shugart, William F. II. “Preserve Baseball’s Antitrust Exemption: Or, Why the Senators Are Out of Their League,” in Stee-Rike Four! What’s Wrong with the Business of Baseball? Daniel R. Marburger, ed. Westport, CT: Praeger Publishers, 1997. Veeck, Bill with Ed Linn. The Hustler’s Handbook. New York: G. P. Putnam’s Sons, 1962. Wilson, Duff. “Bill Would Raise Franchise Value of Sports Teams,” NYTimes.com, August 2, 2004. Zimbalist, Andrew. Baseball and Billions: A Probing Look Inside the Big Business of Our National Pastime. New York: Basic Books, 1992. 549 http://www.gpoaccess.gov/chearings/106scat2.html http://www.gpoaccess.gov/chearings/106scat2.html SECTION 13 Suggestions for Further Reading Surely, I can come up with some. 550 CHAPTER 13 College Sports I still chuckle to some degree when people don’t acknowledge that this is a business. Because it is and it’s a big business. —Joe Castiglione, University of Oklahoma AD Sports Business Journal, June 7, 2004, p. 24. CHAPTER OBJECTIVES After reading this chapter, you should be able to: • Describe the workings of supply and demand in college sports, as well as revenue imbalance results, competition between conferences in the pursuit of TV revenues, and the relationship between the university and its athletic department. • Discuss the role of colleges acting through conferences and the National Collegiate Athletic Association in managing conference memberships, negotiating TV broadcasts, and battling competitive imbalance and cheating. • Understand how the amateur requirement, recruiting regulations, and rules regarding the movement of players between colleges affect college athletes. • Recognize the discrimination problems that plague college sports and the role of Title IX in fighting gender discrimination. • Discuss the special tax and market power status enjoyed by colleges in their sports activities, and explain why rational actor politics helps to explain that special status. SECTION 1 Introduction The Association of Intercollegiate Athletics for Women (AIAW), founded in 1971, was an effective body that promoted women’s sports. It served more than 800 member colleges, providing organizing principles, meetings and conferences for the membership, and a national women’s championship structure. In 1980, the National Collegiate Athletic Association (NCAA) began a relentless move to undo the AIAW. The NCAA is comprised of its member athletic directors and university presidents and governs college sports democratically by this membership. The NCAA established its own competing women’s national championship system and used its relatively greater resources to induce colleges to abandon the AIAW. The NCAA offered free trips to the championships and waived membership fees if the university was already an NCAA member. However, the straw that finally broke the AIAW’s back was the NCAA’s inclusion of the women’s basketball championship tournament in the men’s championship TV package and scheduling it at precisely the same time as the AIAW’s championship. The AIAW folded in 1982. In this chapter, we will examine the same economic issues we did with pro sports—demand and revenue, advertising, costs and profit, market outcomes, talent issues, subsidies and their economic impact, and stadium issues. We will also look at government’s role in the college sports business. Our examination of supply and demand will focus on revenue imbalance results and the relationship between the university and its athletic department. We will see that sports market outcomes are the result of athletic directors (overseen by university administrators) acting through conferences and the NCAA. Their joint ventures include the management of conference membership and TV negotiations. The understanding gained concerning the NCAA will help us understand their behavior toward the AIAW. In this chapter, we will also examine the power of athletic directors, as always overseen by university administrators, acting through the NCAA to restrict earnings and player mobility. We will also see that discrimination problems plague college sports and that federal legislation has been enacted to overcome these problems. Finally, we will take a look at the reasons why the actions of athletic directors, conferences, and the NCAA enjoy special tax and antitrust status and show that rational actor politics provides an interesting explanation of that outcome. 552 SECTION 2 College Sports Really Are Big Business As with our analysis of pro sports, in our discussion of college sports we will look past the character building and other life lessons college sports might offer and instead adopt a clear business perspective on college sports. I think you will find it most revealing if we stay the course set throughout this text and follow the money. As with pro sports in Chapter 1, we can cite many examples that prove that college sports is big business. Most understand the linkage between spending and success in college sports. But then why don’t all athletic directors just spend more to increase their success? And if college sports are not a business, why charge admission? Why sell broadcast and sponsorship rights? And why not pay those players who are worth it more than the in-kind payments they already receive? If all of these questions sound familiar, they should. All come straight from Chapter 1 and are just as appropriate for college sports as they were for the pro version. We’ll see some answers as we continue through this final chapter. We can also take a lesson from a couple of long-time observers of the college sports scene. Nearly 15 years ago, Louisiana State University’s basketball coach Dale Brown had this to say: “When I entered coaching 43 years ago, athletics was a component of education. Today it is about making money and winning big. There is a professional intensity creeping into intercollegiate sports” (Sporting News, 1992). According to Texas Tech basketball coach Bob Knight, not much has changed, “If it isn’t a business, then General Motors is a charity” (Sports Business Journal, December 27, 2004). 553 SECTION 3 Demand and College Sports Revenue College sports fans demand the same scarce items as pro sports fans—athletic grace, commonality, relative and absolute quality, and winning. In terms of commonality, college sports games and their coverage by the media provide a common bond among many people. Most people at least know the tune to their local college fight song—and real boosters know the song to the last verse. In terms of quality, once the absolute level is determined, relative competition becomes the object of fan attention. If the best that fans can support at a particular location is Division III sports, they will still care deeply about how their team does against other Division III opponents. Let’s not forget about winning. Fans love the home team, but they love it even more when it is a winner. DEMAND FOR COLLEGE SPORTS From an economic perspective, commonality, quality, and winning are scarce commodities. Therefore, demand is an operational concept in college sports. Indeed, our first example in Chapter 2 was about demand for college sports— men’s and women’s college basketball. Those demands are reproduced in Figure 13.1. As you can see, demand is simply the relationship between prices and quantities demanded. However, let’s not forget that willingness to pay, and all of the elements that determine willingness to pay, is captured in this deceptively simple concept. MARKET POWER Our first observation is that demand slopes downward for individual college sports. The lack of substitutes required for downward-sloping demand and market power follows purposeful choices by college conferences and their enforcement arm, the NCAA. Remember, the NCAA is an association made up of member colleges that supervises college sports. But it also enhances the economic welfare of its members. College conferences and the NCAA perform the same functions for university administrators that leagues perform for pro team owners. For example, the Washington State University (WSU) Cougars and the University of Idaho Vandals are in different athletic conferences and, until 2001, played NCAA football at different levels. The Vandals, previously Division IAA (now referred to as the Football Championship Subdivision, FCS), completed a drawn-out process proving that they deserved Division IA status in football (Division IA is what the Football Bowl Subdivision, or FBS, used to be called prior to 2006). This involved joining the Sun Belt Conference as its most far-flung northern member (other members of the conference were in Arkansas, Florida, Louisiana, Tennessee, and Texas). Just 554 recently, Idaho was able to join the much more proximate set of teams in the Western Athletic Conference (WAC), although there remain two long road trips to Louisiana Tech and Hawaii. Why couldn’t Idaho just declare their intent and move up a level? The answer is that the NCAA has tight controls over who may and who may not play at the top level of college football. A team must prove itself “worthy” to move up to that top level. Conference commissioners and presidents of the NCAA have offered a number of explanations for this exclusionary practice. Essentially, they all boil down to quality control. Each conference commissioner wants to maintain the quality of its product in the eyes of fans, and the NCAA president helps them to do so. However, regardless of the justification, the result is that the number of close substitutes is reduced, and members of individual conferences are endowed with some market power through this exclusion. We will delve deeper into the role of the NCAA later in the chapter. THE DETERMINANTS OF DEMAND Let’s look at attendance. As you can see in Figure 13.1, if ticket price changes, movement along the given demand function shows changes in quantity demanded. Demand shifters, on the other hand, such as preferences, income, price of other goods, expectation, and population, change demand itself. A few of these demand shifters are particularly interesting in college sports. Experience is one of the determinants of college fan preferences. Not that long ago in the grand scheme of college sports, interest in women’s basketball was so low that there was not even an admission charge. However, judging by the rising importance of women’s sports at the college level, especially basketball, as experience with women’s sports has increased, so has the attention of fans. Thus, the demand for women’s hoops probably was once to the left of its location in Figure 13.1. The demand for women’s basketball has been increasing over time, shifting to the right. We will see how this translates into increased revenues for women’s sports shortly. Expectations are another determinant of the location of demand curves. Expectations play a large role in “lifetime” booster privileges for college sports fans. Boosters put up large sums and are promised their seat at the agreed upon price for many years into the future. Lifetime booster privileges are a common way to raise money for the school’s endowment or for a new stadium or arena. By offering such booster privileges, development officers in athletic departments are trying to convert the avoidance of future price increases into a payment in the present for current capital needs. Because most boosters can just wait and buy seats on an annual basis, we are led to conclude that they must be getting a price break on their seat, given their expectation that prices will rise over time. Population, another demand shifter, is part of the explanation for larger- and smaller-revenue markets in college sports. Consider two universities, one in a densely populated area, 555 the other in a rural area. Suppose their football teams were of equal quality. On a map, think about how large a circle you would draw around the large-population university before you included enough boosters to regularly fill the stadium. Do the same exercise for the rural university. The circle for the rural school typically is many times larger. This is the heart of the issue of defining larger- and smaller-revenue markets in college sports. Preferences, as in our pro analysis, also will play an important role in demand for college sports. The most obvious is team quality. Clearly, fans will pay more for higher quality. In addition, although Rottenberg (1956) never extended the idea to college sports, it is interesting to examine the role that uncertainty of outcome plays in the demand for college sports.

Demand, Total Revenue, and Price Elasticity

The relationship between demand and total revenue that we
see in college sports is similar to the one we saw with pro
sports in Chapter 2. Attempts at shifting demand are behind
the myriad of promotions aimed at changing preferences for a
particular college team. Willingness to pay, buyers’ surpluses,
and price elasticity matter for college sports for the same
reasons they do in pro sports. They are the very foundation of
the pricing decisions that come out of athletic departments.
PRICING AND INELASTIC DEMAND
Looking only at attendance, athletic directors would never
price in the inelastic portion of their gate demand function. By
raising price and lowering attendance, total revenues would
increase and total costs would fall. But, as with pro sports,
other revenues also accompany attendance (parking,
concessions, and merchandise). Consider the case of Arizona
State University. In 2003, despite a losing record, football
attendance increased by 27 percent. Clearly, this was
facilitated by lowering the price of attendance. According to
Senior Associate Athletic Director Tom Collins
(USAToday.com, August 30, 2004), “Instead of looking at
making more money [just on ticket sales], we are trying to fill
in all our seats at a lesser cost. If we get everybody in there
and our team performs, they’ll buy a Coke, a T-shirt, a
program, and they’ll come back.” Indeed, even if the team
doesn’t perform, lowering the price into the inelastic region of
demand still sells the attendance-related parking,
concessions, and merchandise, a key lesson from Chapter 2.
SELLOUTS
A puzzle from Chapter 2 also remains for college sports. Why
are prices for some games and events set clearly below the
market clearing price? In these cases in pro sports, we saw in
Chapter 2 that markets for scalpers are introduced whenever
this happens. That the same occurs for college sports is a
matter of your author’s own experience and investigations you
could perform for yourself. Here is one that I observed,
personally. When Washington State played the University of
Michigan in the 1998 Rose Bowl, as a long-time booster I
received six tickets at $50 ($65.40) face value each. It ended
up that I couldn’t attend after all, so after giving two to family
members, I sold the other four. A “broker” advertised in the
556

local paper and set up in a local motel. The transaction took
just a few minutes. I left quite a few hundred dollars richer
but the tickets left before I did, neatly tucked into a FedEx
envelope and on their way to a Michigan booster.
On the investigation side, I looked into ticket prices for the
Men’s 2005 Final Four (all three sessions) in the Edward
Jones Dome in St. Louis. The face value on the tickets,
available from the NCAA, was $110 ($120) and $130 ($142)
per pair. But these were only available by lottery from the
NCAA on a relatively small number of seats. But have no fear,
other tickets were available. A quick Internet check revealed
that the NCAA prices were far below equilibrium. One could
easily spend $1,200 ($1,310) for a pair. The highest advertised
price I discovered was $14,000 ($15,280) for a pair of luxury
box tickets. Just why it is that the NCAA prices their own
tickets as they do and distributes the rest as they do is a
puzzle. Perhaps some of the explanations from Chapter 2 will
also end up being enlightening here.
PRICE DISCRIMINATION
Price discrimination mechanisms such as variable seat prices
and two-part pricing are common in college sports. Two-part
pricing explains donation seating, one of the mainstays of
college sports pricing. Boosters contribute money to the
athletic department and, in return, they receive access to
priority seating. The boosters also receive membership in
booster organizations. To the extent that these boosters would
have purchased the same seat anyway, the price to the fan is
higher. As long as booster organization membership benefits
do not offset the entire amount of the contribution, athletic
department revenues are higher from the same seats.
It is tempting to chalk this up to price discrimination, but we
must exercise caution. Were regular season tickets renewable
into perpetuity? If not, then donation seating that extends the
period of purchase rights is selling a different good. Do any
other goods and services accompany donation seating that did
not accompany regular season tickets? If so, then, again,
donation seating is a different good than the usual season
ticket seat. Often special hospitality benefits and membership
in special donor clubs goes along with donations. In summary,
even if the seat is located in the same place in the stadium, if
the donation buys additional goods and services, then it
cannot be a price discrimination device.
Interestingly, even some athletic directors with football teams
typically ranked in the Top 25 held off on two-part
contribution seat pricing until very recently. Among the Top
25 in 2004, Michigan, Iowa, and Wisconsin had no such
policy. But all three implemented donation seating for 2005.
A few, like Oklahoma and Minnesota, do not require any
contribution but do use contributions to determine
preferential seating. All of the athletic directors in the
remaining 20 schools in the Top 25 had long used donation
seating, and directors of most of the legendary programs
collect a contribution from every season ticket holder
(Southern Cal, Georgia, Miami, Texas, and Florida). As with
PSLs in the pros, apparently there are other factors that
557

determine just when the right time occurs to implement
contribution seating in college sports. The same holds true for
differential seat prices over the season for the same seat, also
in use in college sports.
THE REVENUE DATA FOR COLLEGE SPORTS
The NCAA sponsors surveys of college team financial
outcomes. Table 13.1 shows the results of this survey for
Division IA (the FBS from 2006 on) from 1985 to 2005 (as an
exception to our usual approach, these data are all in 2009
dollars only because the table would be quite cumbersome).
The data, where possible, are shown both with and without
“institutional support.” Institutional support is money that is
allocated directly from the university’s operating budget to be
spent on athletics. It may or may not be identified separately
in an athletic department’s report of revenues, but it is
revenue nonetheless. This separation will prove crucial later
in the chapter where we try to judge the economic status of
college sports.
As you can see from Table 13.1, if we include institutional
support, modern revenue averages are in the tens of millions
of 2009 dollars, and the largest report was over $120 million
in 2004. The average reported revenues rose over the entire
period, 1985 to 2005, by 167 percent in real terms, for a real
growth rate of 5.0 percent per year. Remember, the typical
real annual growth rate in the U.S. economy is around 3
percent. So the growth of the average of reported revenues is a
remarkable 1.7 times higher than the typical growth rate in
the U.S. economy. The same is true in spades for the largest
reported revenues—178 percent over the entire period for a
real growth rate of 5.2 percent per year. On the other hand,
omitting institutional support shows a much smaller 64
percent increase overall for the average of reported revenues,
a real growth rate of 4.6 percent per year. Clearly, institutional
support is an important revenue source, at least for school
close to the average of reported revenues.
REVENUE INEQUALITY IN COLLEGE SPORTS
Table 13.1 also reveals substantial revenue imbalance in
college sports. The ratio of the top reported revenues to the
average always exceeds two and exceeded three from 1999 to
2004. We must exercise caution in our conclusions about the
behavior of revenue imbalance over time because the survey
data may be affected by better response rates in recent years.
However, Zimbalist (1999, p. 117) also showed that the ratio of
top to average revenues grew from 1.81 to 3.38 over an earlier
period, 1962 to 1985.
Here are some cautious conclusions about revenue inequality.
From Table 13.1, we see that revenue imbalance in college
sports did not change much from 1985 to 1997 but jumped
considerably from 1999 to 2004. The ratio fell substantially in
2005, but only time will tell if this is a return to the levels
prior to 1999 or not. This is dramatically different from the
revenue disparity in pro sports outlined in Chapters 2 and 6.
There, we saw the top-revenue teams gain dramatically
against the bottom, but the average held its own against the
558

top. In college sports, the top teams are outpacing the average
and, by simple arithmetic, the bottom as well. As we will later
see, revenues are just as important for winning in college
sports as they are in pro sports, and this revenue inequality
will help to explain competitive imbalance.
Another interesting observation from Table 13.1 is that the
ratio of the top reported expenditure to the average was fairly
steady from 1985 to 1995. Since then, this expenditure ratio
has taken a bit of a roller-coaster ride, rising 1985 to 1999,
falling from 2000 to 2003, and then rising again in 2004.
While spending and revenue were typically nearly equal to
each other at the average reported levels, quite a gap opened
between the two at the largest reported levels over the period
of 1999 to 2001. The gap started closing after that and appears
to have closed completely as of 2005. However, more data will
arrive only with the passage of time in order to tell if this
remained the case after 2005.
REVENUE AND SPENDING DIFFERENCES: MEN’S AND
WOMEN’S SPORTS
Table 13.2 compares revenues and expenses for men’s and
women’s programs in FBS athletic departments (again, due to
sheer bulk, only the values in 2009 dollars are shown). A
number of things jump out. First, revenues for men’s sports
are dramatically larger than for women’s sports. In every year
in the table, men’s revenues are about 10 times those
generated by women’s sports. Second, while men’s sports
revenues have grown at a healthy 2.8 percent real annual rate,
women’s sports revenues have grown in real terms at 7.1
percent annually, more than 1.5 times higher. As noted
earlier, women’s sports have gained in acceptance over time,
and this is reflected in tremendous revenue growth. Third, the
growth in spending on women’s sports is just over twice as
high as for men’s sports, even though the level of spending
still lags dramatically behind spending on men’s sports.
Finally, we shouldn’t make too much of the result in Table
13.2 that shows losses on women’s sports increasing
dramatically over time. This is the point of Title IX programs
aimed at equalizing college sports opportunities for women, a
topic we’ll cover later in this chapter.
Table 13.3 shows revenue and expenditure percentages for
different men’s and women’s sports. It is no surprise that
football dominates men’s sports in both categories.
Interestingly, basketball generates about the same
percentages of total revenues and expenses for both men’s
and women’s programs. It also is interesting to note those
women’s sports revenues are not dominated by any one sport.
Basketball is important, but the other-sports category swamps
it.
SPONSORSHIP
Sponsorships have become the norm in college sports. Every
athletic director sells space on uniforms for advertising,
typically to athletic shoe and apparel companies such as Nike,
Reebok, and Adidas. Conference championships are
sponsored by corporations for millions of dollars that are then
559

reallocated back to member athletic departments. Lately, as
with the pros, athletic directors are selling stadium and arena
naming rights. Recent naming-rights amounts are shown in
Table 13.4. As you can see, the amounts are typically a few
hundred thousand dollars annually. These amounts pale
compared to postseason bowl game sponsorships, but we’ll
leave that discussion to our next topic, college sports on TV.
560

SECTION 4
The Market for College
Sports Broadcasts
Believe it or not, at the origin of television in the late 1930s,
there was a trade-off between values earned by broadcasting
games and attendance revenue at the gate. By the 1950s, the
NCAA had noticed an 11.4 percent drop in football attendance
from the late 1940s coincident with increased appearance of
games on TV (Brown, 1999). In 1952, the NCAA instituted its
first restrictive broadcast policy, overwhelmingly endorsed by
a referendum of the membership. An interesting difference
between college and pro sports is that this anticompetitive
behavior by the NCAA was ruled illegal under the antitrust
laws. We’ll see just how this treatment under the antitrust
laws occurred, and to what effect, when we examine the NCAA
in detail later in the chapter.
In Chapter 3, we developed an approach for analyzing the
market for broadcast rights that followed the flow of
programming and money. From that perspective, there is one
important difference between college and the pros. In college
sports, conferences negotiate contracts with media providers
under a special three-tier broadcast rights structure.
The first tier is a national contract between a conference and a
major media provider. Any games that the major networks do
not choose to show revert to the second tier, national cable,
primarily ESPN. The games not chosen by the first- or
second-tier providers are available to the third tier, regional
cable or local broadcasters. For sanity’s sake—and for the sake
of some advance notice on advertising slot sales—there is a
time limit for final decisions on which games are chosen for
the first tier. Second- and third-tier broadcasters then make
their choices.
More economic competition exists on the supply side of sports
programming in college sports than in pro sports. In pro
markets, if media providers want sports programming, they
have no alternative but to deal with a single pro league in each
sport. In the market for college broadcast rights, if the
representatives of one conference are unreasonable in
negotiations with media providers (competitively speaking),
there are other college conference representatives that can
provide sports programming. As a result, we cannot be quite
as sure that the bulk of the value of market power flows
primarily to college conferences (and back to member
colleges) as it does to pro sports leagues (and member
owners).
The most recent manifestation of this competitive difference
is in the rise of media provision by college conferences
themselves. Providers like The Big Ten Network and The SEC
Network are competing with other media providers like
ESPN. They “eliminate the middleman” by producing and
561

providing sports and other university programming directly to
cable providers. To date, these offerings include only the
games in major sports that no other traditional media
provider wants, minor sports, and educational programming.
Given the power of this direct provision, it is interesting to
observe how the directors of conference networks behave and
that is all we have been able to do so far with these very new
entities.
In a recent example, directors of The Big Ten Network took on
Comcast over placement of its broadcasts in the Comcast
pricing scheme. Comcast wanted to put the network’s
programming into a special tier offering, available only at an
additional cost over regular subscription. The Big Ten
Network directors demanded that Comcast put the
programming in the basic package at no additional charge
over the subscription price. Now, let’s assume that Comcast
knows how to make the most money from program pricing.
We are left to conclude that, for the short term, The Big Ten
Network directors traded off higher rights fees for some other
objective. One possibility is “growing” a broader base for an
eventual expansion of direct sales—more important games in
major sports may eventually find their way to the Big Ten
Network. Eventually, one wonders if technology will actually
allow complete choice directly beamed from the conference
network themselves to individual media receivers! This is a
truly exciting new area to apply the economic way of thinking
about sports.
BROADCAST REVENUES
Table 13.5 shows the most recent TV contracts for FBS
conferences. The amounts in Table 13.5 make clear the
regional appeal of college sports conferences compared to the
nationwide appeal of professional sports leagues. For
example, NFL TV rights are about $16.8 billion for their
current contract ending in 2014, and NBA TV rights are about
$7.4 billion for their current contract ending in 2015 (Chapter
3). Even if we assigned the highest comparative values for the
unavailable data in Table 13.5 and added in other values
earned by college conferences from postseason TV, college
football and basketball come nowhere close to the level of TV
revenue enjoyed by their professional counterparts.
College conferences do have one thing in common with the
pros—TV revenues are unevenly distributed across
conferences. Comparing national contracts and adjusting for
differences in contract duration, the Big Ten and SEC earn
twice the Big 12’s level, and there simply is no comparison for
conferences like C-USA, the Mountain West, the WAC, and
the MAC. But note that a single school, Notre Dame, has a
football TV contract worth $9 million annually, or about one-
third the football revenue earned by some entire conferences
(like the Atlantic Coast [ACC] and the Pac-10). Further, within
conferences, those teams that appear on television more often
make more than those that do not. In the Pac-10, for example,
TV and radio revenues for the University of Washington
Huskies are typically about twice the amount collected by the
WSU Cougars.
THE INTERESTING CASE OF NOTRE DAME
562

The case of the University of Notre Dame is interesting in this
regard. It remains the only major independent team in the
country. Because its fan base is nationwide, major networks
find it valuable enough to contract separately with Notre
Dame. According to Dick Ebersol, NBC Sports Chairman
(USAToday.com, December 22, 2003), “We covet our
association with Notre Dame, the most powerful brand in
college sports.” Whether you agree with him or not, the result
is $9 million per year for the Notre Dame athletic department.
BOWL GAME REVENUES
As you can see from Table 13.6, the largest sponsorships go
to postseason bowl games. In the past, college football bowl
games were just tourist attractions, drawing people to the
southern United States during the onset of winter in the
North. As such, they typically were organized and run by local
business interests such as the Chamber of Commerce. But
eventually fans became even more interested in bowl games
for their national championship implications.
The “Bowl Coalition”—comprised of some then Division IA
conference commissioners, the athletic director at Notre
Dame, and representatives from the major bowls—was the
first effort to take the job of crowning the national champion
from media and coaches’ polls. Polls were still part of the
formula, but so were more “scientific” elements including
performance and quality of opponents. The Bowl Coalition
tried to do this originally by matching ACC, Big 8 (now Big
12), Big East, SEC, the (now defunct) Southwest conference
champions, and Notre Dame, in the 1992 Cotton, Fiesta,
Orange, and Sugar Bowls. Big Ten and Pac-10 champs were
locked outside of this coalition because the Rose Bowl would
not release their teams to the Bowl Coalition. Note that other
Division IA conferences were left out completely. This
approach was fine-tuned under the Bowl Alliance after the
1994 season for the 1995 New Year’s Day bowls. The same
participants were involved (except that the Southwest
Conference folded, and some of its teams joined the Big 8 to
form the Big 12). Again, the rest were excluded except for a
couple of at-large contingencies.
At present, originally labeled the “Super Alliance,” the modern
Bowl Championship Series (BCS) was further refined for
bowls that followed the 1998 season by adding the Rose Bowl,
and thus the commissioners of the Big Ten and Pac-10
conference, and dumping the Cotton Bowl. ABC retained the
rights to the Rose Bowl, and FOX broadcast the other three
bowls. The game between the number 1 and 2 teams rotated
through the four bowls, and many conferences, except for
unlikely at-large possibilities, remained excluded. Not until
the 2004 postseason games were the remaining five FBS
conferences put into the BCS formula but then only if they
were ranked high enough. The BCS Championship game was
added (also added to the FOX contract) for bowls that
followed the 2006 season (and Division IA became the FBS in
this same year) so that still more teams would be eligible for
BCS play in the determination of the top 10 teams in college
football. The broadcast lineup still remains exclusive to ABC
563

and FOX, but ESPN will take over the former FOX games in
2011. Since ABC owns ESPN and often presents crossover
broadcasts, essentially all BCS games will be the exclusive
property of a single media provider.
Although the organization and promotion of the bowl remain
primarily the job of local officials, large financial injections by
sponsors have dramatically increased the value of these
activities. Especially through the BCS organization of the
national championship, firms have become much more
interested in sponsoring the BCS games. In Table 13.6, note
that all BCS games (except the Rose, for the year in the table)
and many of the non-BCS games are sponsored events.
Unless we make too much of recent BCS sponsorship values,
all of the BCS bowl payouts have grown dramatically over the
last 30 years. Figure 13.2 charts average non-BCS and BCS
payouts to each competing team for about 30 years in 2009
dollars. At this average, BCS bowl payouts have always
dwarfed non-BCS bowl payouts. In addition, while non-BCS
average payouts have remained almost completely steady for
30 years (of course, in real terms), average BCS bowl payouts
have typically grown in leaps and bounds (although they have
fallen a bit in the last few years).
Looking behind the scenes, over the same period depicted in
Figure 13.2, the most startling individual BCS payout growth
occurred for the Orange Bowl—16 percent annually adjusted
for inflation. This is startling compared to the typical real
annual growth rate in the economy of 3 percent. The Fiesta
Bowl payout has risen at 8 percent annually, adjusted for
inflation and the Rose Bowl at 4 percent. Of course, the Rose
Bowl also was historically the highest payout, so its relatively
lower rate of growth is not surprising. Only the Sugar Bowl
payout has remained relatively steady since 1981. However, at
its highest payout to Notre Dame in 2006, the Sugar Bowl had
matched the growth rate of the Fiesta Bowl at 8 percent
annually in real terms. Until the 2006–2007 bowl season
when the BCS Championship game first determined the
national championship, these other BCS games performed
that function. Clearly, fan interest is in the national
championship outcome, and sponsors know it. Table 13.7
lists all conference TV, NCAA distributions that come
primarily from the NCAA basketball championship referred to
as “March Madness,” and bowl earnings aggregated at the
FBS conference level. The point of putting all of these
revenues in one table is to show that postseason money is a
mainstay for all FBS conferences, although less so for the Big
Ten and SEC. For the other nine conferences shown in the
table, postseason revenues almost equal conference TV
revenues for the Big 12, ACC, and Pac-10 and are much larger
than that for the rest. And even for the Big Ten and SEC, these
postseason revenues are over 25 percent of the total in the
table.
NCAA BASKETBALL TOURNAMENT REVENUES
All FBS conferences also earn revenues from participation in
the NCAA basketball tournament, known as March Madness.
So do many other conferences that also play at the top-level of
564

college basketball, referred to for basketball purposes as
“Division I.” All of this money flows from the $10.8 billion
contract for this tournament between the NCAA and CBS and
Turner Broadcasting running from 2010 to 2024. At the
average, that’s $771 million annually. This amount is larger
than the last published NHL contract on an annual basis; a
few weeks of college basketball is worth more on TV than an
entire NHL season.
As with BCS revenues, the March Madness TV contract has
increased dramatically over time. In 2009 dollars, the
previous CBS contract paid about $6.2 billion for 15 years, or
about $413 million annually. For the FBS conferences in
Table 13.7, the NCAA distribution that comes from March
Madness averages just under $20 million, much larger than
regular-season basketball TV revenues for the Pac-10
conference, for example (again, see Table 13.5).
565

SECTION 5
Cost, Budgets, and Quality
In our pro sports analysis, we assumed that owners were
motivated by profit maximization. Given this, we developed
long-run and short-run cost notions and the long-run profit-
maximizing quality choice. We then found that profit pursuits
constrained winning and that profit variation could harm
competitive balance. We then tried to determine what we
could learn from operating revenue statements relative to the
true economic value of owning a team.
We will follow a similar approach for college sports. However,
we need to be careful in choosing our level of analysis. Where
the owner was the unit of analysis in pro team sports, in
college sports, it is university administrators that make
decisions about the level of resources devoted to generating
athletic quality, considering the value of that activity to their
overall pursuits. These administrators are unlikely to act to
maximize profits. Indeed, many are explicitly directed not to
behave that way. Most state colleges, for example, are charged
with many different tasks that have nothing to do with profits.
UNIVERSITY ADMINISTRATOR GOALS
Let’s consider the goals of university administrators and the
typical organization of a university’s many departments.
Then we’ll turn to the relationship between the university and
its athletic department. Who are the actors, and what are their
goals? How is the university organized to pursue those goals?
What does this all suggest about the way that university
administrators fit athletics into their scheme of operations?
Figure 13.3 shows the organization of a typical university. As
with all generalizations, this simple diagram will not hold for
every university, but it will be instructive enough for our
purposes. The university provides the structure and support
for research, teaching, and service to the state. It employs its
inputs, organized into departments and administration, to
generate the funding required for these tasks. The funding can
come directly from the tasks done by its faculty and
administrators or indirectly through political support.
THE UNIVERSITY AND THE ATHLETIC DEPARTMENT
The diagram in Figure 13.3 helps make clear what university
administrators expect of all their university’s component
parts. In this section, we will compare how the athletic
department contributes to the university relative to the college
of business (which might be called the business “school” at
your university). We will stress the similarities of the two
entities, the most important being that the university does not
expect individual units to show profits but to contribute to its
overall goals.
566

THE BUSINESS COLLEGE VS. THE ATHLETIC
DEPARTMENT
Figure 13.4 depicts the organization of a college of business.
Faculty members of like interests are grouped into
departments, and each department has a chairperson. For
example, the economics department in a business college will
have a number of faculty members of different teaching and
research interests organized under its chair. Departments
with a common thread to their disciplines are then organized
into colleges, in this case, the college of business. Thus, the
accounting, marketing, information systems, finance,
management, economics, and decision science departments
might comprise a business college. At the top of the college, in
terms of administration, is the dean. Typically, associate
deans handle the day-to-day tasks of running the college so
that the dean can perform fundraising and external relations
duties. The dean answers to the provost and on up the ladder
through the board of regents to the governor.
Only two departments are shown in Figure 13.4. The
economics department usually cannot generate enough
funding to pay for itself. The information systems department,
on the other hand, is more likely to generate enough grant and
contract activity to more than cover the cost of its faculty. The
ability of the college to grow or at least to avoid cuts in its
budget depends on the success of its departments. Success is
measured along the lines of university administrator’s
objectives—research, teaching, and service. As you would
expect, successful colleges do better during the university’s
budget allocation process.
Interestingly, the athletic department is not organized much
differently, as shown in Figure 13.5. In this case, each sport
is like a department. Each sport has its faculty equivalent,
assistant coaches, and its chair equivalent, the head coach.
The sports are organized around their own dean equivalent,
the athletic director. The athletic director has associate
directors who oversee the day-to-day operation of the athletic
department while they attend to fundraising and other
activities. As with the college of business, the remaining part
of the chart runs right up to the governor (the difference being
that the president, rather than the provost, usually oversees
the athletic director).
Football and women’s volleyball are the only sports depicted
in Figure 13.5. The former typically more than pays its bills,
whereas the latter does not. In a sense, the football program is
like the information systems department in the college of
business; both the football program and the information
systems department pay for themselves and contribute
dramatically to the success of their immediate organization by
performing well along the dimensions that matter to
university administrators. In the information systems
department, the contributions are teaching and research. In
the athletic department, it is service in terms of entertainment
enjoyed by thousands, as well as in terms of producing some
productive sports professionals (either professional players or
future coaches). Although I know of no university that offers a
567

degree in the subject, athletic departments certainly do teach
athletes their sport and how it is coached. Many athletes
make a living from that knowledge through playing or
coaching.
Academic departments contribute to the university’s goals of
research, teaching, and service. Referring back to Figure 13.3,
rewards are dollars directly from students, a budget from
state legislative allocations, and research grants and contracts.
Athletic departments also generate resources for the
university’s pursuits. These contributions will not be direct
dollar amounts (with very few exceptions, athletic
departments do not have any money left over but this is a
complex issue we will return to later in the chapter). However,
the university benefits from indirect contributions that follow
from the activities of the athletic department. One of the most
important types of indirect support is political support.
POLITICAL SUPPORT
Political support is usually analyzed by looking at political
individuals. Some of these individuals are politicians who are
sports fans. Others are boosters who happen to be politically
influential individuals even though they are not elected
officials. These boosters may not be interested in the rest of
the university’s mission, but they will throw their political
support to the university as long as the university supports the
athletic department. In addition, athletics appeals to general
fans who are also taxpayers.
The political support of athletic departments is common
knowledge. In a letter to Sports Illustrated, a Mr. Jim Norton
wrote, “For better or worse, recognition results in more
alumni donations, more student applications and, most
important, more support from state politicians, who control
the university budget” (November 28, 1994, p. 4). The last
part of the quote shows that political support really does
matter. According to Humphreys (2006), institutions fielding
FBS football programs receive 8 precent larger annual state
appropriations than those without such programs. He goes on
to suggest that this helps explain why the number of
institutions fielding FBS college football programs increased
by 10 percent from 1998 to 2002. Thus, it shouldn’t be
surprising that the governor of the state typically attends the
college football intrastate rivalry game every year. However,
the rest of Mr. Norton’s quote hints that there are other
dimensions of support, such as alumni contributions and
increased visibility with potential students. But what does the
evidence show on these spillover benefits?
ATHLETIC DEPARTMENT SPILLOVER BENEFITS
Athletic departments are purported to provide spillover
benefits to the university, including the following:
• Greater giving by alumni and other boosters to the general
university fund
• A larger and better set of student applicants
• Better faculty and administrators
568

• Value added to athletes, many of whom would not be at the
university without athletics
There is scattered evidence showing that college sports
provide some of these spillover benefits. Goff (2004) surveys
the literature in the first two of these areas, updates a few with
his own empirical analysis, and offers some general
conclusions. Major achievements (significant postseason
appearances) appear to spark general giving. But marginal
increases in winning during the season don’t. Major
achievements also appear to spark increased student interest
even at colleges that already have high academic reputations.
But the quality of the pool is better only at very select
institutions. Goff also reminds us that there are actually only a
few works in each area and that those works do not address
the same data. Tucker (2004, 2005) adds that the success
increases the percentage of alumni that give to the general
university fund, increases the SAT scores of entering
freshmen, and enhances graduation rates. Cornell University
economist Robert Frank (2004, p. 33; Reprinted with
permission from author), in a review for the prestigious
Knight Commission on Intercollegiate Athletics, sums up his
reading of nearly the same literature as follows:
The most forceful conclusion that can be drawn about
the indirect effects of athletic success is that they are
small at best when viewed from the perspective of any
individual institution. Alumni donations and
applications for admission sometimes rise in the wake
of conspicuously successful seasons at a small number
of institutions, but such increases are likely to be both
small and transitory. More to the point, the empirical
literature provides not a shred of evidence to suggest
that an across-the-board cutback in spending on
athletics would reduce either donations by alumni or
applications by prospective students.
Much more remains to be done before there is anything
resembling a consensus on these first two of the listed claims,
and I know of no work at all on the last two.
INSTITUTIONAL SUPPORT
The university budget typically includes money for both
academic and athletic departments. Academic departments
receive their share, as do all departments, through a
competitive budgeting process at the university level. The
direct payments to athletic departments out of university
budgets are referred to as institutional support. There are also
other indirect subsidies from the university to the athletic
department. Given our model, institutional support and other
subsidies make perfectly good sense from the perspective of
university administrators since athletic departments
contribute to their research, teaching, and service goals. This
finding sheds light on arguments that college sports do not
belong at the university because they are a losing proposition,
financially.
569

The university administration is willing to pay for both the
economics department and the football program because of
the contribution that each makes to the university’s goals.
Primarily through service to fans, and politically potent
boosters in particular, the athletic department generates
political support for the university. It is not just the bottom
line of this division that determines its worth to the
university. It is its contribution to the bottom line of the
organization, taken as whole, that determines its worth. Here
is where the ultimate similarity between the athletic
department and other university divisions lies: All make their
contribution and receive their reward.
UNIQUENESS OF THE ATHLETIC DEPARTMENT
Faculty in academic departments are teaching and research
oriented with service a distant third in nearly all cases.
Denizens of athletic departments, on the other hand, are
primarily service (entertainment) oriented while also aiding
athletes in their academic pursuits. We should expect
university administrators to allow administrators of each
department to play to their department’s strength in order to
get the highest return from each. After we examine how
athletic departments differ from academic departments, we
will see how the university treats them differently, primarily
through the budgeting process.
DIFFERENCES BETWEEN ATHLETICS AND ACADEMICS
There are three major differences between athletic
departments and academic departments. First, it is much
more difficult to forecast revenues from the athletic
department since success is uncertain and revenues relate to
success. Academic department revenues are well known at
the beginning of each budget session based on enrollments
and known grant and contract income.
Second, even though it is more unpredictable for the athletic
department, success is much more objective in the athletic
department than in academic departments. Contests on the
field with objective scoring pretty much guarantee very little
disagreement on this point. In contrast, analysis of the success
of individual faculty careers is much more subjective, varying
by discipline and even within disciplines. In addition, a
successful academic career may take many years to come to
fruition.
Finally, the relation between quality and the output that
actually is sold at a unit price is clear in the athletic
department. This is not true of academic departments.
Departments do not charge students by the lesson or by the
hour. Research is performed in small pieces that may take a
long time; thus, grants and contracts can run over extended
periods.
These differences mean that university administrators will
treat athletic directors differently than administrators of
academic departments. From the perspective of the university
administrators, differential treatment allows the athletic
director to contribute to the university administration’s goals
in the most beneficial way. Some view this differential
570

treatment as preferential treatment. I think you will see that it
is not; indeed, “it’s just business.”
THE ATHLETIC DEPARTMENT BUDGETING PROCESS
Differential treatment of athletics and academics is most
apparent during the budgeting process. The budgeting
process in research and teaching branches of universities is as
follows: Deans in individual colleges in the university
determine the needs of their departments, put them together
in a budget request, and present their case to the university
administration. During the subsequent budget period, college
faculty and administrators spend what they are given. If
spending is lower or revenues are greater than anticipated,
the university administration retains control over any
balances for future budget hearings.
The situation is quite different for most athletic departments.
Like their academic counterparts, the athletic director
determines his or her department’s needs and receives part of
its budget from the university administration. However, if the
department generates unexpected revenues, revisions can be
made during the budget period to reallocate them into
ongoing expenditures, new and ongoing special facilities
projects, and salary contracts that also pay into the future.
Thus, the athletic director is allowed to spend surplus funds
however he or she sees fit.
This differential treatment is easy to illustrate. Table 13.8
shows the athletic department balance sheets for the 1997–
1998 WSU Rose Bowl season and the year before. The
example is a bit dated, but I think it is instructive because it
uses a department that does receive institutional support and
a decidedly distinct pair of seasons. Revenues increased $2.9
million during the Rose Bowl year, an increase of about 21
percent (the inflation rate was only about 2 percent, so this is
quite a large real increase). The bulk of the revenues came in
identifiable large quantities—about $630,000 from TV
broadcasts, another $391,000 in contributions, $219,000 in
royalties, $350,000 in miscellaneous, and $141,000 in
merchandising. In light of this surge in revenue, how in the
world can it be that the athletic department only managed to
break even? In this case, as in most, the excess revenues were
simply kept by the athletic director and spent.
One way the athletic director distributes the excess revenues
is through salary increases. In our WSU example, salaries rose
after the Rose Bowl. The head coach’s contract, worth about
$253,000 (including all payment sources as well as base
salary) for the 1997–1998 season, was increased to about
$450,000 after the Rose Bowl (Spokane Spokesman-Review,
January 13, 1998, p. C1). Clearly, gains are sent straight back
to the expense side of the ledger.
The university does not have to let this happen, and many
athletic department observers argue that it should not.
However, it makes sense from the perspective of the
university’s goals that the athletic director be allowed this
special budget freedom. Remember, success is uncertain and
revenues are tied to success. When success does occur (as
with a team like WSU going to the Rose Bowl), it is the
571

university administrations best interest to convert the
additional revenues to more of the service done by the athletic
department. The additional spending by the athletic director
is parlayed into additional, unexpected, but welcome political
support for the university.
PAYING THE BILLS
Athletic departments also differ from academic departments
in how the bills are paid. Faculty in academic departments
pay overhead on all grants and contracts. If faculty members
receive outside funding for their work, the university requires
that the granting agency add between 20 and 45 percent of
the grant total to be paid to the university. This is justified
because university facilities and resources are used to pursue
funding and actually carry out the funded research.
Typically, athletic directors do not pay overhead to their
university. At athletic events, the athletic director pays for
game management (the personnel to collect tickets and
control crowds, including armed police presence). They may
also incur cleanup and minor repair costs. The university
picks up the rest of the costs, including rent and maintenance
that other facility users would have to pay. The university also
typically handles student pass production and distribution
and absorbs the cost of doing so. Summer activities, including
income-generating camps run by coaches and training
sessions, receive the same benefits from the university.
An athletic director that paid the same overhead charges that
faculty in academic departments pay would contribute
between $25,000 and $45,000 per $100,000 earned. At
WSU, judging from the revenues in Table 13.8, the athletic
department’s overhead payment would have been as much as
$6.75 million in 1997–1998.
However, this is all just a matter of incentives and rewards.
University administrators do not get their return from the
athletic department in terms of the department’s financial
bottom line. As long as the athletic department contributes
more to the goals of the university than it costs the university,
the athletic department is worth it to university
administrators. The athletic director keeps the department’s
net revenue, and the university helps pay its bills. If the
athletic department does not contribute to the university’s
goals or it becomes costly to the university in embarrassing
ways (covered in a later section), then it may be punished.
Indeed, in one of the most famous cases in the history of
college sports, intercollegiate football was banned at the
University of Chicago in 1939 by the then president Robert
Maynard Hutchins. Under legendary coach Amos Alonzo
Stagg, the Maroons were a storied team winning seven Big
Ten titles and never once losing to Notre Dame. Hutchins is
reputed to have based his decision on unwillingness to cheat
on NCAA rules when (in his opinion) all about him were doing
so. But it also is true that the Maroons lost every game in 1939
and never scored a point (Stagg had left for Pacific two years
earlier). It took 30 years for the game to return—the
university has played Division III football since 1969.
572

THE RESULTS FOR ATHLETIC DEPARTMENTS
College athletic departments typically receive considerable
support from their universities. In general, for FBS athletic
departments, this can be seen referring back to Table 13.1,
where net operating revenue outcomes are shown for
departments with and without institutional support. With
institutional support, in all years but 1985, the vast majority
of athletic departments show some positive net revenue.
Without that support, except for 1993 where the average is
breakeven, the majority of departments would report losses
rather than profits. This leads critics like Murray Sperber
(1991) to argue that, given the other problems with college
sports, they should be abandoned by universities if they can’t
stand on their own financially.
But this view misses the point entirely. First, there are some
athletic departments that do not draw institutional support at
all (my own University of Michigan is one such). But even
more importantly, the university helps cover the athletic
department’s costs through institutional support as long as
that department contributes enough to the goals of the
university administration. Arguing that the athletic
department should be abandoned because it doesn’t cover all
of its costs holds just as true of English departments (like
Professor Sperber’s) at most universities. But the economic
way of thinking suggests that universities view their English
and athletics departments similarly—as long as their
contribution to university administration goals exceeds the
costs, they all are worth the money. Besides, if institutional
support were withdrawn, the athletic department would
simply reduce spending, just as any department at the
university would if faced with such a budget reduction. The
“losses” claimed in the absence of institutional support are
illusory.
Further, breaking even is not always the goal of either the
university or its athletic department. Table 13.9 contains a
comparison of Washington State and its cross-state rival, the
University of Washington. Both universities reported budget
surpluses for 2003–2004 (refer to the far-right totals
column). Net before amortization is just under $1 million for
Washington State and just over $1 million for Washington.
Interestingly, neither athletic department returned any of its
surplus to the university. Instead, long-term capital ambitions
took up the surplus balance. Our diagram of this process
suggest that this occurs because athletic department spending
flexibility helps the university reach its goals.
As a brief summary of our findings about college athletic
departments thus far try to picture the successful athletic
director. Their program is successful on the field and revenues
grow. Under flexible budgeting, net revenues are plowed back
into the program in a variety of ways and… the program just
breaks even every year! Salaries grow. Coaching is one of the
obvious areas of spending focus. Florida football coach Urban
Meyer signed a deal in 2009 paying a reported $4 million
annually—third highest in football behind Pete Carroll at USC
($4.4 million) and Charlie Weiss at Notre Dame ($4.2
million). And let’s not forget the pay to the successful athletic
573

director—Meyer’s Athletic Director at Florida, Jeremy Foley,
earns about $1.2 million per year. But other areas of spending
are also important in competition—facilities and recruitment.
At truly successful places like Florida, net operating revenues
go into capital funds for projects into the future. Other
contributions like participation and gender equity also come
into play (as discussed later in this chapter). Most
importantly, the overseers of the athletic department,
university administrators, reap their political and other
support rewards from this successful program.
POSTSEASON MONEY AND ATHLETIC DEPARTMENT
BUDGETING
We have neglected to discuss one important source of funds.
What about bowl revenues and the revenue sharing from the
championship basketball tournament shown back in Table
13.7? In general, the sharing of postseason revenues already is
built into a team’s expected spending. All athletic directors
know the amounts that will come to the conference and how
much will be shared among the member athletic departments.
In short, athletic directors know how much their conference
will bring in, and make plans to spend it.
But what about those teams that are fortunate enough to
actually go to a bowl or other postseason play? They get a
larger share of the postseason revenues, but they also spend it.
Partly, this spending is expected by the localities that put on
bowl games, especially for lower-level bowls as shown in the
Learning Highlight: Washington State University Copper
Bowl Spending at the end of the section. The university
administrators also see the reinvestment of these revenues as
important to the contribution made by the athletic
department to the university’s overall goals.
The Sports Business Journal (March 12, 2001, p. 40) cites
another example. In 2001, its first trip to the Rose Bowl since
1967, the Purdue University athletic department managed to
retain just $713 after expenses and sharing with the Big Ten
Conference. Net of revenue sharing, by conference rules, the
Boilermakers had $1.8 million, $1.4 million for being
conference representative to a BCS bowl and another
$400,000 estimated from additional ticket handling charges
and private donations related to the trip—and yet they spent it
all except for the meager $713 balance.
574

LEARNING HIGHLIGHT: WASHINGTON STATE
UNIVERSITY COPPER BOWL SPENDING
Bowls were invented as local economic development tools to
lure winter tourists. This means that the bowl hosts expect the
teams to bring spenders and to spend themselves. The result
for participation in lesser bowls is that participating teams
typically do not make any money on their bowl appearances.
Many observers on the academic side of the university think
that spending it all is just gluttony by the athletic department,
but nothing could be farther from the truth. Competition
simply pushes participating teams to spend all of their bowl
payout and to encourage as many people as possible to attend
the game; if the teams chosen do not do that, other teams will.
The participating teams take their reward in terms of
advertising and other types of recruiting exposure and, of
course, a good time had by all in a nice place in December.
When the Washington State University (WSU) Cougars went
to the Weiser Lock Copper Bowl in Tucson in 1992, many got
an education on the finer points of participating in lower-level
postseason bowl games. The Washington press was quick to
point out that the Cougars intended to spend all of the
$650,000 payout. The reasons for spending it all soon were
made very clear. As Copper Bowl executive director Larry
Brown said, “That’s why bowl games are in existence, to
stimulate the local economiesThe economic impact is
expected to be $10 million to $12 million from 5,000 to 6,000
fans that come and spend about $300 a day doing the town
and having a good time.”
WSU athletic director Jim Livengood also pulled no punches
in justifying the spending: “If we were to say we will spend
$325,000 [half the Copper Bowl payout to each team, at that
time], we will never play another bowl game again. Why do
you think the cities have bowls? They wouldn’t have a Copper
Bowl if it wasn’t trying to bring people to Tucson.It’s the
expectation of the Copper Bowl people that WSU bring a lot of
people.” The Copper Bowl people expected fans to spend in
Tucson and bolster that local economy. The opponent
University of Utah Athletic Department finance director,
Peter Hart, echoed this: “The goal of bowl games is to sell
hotel rooms in soft markets. I think they have a sense of
vacancy rates around the time of the bowl.” These statements
make it clear that the payout was to be used to stimulate the
local Tucson economy.
WSU spent all of its $650,000 allotment, bringing
approximately 200 players, coaches, administrators, faculty,
staff, and their families to the Copper Bowl, all expenses paid.
Spending for transportation (chartered airfare), lodging, and
food was budgeted at $345,000. An additional $30,000 was
set aside for gifts and awards to players. Part of the WSU
payment came in the form of 10,000 tickets worth $290,000
face value. WSU sold about 5,000 tickets. The rest were
framed and offered as mementos for those who did not choose
to go to Tucson. The WSU athletic department left itself a
$75,000 buffer, apparently against the contingency that they
575

did not sell all of the tickets. Otherwise, they would have gone
in the hole going to Tucson.
Sources: The Daily Evergreen, December 8, 1992, pp. 1–2;
December 7, 1992, p. 1; The Daily News, December 7, 1992, p.
8A.
576

SECTION 6
Sports Outcomes: Athletic
Departments, Conferences,
and the NCAA
We went into great detail about the relationship between pro
teams and leagues in Chapters 5 and 6. Much of what we
discussed there is also true for college athletic directors and
their conferences. Conferences enable teams to pursue goals
and objectives that they cannot pursue as successfully acting
alone. Coordinated conference activity provides teams with
opportunities along many dimensions, both on and off the
field. Conferences provide play and profit for colleges, just as
leagues do for pro teams.
Conferences produce the competition that fans desire,
including their own championships. In the process,
conferences enhance the economic welfare of member athletic
departments. But college sports fans also want a national
champion, and that involves cooperation across conferences.
To enhance that cooperation, colleges and conferences
invented the NCAA. In the simple organizational role, the
NCAA enforces agreements among member institutions. In
this section, we’ll explore the single-entity and joint-venture
activities of athletic departments, conferences, and the NCAA.
Enforcement of agreements across conferences by the NCAA,
ostensibly to protect athletes and competitive balance, is for
the next section of this chapter. We will see that the NCAA
enhances the economic welfare of member institutions, just as
pro sports leagues do for their member team owners.
SINGLE-ENTITY COOPERATION IN COLLEGE
CONFERENCES
Conferences accomplish the single-entity aims of creating a
schedule, developing rules, and organizing conference
championship play. Everything said about these factors in
Chapter 5 need not be repeated here, and the conclusion is the
same. Colleges that join conferences determine that they are
better off doing so than going it alone. Nearly all colleges have
made this determination in all sports; Notre Dame actually is
the exception that proves the rule (indeed, in all other sports
besides football, Notre Dame is a member of the Big East
Conference).
However, one thing distinguishes college sports from pro
sports. In the pros, there is only one dominant league in each
sport. The determination of the national championship
follows from lower-level champions under league-specified
play-off rules. This is a natural-enough single-entity activity
since a league championship, by definition, must be specified
by the league. This determination does have economic
impacts but cannot be anticompetitive since all members
must agree to the championship structure and nobody is
excluded from participation by this specification.
577

Unlike their pro counterparts, college conferences must
cooperate with each other to make a national championship
happen. In order to cooperate with each other and determine
national champions in all sports except the top level of college
football, college conferences rely upon the NCAA. Originally,
the NCAA came about because individual colleges could not
deal with society’s discontent with college football violence in
the early 1900s. But member institutions now demand that
their NCAA do much more than this. Again, in all other sports
except for the top level of college football, its membership has
absolutely demanded that the NCAA handle the
determination of a national champion, especially college
basketball through March Madness. Again, this is a natural-
enough single-entity activity since, by definition, the NCAA
championship process should be specified by the member of
that organization. In addition, the process is not
anticompetitive since no member is excluded from
participation (if they finish well enough to be chosen).
However, turning toward rival postseason basketball
tournaments, the behavior of the NCAA no longer fits strictly
within the definition of single-entity activity—requiring
members to play in the NCAA basketball championship if
selected is not necessary to make a national championship
happen. And the organization of the top-level national football
championship through the BCS clearly goes far beyond the
idea of just identifying a national champion. I will use both of
these examples as lead-in for the joint-venture activity of
athletic departments.
JOINT-VENTURE COOPERATION IN COLLEGE SPORTS
The NFL crowns its champion through a play-off system that
ultimately ends in the Super Bowl. That definition is sufficient
under the idea of single-entity activity. All of the rest of the
behavior of the league regarding the Super Bowl simply
increases the profits from that game and are properly labeled
joint-venture activity. Similar observations can be made about
the college postseason.
For example, for years, college basketball had competitive
postseason venues, the National Invitational Tournament
(NIT) and March Madness. Believe it or not, early on the NIT
was the more prestigious tournament and the winner was
referred to as the national champion. The NCAA established
the preeminence of March Madness by expanding from an
original field of eight teams, to a field of the “Sweet 16” teams,
and on to a field of 64 teams in 1985. Member institutions
voluntarily chose to move to the NCAA tournament, much to
the chagrin of NIT organizers. But the NIT did continue on,
partly dampening the ability of the NCAA to control
postseason play entirely, that is, until 2005.
In that year, NIT organizers sued the NCAA because an NCAA
rule eventually required members to play in its tournament if
invited. The complaint claimed that this action by the NCAA
was a pretense to the monopolization of college basketball
postseason play. Before a formal verdict was reached, there
was a $56 million settlement. The NCAA compensated some
athletic departments and, more importantly, purchased the
578

NIT and now runs it. The NIT still carries on but (as one
would expect since it is owned by the NCA) with far less
prestige. The result is that the NIT has nearly no revenue
importance to participating athletic departments. The
economics here is straightforward. When championships
happen in a monopoly setting, prices increase and
opportunities decrease to the detriment of fans. Reducing
competition so that the value of March Madness increases is
not essential to the crowning of a champion. So it is joint-
venture activity rather than single-entity activity.
In the top level of college football, the BCS determines the
national champion. Two things are important to observe
about the BCS from the perspective of championship
economic competition. First, it was not always the case that
the BCS determined the national champion. Earlier on,
competitive nationwide polls crowned the champion. Indeed,
if the competing polls chose different teams, the champion
from each poll was referred to as a cochampion. Again,
movement to the BCS doesn’t have to be an economically
anticompetitive determination of the champion as long as
colleges and conferences are free to choose to enter that
particular arrangement.
But the second point here is that colleges and conferences are
not free to choose to enter the BCS. In its modern
configuration, automatic entry into the BCS games is
formulaic for the champions of only a few FBS conferences,
typically referred to as “the big six” (ACC, Big East, Big Ten,
Big 12, Pac-10, and SEC). The remaining FBS conference
champs are only granted at-large opportunities, and there is
nothing akin to the NIT that provides any alternative to the
BCS (there are only lesser bowl alternatives). Unless these
other conference champions meet a set of conditions that do
not bind on the automatic entrants, they cannot participate
for the national championship. In the 2009 BCS games that
followed the 2008 season, this is precisely what happened to
the Utah Utes of the Mountain West Conference. They
finished their season 13-0 having beaten three ranked teams
but placed sixth in the final BCS standings. They then handily
beat Alabama, ranked fourth by BCS going into the bowl
season, 31-17. They ended up ranked second in the nation but
never were seriously considered for the BCS Championship
game.
At this writing, Senator Orin Hatch of Utah is calling for
congressional scrutiny of the BCS for just this type of
exclusionary practice. The BCS formulation is not required in
order to crown a national champion. Thus, it is not simply
single-entity behavior. The formulation does raise the value
generated for those conferences that are perennial
participants in BCS games. Thus, the BCS represents a joint-
venture rather than single-entity activity.
In addition to creating higher revenues from basketball and
football championships, top-level athletic departments would
like to take advantage of many of the other joint-venture
activities undertaken by their professional league
counterparts. Although there is no exclusive franchise
agreement, athletic departments maintain some market
579

power through territory protection through their conferences
and the NCAA. The NCAA has also acted to protect its
members through its behavior toward potential rival
organizations. Most members of college conferences would
also admit that joint-venture negotiations are more valuable
than individual TV broadcast rights negotiations. Finally,
while players are not unionized, joint-venture cooperation
may strengthen the position of conference members relative
to the players. As a reminder, at the pro level, leagues
facilitate joint ventures. However, because there are so many
college conferences, competition among themselves may
hinder the pursuit of the usual goals of concerted joint
ventures. In response, conferences rely upon the NCAA for
much of their joint-venture activity.
TERRITORY PROTECTION
Individual college athletic departments at a particular level of
play do not want economic competition from rival colleges in
their own backyard. If they and their rivals are playing at the
same level, fans perceive more substitute possibilities, and
geographic market power is reduced. Athletic departments
rely on both their conference and the NCAA to protect
geographical territories. Because this type of protection is not
required to “make play happen,” it falls outside of single-
entity requirements, and basic economic intuition suggests
that it occurs because the economic welfare of college athletic
departments is enhanced.
It can make perfectly good sense for a conference to expand to
bring in very tough rivals. When a team enters a conference, it
brings its heritage, tradition, a shot at postseason play, and a
television market. Stronger sports programs bring the most
value along these dimensions (especially TV revenue).
Therefore, it is unlikely that an athletically weak college would
be allowed into a conference over a strong one. Member
colleges must weigh any geographical territory issues against
these benefits when they decide to accept new conference
members.
In addition, member institutions have endowed the NCAA
with the power to determine membership in the hierarchy of
conferences from the FBS down through Division III. No team
can move between divisions without NCAA approval based on
minimum attendance requirements and stadium capacities.
The earlier example of Idaho’s bid for top-level college
football status is a nice case in point. Idaho eventually landed
in the WAC, a distant substitute for Pac-10 football at
Washington State. It should come as no surprise that the
Pac-10 didn’t invite Idaho to join it in the first place,
especially since Idaho was only 8 miles from WSU and
enjoying quite a bit of success against the Cougars on the field
at the time.
RIVAL ASSOCIATIONS AND THE NCAA
Similar to the pro leagues, the NCAA also has acted against
rival associations, resulting in a single dominant association
outcome. At the beginning of the chapter, we mentioned one
580

particularly brutal episode concerning the now defunct AIAW
(Zimbalist, 1999, pp. 59–60). Founded in 1971, that
association served over 800 member colleges and had its own
set of national women’s championships. According to
Zimbalist (p. 59), NCAA members saw a threat coming with
the passage of gender equity laws and moved to curtail the
AIAW. The NCAA took direct and concerted action to win
over AIAW members, and the AIAW folded in 1982.
The NCAA also faces challenges from rivals in basketball. The
case of the NIT has already been covered, but there is more.
The NCAA has a virtual lock on pre-NBA basketball talent.
Rival threats in the production of such talent have included
the Continental Basketball Association, the Collegiate
Professional Basketball League, and recently the NBA’s own
development league. The idea driving these rivals is
unbundling the student-athlete into one part student and
another part athlete. Some basketball players would like to
invest in becoming NBA caliber players, but they would rather
not do so by going to college. These rivals all paid college-aged
players professional salaries and helped those who desired it
with their education. And they posed direct competition for
college athletic departments providing training for future
professional athletes.
NEGOTIATIONS: REGULAR-SEASON BROADCASTING
AND THE NCAA DECISION
Member teams sell their league-play TV rights through
conferences primarily under the three-tier system described
earlier (and for the Big Ten and the SEC through their own
conference networks). But it wasn’t always this way. Before
1984, the NCAA actually negotiated a single football contract
for all of (then) Division IA. The current system was created
after one of the most important court cases in sports history,
NCAA v. Board of Regents of the University of Oklahoma
(otherwise known as the 1984 NCAA Decision).
Beginning with the Television Plan of 1952, the NCAA
negotiated the broadcast rights contracts for all of Division IA
conferences, just as the NFL does for its member team owners
(Brown, 1999). Sponsors had to provide national coverage and
received no more than 12 Saturday games in total. While no
member athletic department had to provide a game, neither
could any member appear more than one time per year. In
Table 13.10, you can see the number of games, the rights
fees paid, and the payment per game that resulted (since this
table is annual and simple, adjustment to 2009 dollars again
appear in parentheses). The gap from 1953 to 1977 is due to
my inability to find any source citing the number of games.
Through 1983, under this centralized approach, the NCAA
limited the number of available broadcasts and the teams that
could be included in this limited number of appearances.
Overall, the results were nothing short of stupendous.
Adjusting for inflation, the total contract value rose 9.6
percent annually. Remember, the U.S. economy typically
increases about 3 percent real annually, so these are truly
impressive rates of growth.
581

Despite this increase shared by all, football powerhouses like
the University of Oklahoma felt that they could be on TV even
more than they already were, dramatically increasing the
revenues of their own individual athletic departments.
Examining Table 13.10 shows why they became particularly
interested at this point in time—the value per game rose 6.6
percent annually up to 1983, and increased 67 percent just
from 1981 to 1982! These powerhouses joined together,
forming the College Football Association, and attempted to
negotiate their own, much more liberal broadcast contract.
The NCAA sued the teams for failing to abide by its
negotiations. Oklahoma countered that the NCAA practice
restricted output and raised prices and that they would not
abide by actions that ran afoul of the antitrust laws.
Ultimately, the U.S. Supreme Court decided that the actions
of the NCAA in the broadcast rights market for college football
did violate the antitrust laws. The court ruled that output was
restricted and prices were raised above the level that would
prevail under a more competitive bidding structure. Note how
this stands in stark contrast to the current practices of pro
sports leagues that sell broadcast rights through national
contracts.
It is easy to see that the courts were correct by looking again
at Table 13.10. After the 1984 NCAA Decision, conferences
were free to make their own TV deals with two very
predictable results. First, the number of games offered on TV
increased immediately in 1984 by almost 30 percent and, in
less than 10 years, by 154 percent (facilitated in part by the
advent of cable TV). Second, adjusted for inflation, the price
per game fell immediately by 75 percent. Indeed, although it
fell for a few years in the late 1980s, in real terms, the price
per game essentially stayed at this lower level from 1984 to
1995. Plain and simple, the court decision invoked
competition, resulting in lower prices and increased output
for fans to enjoy. In addition, powerhouses were broadcast
more often, and more schools were on TV than ever before.
NEGOTIATIONS: POSTSEASON PLAY, THE BCS, AND
THE NCAA
Postseason rights in football are handled through the BCS,
nothing more than the formula agreed upon by the big six FBS
conference commissioners, Notre Dame, and media providers
ABC (which owns ESPN) and FOX. As we saw back in Table
13.6, the 10 teams in 2009 BCS games earned $121.5 million
for FBS conferences to distribute to member athletic
departments ($22.5 million each for the Big 12, Big Ten, and
SEC; $18 million for the ACC and Big East, and another $18
million to be split between C-USA, the MAC, Mountain West,
Sun Belt, and WAC). This has been going on for years under
the BCS, and these amounts are significantly larger than any
other payments, historically, from bowl games. So it seems
clear that the big six FBS conferences with automatic bids
have done well for their member athletic departments
through the BCS relative to what they could have done
separately and individually. The BCS web page
(www.bcsfootball.org/ and click on “Chronology” under the
History menu) also points out that “In the first ten years of
582

http://www.bcsfootball.org

http://www.bcsfootball.org

the BCS system, more than $90 million was distributed to
conferences that do not have an annual automatic berth in the
system.” So the remaining FBS conferences (C-USA, the MAC,
Mountain West, Sun Belt, and WAC) haven’t done so badly,
either.
Apparently, the members do not harbor the same feelings
about the NCAA management of postseason basketball as they
did about football. The NCAA manages and sells the broadcast
rights to March Madness. This appears to satisfy NCAA
members because they have never organized anything similar
to the BCS in football to strike out on their own in the
basketball postseason. Small wonder. As stated earlier, the
$10.8 billion March Madness TV rights contract with CBS and
Turner Broadcasting will average $771 million annually. Table
13.7 showed how more than $215 million ($221 million 2009
dollars) from March Madness was distributed to just to the 11
FBS conferences for 2007–2008. In all, across 31 conferences
that are eligible to participate in March Madness, the NCAA
distributed $1.7 billion ($1.75 billion 2009 dollars). The rest
goes to fund the operations of the NCAA itself.
Harkening back to the determinants of demand, just as pro
leagues manage the structure of their play-offs, the NCAA has
carefully managed the structure of March Madness. The
number of teams in the postseason basketball championship
increased from the Elite 8 (1939–1951) to the Sweet 16 in
1952. From 1953 to 1974, between 22 and 25 teams were
selected to the tournament. The field of 32 made its first
appearance in 1975, but the number rose from 40 to 53 from
1979 to 1984. The first field of 64 appeared in 1985, and the
current number of 65 teams began play in 2001.
COLLEGE SPORTS, THE NCAA, AND ATHLETES
All dealings between athletic departments and college athletes
are handled under the watchful eye of the NCAA. This
provides a unified front for athletic departments, and it isn’t
surprising that they need one. The relationship is very one-
sided. First and foremost, amateur requirements reduce
payments to some players, particularly stars. In addition,
players must remain at their original college of choice (or lose
eligibility) except for a very few special circumstances. This
rule makes player movement between teams fairly difficult.
Whenever there are rules, there can be a value to cheating on
the rules, so member institutions give the NCAA the
important role of policing compliance and punishing
violators. This relationship is so important to college sports
that we will examine it in great detail later in this chapter.
ALL THINGS CONSIDERED: THE CHANGING FACE OF
CONFERENCE MEMBERSHIP
College sports teams cannot relocate (universities cannot be
moved to a new location for the sake of enhancing the athletic
department’s welfare). Thus, there is no college equivalent to
a pro league’s careful balancing act between expansion and
believable threat locations. However, there are some
similarities between pro and college sports in terms of
changing conference membership. This decision roughly
583

parallels the example in Chapter 5 about professional
basketball teams choosing between the National Basketball
League (the forerunner of the modern NBA) and the
Basketball Association of America in the 1940s. Another
parallel is division realignment as occurred most recently in
the NFL.
In a broad sense, all college conferences are rivals. In the face
of this competition, conferences must provide services to their
members or risk losing them to another conference. Athletic
departments will simply find another organization that will
provide the desired services. This has been true since the
beginning of big-time college sports. For example, in 1928,
Iowa State, Kansas, Kansas State, Missouri, Nebraska, and
Oklahoma left the Missouri Valley Conference to form the
“Big 6” in order to be able to pursue postseason play. The
conference became the “Big 7” when Colorado jumped from
the Skyline Conference in 1948 and the “Big 8” when
independent Oklahoma State joined in 1960. The saga of the
Big 8 then became entwined with perhaps the most famous
conference change in the history of college sports.
Although the Southwest Conference was once one of the most
storied conferences in college football history, Texas, Texas
A&M, Texas Tech, and Baylor bolted to join the Big Eight
Conference and form the Big 12 in 1996. The reasons mainly
concerned dissatisfaction over revenue sharing and the
chance to form a conference with two divisions and a
postseason championship game. As Bill Carr, Houston’s
athletic director at the time put it (Sports Business Journal,
December 8, 2003, p. 32), “The economics of the Southwest
Conference weren’t working, so we no longer had a Southwest
Conference.”
Conferences clearly run the risk of losing members. The SEC
formed shortly after the original “Big 6” when Alabama,
Auburn, Florida, Georgia, Kentucky, LSU, Mississippi,
Mississippi State, Tennessee, and Vanderbilt all abandoned
the Southern Conference again due to postseason restrictions
in that conference. The SEC remained stable until adding
Arkansas from the Southwest Conference and independent
South Carolina, both in 1991, facilitating a two-division
conference with its own championship game. Indeed, the only
truly “original” conference made up almost entirely of
independents that has stood the test of time is the Big Ten.
Illinois, Michigan, Minnesota, Northwestern, Purdue, and
Wisconsin formed the “Western Conference” in 1896 all as
independents. Independents Indiana and Iowa joined in 1899.
Ohio State left the Ohio Athletic Conference to join in 1912. It
stood that way until independent Michigan State made it the
Big Ten in 1950. And the Big Ten stood another 40 years until
Penn State, another independent, joined in 1990.
Losing members to a competitive conference can spell the
end. Related to the examples just given, the Southern
Conference also lost the seven original founding schools of the
ACC (Clemson, Duke, Maryland, North Carolina, North
Carolina State, Virginia, and Wake Forest). Shortly after
losing Arkansas to the SEC, the Southwest Conference died
when it lost the rest of its teams to the Big 12 and the WAC.
584

The Big West Conference, originally formed in 1969, died in
2000 losing all of its teams to the Sun Belt and WAC. Indeed,
Hawaii is the only remaining original WAC team because all
of the replacements in the WAC defected to form the
Mountain West Conference. Since joining the WAC in 1979,
Hawaii has seen fully 21 different comembers. The details of
the fate of the Big East and the ACC is covered at the end of
this section in the Learning Highlight: Miami Joins the Big
East … No, Make that the ACC.
585

LEARNING HIGHLIGHT: MIAMI JOINS THE BIG
EASTNO, MAKE THAT THE ACC
While technically no college sports team can change its
physical location, the University of Miami has changed its
conference location twice in the last 20 years. In 1990, Miami
(at that time an independent in football like Notre Dame)
joined the Big East, a conference more noted for basketball
than for football. This occurred just after Penn State joined
the Big Ten; Arkansas, the SEC; and Florida State, the ACC.
With so many teams changing conferences, the Big East
feared that its stronger football schools might bolt and take
their basketball programs with them. To strengthen its
football offerings, the conference recruited Miami as a football
anchor and also added West Virginia.
Here is how valuable Miami was to the goals of the Big East.
The Big East allowed Miami to keep its TV and bowl revenue
for a few years, and after that period elapsed, only schools in
the Big East that played (then) Division IA (now FBS) football
would share the revenues. To top it all off, the eventual
makeup of the Big East for football featured only eight teams.
With such a small number of conference games to play to
determine the Big East championship, a powerhouse like
Miami was free to schedule nonconference games and
command large appearance fees. In essence, to save a solid
basketball conference, the Big East created a solid top-level
football conference and essentially gave Miami all its football
money in order to do so.
It worked until 2003. Along with Boston College and Virginia
Tech, Miami left the Big East for the ACC. The ACC didn’t
expand to avoid catastrophe. As Commissioner John Swofford
puts it, “But as the saying goes, ‘It wasn’t raining when Noah
built the ark.’ Our feeling within the majority of our schools
was that we needed to be a bit larger to maintain our relative
place in the future of college athletics.” (Sports Business
Journal, December 8, 2003, p. 25). As you can see looking
back to Table 13.7, after its expansion the ACC ranked fourth
in total conference TV and post-season revenues. When
Boston College joined the conference in 2005, the ACC split
into divisions and held its first postseason conference football
championship (adding a few more million conference TV
dollars).
Economically speaking, did it make sense for the ACC to
expand in this way? Just prior to expansion, the ACC
distributed around $88 million ($99 million) to its nine
members, or about $9.8 million ($11 million) each on average
(even though equal shares do not occur). To keep that amount
constant when Miami and Virginia Tech joined, the
conference needed to generate another $19.6 million ($21.4
million). But immediately, it’s clear that this amount was met
and then some. The ACC drew 3 million in total attendance
for the first time in its history. Season ticket sales were at
record levels at both Maryland and Clemson. And the TV
contract with ABC doubled relative to the value of the contract
signed in 1998, adding another $37 million ($40 million).
586

None of this was lost on the old sage, Florida State coach
Bobby Bowden, who noted, “[The East Coast] is where
everyone lives, you know. I don’t care what the SEC says,
there’s not that many people over in Starkville [Mississippi].”
Duke’s coach Ted Roof put it succinctly, “I think we’re the
premier football conference in America.” Admittedly, a
conference once dominated in football by Florida State will
now be more competitive with the addition of Miami. But over
$2 million more for each of the original nine members, plus a
conference championship game in 2005, and a future as a
premiere conference should soften the blow.
Sources: Sports Illustrated, October 22, 1990, pp. 64–66;
Sports Business Journal, December 8, 2003, pp. 25, 32–33;
Washington Post, August 25, 2004, p. H01; and Citizen-
Times.com, accessed August 7, 2005.
587

SECTION 7
Sports Outcomes:
Competitive Balance
Exclusive territories, based on the historical happenstance of
college location, can lead to conferences with competitive
imbalance and lower fan interest for some conference teams.
The same set of historical accidents can lead to competitive
imbalance across conferences. The Temple Owls of the Mid-
American Conference are 1-1 in postseason bowl games, losing
to Tulane in the 1939 Sugar Bowl and defeating the University
of California in the 1979 Garden State Bowl. By comparison,
the University of Michigan has played in 39 bowl games (from
the 1902 Rose Bowl to the 2008 Capital One Bowl), compiling
about a 0.500 record, and missing a bowl trip only 10 times
since 1970. As a leading indicator of this type of imbalance,
Temple coach Al Golden makes $575,000, while Michigan
coach Rich Rodriguez makes $2.5 million.
Whether Rottenberg’s (1956) uncertainty of outcome
hypothesis is as important in college sports as in pro sports
awaits detailed investigation in the future. However, the
stated goal of NCAA compliance rules is the preservation of a
“level playing field,” that is, enhancing competitive balance.
Let’s look at some competitive balance data and then explore
how colleges, acting through conferences and the NCAA, have
addressed it.
COMPETITIVE BALANCE DATA IN COLLEGE SPORTS
As you can imagine, with so many college conferences even
for just football and basketball, we would have to spend a lot
of time and space for a complete analysis of competitive
balance. Instead, let’s opt for a look at revenue imbalance in
general for FBS athletic departments and examine winning
percent imbalance and championship imbalance for just a
couple of major conferences.
Revenue imbalance was already presented back in Table 13.1.
It is worth a reminder that revenues are imbalanced and have
become increasingly so through the 1990s in FBS college
sports. The top teams are gaining against teams at the average
and at the bottom.
Our winning percent imbalance observations come from
Table 13.11. The table shows our usual measure, the ratio of
the actual to idealized standard deviations, for the Big Ten
and the Pac-10 (the conferences I follow the closest). Winning
percent imbalance is the rule in football in these two
conferences. Over the last 35 years, the decade average
standard deviation ratio is always greater than 1.5 (except for
the Pac-10 in the 1980s but still much greater than one). Also
by decade averages, the level of winning percent imbalance is
larger in the Big Ten than it is in the Pac-10, except for the last
eight years, when they are essentially equal. From the data in
588

Table 13.11, there does not appear to be any trend in
imbalance over time for either conference. But using more
sophisticated techniques and all of the historical data on all
Division IA (FBS) teams prior to 2000, Depken and Wilson
(2004b) find a negative trend in regular season competitive
balance in college football. By the data for the 2000s in Table
13.11, things have stayed the same in the Pac-10 and evened
up a bit in the Big Ten.
The data on championship imbalance for football in the same
two conferences are also in Table 13.11. Through 2008, in the
Pac-10, Southern California won or tied for the championship
19 times. UCLA and Washington tied for second with eight
first-place finishes, although none of them are very recent.
Including ties, there were 51 first-place finishers (including
ties) in total in the Pac-10 in the last 38 years. Thus, these
three teams were responsible for 69 percent of first-place
finishes. In the Big Ten, Michigan (20 times), Ohio State (19
times), and Iowa (five times) had the most first-place finishes.
Because there were 60 first-place finishers (including ties)
over the period in Table 13.11, these three teams garnered 73
percent of the total. Thus, a distinct minority of teams in each
of these conferences won the majority of the championships
(by a landslide in the Big Ten). Therefore, it appears that
winning percent imbalance and championship imbalance go
hand in hand with revenue imbalance in college sports.
But an interesting occurrence over the decades partially
mitigates this view of championship imbalance. Compared to
the 1970s and 1980s, far more teams began to finish first in
both the Big Ten and the Pac-10 starting in the 1990s. In the
Pac-10, five different teams finished first (including ties) in
the 1970s and four in the 1980s. Then twice as many finished
first in the 1990s and there already have been seven different
first-place finishers in the 2000s. The story is nearly identical
in the Big Ten—three first-place finishers in the 1970s, five in
the 1980s, compared to eight in the 1990s and seven already
in the 2000s. Thus, while it is true that just a few teams
dominate, historically, more teams are finding their way into
first-place finishes.
COMPETITIVE IMBALANCE REMEDIES IN COLLEGE
SPORTS
Competitive imbalance exists in college sports, and the same
conclusion about pro sports appears to hold for college sports.
There are different revenue potentials in different
geographical territories, and successful teams collect on those
higher revenues more than unsuccessful teams. Winning
percent and championship imbalances follow, as we would
expect. As with their pro counterparts, college conferences
and their coordinating organization, the NCAA, have adopted
measures to deal with revenue imbalance and competitive
imbalance.
Like their pro counterparts, members of college conferences
typically share revenues. However, their primary approach to
solving competitive imbalance is through restrictions on
players. Just as with pro leagues, a cursory analysis reveals
that none of the restrictions on players actually have anything
589

to do with competitive imbalance. Instead, they transfer value
produced by players away from the players and toward the
athletic department.
REVENUE SHARING
Conference sharing rules cover the conference TV contract,
conference tournament revenues, and postseason revenues.
Referring back to Table 13.7 shows the total amount of these
types of revenues available for revenue sharing for major
college conferences. If distributed evenly in each conference,
there would still be substantial variation in shared amounts
across conferences—a range of around $1 million (Sun Belt
and MAC) to almost $30 million (Big Ten) with an average of
just over $9 million. But remember, the differences among
athletic departments, compared to conferences, would be
magnified because revenue sharing is not equal within a
conference. Only three conferences have what might be
termed “full” revenue sharing—the Big Ten, SEC, and ACC.
For the major revenue-producing sports, revenues are pooled
and shared evenly net of expenses like travel and hosting at
bowl games.
The Pac-10 conference again proves instructive on revenue
sharing in conferences besides the three just listed (it is the
only conference about which I have any working knowledge,
backed by spending 23 years of my career at Washington State
University). The Pac-10 has what its Commissioner Tom
Hansen calls “hybrid” sharing. Like the Big Ten, SEC, and
ACC, Pac-10 athletic departments share postseason football
bowl and March Madness revenues. However, the Pac-10
parts ways on sharing gate and TV revenue.
At the gate, for all games except rivalry games, the Pac-10
operates on a “guarantee” system in place since 1985. The
visitor is guaranteed at least $125,000 and at most $200,000,
but virtually every recent result has been at the maximum.
Thus, it is strictly the home team share that dominates gate
revenue disparity in the conference. Rivalry game gate
revenues are split evenly. For a feel of just how these rules
drive revenue disparity, the University of Washington
received just under $450,000 for its 2008 rivalry game at
WSU. But at the last rivalry game held at the University of
Washington in 2007, WSU received just over $1 million.
Turning to TV, the revenue-sharing rule is different for
conference and nonconference games. For conference games,
the athletic departments of the teams that actually appear on
TV split 59.5 percent. Only the remaining 40.5 percent is
shared evenly among the other departments in the
conference. For nonconference games, the participating team
keeps its entire share of the money. Sportsbusinessnews.com
(June 1, 2009) reported that total conference TV revenue for
2008 was $43.2 million. Equal shares would have each team
getting $4.3 million. But because of unequal sharing, USC
took home $6.5 million, while WSU took home $3.0 million.
The conference contract typically promises a minimum
number of appearances for each team so that even bottom-
dwellers get some TV money. Better teams appear much more
often and keep a higher share of their appearance money. The
590

TV sharing system compounds competitive imbalance
because TV money is strictly allocated according to winning.
Of course, this type of unequal sharing has predictable
consequences for competitive balance.
Returning to the logic of the two-team league from Chapter 6,
suppose a two-team college conference with both a larger-
revenue athletic department member and a smaller-revenue
athletic department member. As with the pro version, college
revenue sharing decreases the marginal value of college talent
for both athletic departments. But allowing the larger-
revenue department to share less than the smaller-revenue
department means that the larger-revenue department will
cut back less on talent than the smaller-revenue department.
As a result, the larger-revenue market team actually wins
more than it would under no sharing at all.
PLAYER RESTRICTIONS AND COMPETITIVE BALANCE
Restrictions on players are highly touted by athletic directors
and NCAA presidents as good for players and good for
competitive balance. Many arguments are offered in favor of
these restrictions, from paternalistic ones about caring for
players to the idea that players should have a sense of
obligation to the athletic departments that help pay for their
education. In actuality, only some of these restrictions may
enhance competitive balance, but another thing is certain.
Player restrictions all transfer value from players to the
athletic department budget. Because there are so many
conferences and schools, it is unlikely that any joint venture
aimed at restricting the choices of athletes would be self-
enforcing. Enforcement of these restrictions is the job of the
NCAA.
RECRUITING RESTRICTIONS
There is no draft in college sports. However, recruiting
restrictions preserve part of the value of talent to colleges in
the same way that the draft does for pro teams. Athletic
departments are restricted in how much money they can
spend on recruiting and how they spend it. For example,
coaches can only make a specific number of visits to any given
potential recruit, and potential recruits can only make so
many visits to a given university. Additionally, payment to
players or anybody with influence over players is not allowed.
The athletes must also comply with various restrictions on
their behavior during the recruiting process.
At the end of the recruiting period, athletes must commit to a
particular team on a particular date by signing a national
letter of intent. Once signed, these letters bind a player to a
particular college athletic department. Only if that athletic
department agrees (or if there is a coaching change) can a
player walk away from that commitment. All of these
restrictions reduce the level of compensation that goes to
players and stops athletic departments from burning up the
value players produce for them. As R. C. Slocum, head football
coach at Texas A&M University, put it after losing a prize
recruit to a lucrative contract offer by MLB’s Florida Marlins’,
“I have a lot of rules and no money. Baseball has a lot of
591

money and no rules” (Sports Illustrated, August 24, 1998, p.
20).
Interestingly, many believe that these recruiting restrictions
create a level playing field that enables poorer athletic
departments to compete with richer ones for talent. Nothing
could be farther from the truth. Instead, these restrictions
entrench power at departments at the top end of the winning
percent distribution. Imagine two athletic departments at two
different colleges, A and B. The department at college A
provides a better education and a better chance at a shot at
the pros than the department at college B. All else constant,
competitive recruiting finds the athletic department at A
getting better talent than the one at B—it is a no-brainer for
athletes to choose the department and college that provide
more of both opportunities. In such a case, the only way that
college B might be able to entice recruits is with other types of
compensation.
In this case, what happens when restrictions are put in place?
Some of the ways that the athletic department at college B
might compete are eliminated. Even if they wanted to try to
“recruit harder” than the department at college A, they are
restricted from doing so. College B’s athletic department
cannot win under this system, so why do they stand for it? The
reason this system persists is that players’ contributions to
some teams’ values are preserved, and richer teams share
(albeit unequally) with weaker teams through conference
revenue sharing. Quite simply, it pays to live under these
restrictions even though there may be a chance for some
teams to be better without the restrictions. Because stronger
teams have an advantage under recruiting restrictions, such
restrictions cannot increase competitive balance.
RESTRICTIONS ON PLAYER MOVEMENT
College players are not free to move and play elsewhere after
they sign their letter of intent, except for a very few special
situations (e.g., the coach that signed them moves before the
season starts). They cannot move between teams without
forfeiting an entire year of eligibility unless they move to a
lower division. Having to sit out a year is a huge cost to college
players, so most of them stay put.
Because players cannot be traded in college sports, sit-out
restrictions definitely affect competitive balance. Unless all
players choose the college where they are most valuable,
without error, then some athletic departments get better
players than they would be able to keep in a system where
players either are free to move between colleges or could be
traded. Because they do not pay their athletes, athletic
departments get to keep the benefits of mobility restrictions.
We would expect the impact of players selecting the wrong
school to have an asymmetric effect. Strong programs will not
accept weak recruits, so the only errors that can happen for
strong programs involve which strong athletic department is
chosen by strong players. However, weak programs will
certainly accept strong players who err in their consideration
of where to play. Thus, restrictions on player movement
enhance competitive balance as strong players who
592

mistakenly select athletic departments of lower quality
enhance the quality of those lower quality departments.
The second type of restriction on college player movement
concerns players trying their chances in the pro draft. In all
college sports except baseball, players who declare their entry
into the draft must forfeit remaining playing eligibility. This
dampens players’ incentive to enter the draft and keeps some
players earning revenue streams for their athletic
departments longer than they otherwise would. But because
this restriction only helps maintain restricted mobility, it
cannot enhance competitive balance. The players were already
stuck with their college to begin with, and forfeiting eligibility
just keeps them there longer across all of college sports.
COMPETITIVE BALANCE IMPACTS OF THE NCAA
AMATEUR REQUIREMENT
Through the NCAA, athletic directors impose an amateur
requirement on college athletes. Athletes may not take pay for
their sport. Instead, players are compensated with what is
called a grant-in-aid, or GIA. Their tuition is paid (in some
sports, only partial tuition GIAs are allowed), plus the
calculated amount for room and board and books. In addition,
they can obtain limited need-based grants and can earn a few
thousand dollars working in the off-season. From an
economic perspective, the amateur requirement quite simply
is a salary cap (we will discuss this thoroughly in the next
section). It should come as no surprise that the cap is chosen
so that the actual level of compensation just equals the
opportunity cost of players who can make it into college—
essentially, tuition, room and board, and books. (See Byers
[1995] for details on the evolution of this system.)
Clearly, the amateur requirement cannot enhance competitive
balance. The GIA cap is higher at colleges with higher tuition.
Thus, the athletic departments at higher-priced colleges have
a talent recruiting advantage along this dimension. The source
of this recruiting advantage is that higher tuition typically
signals higher quality of education and degree value. Because
the cap imposed by the NCAA amateur rules does not impose
equal spending by all athletic departments (i.e., there is no
minimum requirement), this cannot improve competitive
balance. Finally, as with all caps, players make less than they
would in a competitive system.
ENFORCEMENT PROBLEMS AND THE NCAA
The NCAA was invented to enforce the rules designed by
member athletic directors (who are overseen by university
administrators). All of the restrictions on players are carried
out through rules that must be enforced. Because there is a
value to violating the rules, the NCAA has an enforcement
problem. Often the setting for cheating on the rules is
misconstrued as a battle between athletic departments and
the NCAA. But it is important to remember that “the athletic
department” doesn’t really do anything when it comes to
recruiting. It is athletic directors, coaches, boosters, and
players that are bound by the rules. Athletic directors bear the
costs of cheating, while boosters, coaches, and athletes seldom
593

do. Thus, the setting is really a battle between the NCAA and
athletic directors on one side and coaches, boosters, and
players on the other.
THE LOGIC OF CHEATING
Suppose that athletic directors agree not to pay athletes and
codify this agreement in NCAA rules. If all the coaches and
athletes in all athletic departments follow the agreement, but
one coach breaks the bargain and makes payments to athletes
(using, say, money supplied by a few boosters), then a talent
advantage is gained resulting in additional television
appearances and postseason play. Note that little is gained for
the athletic department since most of the values created are
shared across the conference. But a brighter future is gained
for a given coach (if they are not caught!). Further, the player
makes money, and the boosters enjoy better talent. Because
cheating on NCAA rules can generate value, we should expect
cheating to occur unless expected punishment, or NCAA
sanctions, prevent it.
If cheating is universal, the value of the NCAA agreement is
completely undone, and all coaches and boosters are worse off
because nobody gains any advantage from an amateur
requirement (players remain better off because some of them
are getting paid more, after all). Athletic departments would
no longer keep the money that now is spent by all on players.
Ultimately, fans who cherish the amateur aspect of college
athletics would be betrayed if players were paid. Therefore,
athletic directors must be able to monitor their own coaches,
players, and boosters and each other; preempt cheating; and
punish offenders.
James Quirk and I (1999) analyzed cheating with the
following simple model. Let P denote the probability of
getting caught cheating on NCAA rules. G is the gain achieved,
and L is the loss under NCAA sanctions if caught. The
expected value of cheating is

Although simple, this expression helps organize some
thoughts about cheating. First, the lower P or L is and the
higher G is, the higher the expected value of cheating and
cheating is more likely. Those with high potential gains from
additional winning are most likely to have higher gains, G,
from cheating. These likely cheaters might have low winning
percents, large unused stadium capacity, and large potential,
but low actual gate and TV revenue. Possible losses if caught
cheating, L, are largest for the most successful programs
(lucrative TV values, wealthy donors, and perennial
postseason play). Cheating would be less likely for older
established coaches, whereas younger coaches trying to build
a team would be more likely to cheat. Again, note that L is low
for coaches, players, and boosters, but not for the athletic
department, itself. Success against cheating will have to be the
product of tireless monitoring by athletic directors of coaches,
boosters, and players. If P were the same for all colleges, then
a program with a currently weak record and high potential for
financial gain with a young coach would be a prime suspect.
594

Figure 13.6 uses the same graphical technique you saw for
the expected value of a lockout in Chapter 9 and the expected
value of a change in the tax law in Chapter 12. Figure 13.6
portrays the “y-intercept” version of the cheating formula,
graphing EV against P:

Our old friend the break-even probability PBE also appears in
Figure 13.6 (remember, PBE is such that EV = 0). For the case
at hand, PBE = G/(G+L). The figure shows that the trade-off is
between the gain from cheating—the y-intercept G—and the
relative loss from cheating—the slope −(G+L). Any probability
of getting caught below PBE yields EV > 0. Since policing
cheaters is costly, the NCAA may have to settle for the case
where EV > 0. So the NCAA must worry about both its
policing activity that determines P as well as the penalty to
impose on offenders that determines L. While it can use the
same logic just above to identify the most likely cheaters, the
NCAA also must choose its behavior toward cheaters
carefully.
THE VERDICT ON CHEATING AND ENFORCEMENT
Like all crime analyses, ultimately we only have observations
about those cheaters who are caught. Care must be exercised
in extending any findings based on actual violations and
sanctions because those violations that are not caught may be
quite different from those that are. However, some analysis of
NCAA enforcement and sanctions has been conducted, and
the results are insightful.
Fleisher, Goff, and Tollison (1992) found that “crime pays,” so
that violators had higher revenues and profits, but they also
noted that the members of competing athletic departments
monitor each other’s activities. So apparently some athletic
directors find value in aiding the NCAA. It is also in the
university administrators’ interests to watch for cheating by
its own coaches, boosters, and players, as donations to
academics appear to decline for colleges hit with NCAA
sanctions (Grimes and Chressanthis, 1994). Winfree and
McCluskey (2008) also find that the level of self-imposed
punishments offered by athletic directors can significantly
affect the form or type of the final punishment their
departments receive from the NCAA. This may help explain
why athletic directors often self-report violations and suggest
their own punishment.
Padilla and Baumer (1994) also found that programs with a
history of violations had higher revenues and profits than
those that were “clean.” However, they also noted that the
negative impacts of sanctions on programs did not seem to
last long. In a related vein concerning the pattern of
punishment, Zimbalist (1999, p. 179) noted that the number
of infractions detected has risen steadily over the decades
from the 1950s to the 1990s, but the pattern of punishment
has eased over time. The length of probation has risen from
the decade of the 1970s to the 1990s (from 1.11 to 1.73 years),
but the length of prohibitions on postseason play has fallen
since the 1960s (from .85 to .49 years). Further, the length of
prohibitions on TV appearances (implying reduced TV
595

revenues) has fallen steadily from the decade of the 1950s to
the 1990s (from .73 to .14 years).
Another interesting finding on NCAA enforcement concerns
whether it even has had the desired effect of leveling the
playing field. In essence, warts and all, does NCAA
investigation and punishment improve competitive balance?
According to Depken and Wilson (2004a), a sophisticated
view of how athletic directors respond to the success of NCAA
enforcement activity suggests enhanced competitive balance.
If a higher share of departments are caught cheating, the
remaining athletic directors would decrease cheating because
the chances of being caught have increased. Competitive
balance would increase. Depken and Wilson find just such a
result on a sample of major college programs.
So if one looks at the measurable impacts on athletic
departments, there is cause for pessimism about NCAA
enforcement. Long ago, Noll (1991) offered the following,
“Cheating against the NCAA rules will continue—indeed,
increase—because it is the profit-maximizing strategy for
nearly all universities and the income-maximizing strategy for
coaches” (p. 198). But if one looks at the outcome on
competitive balance, there is cause for optimism in the
Depken and Wilson findings.
596

SECTION 8
The Value of College Sports
Talent
As in pro sports, marginal revenue product (MRP) is the
primary concern when discussing the value of college sports
talent. A common question at many universities is why the
team cannot become more competitive. Why doesn’t the
athletic department just go out and hire a million-dollar coach
and build state-of-the-art facilities? Better recruiting and an
improved team would surely follow.
These questions echo those from Chapter 4 by those
dissatisfied with the level of team quality chosen by pro
owners. The answer to these questions boils down to the same
comparisons at the margin that are made by pro team owners.
What good is hiring a million-dollar coach or pouring millions
into facilities if enough revenues cannot be generated to cover
their costs?
In this section, we will examine the value generated by college
sports talent. First, the MRP idea will be presented for college
athletes along with an analysis of the value of restricting
payment below MRP. Second, we will examine some estimates
of what college athletes’ MRPs actually are, compared to their
level of compensation. Third, we will touch on the play-for-
pay issue. Finally, we will address discrimination in college
sports, primarily, gender issues in student-athlete
opportunities and pay and hiring for coaches.
WHAT IS THE VALUE OF RESTRICTING PLAYER
COMPENSATION?
MRP theory provides important insight into the value of
college players. Players contribute to winning outcomes for
their team, that is, their marginal product (MP). The athletic
department is able to sell that contribution to generate
marginal revenue (MR). But selling the sports output is not
the only contribution that athletes make. As we saw before in
the relationship between the athletic department and its
university administration, other contributions have to do with
helping administrators pursue their goals. So MR also
includes contributions to political support that typically
generate institutional support for the athletic department. As
is always the case by definition, the MRP of college players is
equal to MP × MR.
In pro sports, players are paid in cash. In college sports,
athletes receive a GIA. It is possible that the value of this type
of compensation to college athletes approximates their MRP,
but it is also possible that it does not, especially for star
college players who may contribute significantly to athletic
department revenues and political support for the university.
If the latter is the case, so that compensation is less than
MRP, then the amateur requirement allows the athletic
department to retain value created by players.
597

THE ECONOMICS OF AMATEURISM
Amateur requirements, simply put, reduce payment to players
by definition. Historian Ronald A. Smith (1988) emphasized
that the implementation of amateurism and the restriction of
player movements were driving forces behind early NCAA
actions. These restrictions have evolved over time and
continue almost untouched today. The NCAA has enforced the
extraction of the value of player talent for about 90 years.
In practice, recruiting restrictions reduce payments to players
as they enter college sports. In the absence of such
restrictions, the history of pro scouting suggests that players
would be paid to sign with college athletic departments. Once
players have signed their letters of intent, mobility restrictions
also reduce payments to players. If players were free to seek
their highest-valued alternative, they might move about a bit
between teams. That they cannot means that they are not free
to pursue higher-valued alternatives. These values, denied to
athletes, do not just go away, and they are not given back to
the ultimate source of the funds—the fans who buy tickets and
advertised products and taxpayers who fund the university
budget, part of which becomes institutional support of the
athletic department. Athletic departments keep the money.
GRAPHICAL ANALYSIS OF THE VALUE OF
AMATEURISM TO ATHLETIC DEPARTMENTS
Let’s explore the value of amateurism a little further. Suppose
that a competitive market for college players existed, just as it
does for pro players. In such a setting, players are recruited
and hired to play for college athletic departments. Once in
college sports, if there were truly a competitive market, they
would be free to move from one athletic department to
another. Further, imagine a two-team college conference with
a larger-revenue college and a smaller-revenue athletic
department. This situation results in a “free-agent” college
player market shown in Figure 13.7.
Figure 13.7 shows the competitive equilibrium outcome for a
two-team college conference. We need to generalize on the
idea of the added value of talent and winning in Figure 13.7
because athletic departments need not care about profit. So
let’s just think of MRL and MRS as the added value of winning.
The equilibrium price of talent is P*, and the level of winning
percents is W*L = 1 −W*S. What is it worth to athletic directors
to reduce the payment to talent? First, we need to recognize
that there is a next-best opportunity available to athletes.
Suppose the opportunity cost outside the sport is N. If teams
acting together through the NCAA can reduce the price paid to
N, they get to keep the rest. Talent bills are reduced to the
much smaller shaded areas, and the league keeps the amount
shown by the top shaded rectangle, namely, (P* − N) × (W*L +
W*S. (Note that this savings is actually P* − N because the
summed winning percents are equal to 1 for a two-team
league.)
THE COLLEGE VERSION OF THE INVARIANCE
PRINCIPLE
598

We gain an extremely important insight from our analysis of
Figure 13.7. The largest possible amount that athletic
directors can keep is generated when the distribution of talent
between the two teams is such that MRL = MRS. At any other
outcome than W*L = 1 −W*S, the price of talent would have to
be lower so that the total area would be less than under the
price P*. Therefore, if the athletic directors want to make as
much as possible from players, both directors want to play at
their revenue-maximizing level and hire the revenue-
maximizing level of talent. Thus, the level of college athletic
talent hired and its distribution between the two athletic
departments in this conference is the same regardless of
whether amateurism is imposed.
This is our old friend from Chapter 8, Rottenberg’s (1956)
invariance principle (even though Rottenberg never extended
the logic to college). Let’s state the college version of the
invariance principle clearly:
The distribution of talent in a college sport is invariant
to who gets the revenues generated by players; talent
moves to its revenue-maximizing use in the sport
whether players or athletic departments receive players’
MRPs.
Athletic directors and NCAA presidents alike argue that
competitive balance would be harmed if competition were to
dictate college player compensation. However, the invariance
principle suggests that competitive balance is invariant as to
who gets to keep the MRP of players. Under the amateur
requirement, players simply earn less but go to the same
athletic departments they would otherwise choose.
WHAT IS THE MRP OF COLLEGE PLAYERS?
The MRP of college stars can be quite high. You already have
some insight into this idea if you recall our earlier example of
the WSU Cougars and their 1998 Rose Bowl appearance. In
that year, revenues rose $2.9 million ($3.8 million) for the
WSU athletic department. Even though football is a team
sport, it is reasonable to say that most of this value was
created by star quarterback Ryan Leaf. Leaf already had an
MRP even if the Cougars did not go to the Rose Bowl.
However, because revenues rose $2.9 million ($3.8 million)
with the WSU Rose Bowl appearance, Leaf’s MRP was at least
$2 to $3 million ($2.6 to $3.9 million).
ECONOMIST’S ESTIMATES OF MRP
Following Scully’s procedure laid out in Chapter 7, Brown and
Jewell (2004) estimated that a star football player could have
generated about $406,914 ($569,252) for his athletic
department for the 1995–1996 season. On average, net of the
tuition paid to the university (remember, universities bill the
athletic department for the full scholarship cost), they
estimate that the athletic department extracted close to
$400,000 ($559,580), or 90 percent, of this amount. Turning
to basketball, these economists estimate that many star
college players are easily worth $871,310 ($1.2 million) and
could be worth up to $1,283,000 ($1.8 million) to their
athletic departments. Again, after tuition and other support,
599

athletic departments could easily extract $1 million ($1.4
million) from these athletes.
Roger Noll (1991) reached a similar conclusion. He noted that
quarterback John Paye was the only major roster difference
on the Stanford football team in 1987. Revenues increased by
$200,000 ($375,000) for the 1987 season. Net of the
scholarship payment to the university of around $17,000
($32,000) per year, the athletic department kept about 92
percent of Paye’s value, or just under $199,000 ($373,000),
and surely in the ballpark compared to the estimates by
Brown and Jewell.
The clear conclusion is that the GIA of star athletes are
nowhere near their MRP to the athletic department. This does
not mean that all college athletes receive less than their MRP
to their athletic department. For example, nonscholarship
football players probably earn close to their value, namely,
close to zero. For some players, the in-kind value of their GIA
probably comes close to their contribution. But star players do
receive far less than their MRP.
WHAT IF COMPETITION RULED THE COLLEGE PLAYER
MARKET?
If competition ruled the market for college players rather than
athletic departments working through the NCAA, there would
be no amateur requirement, no recruiting restrictions, and
players would be free to move between teams subject to any
contractual obligations with their current college athletic
department. The absence of an amateur requirement would
mean that players would be free to negotiate whatever
compensation arrangement was mutually agreeable to them
and their athletic department. For some players, the current
GIA arrangement might suffice. For others, it would not.
Few topics evoke the same level of gut-reaction disapproval
than play for pay, that is, paying salaries to college players in
addition to, or instead of, a GIA. Economists have commented
on this issue for nearly 20 years (Gary Becker in Business
Week, September 30, 1985, p. 18; Robert E. McCormick in
Wall Street Journal, August 20, 1985, p. 27), and it is a hot
topic even today. According to University of Minnesota
Athletic Director Joel Maturi (Twin Cities Pioneer Press,
www.twincities.com, April 24, 2009), “If indeed we were to
pay our student-athletes in those sports (he means high
revenue sports), we probably would eliminate maybe 22 of the
other sports that we have at the University of Minnesota
because only three of them generate enough money to pay for
themselves.” I added the parenthetical for clarity. Here we will
use our model of a competitive player market to make two
important observations about implementing an economically
competitive college player market. We will then address a few
of the arguments against doing so.
IMPLICATIONS FROM THE THEORY ON AMATEURISM
Our theoretical development of the amateur requirement
yields two important observations about play for pay. First,
money that is currently kept by the athletic department would
600

http://www.twincities.com

http://www.twincities.com

flow back to players, although the amount would be no more
than player MRP dictates. Returning to Figure 13.7, the
shaded area defined by a competitive price, P*, would go back
to players but no more than that. Thus, denizens of the
athletic department would have less and players would have
more, but no more product would be created and none would
be lost.
The second observation from Figure 13.7 is that the level of
talent hired by different athletic departments would not
change if players earned the competitive rate, P*, rather than
something less. This is the college version of Rottenberg’s
invariance principle. Again, in Figure 13.7, talent choices so
that W*L = 1 −W*S were the ones that generated the most
money for redistribution away from players in the first place.
Further, that level of talent choice by each athletic department
is the level of quality that makes athletic departments as
successful as possible. The distribution of college talent is
invariant to who gets to keep player MRP. With these two
observations, let’s have a look at some of the arguments
against play for pay.
TOPICAL ARGUMENTS AGAINST PLAY FOR PAY
If you think that college players should not be paid more than
their current level of compensation, you are in the majority.
Surveys have revealed that nearly 64 percent of those polled
were against play for pay (Sporting News, April 18, 1994, p.
54). Let’s look at four arguments against play for pay and
make of them what we can from an economic perspective.
WHERE WILL THE MONEY COME FROM?
One argument against play for pay is that athletic
departments barely break even in the first place, as shown
back in Table 13.1. A few make money, mostly on football and
men’s basketball, but not the vast majority. So where will the
money to pay players come from?
This argument is confused for two reasons. First, it really is
not the case that athletic departments break even in any
meaningful sense of that term. University administrators
allow athletic directors to spend all excess revenues on an
updated basis during any given budget period. Thus, a
department whose costs do not rise over budgeted amounts
but whose revenues are higher than expected will appear to
break even because it is allowed to spend the excess. So there
can be plenty of revenue to be rearranged.
The second confusion is that one does not need to worry
where the money will come from because it is already there!
Under the current amateur requirements, players generate
their MRP. And athletic departments collect it from boosters,
sponsors, media providers, and the university administration
through institutional support. Athletes see some of it in their
GIA but the amateur requirement reduces compensation
below their MRP, especially for star players. The difference
between MRP and what is actually spent on athletes remains
under the athletic director’s discretion. But competition is
relentless; the next most valuable use for these funds is where
the money will be spent if the athletic director is to be judged
601

successful. Likely candidates are athletic department salaries
(including the athletic director), coaches’ salaries, recruiting
expenses, and facility investment. So player MRP does not go
away, it is just spent elsewhere in the athletic department
rather than on players. If players were paid their MRP, funds
currently spent elsewhere in the athletic department budget
would just be reallocated to players. Once again, the money is
already there.
Athletic director compensation is an interesting case in point.
Using data on athletic director base salaries (not including
other compensation) found at Bloomberg.com (January 9,
2009), the top 10 athletic directors average $700,000
annually. Even for his base, Florida athletic director Jeremy
Foley tops the list at $965,000 (and other opportunities
generate the actual $1.2 million stated earlier). The bottom 10
base salaries are about the same as earned by your textbook
author, about $145,000. From 2004 to 2008, adjusted for
inflation, annual rates of growth in athletic director base
salaries ranged from about 2 percent in C-USA to 12.6 percent
in the SEC, averaging 6.2 percent across all FBS conferences.
So the rate of growth at the average doubles the typical 3
percent real growth in the economy at large, and SEC athletic
director base salaries have grown at over 4 times that rate! No
doubt this type of growth partly covers the fact that the job is
very demanding. But it also includes portions of the increase
that fans are willing to spend to watch college athletes who do
not receive their MRP.
Turning to coaches, data from KansasCityStar.com
(September 16, 2007) allow us to make the following
interesting comparison. Total compensation (over and above
just base salary) for Alabama football coach Nick Saban is
widely reported at about $4.2 million. That makes him both
the highest-earning coach in the nation and, so, in the state of
Alabama. The governor of the state, Bob Riley, earns
$118,650. You read correctly—Coach Saban makes 35.4 times
more than Governor Riley. The only states where the top-paid
coaches earn on a par with their state governors are South
Dakota (SDSU football coach John Stiegelmeier and Governor
Mike Rounds) and Alaska (Alaska-Anchorage hockey coach
Dave Shyiak and Governor Sarah Palin). Once again, college
coaching is a demanding job. Furthermore, state governors
also have other outside earning opportunities, especially upon
leaving office. Nonetheless, the differences can be startling.
Part of this result is driven by the fact that fierce competition
leads athletic directors to spend portions of the growth in
MRP produced by players procuring top coaches.
WON’T NON–REVENUE-GENERATING SPORTS BE
HARMED, ESPECIALLY WOMEN’S SPORTS?
The argument that play for pay would harm non–revenue-
generating sports recognizes that the money that would go to
players under play for pay would come from other parts of the
athletic department. But the argument conveniently avoids a
complete list of sources within the athletic department.
Setting the argument in this way pits play for pay against
allocations to other sports in the athletic department but not
602

against administrative salaries, coaches’ salaries, recruiting,
and facilities. Opponents of play for pay forecast dire
consequences for low-revenue sports. A common fear is that
the ability of the high-revenue sports, typically football and
men’s basketball, to support the rest would be reduced. This
fear is especially worrisome to those supporting gender equity
if the high-revenue sports are supporting women’s sports
(actually, they aren’t, but more on that in the following
section).
So the clue to the confusion here is in the way the argument
sets revenue sports against so-called nonrevenue sports but
omits all of the other lines in the athletic department budget
from consideration. If the world changed to play for pay,
where would play for pay funds come from? The answer, of
course, is from those areas in the athletic department budget
that currently are overpaid relative to their actual MRP.
Suppose it is other nonrevenue men’s sports that are currently
overpaid. This would mean that prior to play for pay, the
athletic director was investing in those sports at a higher rate
than their MRP would support. The same goes for women’s
sports, with one additional feature. Federal and state laws
require athletic directors to ensure gender equity; among the
choices that the athletic director could make would be to
spend on women’s sports until that margin was met. The
athletic director would have no reason to wastefully
overspend on either men’s or women’s sports.
If athletic directors cut a sport or reduce spending on a sport,
they are cutting their department’s revenue position. If sports
are cut or reduced, the success of the department falls below
the best level the athletic department can obtain is the current
level. Athletic directors might wish they could rearrange
money in this fashion, and they may use such threats to try to
sway others against play for pay, but that option simply is not
believable. At least it is not believable as long as athletic
directors care about the bottom line of their department.
This suggests that the portion of player MRP that goes to
athletic departments under NCAA amateur requirements goes
mostly to salaries, recruiting, and facilities. Some would say
that salaries are just a market outcome and that there is
nothing that colleges can do about it. On this issue, Duke
University law professor John Weistart cuts to the heart of the
matter:
It is sometimes suggested that the coaches’ high incomes
are simply the product of natural market forces. But the
market in which coaches function is neither free nor
natural. It is more likely that college coaches are able to
receive their extraordinary rewards because potential
competitors are tightly constrained. Most important, star
players are absolutely prohibited from claiming any of
the commercial value of their athletic achievement. (Wall
Street Journal, September 11, 1992, p. A12; Reprinted
with permission from the Wall Street Journal)
The clear corollary is that if athletic department
administrators and coaches had to compete for salary in a
603

play-for-pay world, there simply would be less money around
to bid up their own salaries.
Figure 13.8 depicts the department’s willingness to pay for
salaries. Under NCAA amateur requirements and when all
sports are operated close to their MRP, the athletic
department’s willingness to pay for coaches and
administrators includes an amount that would otherwise be
payments to athletes. This is portrayed as the MRP +
Extraction function in Figure 13.8. In a play-for-pay world,
players would be paid what is currently extracted from them
under NCAA rules. This means that the willingness to pay for
coaches and administrators would fall back to the MRP
function in Figure 13.8. As a result, pay to these other inputs
would fall, and less of those services would be purchased.
Market forces would send payments to players that currently
go to coaches and administrators. Nobel Prize–winning
economist Gary Becker sums it up well:
Why should the viability of athletic programs depend on
what amounts to subsidies from athletes, rather than on
such resources as larger gifts from alumni, higher
tuition from students, bigger subsidies from taxpayers,
or higher ticket prices for spectators?The NCAA’s efforts
to justify its restrictions on competition for athletes
should be viewed with suspicion because they increase
the financial benefits colleges receive from football,
basketball, and other sports. (Business Week, September
30, 1985, p. 18)
WON’T WEALTHY ATHLETIC DEPARTMENTS JUST BUY
ALL THE BEST PLAYERS?
In a play-for-pay world, some argue that wealthy schools
would buy the best players. The idea behind this argument is
that rich, successful athletic departments would be able to
outbid their competitors for athletic talent and would buy up
all the top-quality players. The result, according to this
argument, would be competitive imbalance that would ruin
college sports for fans (note the similarity to Rottenberg’s
uncertainty of outcome hypothesis). Fans of all but the top
teams would lose interest, and their teams would suffer even
further revenue declines. Ultimately, those colleges might
simply drop out of the FBS. Fans at all but a few colleges
would be as bad off as they possibly could be.
There is a single response to this misunderstanding—
Rottenberg’s invariance principle. As our theory shows, there
is one and only one winning percent combination that creates
the largest possible level of revenue to share. If the athletic
departments in a conference are at that combination, players
already go to the athletic departments where they are most
highly valued. This means that the richer, stronger schools
already get the bulk of the most talented players. Play for pay
will not change the distribution of talent because talent
already is distributed in a way that generates the best bottom
line for athletic departments. Coupled with our answer to the
last question, play for pay would just make coaches and
administrators in athletic departments poorer, but talent
would still be distributed in approximately the same way.
604

WON’T DIFFERENT PLAYERS IN DIFFERENT SPORTS
BE PAID DIFFERENTLY (ESPECIALLY WOMEN)?
Another argument against play for pay rests on an inescapable
observation about markets and pay: People are paid different
amounts. The issue raised by this question is the usual one of
the fairness of such an outcome. How can differential
payment to athletes be fair? Also, in an era where athletic
departments are wrestling with gender equity, how can it be
fair to pay those in higher-revenue-producing sports,
primarily men, more than those playing lower-revenue sports,
primarily women?
In a play-for-pay world, there would be differential pay.
Players would be paid just enough more than their second
highest college sport MRP to keep them at their current
athletic department. Because MRP = MP × MR, players will
be paid differently if either their contribution to team success
(MP) is different from that of other team members or if fans
and university administrators are willing to pay different
amounts for the contribution to winning by different players
(MR). Therefore, we would expect that quarterbacks would be
paid more than second-string offensive linemen. Figure 13.1
speaks directly to the difference in pay between men’s and
women’s sports. The figure shows that a sample of
undergraduate students was willing to pay more for men’s
basketball than for women’s basketball. This difference in
willingness to pay would certainly result in higher payment
for men’s team members.
The argument against differential pay at the level of society in
general, as with athletic departments, is one of fairness.
However, in national debates on differential pay, nobody
argues that people should not be paid at all! Besides, even if
you want a fairer way of paying players, then you still want
them to be paid. In short, different people are paid different
amounts in all employment markets. College sports would be
no different. Again, coaches and administrators would see
their pay reduced under play for pay, but it is hard to imagine
that any athlete can be worse off if all are getting paid closer
to their MRP (albeit different amounts).
ENOUGH ALREADY! THEY’RE ALREADY GIVEN A FREE
EDUCATION
The final argument against play for pay centers on the fact
that scholarship athletes are given a subsidized chance to earn
the future income returns and quality of life that go along with
a college education. Some judge that to be more than enough.
In addition, a select few are given the chance to pursue the
mammoth incomes earned by professional athletes. However,
this argument has two shortcomings.
The first shortcoming is that some (I repeat, some) athletes do
not value the educational component of their time spent in
college very highly. This irritates many people who hold a
college education in high regard. However, the preferences of
athletes cannot be invalidated just because some observers
wish they had a greater appreciation for their educational
opportunity. In fact, as with all things, the value of this type of
605

payment is an entirely subjective exercise. Because it is value
received that matters when discussing compensation, the
athletes are the only persons who can know the value of their
individual scholarships. This means that for some athletes
compensation in the form of educational opportunity is worth
little.
This actually is what drives some of the fortunate few players
to leave early, declaring themselves eligible for professional
drafts before they run out of college eligibility. They simply
assess the additional value of staying in school, including their
personal assessment of the value of education, versus the
value of moving on to the pro level. Roger G. Noll of Stanford
University showed, as discussed in the Learning Highlight:
Investing in an Athletic Career (Chapter 7), it can certainly
pay to invest in a shot at the pros by playing college sports.
Winfree and Molitor (2007) show a similar result—players
drafted in higher rounds have higher expected earnings if they
enter professional baseball, but players drafted in lower
rounds have higher expected earnings if they attend college.
Given this result, it is not surprising that players’ decisions
were highly correlated with the difference in expected lifetime
earnings. The simple corollary is that it can also pay to take
that reward before finishing an education. It also is clear that
this logic confronts players who do value their education
highly as well.
Here is a quote from another person from Stanford, All-
American and Pro Bowl offensive lineman Bob Whitfield
(Sports Illustrated, May 4, 1992, p. 78). “You’re asking a
player who has a chance to make money he never had before—
a young, desperate, black male—to stay at a school that can’t
really help me out as much as I can help myself? That’s
ludicrous. I may be young, but I’m not stupid.” Whitfield left
Stanford early and was chosen eighth overall in the 1992 NFL
draft by Atlanta. His career lasted 14 years, including 12
straight years with Atlanta, where he missed only one game
from 1993 to 2001. The Pro-Bowler earned over $2.3 million
(2009 dollars) in eight of those years and also earned a high
of $3.8 million in 2000. Whitfield played from 1992 to 2006,
while data on his NFL compensation begin only in 1995. But
from 1995 to 2006 he earned a total of $34.5 million (2009
dollars).
Now, only Whitfield knows how much his education was
worth, and not all players necessarily assess their pro
potential correctly. But apparently for Whitfield, finishing
even a prestigious and economically valuable Stanford degree
couldn’t compare to these types of earnings plus his
enjoyment from competing at the professional level. The value
of an education is a subjective thing even for those not as
fortunate as Whitfield.
A second shortcoming to the argument that an education is
sufficient compensation is that some athletes contribute more
revenue to the athletic department than the tuition value
included in their GIA. As we saw earlier in this chapter, the
MRP of star athletes can be between $400,000 and $1 million
(and possibly more at times). Players do receive
compensation, but it can be much lower than their MRP. In
606

this context, arguing that athletes are paid enough is now a
value judgment, and any person is free to pass such
judgments. But consider this: Do you think that you should be
paid your MRP at your job, even though you might be willing
to work for less if you had to?
POSTSCRIPT: WHY DON’T COLLEGE ATHLETES JUST
ORGANIZE?
In the face of monolithic owners determined to extract the
MRP of players, professional athletes organized under the
labor laws of the U.S. and Canada (Chapter 9). College
athletes face the same monolithic approach by athletic
directors through their enforcement arm, the NCAA. So why
don’t college athletes organize or at least challenge the
amateur requirement in court? The answer actually is quite
simple. To date, courts have typically decided that college
athletes are not employees of their institutions and labor law
only guarantees the rights of employees to organize. This also
affects their standing to sue in the court system. Until athletes
are employees, they will have to be satisfied with the access to
the process that they currently have—an advisory role granted
by the NCAA and efforts to organize outside the protection of
the labor laws (see the section-end Learning Highlight: The
College Athletic Revolution?).
607

LEARNING HIGHLIGHT: THE COLLEGE ATHLETIC
REVOLUTION?
One way to think about college athletes is as full-time
students who have full-time jobs in order to cover the costs of
attending their university. The NCAA has an official limit of
20 practice hours per week. Including Saturdays for a six-day
week, that’s three hours and 20 minutes per day. But FBS and
Division I athletes typically also have many hours per week in
other “suggested” sport-related activities and also travel
extensively during their season. It’s easy to see how this is a
full-time job. Further, these days it is a year-round activity;
outside of their season, all athletes attend “voluntary”
workout sessions and other activities.
Now, the full cost of attendance may be higher, but suppose
tuition, room and board, and books for an in-state student-
athlete are about $8,000. Further, suppose the same coverage
for a nonresident is about $12,000. At the NCAA limit of 20
hours, for a full year resident athletes would earn about $7.69
per hour and nonresidents about $11.54 per hour. But if
athletes spend fully 40 hours per week on their sports, the
numbers drop to $3.85 and $5.77 per hour for resident and
nonresident athletes, respectively. This latter case clearly
violates the minimum wage laws in effect in every state in the
country. For this, a very few athletes will have pro chances in
addition to their education, but the vast majority are just like
their nonathlete counterparts. The education is what they get
for their toil (plus the satisfaction of competing).
Now picture this: The fans in the arena and those watching on
TV wait breathlessly for the college basketball national
championship game to begin, but the CBPA (College
Basketball Players Association) has not reached an agreement
with the NCAA on seasonal wage structures or the share of
postseason revenues that will go to players (perhaps they’ve
seen Table 13.7). In protest, the CPBA brings a work stoppage,
and there is no NCAA championship that year.
Of course, there is no CPBA. Unlike their professional
counterparts, college athletes have not organized themselves
in any meaningful way to offset the power their athletic
departments have over them. Why haven’t college athletes
organized? Technically, the answer is that athletes are not
employees, so labor laws do not apply, and their right to
organize is not guaranteed. If athletes want to organize, they
must do it without the protections afforded by labor law.
The obstacles are many. The usual obstacles to organization
apply, such as educating the athletes, overcoming
organization costs and free riding, and ultimately winning
something for the membership. There is good reason to
suspect these obstacles to be difficult to overcome. Remember
we are talking about people who do not reach the age of
majority until well into their college careers. In addition, they
have stars in their eyes; they finally have their college shot
and might be reluctant to “rock the boat” (remember the
Learning Highlight: Pro Athletes as “Company Men” back in
Chapter 9).
608

Further, without the protection of labor law, the history of
those who organize labor suggests that retribution would be
expected. Indeed, athletic scholarships are an annually
renewable affair. In some cases, coaches have pressured
athletes to give up their scholarships. In a survey that
deserves to be run again (Sporting News, April 25, 1994, p.
50), 25 percent of juniors and seniors stated that their coaches
had threatened to terminate their scholarships.
However, despite these obstacles, athletes are organizing.
Formed in January 2001 by a group of former UCLA football
players, and formerly known as the Collegiate Athletes
Coalition, the National College Players Association (NCPA,
www.ncpanow.org) is backed by the United Steelworkers of
America. Rather than wages and shares of postseason
contracts, the NCPA seeks much more modest benefits. They
seek year-round medical coverage (not just during the
season), insurance coverage above student policy minimums,
scholarships at the full cost of attendance (typically adding
around $2,000 annually), and more flexibility to earn money
during the off-season (currently, the NCAA limits such
earnings to $2,000).
As Stanford All-American guard Casey Jacobsen put it, “We
feel that because of the success of our sport and all the money
that’s being generated, we want to see more of it coming back
to us.” But to pull it off may take something like a strategic
holdout at a crucial time, like the NCAA Final Four. It is
understandable that college athletes might not find the costs
worth the benefits.
The chances that college players will bring a work stoppage
are slim to none. [Photo Final Four Tip Off.]
Sources: Sports Illustrated, November 23, 1992, p. 124;
Seattle Times, April 25, 1995, p. L1; Jack Scott, The Athletic
Revolution (New York: Free Press, 1971); Sports Business
Journal, January 29, 2001, p. 23; ESPN.com, May 21, 2001.
609

http://www.ncpanow.org

http://www.ncpanow.org

SECTION 9
Synopsis of Remaining
Issues
A litany of issues remain, including race and sex
discrimination, subsidies and economic impact assessment
for college sports, stadium issues, and the relationship
between government and the college sports business
concerning taxes and market power. Because the economic
insights are a bit broader for two of these, we’ll pay particular
attention to gender discrimination under Title IX and
government relations. But let’s note the following in passing.
Pay differences for college coaches appear to be mostly
explained by differences in revenue potential across men’s
and women’s sports (Humphreys, 2000). Where women’s
sports are extremely successful, coaches of the women’s teams
have been paid more than their male counterparts. The Racial
and Gender Report Card, cited heavily in Chapter 7, reports
on college sports as well as pro. In the 2008 edition, it is clear
that women and minorities are underrepresented in upper
administration, especially at the level of athletic director.
Not much work has been done on the size of the subsidies
paid to college sports, but there is work on economic impact
assessment for major college sports events. As we would
expect given our coverage in Chapter 10, economic activity
value estimates for college sports and events appear grossly
overstated (Baade and Matheson, 2004). But there are
consumers’ surpluses and externality values from the
construction of college facilities worth a few million dollars
(Johnson and Whitehead, 2000). Further work on surpluses
and externalities will prove enlightening.
Finally, it isn’t clear whether there is a college equivalent to
concerns over venue subsidies (Chapter 11). Athletic
departments appear not to be constrained by the university as
long as they build within the means of their booster
supporters through fund-raising activities. At a few of the
most successful departments, accrued annual net operating
revenues are also used to save overtime for capital projects.
But it is clear that athletic departments are seeking more
lavish surroundings for their fans in attendance in order to
capture higher revenue streams, and lately, the argument is
being made that capital projects also deserve public subsidy.
The University of Washington athletic department recently
commissioned a study of the economic impact of Husky
athletics on the state of Washington (the study was done by
the same person who produced the Mariners and Seahawks
impact statements discussed in Chapter 10). In conjunction
with the release of that study, University of Washington
president Mark Emmert released his “Husky Stadium
Renovation: Making the Case for Public
Subsidy” (www.washington.edu/research/industry/
newsletter/0908insidestory.html). In it, he states,
“Specifically, we will ask the state to redirect $150 million in
610

http://www.washington.edu/research/industry/newsletter/0908insidestory.html

http://www.washington.edu/research/industry/newsletter/0908insidestory.html

http://www.washington.edu/research/industry/newsletter/0908insidestory.html

http://www.washington.edu/research/industry/newsletter/0908insidestory.html

revenue from existing ‘tourism taxes’ that previously have
been used to pay for the construction of Safeco and Qwest
fields and that are paid primarily by visitors to King County.”
After our coverage in Chapter 10, we shouldn’t be surprised
that similar logic is behind this request.
ECONOMIC INSIGHTS INTO DISCRIMINATION POLICY
IN COLLEGE SPORTS: TITLE IX
Title IX of the Educational Amendments of 1972 (Title IX for
short) recognizes that discrimination is much more than just a
pay or hiring problem. It is also about access. Title IX states,
“No person in the United States shall, on the basis of sex, be
excluded from participation in, or denied the benefits of, or be
subjected to discrimination under any educational program or
activity receiving Federal financial assistance.” Because
almost all schools receive federal funds, Title IX applies to
nearly all education institutions.
The change mandated by Title IX has been slow in coming,
and participation gaps still remain. Spending parity isn’t
required under Title IX, but relative spending on men’s and
women’s sports is one index of equality. The example of the
major universities in the state of Washington back in Table
13.9 is representative of Title IX outcomes across college
sports partly captured in Table 13.2. In Table 13.9, three
things are quite clear concerning relative spending. First, the
amount of money spent on football swamps all other
considerations. But so do football revenues. This gives rise to
an interesting debate (covered next) about whether football
subsidizes other sports, including women’s sports. Second, in
terms of revenues generated, women’s sports pale in
comparison; men’s sports generate 8.7 times and 11.9 times
the revenue of women’s sports at Washington State and
Washington, respectively. Third, the athletic department at
Washington State spends twice as much on men’s sports, and
the multiple at Washington is 2.5. Interestingly, the two
departments spend almost identical shares of total revenue on
women’s sports, about 18 percent.
Even 30 years after the passage of Title IX, equity remains
elusive. At average reports for FBS conferences, this is clear
from Table 13.2. Inequality has evolved over a long time, and
very well-established interest groups like the way that money
has traditionally been spent. Boosters, athletic directors, and
coaches all are doing quite nicely under the traditional
arrangement. It should come as no surprise that the shriek
from the athletic department is clearly audible when
university presidents, who fear reductions in their federal
funds, require athletic directors to meet the requirements of
Title IX. Addressing some of the arguments against Title IX
helps to clarify why this remains the case.
WON’T GENDER EQUITY GAINS HARM FOOTBALL AND
KILL MINOR MEN’S SPORTS?
The main justification for how slow institutions have been to
respond to the requirements of Title IX follows a line of logic
that is practically a perfect echo of arguments against play for
pay. Athletic administrators argue that they barely break even
611

as it is. Therefore, to pay for gender equity, either nonrevenue
men’s programs will have to suffer or revenues will have to
come out of men’s revenue sports, primarily football. Some
argue diverting money from football could so weaken the
ability of the football program to continue to prosper and
subsidize other sports that all of the sports at any college
could be worse off.
It is true that on occasion non-revenue men’s sports have
been cut. In one notable case, the FBS University of Kansas
athletic director cut men’s swimming and tennis to stay in the
black, saving about $600,000 per year and reducing
participation by 50 male athletes (Sports Business Journal,
March 12, 2001, p. 20). Among the reasons cited for the cuts
were increasing scholarship costs, a 115 percent increase in
team travel costs for the other sports, and increases in
coaches’ and administrators’ salaries. It was also noted that
the cuts were made in order to meet gender equity
requirements; cutting men’s sports may free some funds to
spend on women’s sports, and the arithmetic of increasing the
ratio of women is obvious if participation by men is reduced.
In an interesting twist of fate, effective July 1, 2007, the James
Madison University athletic director cut seven men’s sports
and three women’s sports, stating flatly that the cuts were
made for the sake of Title IX compliance (USAToday.com,
November 1, 2006). At the end of the 2007 season Ohio
University’s athletic director ended men’s outdoor track and
field, men’s indoor track and field, men’s swimming and
women’s lacrosse for the same reasons
(ColumbusDispatch.com, May 3, 2007). Just how it is that
cutting women’s sports helps contribute to Title IX was never
addressed in either reported case.
But do these cuts happen because athletic directors have no
other way to meet gender equity requirements? Or do they
happen because cutting minor men’s and women’s sports
preserves other spending that athletic directors value more. It
is extremely interesting that, as in the case of play for pay,
athletic administrative and coaches salaries never seem to
make it into the discussion of where to find money for Title IX
compliance.
Let’s look first at the premise that the ability of football to
subsidize other sports will be harmed. This can only be true if
football indeed even generates enough net revenue to carry
out this subsidization. There is some startling evidence that
this simply is not true. Leeds, Suris, and Durkin (2004) find
that colleges have continued to underfund women’s athletics,
but more importantly only a few of the most profitable
football programs provide any actual subsidy to women’s
sports. Almost all other athletic departments, including some
with highly profitable football programs, actually drain money
from women’s sports. So not only have Title IX gains to date
proven no danger to big-time college football, but just the
opposite appears to be true. The authors conclude, “One is
forced to conclude that many collegiate athletic departments
view violating Title IX as an optimal strategy” (p. 150).
612

The case of universities in the state of Washington yields
another look at the relationship between football and
spending on gender equity. In Table 13.9, take Washington
State first. Football generated $4.9 million over its own
expenses, and the net sum across all men’s sports was $4.0
million. Some $3.8 million in excess of women’s sports
revenues were spent on those sports. So men’s sports,
including football, just cover the excess spending on women’s
sports. But would any increment to gender equity have to
come from football? The answer is clearly no. In addition to
all of the other amounts, $13.5 million, or 44 percent of the
budget, was spent on the rest of the athletic department. If the
athletic director spent optimally on men’s sports, additional
spending on women’s sports would have to come from this
other spending. Similar calculations for the University of
Washington show that football generated $16.0 million after
its costs. The sum across all men’s sports is $14.4 million.
After covering women’s sports, men’s sports including football
yield $9.8 million on net for other spending in the athletic
department. Clearly, in this case, additional spending on
women’s sports could be had just out of the current men’s
surplus.
Our observations on this argument when it was encountered
against play for pay serve us just as well in the case of Title IX.
Athletic departments do not really break even because they
adjust their budgets to spend any excesses within budget
periods. The money is already there. It is just being spent on
other parts of the athletic department budget. At a place like
the University of Washington, a highly successful athletic
program, enough surplus from men’s sports exists to increase
spending on gender equity by nearly $9.8 million, in fact, 1.3
times more than is currently spent on women’s sports there.
At a more typical athletic department, Washington State,
additional spending on gender equity would need to come out
of money not spent on men’s sports, but there are quite a few
million dollars there to draw from.
Returning to Table 13.2, prior to 2004, the same story was
typically true (at the average of reported revenues and
expenses) across the FBS. The average net operating revenue
for men’s sports was large enough to cover the women’s sports
deficit. No sports would have been needed to have been cut as
long as the men’s net amount were diverted to women’s sports
rather than salaries and facilities. From 2004 on, it is no
longer the case the men’s surplus revenue can cover the
women’s deficit. But, again, in determining how to cover that
deficit, all spending across the athletic department can be
under consideration, not just spending on sports.
If discrimination has reduced investment in women’s sports
below the amount preferred by those taxpayers supporting
Title IX, then there may have been overinvestment in other
areas in these departments. Equalizing the resources devoted
to men and women will just reduce these overpayments. Part
of the overinvestment is in terms of salaries and
administration, but some of the overinvestment will also be in
terms of spending on men’s sports. Therefore, Title IX will
prompt athletic directors to correct for past overinvestment in
613

men’s sports. As in the case of play for pay, we should expect
that those who currently benefit from past distortions in
spending will not be happy about reallocations that will come
out of their paychecks. However, their arguments only thinly
veil the obvious reason they have dragged their feet on Title
IX compliance.
SOME UNINTENDED CONSEQUENCES OF TITLE IX
When Title IX dramatically increased the value of offering
more opportunities for women to participate in sports, inputs
to that process became more valuable to athletic departments.
This is easy to see by looking at the gender balance of coaches
for women’s sports. In 1972, women coached more than 90
percent of women’s college teams. By 1996, that had fallen to
47.7 percent (Zimbalist 1999, p. 60). Through the end of the
1990s, women coached about 51 to 52 percent of women’s
college teams (Center for the Study of Sports, 2001).
Apparently, as the value of women’s basketball rose because
of Title IX compliance, the price of coaches was bid up. This
led many male coaches to enter the market to coach women’s
teams. The final result of that market outcome was a larger
number of male than female coaches. Sticking to the MRP
theory of pay, this would tend to indicate that women coaches
at the time were outdone by an influx of men who had higher
demonstrated abilities to produce winning for college
programs. In terms of our discussion of discrimination in
Chapter 7, women’s coaching resumes at the time were less
impressive than their male counterparts because they did not
have the same access to training and experience. Of course,
that raises the more important issue of why that access was
limited in the first place. This is especially true since nearly all
of the true coaching superstars in women’s sports are women.
But this is a topic economics cannot address.
GOVERNMENT AND THE COLLEGE SPORTS BUSINESS
We can take a look at the relationship between government
and college sports through the same lens we used in pro
sports. College sports enjoy some special tax advantages and
have had run-ins with the antitrust laws. The NCAA provides
market power for its member colleges in their sports
endeavors, and our model of the government process may
lead us to wonder just why that occurs.
TAXES
The tax laws are an important contributor to both athletic
department welfare and the welfare of the NCAA (a
corporation granted not-for-profit status under the tax laws).
Public universities and private not-for-profit universities must
be careful with revenue-generating activities. If revenues
exceed the percentage specified under the law, tax liability
looms. From this perspective, an important behavior from
earlier in this chapter safeguards universities from tax
scrutiny. The accounting practices of athletic departments
allow them to plow all of their net revenues back into the
department during the year that they are earned. Under this
practice, no profits ever are shown to draw the scrutiny of tax
authorities.
614

Another form of tax advantage to universities concerns tax-
deductible contributions to the athletic department. The idea
of such a contribution is that the contributor expects nothing
in return; the deduction is granted for true acts of charity. The
estimated value of such contributions is tax deductible. To
date, the IRS has treated contributions to athletics
departments just like any other charitable contribution to any
of the university’s academic programs—they are tax
deductible to the contributor. However, this is a questionable
practice by the IRS for nearly all contributions to athletics
departments. Most contributions come from boosters and
sponsors, and these “contributors” definitely expect and
receive something in return. Senator Charles Grassley, then
Chair of the Senate Finance Committee, sums up the point of
contention nicely (The Chronicle of Higher Education,
chronicle.com, October 5, 2007), “When I hear stories about
top donors to college athletic programs getting a free seat on
the team plane, I wonder what the public gets out of that. We
need to make sure that taxpayer subsidies for college
athletics-program donations benefit the public at large.”
Boosters are rewarded with prime seating and other types of
services (guest tent privileges and close proximity to players
and coaches) by the athletic department. Indeed, prime
seating is available only if a booster makes a contribution.
Typically, the higher the contribution, the better the seating
and the greater the other benefits. This clearly is payment for
benefits received and, in all cases except college sports,
eliminates tax-deductible status.
Business sponsors clearly get something in return for their
contribution. During breaks in the action, sponsors are
announced and broadcast on scoreboard TV screens. Sponsor
signage is all around the stadium. So, their “contribution”
generates recognition. Even though this bears a striking
resemblance to a fee paid for services rendered, sponsorship
contributions are, by and large, tax deductible. And the
justification, again, is that the athletic department rides on
the tax-deductible coattails of the academic mission of the
university.
MARKET POWER AND ANTITRUST
Member athletic directors, acting through conferences and
the NCAA, create market power. At times, they act outside of
the NCAA and create market power as well, as in the case of
the formation of the BCS by the big six FBS conference
commissioners and media providers. At the basic level, joint
venture activity is not required, so economic intuition leads us
to believe that college athletic directors (overseen by
university administrators) are better off pursuing conference
membership restrictions, TV negotiations, and player
restrictions through conferences, the NCAA, and the BCS than
going it alone.
Just like the Federal Baseball decision did in pro sports, the
NCAA Decision set the stage for most of these issues in college
sports. Interestingly, Congress has chosen the same approach
to both decisions, namely, to do nothing. Congress has not
overturned the ability of pro leagues to completely control
615

team location; it has encouraged mergers and the single
dominant league outcome; and it has codified joint venture
TV negotiations for pro sports leagues. Likewise, Congress has
done nothing to interfere with the operations of college
conferences or the NCAA on a similar array of issues.
Conferences tightly control membership; the NCAA takes
actions to guarantee its sole authority against potential rivals;
joint venture TV negotiations are the rule; and players’
earnings are restricted. The one behavior that was ruled illegal
by the courts in the NCAA Decision, joint venture sports
management by the NCAA in college football, continues in
basketball as the NCAA manages March Madness in both
men’s and women’s basketball.
Government intervention clearly has benefits for college
sports consumers. This can be seen in the analysis of the
competitive impacts of the NCAA Decision. Bennett and Fizel
(1995) reported three main findings on the impact of this
decision. First, in the long run, the NCAA Decision had no
overall impact on the standard deviation of winning percent
(compared to the idealized standard deviation in the usual
way). Second, winning percents of strong teams declined,
whereas those of weak teams increased (enough to reduce the
difference by two games per year). Finally, there was a decline
in the difference in playing strength between traditional
powers and nonpowers. In summary, the NCAA Decision, by
imposing more economic competition among NCAA athletic
departments, enhanced competitive balance.
James Quirk and I (1999) also looked at a few of these
measures and some others for college football. On average,
top 10 finishers before the NCAA Decision finished higher
after the decision. However, they did so much less often.
Quirk and I also examined the Pac-10 and Big Ten in some
detail. We found that the conference winning percents became
more balanced after the NCAA Decision. Inequality in
winning percents declined nearly 24 percent in the Pac-10 and
almost 6 percent in the Big Ten. Finally, literally the same
teams won all of the conference championships just prior to
the NCAA Decision (USC, UCLA, and Washington in the
Pac-10, and Michigan, Ohio State, and Iowa in the Big Ten).
However, these same teams won 20 percent fewer
championships in the Pac-10 and almost 10 percent fewer in
the Big Ten after the decision (harken back to Table 13.11).
The lesson from these analyses is that a little economic
competition appears to be good for competition on the field.
We also saw in Table 13.10 that a little economic competition
reduced the price of games to media providers, which
increased the viewing alternatives for sports fans at home.
These lessons cannot be lost on Congress, so we are left to try
to explain why it remains noncommittal in its treatment of
college conferences and the NCAA.
THE RATIONAL ACTOR POLITICS BEHIND THE STATUS
OF COLLEGE SPORTS
In college sports, so-called booster and sponsorship
“contributions” are tax deductible as charitable donations. It
616

is a funny kind of donation that earns boosters prime seat
locations or firms a service in return for the contribution.
However, this practice is permitted by the government.
Further, Congress has allowed colleges acting through
conferences and the NCAA completely free reign over college
sports output and players.
I think you can probably see how a rational actor politics
model of college sports would explain the special tax and
market power status afforded athletic departments and their
sports activities. Here, the concentrated interest occurs in
every single state. Any revision of the special tax or market
power status enjoyed by colleges would hurt all college
athletic departments. However, larger, richer college athletic
departments would probably be harmed more than smaller
ones. Therefore, politicians pursue the welfare of colleges
under the IRS interpretation of charitable gifts and allow
college athletic directors to do as they will in running sports
through the NCAA. Athletic directors, working to further the
goals of university administrators, favor this outcome because
their boosters and sponsors enjoy greater net benefit from
their contributions than if those contributions were taxed. In
addition, they are also able to price according to their market
power position and keep nearly all of the value produced by
college athletes.
617

SECTION 10
Chapter Recap
For the most part, there is very little new in terms of
economics when we turn from pro sports to college sports
until we get to college sports market outcomes. Demand
concepts provide similar insights into college sports,
especially the impact of preferences for women’s and men’s
sports and the importance of population as a revenue
determinant. Market power exists for individual athletic
departments, but athletic directors need the NCAA to help
maintain that power because so many conferences compete
with each other for talent and TV dollars.
Revenues are much smaller for college athletic departments
than for pro sports teams. However, revenue inequality is just
as apparent for college sports, and it appears to be growing
over time. The top revenue schools are gaining against the
middle as well as the bottom. In terms of TV revenues,
conferences are much more competitive in providing
programming to media providers than are pro leagues. Total
college broadcast revenues are much lower than even the
NHL but are unequally distributed among teams in a
conference even though some revenue sharing does occur.
Bowl payouts to athletic departments are simply spent, with
the directors of those departments taking their value out in
terms of exposure and advertising for recruiting purposes.
On the supply side, comparisons to other parts of the
university reveal that the value of athletic departments is in
terms of money and political support. From that perspective,
it is clear that athletic departments are worth having. This
simple logic explains institutional support of college athletic
departments. Further, nonstandard treatment of the athletic
department relative to other departments on campus, such as
budgeting process differences and cost forgiveness, are
understandable given the difficulty in predicting when a
department will be successful and the values they provide to
university administrators.
Turning to college sports market outcomes, athletic
departments are in conferences for the same reasons that pro
owners join leagues—they are better off financially in
conferences than going it alone as independents. In college
sports, the NCAA performs the joint entity cooperation
function for athletic directors. The NCAA maintains
exclusivity for the FBS in football and Division I in basketball
and negotiates the TV deal for college basketball’s national
championship. The famous NCAA Decision (1984) precludes
the NCAA from doing the same for anything but the
postseason in football (outside of the BCS) and basketball.
Competitive imbalance follows from revenue imbalance and
exclusive conference membership. Postseason revenues are
shared by members of a conference, but better teams get a
618

larger slice. And the vast majority of FBS conferences do not
share any other revenues equally. Competitive balance is,
instead, maintained primarily through restrictions on players
(the amateur requirement, recruiting restrictions, and
mobility restrictions). The main result of those restrictions is
to redistribute money from college players to their athletic
departments. Of course, all of these requirements generate an
incentive to cheat by individual coaches, boosters, and
players, and it does not appear that the NCAA’s heart is in its
enforcement job.
Theoretical analysis of the amateur restriction suggests the
college version of Rottenberg’s invariance principle. Athletic
directors can make the most of their athletes at the same
distribution of talent between athletic departments whether
athletes earn their MRP or it goes mostly to athletic
departments. The range of estimated values of individual star
players to their athletic departments is from $400,000 to $1
million. Following the theory, paying college players their
MRP, rather than just paying their tuition, room and board,
and books, should not change anything except to reduce the
level of spending on administrative salaries, coaches’ salaries,
and facilities. However, different athletes would be paid
different amounts.
For two remaining items, economics offers substantial
insights. Title IX (1972) was passed to bring about gender
equity in terms of participation for male and female athletes
in college. From an economic perspective, all of the fears
voiced about paying for gender equity are as groundless as
those concerning play for pay. However, gender equity has
proven elusive, even after more than 30 years since passage of
Title IX. Turning to government and the college sports
business, athletics enjoys tax advantages, primarily the tax-
deductible status of donations despite the fact that these
“donations” typically provide goods and services to the
contributor. Congress has made no moves to investigate the
market power enjoyed and promulgated by the NCAA for
member athletic departments. The same rational actor politics
model used to explain pro sports advantages also explains this
college sports outcome.
619

SECTION 11
Key Terms and Concepts
You should have run into each of these in pop-ups in the text
of this chapter:
• National Collegiate Athletics Association (NCAA)
• Donation seating
• Institutional support
• Revenue imbalance in college sports
• Three-tier broadcast rights structure
• Spillover benefits to the university
• Institutional support
• NCAA Decision
• Winning percent imbalance
• Championship imbalance
• Recruiting restrictions
• National letter of intent
• Sit-out restrictions
• Forfeit remaining playing eligibility
• Amateur requirement
• Cheating on NCAA rules
• College version of the invariance principle
• Play for pay
• Title IX
• Tax-deductible contributions to the athletic department
• Rational actor politics model of college sports
620

SECTION 12
Review Questions
1. Which has higher absolute quality of play, the FBS
(previously, Division IA) or the FCS (previously, Division
IAA)? Explain. Give an example of relative competition in
college sports.
2. Explain why it must be preferences that explain the
difference between the demand functions in Figure 13.1.
3. Explain the effect that each of the following would have
on the demand for University of Georgia college football
games:
a. All else constant, some UGA fans do not like the
cheating that occurs in college football.
b. All else constant, incomes of residents in the state of
Georgia increase.
c. All else constant, the price of a season ticket to the
MLB Atlanta Braves falls to equal the price of a UGA
season ticket.
d. All else constant, UGA fans expect season ticket prices
to rise next year.
4. Describe donation seating. Is it price discrimination?
Explain.
5. What does “institutional support” mean when describing
college sports revenues? Describe the behavior of
revenues in college sports over time using Table 13.1.
Does it grow over time nearly as much without
institutional support?
6. Describe the three-tier broadcast rights structure used in
college sports. Explain each tier fully.
7. Why is there more competition on the supply side of
sports programming in college sports than in pro sports?
8. Describe the four types of sponsorship currently
occurring in college sports.
9. What was the original purpose of college bowl games?
Has this changed for all bowl games or just for some? If
just for some, which ones? Explain.
10. Why isn’t a focus on the athletic department the most
insightful level of analysis when discussing college sports?
What is the object of analysis that is most insightful?
What are the goals of that decision maker? Explain, with
special attention to Figure 13.2.
11. List the spillover benefits that are purported to flow from
the athletic department to the university. What actually is
known about these spillover benefits?
621

12. Why do colleges join conferences? How does the
example of Notre Dame’s remaining independent help
explain this motivation?
13. What is the NCAA? What purpose does it serve for
college sports programs?
14. What is Title IX? What does it require?
15. What special tax advantages are enjoyed by college
sports? Provide examples.
622

SECTION 13
Thought Problems
1. Has institutional support grown over time in college
sports (use Table 13.1)? Why or why not? Provide an
explanation for your answer using Figure 13.2.
2. Examine the revenue disparity data in Table 13.1. What
has happened to the ratio of highest to average revenue
over time? What does this suggest about the level of
competitive balance in the FBS?
3. How does just looking at Notre Dame’s broadcast rights
value indicate that broadcast rights revenues are
unequally distributed throughout the FBS? (Hint:
Remember that there are 120 FBS schools.)
4. A clever sports economics student asked the athletic
director at that university to dig back into the records for
data on attendance and prices at the college’s tennis
matches. The student’s research revealed the following
attendance demand function for alumni and other
boosters:

PB = booster price and AB = booster attendance. Draw this
attendance demand function. Does this tennis team have
market power over boosters? Why or why not? What is the
highest possible total revenue that the team can hope to
collect? At what level of attendance? At what price?
5. In what ways is the football program just like any
academic department at the university? In what ways is it
different? How does this help determine the different
ways that the university administration treats each?
Explain fully using Figures 13.3 and 13.4.
6. College athletic departments are rarely economically self-
sufficient. Why is this the case? If athletic departments
cannot break even, why do coaches’ salaries increase year
after year? Explain fully.
7. FBS athletic departments are required to have a
minimum number of varsity sports. What role does this
restriction play in the welfare of conference members?
8. What essential service did the Big East provide for Miami
to get it to join that conference? Why was the Big East so
easy on Miami in terms of it sharing revenues with the
rest of the conference? (Refer to the Learning Highlight:
Miami Joins the Big EastNo, Make That the ACC.)
9. Why do colleges need the NCAA? What does the NCAA
accomplish for colleges that their conference cannot
accomplish? Give examples.
623

10. Give an overview of revenue, winning percent, and
championship imbalances in college sports using the data
in Tables 13.1 and 13.11. What is the ultimate source of
these imbalances?
11. List the ways that athletic departments acting through
conferences and the NCAA try to reduce competitive
imbalance. Are these methods effective? Why or why not?
12. Using the Fort and Quirk (1999) expression for the
expected value of cheating, if the probability of getting
caught is the same across all colleges, explain why
cheating would be more likely in a program with a
currently weak record and high potential for financial
gain with a young coach. Using Figure 13.6, demonstrate
the effect of an increase in the gains from cheating.
13. What is the main lesson from the graphical analysis of
the NCAA’s amateur requirement? How does this
generate the college version of the invariance principle?
Be sure to use Figure 13.7 in your explanation.
14. About 49 percent of the head coaches of women’s sports
were women according to the Race and Gender Report
Card, 2001. Does this mean that there is hiring
discrimination against women at that position? State the
arguments against that position. State the arguments
supporting that position. Can you draw any conclusions
about this problem from economic analysis? Explain.
15. How is it that special tax advantages and market power
for the NCAA persist in college sports? Detail a rational
actor politics response.
624

SECTION 14
Advanced Problems
1. Performance-enhancing drugs are an issue in college
sports. Are players looking to gain absolute or relative
advantage through these drugs? What is the outcome on
the field if all players use performance-enhancing drugs?
What are the benefits and costs of outlawing these drugs?
2. A common story is that cost escalation plagues college
athletic departments. However, over the last few years,
more athletic departments are showing a positive balance
(Table 13.1). Reconcile these conflicting stories. (Hint:
Which elements on athletic department balance sheets
are increasing in cost? Rent? Players?)
3. In addition to the booster demand function in Thought
Problem 4, our clever student also discovers that the
demand by students for attendance at the same tennis
matches and the same seat types is as follows:

PS = student price and AS = student attendance. Show
why students will be charged a lower price for the same
event and seat type than boosters. What is the ratio of
booster to student price if the athletic director cares about
the bottom line? Is this price discrimination?
4. Pro owners, due to the imputed profit charge when they
obtain a team either through purchase or expansion, can
be expected to behave like rapacious monopolists
(Chapter 5). Does the same force operate in college
sports? What drives college athletic departments to
behave the same way?
5. Suppose that the NCAA was not allowed to run March
Madness. What do you expect would happen to broadcast
rights fees for the championship series? Who would host
it and set it up?
6. Patrick Ewing played college basketball at Georgetown
University. According to the Chronicle of Higher
Education (January 8, 1986), Ewing’s presence was worth
about $12 million over his four-year college career—the
value of attendance at home games tripled; sales of
memorabilia skyrocketed; and additional TV appearances
went with the success of the team. The athletic
department paid for the cost of Ewing’s scholarship over
the period, about $48,600. What percent of Ewing’s MRP
did the Georgetown athletic department keep? Compare
this to the other estimates in the text, and explain any
discrepancy between the Ewing case and those estimates.
7. Suppose through a legal finding that college athletes were
deemed employees of the athletic department. How
625

would this change the structure of the relationship
between college athletes and their athletic departments?
Is it possible that the amateur requirement could be in
danger? How? (Hint: As employees, athletes would enjoy
the same labor rights under federal law as any other
group of employees.)
8. Why is it important to athletic departments that the
competitive equilibrium level of talent on all teams be
maintained when the amateur requirement is in force?
(Use Figure 13.7 in your explanation.) What is the
implication of this important observation for arguments
against play for pay? Examine each argument one at a
time in your explanation.
9. Give the MRP explanation for why most male college
basketball coaches make more than most female coaches.
Again, using the MRP explanation, when will female
coaches make more than male coaches? What is the most
obvious evidence that there is discrimination against
women college basketball coaches? (Hint: How many
female coaches of men’s programs are there?)
10. Why is participation in college sports still 60–40 in favor
of men over women, even though federal law has dictated
equality for more than 30 years under Title IX?
626

SECTION 15
References
Baade, Robert A., and Victor A. Matheson. “An Economic
Slam Dunk or March Madness? Assessing the Economic
Impact of the NCAA Basketball Tournament,” in The
Economics of College Sports. John Fizel and Rodney Fort,
eds. Westport, CT: Praeger Publishers, 2004.
Bennett, Randall W., and John L. Fizel. “Telecast
Deregulation and Competitive Balance: Regarding NCAA
Division I Football,” American Journal of Economics and
Sociology 54 (1995): 183–198.
Brown, Gary T. “The Electronic Free Ticket: Early NCAA
Feared Televised Football Would Harm Attendance,” NCAA
News, November 22, 1999.
Brown, Robert W., and R. Todd Jewell. “Measuring Marginal
Revenue Product in College Athletics: Updated Estimates,” in
The Economics of College Sports. John Fizel and Rodney Fort,
eds. Westport, CT: Praeger Publishers, 2004.
Byers, Walter with Charles Hammer. Unsportsmanlike
Conduct: Exploiting College Athletes. Ann Arbor, MI:
University of Michigan Press, 1995.
Center for the Study of Sports. Race and Gender Report Card.
Northeastern University, Boston, MA, 2001.
Depken, Craig A. II, and Dennis P. Wilson. “The Impact of
Cartel Enforcement in NCAA Division I-A Football,” in The
Economics of College Sports. John Fizel and Rodney Fort,
eds. Westport, CT: Praeger Publishers, 2004a.
Depken, Craig A. II, and Dennis P. Wilson. “Institutional
Change in the NCAA and Competitive Balance in
Intercollegiate Football,” in The Economics of College Sports.
John Fizel and Rodney Fort, eds. Westport, CT: Praeger
Publishers, 2004b.
Fleischer, Arthur A., Brian L. Goff, and Robert D. Tollison.
The National Collegiate Athletic Association. Chicago, IL:
University of Chicago Press, 1992.
Fort, Rodney, and James Quirk. “The College Football
Industry,” in Sports Economics: Current Research. John Fizel,
Elizabeth Gustafson, and Lawrence Hadley, eds. Westport,
CT: Praeger, 1999.
Frank, Robert H. “Challenging the Myth: A Review of the
Links Among College Athletic Success, Student Quality, and
Donations,” Prepared for the Knight Foundation Commission
on Intercollegiate Athletics. May, 2004.
Fulks, Daniel L. Revenues and Expenses of Divisions I and II
Intercollegiate Athletic Programs: Financial Trends and
627

Relationships. Overland Park, KS: National Collegiate Athletic
Association, 2003, 2008.
Goff, Brian. “Effects of University Athletics on the University:
A Review and Extension of Empirical Assessment,” in The
Economics of College Sports. John Fizel and Rodney Fort,
eds. Westport, CT: Praeger Publishers, 2004.
Grimes, Paul W., and George A. Chressanthis. “Alumni
Contributions to Academics: The Role of Intercollegiate
Sports and NCAA Sanctions,” American Journal of Economics
and Sociology 53 (1994): 27–40.
Humphreys, Brad R. “Equal Pay on the Hardwood: The
Earnings Gap Between Male and Female NCAA Division I
Basketball Coaches,” Journal of Sports Economics 1 (2000):
299–307.
Humphreys, Brad R.. “The Relationship between Big-Time
College Football and State Appropriations for Higher
Education,” International Journal of Sport Finance 1, no.
2(May 2006): 119–128.
Johnson, Bruce, and John Whitehead. “Value of Public Goods
from Sports Stadiums: The CVM Approach,” Contemporary
Economic Policy 18 (2000): 48–58.
Leeds, Michael A., Yelena Suris, and Jennifer Durkin. “College
Football and Title IX,” in The Economics of College Sports.
John Fizel and Rodney Fort, eds. Westport, CT: Praeger
Publishers, 2004.
Noll, Roger G. “The Economics of Intercollegiate Sport,” in
Rethinking College Athletics. Judith Andre and David N.
James, eds. Philadelphia, PA: Temple University Press, 1991.
Padilla, Arthur, and David Baumer. “Big-Time College Sports:
Management and Economic Issues,” Journal of Sport and
Social Issues 18 (1994): 123–143.
Rottenberg, Simon. “The Baseball Players’ Labor Market,”
Journal of Political Economy 64 (1956): 242-258.
Scott, Jack. The Athletic Revolution. New York: Free Press,
1971.
Smith, Ronald A. Sports and Freedom: The Rise of Big-Time
College Athletics. New York: Oxford University Press, 1988.
Sperber, Murray. College Sports, Inc.: The Athletic
Department vs. the University. New York: Henry Holt and
Company, 1991.
Tucker, Irvin B. “A Reexamination of the Effect of Big-Time
Football and Basketball Success on Graduation Rates and
Alumni Giving Rates,” Economics of Education Review 23
(2004): 655–661.
Tucker, Irvin B. “Big-Time Pigskin Success: Is There an
Advertising Effect?” Journal of Sports Economics 6 (2005):
222–229.
Winfree, Jason A., and McCluskey, Jill J. “Incentives for Post-
Apprehension Self-Punishment: Univrsity Self-Sanctions for
628

NCAA Infractions,” International Journal of Sport Finance 3
(2008): 196–209.
Winfree, Jason A., and Molitor, Christopher J. “The Value of
College: Drafted High School Baseball Players,” Journal of
Sports Economics 8 (2007): 378–393.
Zimbalist, Andrew. Unpaid Professionals: Commercialism
and Conflict in Big-Time College Sports. Princeton, NJ:
Princeton University Press, 1999.
629

SECTION 16
Suggestions for Further
Reading
Surely, I can come up with some.
630