For years, many working individuals in the United States have counted on Social Security as their primary retirement income. With the deductions from every paycheck to fund this future benefit, Americans looked forward to a comfortable retirement based on the assumption that these funds were being invested wisely. Is this a factual assumption, or is it another myth of the U.S. social welfare system?

Relying on Social Security as a sole means of support in retirement is uncommon. As society’s work habits and life expectancies change, Social Security has come to mean different things to different people. For those who are unable to save and/or invest on their own, it functions like a forced retirement program that provides a financial safety net for the future. For those who become unable to work, social security often is viewed by others as a public assistance program. For others who are financially secure and prepared for retirement, it is an entitlement program where individuals draw Social Security at retirement in an effort to recoup what they paid into the program by deductions from their wages. How might these perceptions of Social Security relate to the perceptions of public assistance programs that were discussed in last week’s resources?

For this Discussion, review this week’s resources. Then, consider how Social Security is different from public assistance programs. Finally, think about how these differences are important in terms of the general perception of Social Security benefits.

By Day 3

Post an explanation of how Social Security is different from public assistance programs. Then, explain how these differences are important, in terms of the general perception of Social Security benefits.

Popple, P. R., & Leighninger, L. (2019). The policy-based profession: An introduction to social welfare policy analysis for social workers (7th ed.). Upper Saddle River, NJ: Pearson Education.

• Chapter 7, “Aging: Social Security as an Entitlement” (pp. 138-160)

CONGRESS OF THE UNITED STATES
CONGRESSIONAL BUDGET OFFICE

CBO

Social Security

Policy

Options

JULY 2010

Pub. No. 4140

A

S T U D Y

CBO

Social Security Policy Options

July 2010

The Congress of the United States O Congressional Budget Office

CBO

Notes

Unless otherwise noted, all years are calendar years.

Numbers in the text and tables may not add up to totals because of rounding.

Preface

Social Security is the federal government’s largest single program, and as the
U.S. population grows older in the coming decades, its cost is projected to increase more
rapidly than its revenues. As a result, under current law, resources dedicated to the program
will become insufficient to pay full benefits in 2039, the Congressional Budget Office (CBO)
projects. Long-run sustainability for the program could be attained through various
combinations of raising taxes and cutting benefits; such changes would also affect the
Social Security taxes paid and the benefits received by various groups of people. This CBO
study examines a variety of approaches to changing Social Security, updating an earlier work,
Menu of Social Security Options, which CBO published in May 200

5.

In keeping with CBO’s
mandate to provide objective, impartial analysis, the current study makes no
recommendations.

The study was written by Noah Meyerson, Charles Pineles-Mark, and Michael Simpson of
CBO’s Health and Human Resources Division, under the direction of Joyce Manchester and
Bruce Vavrichek. Research assistance was provided by Philip Armour, Sarah Axeen, and
L. Daniel Muldoon. James Baumgardner, Sheila Dacey, Benjamin Page, David Rafferty,
Jonathan Schwabish, and Julie Topoleski provided helpful comments on earlier drafts.
Andrew Biggs of the American Enterprise Institute and Paul Van de Water of the Center for
Budget and Policy Priorities also provided useful comments. (The assistance of external
reviewers implies no responsibility for the final product, which rests solely with CBO.)

Kate Kelly edited the manuscript, and Leah Mazade and Sherry Snyder proofread it.
Maureen Costantino took the cover photograph and designed the cover, and Jeanine Rees
prepared the study for publication. Jonathan Schwabish provided help with graphics.
Monte Ruffin produced the initial printed copies, Linda Schimmel coordinated the print
distribution, and Simone Thomas prepared the electronic version for CBO’s Web site
(www.cbo.gov).

Douglas W. Elmendorf
Director

July 201

0

CBO

www.cbo.gov

http://www.cbo.gov/doc.cfm?index=637

7

MaureenC
Doug Elmendorf

Contents

Summary

ix

Introduction 1

An Overview of Social Security 1

Social Security Projections

4

Assessing Options for Changing Social Security 7

Key Elements of Social Security

8

Scope of the Options 9

Effects of the Options on the System’s Finances 11

Effects of the Options on Payroll Taxes Paid and Benefits Received by
Various Groups 13

Effects of the Options on Work and Saving 1

5

Options That Would Change the Taxation of Earnings 1

6

Option 1: Increase the Payroll Tax Rate by 1 Percentage Point in 2012 17

Option 2: Increase the Payroll Tax Rate by 2 Percentage Points Over 20 Years 17

Option 3: Increase the Payroll Tax Rate by 3 Percentage Points Over 60 Years 17

Option 4: Eliminate the Taxable Maximum 18

Option 5: Raise the Taxable Maximum to Cover 90 Percent of Earnings 18

Option 6: Tax Covered Earnings Above the Taxable Maximum;
Do Not Increase Benefits 18

Option 7: Tax Covered Earnings Up to $250,000; Do Not Increase Benefits 19

Option 8: Tax All Earnings Above the Taxable Maximum at 4 Percent;
Do Not Increase Benefits 19

Option 9: Tax All Earnings Above $250,000 at 4 Percent;
Do Not Increase Benefits 19

Options That Would Change the Benefit Formula

20

Option 10: Raise from 35 to 38 the Years of Earnings Included in the AIME 20

Option 11: Index Earnings in the AIME Formula to Prices 20

Option 12: Reduce All PIA Factors by 15 Percent 21

Option 13: Reduce the Top Two PIA Factors by Roughly One Third 21

Option 14: Reduce the Top PIA Factor by One Third 2

2

Option 15: Reduce All PIA Factors by 0.5 Percent Annually 22

Option 16: Index Initial Benefits to Changes in Longevity 22

Option 17: Reduce PIA Factors to Index Initial Benefits to Prices Rather
Than Earnings 23

CBO

VI SOCIAL SECURITY POLICY OPTIONS

CBO

Option 18: Lower Initial Benefits for the Top 70 Percent of Earners 23

Option 19: Lower Initial Benefits for the Top 50 Percent of Earners 24

Option 20: Index the Bend Points in the PIA Formula to Prices 26

Option 21: Index Earnings in the AIME and Bend Points in the
PIA Formula to Prices 26

Option 22: Replace the Current PIA Formula with a New Two Part Formula 26

Options That Would Increase Benefits for Low Earners 28

Option 23: Modify the Special Minimum Benefit and Index It to
Growth in Earnings 28

Option 24: Introduce a New Poverty Related Minimum Benefit 29

Option 25: Enhance Low Earners’ Benefits on the Basis of Years Worked 29

Options That Would Raise the Full Retirement Age

30

Option 26: Raise the FRA to 68 31

Option 27: Raise the FRA to 70 31

Option 28: Index the FRA to Changes in Longevity 31

Options That Would Reduce Cost-of-Living Adjustments 31

Option 29: Reduce COLAs by 0.5 Percentage Points 32

Option 30: Base COLAs on the Chained CPI U 32

Appendix: Distributional Effects of Options with Similar Effects on the
System’s Finances 47

Glossary

51

Options That Would Change the Benefit Formula (Continued)

CONTENTS SOCIAL SECURITY POLICY OPTIONS VII

Tables

1.

Social Security’s

Revenues

and Outlays Under Current Law with
Scheduled Benefits 6

2.

Changes to Social Security’s Finances Under Various Options with
Scheduled Benefits 33

3.

Changes to Social Security’s Scheduled Benefits and Payroll Taxes for
Different Groups Under Various Options 39

4.

Changes to Social Security’s Payable Benefits and Payroll Taxes for
Different Groups Under Various Options 43

A 1.

Changes to Social Security’s Scheduled Benefits and Payroll Taxes for
Different Groups Under Various Options That Have Similar
Effects on the System’s Finances 49

Figures

S 1.

Effects of the Policy Options on the OASDI Trust Fund Actuarial Balance xi
1.
Distribution of Social Security Beneficiaries, by Type of
Benefits Received, 2010 2
2.
U.S. Population Age 65 or Older as a Percentage of the Population
Ages 20 to 64, 1962 to 2080 4
3.
Social Security’s Revenues and Outlays with Scheduled and Payable Benefits 7
4.
Projected Outlays for Social Security Under Current Law and with a
Gradual Reduction in Benefits Starting in 2012 8
5.
Effect of Delaying a Payroll Tax Increase or Benefit Reduction on
Social Security’s Finances 12

6.

Calculating the PIA in 2010 Under the Current Social Security System 21

7.

Calculating Initial Benefits with Progressive Price Indexing

25

8.

Calculating Initial Benefits with Indexing of Bend Points to Prices 27
CBO

Summary

Social Security, the federal government’s largest
single program, provides benefits to retired workers
(through Old Age and Survivors Insurance, OASI), to
people with disabilities (through Disability Insurance,
DI), and to their families as well as to some survivors of
deceased workers. Those benefits are financed primarily
by payroll taxes collected on people’s earnings. In 2010,
for the first time since the enactment of the Social Secu
rity Amendments of 1983, Social Security’s annual out
lays will exceed its annual tax revenues, the Congressional
Budget Office (CBO) projects. If the economy continues
to recover from the recent recession, those tax revenues
will again exceed outlays, but only for a few years. CBO
anticipates that starting in 2016, if current laws remain in
place, the program’s annual spending will regularly exceed
its tax revenues.

Social Security’s dedicated revenue stream sets it apart
from most other federal programs in that the dedicated
revenues are credited to trust funds that are used to
finance the program’s activities. Interest on the balances
of those funds also is credited to the funds (which often
are treated collectively as the OASDI trust funds). CBO
estimates that, unless changes are made to the system, the
trust funds combined will be exhausted in 2039. At that
point, the resources available to the Social Security
program will be insufficient to pay full benefits as they
are currently structured.1

This CBO study first provides an overview of Social
Security and discusses some criteria for evaluating
proposals to change the system. It then presents a variety

1. See Congressional Budget Office, The Long Term Budget Outlook
(June 2010), Chapter 3.

of options for changing the Social Security system and
analyzes the financial and distributional effects of those
options—that is, how they would affect Social Security’s
finances and how they would alter the benefits paid to
people in various earnings categories and people born in
various decades.

The Outlook for Social Security’s
Finances
As the population of the United States continues to
grow older, the number of Social Security beneficiaries
will continue to rise, and the program’s outlays will
increase faster than its revenues. Long term projections
are unavoidably uncertain but, under a broad range of
assumptions, benefits that are scheduled under current
law will consistently exceed revenues.

CBO projects that beginning in 2039 the Social Security
Administration will not be able to pay those scheduled
benefits, however. If revenues were not increased, benefits
would need to be cut by about 20 percent in 2040 to
equalize outlays and revenues. Those proportionately
lower payments, which would be made to all Social
Security recipients once the trust funds were exhausted,
are known as payable benefits.

A commonly used summary measure of the system’s long
term financial condition is the 75 year actuarial bal
ance—a figure that measures the long term difference
between the resources dedicated to Social Security and
the program’s costs under current law. The actuarial bal
ance is the value of Social Security’s revenues over the
75 year period, discounted to their value in current dol
lars, plus the current balance in the OASDI trust funds,

CBO

http://www.cbo.gov/doc.cfm?index=11559

X SOCIAL SECURITY POLICY OPTIONS

CBO

minus the present value of future Social Security outlays,
minus the value of a year’s worth of benefits as a reserve at
the end of the period.2 CBO estimates the 75 year actuar
ial balance to be 0.6 percent of gross domestic product
(GDP); that is, under current law, the resources dedicated
to financing the program over the next 75 years fall short
of the benefits that will be owed to beneficiaries by
about 0.6 percent of GDP.3 That figure is the amount
by which the Social Security payroll tax would have to
be raised or scheduled benefits reduced for the system’s
revenues to be sufficient to cover scheduled benefits. In
other words, to bring the program into actuarial balance
over the 75 years, payroll taxes would have to be
increased immediately by 0.6 percent of GDP and kept
at that higher rate, or scheduled benefits would have to
be reduced by an equivalent amount, or some com
bination of those changes and others would have to be
implemented.

The actuarial balance averages the smaller deficits that
would occur near the beginning of the projection period
and the larger ones that would occur near the end. In
2084, scheduled outlays would exceed revenues by
1.4 percent of GDP.

Policy Options
In this study, CBO analyzes 30 options that are among
those that have been considered by various analysts and
policymakers as possible components of proposals to pro
vide long term financial stability for Social Security. The
options follow the convention of not reducing initial ben
efits for people who are currently older than 55, and all
would directly affect outlays for benefits or federal
revenues dedicated to Social Security.

2. CBO discounts those values using a real (inflation adjusted)
discount rate of 3 percent, equal to CBO’s estimated long term
interest rate used to compute interest credited to the Social
Security trust funds. The actuarial balance is calculated on the
basis of Social Security’s scheduled benefits, which are the benefits
specified under current law without regard to the balances in the
system’s trust funds. Scheduled benefits are used in this study’s
analysis of the system’s finances because, by definition, the system
is in financial balance with payable benefits, which would be set so
as to eliminate any system deficit.

3. The projected actuarial balance can also be expressed as
1.6 percent of taxable payroll.

The options fall into five categories:

B Increases in the Social Security payroll tax,

B Reductions in people’s initial benefits,

B Increases in benefits for low earners,

B Increases in the full retirement age, and

B Reductions in the cost of living adjustments that are
applied to continuing benefits.

Each option is analyzed in isolation, although most pro
posals to make substantial changes to Social Security
combine several provisions. Many options would interact
with one another, so combining them might cause
changes to the overall finances of the system that are
larger or smaller than would be produced by a simple
sum of the effects of several discrete options.

This list of options is far from exhaustive. It does not
include changes that would draw on general government
revenues, create individual accounts, or change the trust
funds’ investments. Other than an increase in the Social
Security payroll tax, changes to federal tax policy are not
considered. The options do not include any that apply
only to people who receive DI benefits, although some of
the options would affect OASI and DI beneficiaries alike.

Effects of the Options
This study analyzes the overall effect of each option on
the finances of the Social Security system. Some options,
such as those that would apply the payroll tax rate to all
earnings or those that would index initial benefits to
prices, would more than eliminate Social Security’s actu
arial deficit; others would have far smaller financial effects
(see Summary Figure 1).

This study also analyzes the options’ effects on taxes that
would be paid and benefits that would be received by var
ious groups of program participants. For that distribu
tional analysis, participants are grouped by the amount of
their lifetime household earnings and by their birth
cohort (that is, by the decade in which they were born).
Those distributional effects of the options are measured
relative to the outcomes that would result both from
scheduled benefits and from payable benefits under
current law.

SUMMARY SOCIAL SECURITY POLICY OPTIONS XI

Summary

Figure 1.

Effects of the Policy Options on the OASDI Trust Fund Actuarial Balance

Source: Congressional Budget Office.

Notes: The actuarial balance is the present value of revenues plus the OASDI trust fund balance at the beginning of 2010, minus the present

value of outlays from 2010 to 2084, minus a year’s worth of benefits as a reserve at the end of the period, expressed as a percentage
of the present value of GDP over the period.
The AIME for a retired worker who reaches age 62 after 1990 is calculated on the highest 35 years of earnings on which that worker
paid Social Security taxes (up to the taxable maximum, $106,800 in 2010). Earnings before age 60 are indexed to compensate for
inflation and for real (inflation-adjusted) growth in wages; earnings after age 59 enter the computations at nominal values. Dividing
total earnings by 420 (35 years times 12 months) yields the AIME.
The PIA is the monthly payment to a worker who begins receiving retirement benefits at the full retirement age or to a disabled
worker who has never received a retirement benefit reduced for age. For workers who turn 62, become disabled, or die in 2010 (for
calculation of survivor benefits), the PIA formula is 90 percent of the first $761 of the AIME plus 32 percent of the AIME between
$761 and

$4,586

plus 15 percent of the AIME over $4,586. Those percentages constitute the PIA factors.
A COLA is an annual increase in benefits indexed to consumer prices. Under current law, the COLA equals the percentage increase in
the CPI-W; the chained CPI-U is an alternative measure of inflation.
OASDI = Old-Age, Survivors, and Disability Insurance; GDP = gross domestic product; AIME = average indexed monthly earnings;
PIA = primary insurance amount; FRA = full retirement age; COLA = cost-of-living adjustment; CPI-W = consumer price index
for all urban wage earners and clerical workers; chained CPI-U = chained CPI for all urban consumers; * = between -0.05 percentage
points and zero.

Base COLAs on the Chained CPI-U

Reduce COLAs by 0.5 Percentage Points

Index the FRA to Changes in Longevity

Raise the FRA to 70

Raise the FRA to 68

Enhance Low-Earners’ Benefits on the Basis of Years Worked

Introduce a New Poverty-Related Minimum Benefit

Modify the Special Minimum Benefit and Index It to Growth in

Earnings

Replace the Current PIA Formula with a New Two-Part Formula

Index Earnings in the AIME and Bend Points in the PIA Formula to Prices

Index the Bend Points in the PIA Formula to Prices

Lower Initial Benefits for the Top 50% of Earners

Lower Initial Benefits for the Top 70% of Earners

Reduce PIA Factors to Index Initial Benefits to Prices Rather Than Earnings

Index Initial Benefits to Changes in Longevity

Reduce All PIA Factors by 0.5% Annually

Reduce the Top PIA Factor by One-Third

Reduce the Top Two PIA Factors by Roughly One-Third

Reduce All PIA Factors by 15%

Index Earnings in the AIME Formula to Prices

Raise from 35 to 38 the Years of Earnings Included in the AIME

Tax All Earnings Above $250,000 at 4%; Do Not Increase Benefits

Tax All Earnings Above the Taxable Maximum at 4%; Do Not Increase Benefits

Tax Covered Earnings Up to $250,000; Do Not Increase Benefits

Tax Covered Earnings Above the Taxable Maximum; Do Not Increase Benefits

Raise the Taxable Maximum to Cover 90% of Earnings

Eliminate the Taxable Maximum

Increase the Payroll Tax Rate by 3 Percentage Points Over 60 Years

Increase the Payroll Tax Rate by 2 Percentage Points Over 20 Years

Increase the Payroll Tax Rate by 1 Percentage Point in 2012

0.3

0.6

0.5

0.6

0.2

0.9

0.5
0.3

0.1

0.1
0.2
0.5

0.7

0.1

0.4

0.2
0.5
0.4
0.5
0.6
0.2

-0.2

-0.3

0.1
0.3
0.2
0.3
0.2

1.0

*

Change the
Taxation of

Earnings

Change the
Benefit Formula

Increase Benefits
for Low Earners

Raise the Full
Retirement Age

Reduce Cost-of-Living
Adjustments

0.3
0.6
0.5
0.6
0.2
0.9
0.5
0.3
0.1
0.1
0.2
0.5
0.7
0.1
0.4
0.2
0.5
0.4
0.5
0.6
0.2
-0.2
-0.3
0.1
0.3
0.2
0.3
0.2
1.0
*
Change the
Taxation of
Earnings
Change the
Benefit Formula
Increase Benefits
for Low Earners
Raise the Full
Retirement Age
Reduce Cost-of-Living
Adjustments
CBO

XII SOCIAL SECURITY POLICY OPTIONS

CBO

Some options, such as an across the board increase in the
payroll tax rate or a flat reduction in benefits, would
affect all participants proportionately, but some options
would have disparate effects on people in different earn
ings groups. For example, some options would primarily
affect people with higher lifetime earnings by placing an
additional tax on earnings above a threshold or by
increasing the progressivity of the Social Security benefit
formula.

Many options with similar financial effects in the
aggregate would affect older and younger generations
differently. In particular, the timing of the changes would
affect their impact on different generations (as well as the

magnitude of the change necessary to bring the system
into balance). Some options, such as one that would
reduce benefits by a flat 15 percent, would take effect in a
single year and would affect all future beneficiaries the
same way. Others would be phased in and, initially,
would have only small effects. For example, a policy that
gradually reduced benefits would have a much larger
effect on people whose benefits began in 2040 than it
would on those whose benefits began in 2020. Raising
tax rates would increase the amounts paid by younger
people but make little difference in the sum of taxes paid
over a lifetime by people who already have left or are
about to leave the workforce.

Social Security Policy Options

Introduction
The federal government levies taxes on workers to pro
vide Social Security benefits to the elderly, to disabled
people, and to their families as well as to some survivors
of deceased workers. Although the program is part of the
overall federal budget, its funding differs from that of
many other programs in the budget: Its spending is
financed from two trust funds that are credited with the
dedicated tax revenues and from which benefits may be
paid, without further legislative action, as long as the
trust funds have sufficient balances.1 The balances that
exist today—more that $2 trillion—have accumulated
over many years, during which tax revenues credited to
the trust funds exceeded the benefit payments from
those funds. Interest on the balances is credited to the
trust funds.

In 2010, for the first time since the enactment of the
Social Security Amendments of 1983, the Congressional
Budget Office (CBO) projects that Social Security’s
annual outlays will exceed its annual revenues, excluding
interest credited to the trust funds. As the baby boom
generation (the group of people born between 1946 and
1964) continues to age, the number of Social Security
beneficiaries will increase, and outlays will rise faster than
revenues. In part because of this growth, the federal bud
get is on an unsustainable path: Without significant
changes in government policy, in coming decades the
aging of the population and rising health care costs will
boost federal outlays sharply relative to the size of the
economy under any plausible assumptions about future
trends in the economy, demographics, and health care
costs.2 Also because of the growing number of Social
Security beneficiaries, CBO projects that, under current
law, the Social Security trust funds will be exhausted in

1. Spending for Social Security benefits and receipts from Social
Security taxes are part of the unified federal budget but are
categorized as “off budget” for certain budget enforcement
procedures.

2039. Thereafter, the Social Security Administration
will not have the legal authority to pay the full benefits
specified in law.

This study analyzes the effects of 30 options for changing
Social Security. The options are among those commonly
proposed by policymakers and analysts for bringing long
term financial stability to the program.3 This study
describes the options’ effects on the finances of the Social
Security system, on the taxes the program’s participants
pay, and on the benefits participants receive. Participants
are grouped by their lifetime household earnings and
birth cohort (that is, by the decade of their birth).

An Overview of Social Security
The Social Security Act of 1935 created the federal gov
ernment’s largest single program. Currently, 53 million
people receive Social Security benefits, and, although
Social Security is commonly thought of as a retirement
program, only 69 percent of its beneficiaries are retired
workers, their spouses, and children (see Figure 1).
Another 12 percent of beneficiaries are survivors of
deceased workers, and the remaining 19 percent are peo
ple who are receiving Disability Insurance (DI) benefits

2. See Congressional Budget Office, The Long Term Budget Outlook
(June 2010).

3. CBO presented an analysis of various long term Social Security
options in Menu of Social Security Options (May 25, 2005).
Versions of some of the options presented in this study also were
discussed in Congressional Budget Office, Budget Options,
Volume 2 (August 2009). The Chief Actuary of the Social Security
Administration has published a list of policy options that would
address the solvency of the Social Security trust funds and other
issues related to Social Security benefits and financing. That
document, Individual Changes Modifying Social Security, is
available at www.ssa.gov/OACT/solvency/provisions/index.html.
The Social Security Administration’s Office of Retirement and
Disability Policy has published a series of policy briefs that analyze
the distributional effects of various options, available at
www.ssa.gov/policy/docs/policybriefs/index.html.

CBO

http://www.cbo.gov/ftpdocs/63xx/doc6377/Social_Security_Menu-CBO_baseline

http://www.cbo.gov/ftpdocs/102xx/doc10294/08-06-BudgetOptions

http://www.ssa.gov/OACT/solvency/provisions/index.html

http://www.ssa.gov/policy/docs/policybriefs/index.html

http://www.cbo.gov/doc.cfm?index=11559

2 SOCIAL SECURITY POLICY OPTIONS

CBO
Figure 1.

Distribution of Social Security
Beneficiaries, by Type of Benefits
Received, 20

10

Sources: Congressional Budget Office; Social Security
Administration, data for May 2010.

as disabled workers or who are the spouses and children
of disabled workers. Social Security is an important
source of income for the elderly. In 2008, almost 90 per
cent of people over age 65 received Social Security bene
fits. Among the population age 65 or older, those benefits
were the major source of income (providing at least
50 percent of total income) for 57 percent of families and
90 percent or more of income for almost a third of such
families.4 Consequently, if Social Security benefits were
reduced, many people would respond by working and
saving more. The responses would be greater if such
reductions were announced well in advance of the
changes.

Social Security now consists of two parts: Old Age and
Survivors Insurance (OASI), which pays benefits to
retired workers and their dependents and to survivors of
deceased workers; and Disability Insurance, which pays
benefits to workers who become disabled when they are

4. See Social Security Administration, “Income Sources,” in Income
of the Aged Chartbook, 2008, www.ssa.gov/policy/docs/
chartbooks/income_aged/2008/iac08.html#income.

Retired Workers

Disabled Workers

Spouses and
Children of

Disabled Workers
Survivors of
Deceased
Workers

Spouses and
Children of
Retired Workers

(5%)

(12%)

(15%)

(4%)

(64%)

younger than the full retirement age (FRA, the age at
which people can receive unreduced retirement benefits)
and to their dependents. Old Age and Survivors Insur
ance currently accounts for 82 percent of benefits, and
Disability Insurance accounts for 18 percent. CBO pro
jects that outlays for the program in fiscal year 2010 will
total $708 billion, roughly one fifth of the federal bud
get. Since 1989, administrative expenses have totaled
1 percent or less of program outlays.

During the program’s first four decades, spending for
Social Security benefits increased significantly relative to
the size of the economy, reaching about 4 percent of gross
domestic product (GDP) in the mid 1970s. The costs
spiked to nearly 5 percent of GDP in the early 1980s, the
period that saw the most recent major legislative changes
to the program. In the 1990s and early 2000s, spending
for Social Security benefits fluctuated between 4.1 per
cent and 4.6 percent of GDP. During the recent eco
nomic downturn, GDP contracted and Social Security
outlays increased more rapidly than they would have with
stable economic growth because the number of OASI and
DI claimants increased as the job market deteriorated.
Social Security’s outlays rose to 4.8 percent of GDP in
2009, and CBO projects they will remain at that level
in 2010.

Taxes. Social Security has two primary sources of dedi
cated tax revenues: payroll taxes and taxes on benefits.
Roughly 97 percent of dedicated tax revenues are col
lected from a payroll tax of 12.4 percent that is levied on
earnings and split evenly by workers and their employers
at 6.2 percent apiece. Self employed workers pay the
entire 12.4 percent tax on earnings themselves. The pay
roll tax applies only to taxable earnings—earnings up to a
maximum annual amount ($106,800 in 2010).

Taxable

earnings are about 83 percent of total covered earnings,
which are all earnings—from wages and from self
employment income—for employment covered by Social
Security. In addition, some Social Security benefits are
subject to taxation: In 2009, about 3 percent of Social
Security’s dedicated tax revenues came from the income
taxes that higher income beneficiaries paid on their Social
Security benefits.

Benefits. In general, workers are eligible to receive
Social Security retirement benefits if they are age 62 or
older and have paid a sufficient amount of Social Security

http://www.ssa.gov/policy/docs/chartbooks/income_aged/2008/iac08.html#income

SOCIAL SECURITY POLICY OPTIONS 3

taxes for at least 10 years.5 Retirement benefits are
reduced for workers who begin to collect Social Security
before reaching the full retirement age, currently 66.
Workers who are judged unable to perform substantial
work because of a physical or mental disability can
become eligible for DI benefits at an earlier age and, in
many cases, with a shorter employment history. Various
rules for determining eligibility and benefits apply to
family members of retired, disabled, or deceased workers.

When retired or disabled workers first claim Social
Security benefits, payments are based on their average
lifetime earnings. The formula used to translate average
earnings into benefits is progressive; that is, the replace
ment rate—the ratio of benefits received to a worker’s
past earnings—is higher for people with lower average
earnings than for people with higher earnings. The Social
Security Administration estimates that workers who had
average annual earnings throughout their careers and who
claim benefits in 2010 at age 65 will be eligible for an
annual benefit of about $16,500, an amount that will
replace about 40 percent of their average preretirement
earnings.

For the purpose of calculating average earnings to deter
mine the initial benefit, the amounts earned in earlier
years are converted to current year values based on
changes in average annual earnings in the economy as a
whole. Because average national earnings are projected to
grow faster than inflation, that indexation will cause aver
age initial benefits to grow in real (inflation adjusted)
terms and will keep the average replacement rate stable.
(In later decades, the replacement rate will be slightly
lower for workers with average earnings who claim bene
fits at age 65, mainly because of the scheduled increase in
the full retirement age.)

An adjustment is made to retirement benefits on the basis
of the age at which a recipient chooses to start claiming
benefits: The longer someone waits (up to age 70), the
higher the benefits will be. That adjustment is intended
to be actuarially fair, so that a person’s total lifetime bene
fits will have an approximately equal value regardless of
the age at which he or she begins collecting them. For all

5. Most workers need to earn 40 credits (each credit is known as
“a quarter of coverage”) to be eligible for retirement benefits.
Workers can earn up to four credits per year on the basis of the
amount they earn for employment that is covered under the
program. In 2010, one credit is earned for each $1,120 in wages,
so any worker who earns at least $4,480 will receive four credits
for the year.

types of benefits, a cost of living adjustment (COLA) is
made each year after the initial benefits are received to
keep pace with annual changes in consumer prices.6

Trust Funds. Revenues from payroll taxes and from taxes
on benefits, along with intragovernmental interest pay
ments, are credited to the two Social Security trust
funds—one for OASI and one for DI. The program’s
benefits and administrative costs are paid from those
funds. Legally, the two funds are separate, but they often
are described collectively as the OASDI trust funds.

Federal trust funds, including those for Social Security,
essentially constitute an accounting mechanism. In a
given year, the sum of receipts to a fund along with the
interest that might be credited on previous balances, less
spending for benefits and administrative costs, constitutes
a fund’s surplus or deficit. The cash generated by a sur
plus in any year is turned over to the Treasury in
exchange for special Treasury securities. The Treasury
uses the cash to finance the government’s ongoing activi
ties. If the trust funds’ cash receipts are less than their
outlays, the Treasury securities they hold are redeemed for
cash as needed. The Treasury obtains that cash from
other revenues or by borrowing from the public.

The trust funds are part of the federal government, so
transactions between Social Security and the Treasury are
intragovernmental and have no net effect on federal bor
rowing from the public or on the unified budget. Any
increase in revenues credited to the trust funds or
decrease in outlays from the funds makes available addi
tional cash that can be used to finance other government
activities without requiring new government borrowing
from the public; the trust fund surpluses that were gener
ated in previous years have been used in that way. Simi
larly, any increase in outlays or decrease in revenues for
the OASDI trust funds in some future year will represent
a draw on the government’s cash in that year. Thus, the
balances in the OASDI trust funds (in the form of
government securities) are an asset to the Social Security

6. Social Security benefits are indexed to inflation as measured by the
consumer price index for urban wage earners and clerical workers
(CPI W). The Social Security Administration generally adjusts
benefits paid in January on the basis of the change in the CPI W
through the third quarter of the previous calendar year. If the
resulting adjustment is negative, no COLA is given. The next
COLA is made when the CPI W for the third quarter of the
calendar year exceeds the CPI W for the third quarter of the last
year in which an adjustment occurred.

CBO

4 SOCIAL SECURITY POLICY OPTIONS

CBO

Figure 2.

U.S. Population Age 65 or Older as a Percentage of the Population Ages 20 to 64,
1962 to

2080

(Percent)

Sources: Congressional Budget Office; Social Security Administration.

1962 1970 1978 1986 1994 2002 2010 2018 2026 2034 2042 2050 2058 2066 2074

0
5
10

15

20
25
30

35

40

45

Actual Projected

system but a liability to the rest of the government. The
resources to redeem government securities in the OASDI
trust funds and thereby pay for Social Security benefits in
some future year must be generated from taxes, other
government income, or government borrowing in that
year.

Social Security Projections
Under current law, the cost of Social Security benefits will
escalate in coming decades.7 Economic growth leads to
higher average benefits because benefits are based on past
earnings. In addition, changes in the nation’s demo
graphic structure will cause total benefits to grow faster
than the economy: As the baby boom generation reaches
retirement age, and as decreasing mortality leads to

7. This study uses projections for Social Security as published in
Congressional Budget Office, The Long Term Budget Outlook,
Chapter 3, based on An Analysis of the President’s Budgetary
Proposals for Fiscal Year 2011 (March 2010). Future revenues from
income taxes on benefits will depend on future income tax rates.
The projections used here reflect the assumptions underlying the
extended baseline scenario published in The Long Term Budget
Outlook, namely, that income tax law does not change and income
taxes on benefits grow as a share of Social Security benefits
throughout the 75 year projection period. Under that report’s
alternative fiscal scenario, in contrast, income taxes on benefits are
assumed to remain a constant share of benefits after 2020. As a
result, projected Social Security revenues are slightly lower under
the alternative fiscal scenario.

longer lives and longer retirements, a larger share of the
population will be drawing Social Security benefits.8

Rising Cost of Benefits. Between now and 2035, the
number of people age 65 or older will increase by about
90 percent, compared with an increase of more than
10 percent in the number of people between the ages of
20 and 64, CBO projects. Today, that older population is
one fifth the size of the younger population; at those
growth rates, it will be more than one third the size of the
younger group by 2035 (see Figure 2). In 2035, about
93 million people will collect Social Security benefits,
compared with 53 million today, and the average benefit
will have grown nearly as rapidly as GDP per person.

As more baby boomers begin collecting benefits, spend
ing for the program will climb from 4.8 percent of GDP
in 2010 to 6.2 percent of GDP in 2035, CBO projects.
Spending as a share of GDP will decline slightly over the
15 years after that, to 5.9 percent of GDP, as an
increasing number of baby boomers die. However,

8. Expectations regarding how the baby boomers will fare financially
in retirement are summarized in Congressional Budget Office,
The Retirement Prospects of the Baby Boomers, Issue Brief (March
18, 2004); for additional details, see Baby Boomers’ Retirement
Prospects: An Overview (November 2003) and Will the Demand for
Assets Fall When the Baby Boomers Retire? Background Paper
(September 2009).

http://www.cbo.gov/doc.cfm?index=11559

http://www.cbo.gov/ftpdocs/112xx/doc11280/03-24-apb

http://www.cbo.gov/ftpdocs/51xx/doc5195/03-18-BabyBoomers

http://www.cbo.gov/ftpdocs/48xx/doc4863/11-26-BabyBoomers

http://www.cbo.gov/ftpdocs/105xx/doc10526/09-08_Baby-Boomers

SOCIAL SECURITY POLICY OPTIONS 5

demographers generally anticipate that life expectancy
will continue to increase, and CBO projects that Social
Security outlays will resume their upward trajectory rela
tive to GDP after 2050, reaching 6.3 percent in 2080.9

The aging of the population is primarily responsible for
the growth in Social Security’s outlays as a percentage of
GDP. If the age distribution of the population remained
constant, Social Security outlays would decline slightly,
from 4.8 percent of GDP today to 4.3 percent of GDP in
2035 and remain approximately at that level thereafter;
the decline would occur because the full retirement age
will continue to rise under current law, effectively reduc
ing benefits. Social Security’s total benefits would remain
a generally constant share of GDP in the absence of aging
because scheduled benefits are indexed to growth in earn
ings and (after initial benefits are received) to inflation. If
that indexation were changed in certain ways, average
benefits could be significantly lower than those under
current law. For example, if initial benefits grew at the
same rate as average prices (rather than at the same rate as
average wages as they do under current law), average ben
efits—and thus total outlays—would be one third lower
by 2060 and one half lower by 2080. Alternatively, if
continuing benefits were increased by a smaller COLA
than provided under current law, average benefits would
be smaller. CBO estimates that, in years after 2040, more
than 25 percent of the benefit payments under current
law will be the result of COLAs provided between now
and then.

Worsening System Finances. CBO projects that, in 2010,
for the first time since the Social Security reforms of the
early 1980s, benefit payments from the trust funds will
exceed trust fund receipts from the public. Receipts from
the public consist mostly of revenues from payroll taxes
and exclude interest on Treasury securities held by the
trust funds. As the economy recovers from the recent
recession, receipts will again exceed benefit payments, but
only until 2016. If benefits are paid as specified under
current law, outlays will exceed revenues by 0.3 percent of

9. For details on CBO’s methodology for projecting Social Security’s
revenues and outlays, see Congressional Budget Office,
CBO’s Long Term Projections for Social Security: 2009 Update
(August 2009), p. 4.

GDP in 2020, CBO projects, and by 1.0 percent to
1.3 percent between 2040 and 2080 (see Table 1).

Trust Fund Exhaustion. CBO projects that the Disability
Insurance Trust Fund will be exhausted in fiscal year
2018, with the sum of the balance in the fund at the
beginning of the year and projected revenue in that year
falling $15 billion below projected expenditures. Once
the trust fund balance has fallen to zero and current reve
nues are insufficient to cover the benefits that are speci
fied in current law, the DI program will be unable to
meet its obligations fully without changes in law. CBO
projects that the Old Age and Survivors Insurance Trust
Fund will be exhausted in 2042.

The DI trust fund has been close to exhaustion before.
The 1994 Annual Report of the Social Security Board of
Trustees projected that the DI trust fund would be
exhausted in 1995.10 That outcome was prevented by leg
islation that redirected revenue from the OASI trust fund
to the DI trust fund. In part because of that experience, it
is a common analytical convention to consider the DI
and OASI trust funds as combined, and CBO projects
that, if legislation to shift resources from the OASI trust
fund to the DI trust fund was enacted, the combined
OASDI trust funds would be exhausted in 2039.

Scheduled and Payable Benefits. Benefits as calculated
under the Social Security Act, regardless of the balances in
the trust funds, are known as scheduled benefits.11 How
ever, the Social Security Administration lacks authority
to pay scheduled benefits if those payments would
exceed the available balances. If the trust funds became
exhausted, payments to beneficiaries would be reduced or
otherwise modified as necessary to make outlays from the

10. See Board of Trustees of the Federal Old Age and Survivors
Insurance and Disability Insurance Trust Funds, 1994 Annual
Report, 78 187 (April 11, 1994), www.ssa.gov/history/reports/
trust/1994/1994 .

11. CBO prepares cost estimates for legislation under the assumption
that scheduled payments will be made, which is consistent with a
long standing statutory requirement that CBO, in its baseline
projections, assume that laws are implemented as specified and
that funding for entitlement programs is adequate to make all
payments. See section 257 of the Balanced Budget and Emergency
Deficit Control Act of 1985, Public Law 99 177, as amended;
2 U.S.C. 907.

CBO

http://www.cbo.gov/ftpdocs/104xx/doc10457/08-07-SocialSecurity_Update

http://www.ssa.gov/history/reports/trust/1994/1994

6 SOCIAL SECURITY POLICY OPTIONS

CBO

Table 1.

Social Security’s Revenues and Outlays Under Current Law with
Scheduled Benefits
(Percentage of GDP)

Source: Congressional Budget Office.

Notes: The 75-year period is 2010 through 2084. Revenues consist of payroll taxes and income taxes on benefits (but not interest credited to
the trust funds) in the specified year. Outlays consist of Social Security benefits and administrative costs. The balance is the surplus or
deficit, which is the difference between revenues and outlays. The 75-year present value of revenues includes the current Old-Age,
Survivors, and Disability Insurance trust fund balance. The 75-year present value of outlays includes a year’s worth of benefits as a
reserve at the end of the period.

GDP = gross domestic product.

Revenues 4.9 4.9 4.9 4.9 5.0 5.2 14.4
Outlays 4.8 5.2 6.2 6.0 6.3 5.8 16.0
Balance 0.1 -0.3 -1.3 -1.1 -1.3 -0.6 -1.6

75-Year Present Value

Taxable
2060

Payroll

as a

Percentage of

20802020
Actual
2009 GDP2040

funds equal revenues to the funds.12 Benefits reduced or
modified in that way are known as payable benefits. Under
those circumstances, all receipts to the trust funds would
be used and the trust fund balance would remain essen
tially at zero.

In 2040, CBO projects, payroll tax revenues and revenues
from the taxation of benefits will be 80 percent of sched
uled benefits for OASI and DI. At that time, payable
benefits will be 20 percent lower than scheduled benefits
for all beneficiaries. By 2084, the gap between scheduled
and payable benefits will be 24 percent, CBO estimates
(see Figure 3).

Cutting scheduled benefits or raising Social Security taxes
during the next few decades would extend the solvency of
the OASDI trust funds. For example, under a policy
option with a gradual reduction in benefits such as a
reduction in cost of living adjustments by 0.5 percent
annually starting in 2012, the trust fund exhaustion date
would be extended by nine years to 2048—such that

12. See Kathleen Romig, Social Security: What Would Happen If the
Trust Funds Ran Out? Report for Congress RL33514
(Congressional Research Service, Updated April 25, 2008). The
report notes the entitlement created under the Social Security Act,
cites other law that prohibits officials from making expenditures
in excess of available funds, and acknowledges that the two create
a potential conflict that must be resolved by the Congress or in the
courts.

benefits from 2039 through 2048 would be higher than
payable benefits under current law (see Figure 4).

Actuarial Balance. A commonly used measure of the sus
tainability of a program that has a trust fund and a dedi
cated revenue source is its actuarial balance, a single
measure of the difference between the trust fund’s
resources and projected expenditures over a specified
period. The actuarial balance is calculated as the present
value of projected revenues, plus the trust fund balance at
the beginning of the period, minus the present value of
projected outlays and the cost of maintaining a reserve
equal to a year’s worth of benefits at the end of the
period, expressed as a percentage of the present value of
GDP or (in the case of Social Security) as a percentage
of taxable payroll over the same period. CBO estimates
that over the 75 year period from 2010 to 2084, dedi
cated revenues for Social Security will fall short of sched
uled benefits by 0.6 percent of GDP, assuming a real
interest rate of 3 percent (see Table 1). (As a percentage of
taxable payroll—the amount of earnings subject to the
Social Security payroll tax—the shortfall is 1.6 percent.)

Thus, to bring the program into actuarial balance over
the 75 year period—that is, for the system’s projected rev
enues to be sufficient to cover the benefits prescribed by
current law—payroll taxes could be increased immedi
ately by 0.6 percent of GDP and kept at that higher rate,
or scheduled benefits could be reduced by an equivalent
amount, to give two examples. (Such an increase in

SOCIAL SECURITY POLICY OPTIONS 7

Figure 3.

Social Security’s Revenues and Outlays with Scheduled and Payable Benefits
(Percentage of gross domestic product)

Source: Congressional Budget Office.

Note: Revenues consist of payroll taxes and income taxes on benefits (but not interest credited to the trust funds). Outlays consist of Social
Security benefits and administrative costs. Benefits as calculated under the Social Security Act, regardless of the balances in the trust
funds, are known as scheduled benefits. If the trust funds became exhausted, payments to beneficiaries would be reduced to make
outlays from the funds equal revenues to the funds; such benefits are known as payable benefits. In that case, total revenues would
decline slightly because revenues from taxation of benefits would decline.

1985 1990 1995 2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050 2055 2060 2065 2070 2075 2080
0
4

5
6
7

Outlays with Payable Benefits

Actual Projected

Outlays with Scheduled Benefits

Outlays

Revenues

payroll taxes would generate about 13 percent more reve
nue than the amount projected under current law.) More
generally, a combination of tax and benefit changes that
in combination improved the 75 year actuarial balance
by 0.6 percent of GDP could be implemented over time.

Trust Fund Ratio. Another common measure of Social
Security’s finances is the ratio of the trust fund balance to
the program’s annual outlays. That calculation indicates
how many years’ worth of benefits could be financed by a
given balance if outlays per year remained the same. The
trust fund ratio for 2010—the balance in the Social
Security trust funds at the beginning of the year divided
by projected 2010 outlays for the program—was 3.6,
CBO estimates. The ratio is projected to peak in 2010
and then to decline quickly until the trust funds are
exhausted in 2039.

Sustainable Solvency. Some analysts suggest that changes
to Social Security should have two financial objectives:
balancing the system’s finances over the 75 year projec
tion period and putting the system on a sustainable path
thereafter, a goal known as sustainable solvency.13 The
actuarial balance, as a single number, usefully summarizes
the entire stream of revenues and outlays over the 75 year
period (after adjusting for the starting balance in the trust

funds), but it does not convey any information about
whether the pattern of annual finances is sustainable
beyond 75 years. A proposal that would attain sustainable
solvency would produce positive trust fund ratios
throughout the 75 year projection period as well as stable
or rising ratios at the end of the period.

Assessing Options for Changing
Social Security
In this study, CBO examines 30 options for altering vari
ous elements of Social Security that have been considered
by various analysts and policymakers. The options mostly
involve changes to the system’s current structure that
would have a marked influence on Social Security’s
finances. Several criteria can be applied to analyze the var
ious options’ effects. This study considers how the
options would affect Social Security’s finances over time
and discusses distributional outcomes, such as the
amount of taxes collected from or the amount of benefits

13. See, for example, Social Security Administration, Report of the
1994–1996 Advisory Council on Social Security, Volume I: Findings
and Recommendations (January 1997), www.ssa.gov/history/
reports/adcouncil/report/toc.htm.

CBO

http://www.ssa.gov/history/reports/adcouncil/report/toc.htm

8 SOCIAL SECURITY POLICY OPTIONS

CBO

Figure 4.

Projected Outlays for Social Security Under Current Law and with a
Gradual Reduction in Benefits Starting in 2012
(Percentage of gross domestic product)

Source: Congressional Budget Office.

Note: Outlays consist of Social Security benefits and administrative costs. Benefits as calculated under the Social Security Act, regardless of
the balances in the trust funds, are known as scheduled benefits. If the trust funds became exhausted, payments to beneficiaries
would be reduced to make outlays from the funds equal revenues to the funds; such benefits are known as payable benefits. The
gradual reduction in benefits would begin in 2012, reducing cost-of-living adjustments by 0.5 percent annually.

1985 1990 1995 2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050 2055 2060 2065 2070 2075 2080
0
4
5
6

7
Actual Projected

With Payable Benefits Assuming
Gradual Reduction in Benefits

With Scheduled Benefits
Under

Current Law

With Payable Benefits
Under Current Law

paid to people in different groups according to lifetime
earnings or year of birth. Changes in economic incentives
also are important; they are discussed below, although
they are not analyzed in detail.

Key Elements of Social Security
Each policy option would alter at least one significant
element of Social Security that determines its revenues or
outlays under current law. Descriptions of the key
elements appear below.

Payroll Taxes. Several options would affect the amount of
Social Security payroll taxes people would pay. The cur
rent system operates as follows:

B Tax Rate. Employers and employees each pay half of
the 12.4 percent payroll tax (self employed people pay
the entire amount).

B Taxable Maximum. The payroll tax is imposed on
earnings up to a maximum that increases as average
earnings rise. In 2010, that taxable maximum is
$106,800.

Benefit Formula. Social Security benefits are determined
by a formula that constructs summary measures of life
time earnings. That formula has several key elements:

B Average Indexed Monthly Earnings. Social Security
benefits are determined by earnings over a person’s
lifetime, expressed as average indexed monthly earn
ings (AIME). For anyone who reaches age 62 after
1990, the total earnings amount is calculated based on
earnings that are subject to Social Security taxes, using
the highest 35 years of those earnings. For retirees,
earnings before age 60 are indexed to compensate both
for inflation and for economywide real growth in
earnings; earnings at age 60 and later enter the com
putations at their actual amounts. For disabled work
ers, earnings in the two years before initial benefit
computation enter at their actual amounts, and earlier
earnings are indexed. For retirees, dividing the result
ing value for total earnings by 420 (35 years multi
plied by 12 months) yields the AIME.

B Primary Insurance Amount. The primary insurance
amount (PIA) is the monthly amount payable to a
worker who begins to receive Social Security retire
ment benefits at the age at which he or she is eligible

SOCIAL SECURITY POLICY OPTIONS 9

for full benefits or the amount payable to a disabled
worker who has never received a retirement benefit.
For workers who turn 62 or become disabled in 2010,
for all of their dependents, and for dependents of
workers who die in 2010, the PIA is calculated as
90 percent of the first $761 of the AIME, plus 32 per
cent of the AIME between $761 and $4,586, plus
15 percent of the AIME above $4,586. Actual
monthly benefits paid to retired workers and their
dependents differ from the PIA if an individual claims
retirement benefits prior to or later than his or her full
retirement age.

B PIA Factors. The rates by which the components of the
AIME are multiplied—90 percent, 32 percent, and
15 percent—are the PIA factors. The PIA formula is
progressive; it replaces a larger share of preretirement
earnings for people with lower average earnings than it
does for people with higher earnings.

B Bend Point. The threshold at which a PIA factor
changes is called a bend point. Under current law,
there are two: In 2010, they are $761 and $4,586. The
bend points change every year with changes in the
average annual earnings for the workforce as a whole.
Consequently, bend points occur at approximately the
same place in the distribution of the AIMEs each year
(the 11th and 71st percentiles, respectively, of AIMEs
for people who are age 62 and eligible for OASI bene
fits in 2010), and average initial benefits rise at a pace
that matches the increase in average earnings over
time.

Special Minimum Benefits for Low Earners. Beneficiaries
receive the larger of the standard benefit or a special min
imum benefit. For people who had very low earnings for
more than 10 years, the special minimum benefit is
sometimes larger than the standard benefit. The special
minimum benefit increases over time, keeping pace with
prices. Because the standard benefit formula increases
with earnings, which tend to grow faster than prices, the
special minimum benefit affects fewer people every year;
it is projected to have no effect on beneficiaries who
become eligible to collect benefits after 2010.

Full Retirement Age. Social Security’s full retirement age,
also called the “normal retirement age,” is the age at
which someone is eligible to receive full retirement bene
fits. For workers born before 1938, the FRA is 65. Under
current law, the FRA is increasing gradually; it will be

67 for people born in 1960 or later. The age at which
workers may start receiving reduced benefits—age 62—
remains the same. For each year that a worker claims ben
efits before reaching the FRA, benefits are reduced by an
amount that ranges from 5 percent to 6 2/3 percent. For
most current new beneficiaries, benefits are increased by
8 percent for each year after the FRA that initial receipt
of benefits is delayed, until age 70. (The increase is less
than 8 percent for people born before 1943.)14

Cost-of-Living Adjustments. At the end of each year, the
Social Security Administration adjusts each beneficiary’s
PIA by an amount that is equal to any increase in the
consumer price index for urban wage earners and clerical
workers (CPI W). (When prices decline, the COLA is set
at zero, as occurred in 2010.)

Scope of the Options
This study focuses on options that would directly affect
outlays for Social Security benefits or federal revenues ded
icated to Social Security. Most would increase the trust
fund balances, but a few would reduce balances because
they would raise benefits for people with low lifetime earn
ings. Each option is considered in isolation even though
any substantial proposal to change the Social Security pro
gram probably would include several provisions.

Options that would reduce initial benefits are assumed to
take effect in 2017; other options are assumed to take
effect in 2012. Although some would affect all beneficia
ries, including those who receive disability insurance,
CBO did not examine any options that are specific to DI.
Options that are outside the current system’s structure
and those that would not have sizable effects on the sys
tem’s finances are excluded from the study.

Timing of Implementation. The options that CBO has
analyzed are not detailed legislative proposals and are
generally presented in a simplified form. Several, for
example, would involve abrupt reductions in benefits, so
workers born a year apart would receive substantially dif
ferent amounts. In practice, policymakers might choose
to introduce major changes gradually, as they have done
in the past. Introducing changes incrementally ensures
that people of similar ages and circumstances will be sub
ject to similar tax and benefit rules. (For all the options,

14. See Social Security Administration, “Effect of Early or Delayed
Retirement on Retirement Benefits” (November 12, 2009),
www.ssa.gov/OACT/ProgData/ar_drc.html.

CBO

http://www.ssa.gov/OACT/ProgData/ar_drc.html

10 SOCIAL SECURITY POLICY OPTIONS

CBO

as under current law, benefit rules that are applicable to
an individual are those in force in the year in which that
person becomes entitled to benefits, not the year he or
she chooses to begin receiving benefits.)

Proposals to change Social Security commonly would not
reduce benefits for people older than 55 in the year a
reform proposal is considered. Therefore, options in this
study that would reduce initial benefits would not affect
anyone older than 55 in 2010. Because retired workers
become entitled to benefits at age 62, the changes in
initial benefits would generally apply to people in birth
cohorts that will become entitled to benefits in 2017
or later.

Options That Would Affect Recipients of Disability
Insurance. Some of the options would affect all Social
Security beneficiaries, including people who receive DI
benefits (see, for example, Options 12, 13, 14, and 15).
If policymakers wanted to offset some of the effects on
DI benefits of an option that changed retirement and dis
ability benefits alike, they could add an offsetting policy
change that increased DI benefits.

Disability Insurance accounts for one sixth of Social
Security benefits, and DI expenditures have increased
rapidly over the past 20 years.15 Consequently, options
specific to Disability Insurance could have substantial
consequences for the system’s finances by reducing or
increasing DI outlays. However, this study does not
include options (such as those that would affect eligibility
for benefits, adjust initial DI benefits, or apply different
cost of living adjustments to DI benefits) that would
affect DI beneficiaries only.

Options Not Encompassed by This Study. This study
focuses on changes to the current structure of Social
Security that could have sizable effects on the program’s
finances. Thus, two main categories of options are
excluded: options outside the context of the existing pro
gram and options within the program’s existing structure
that would not have sizable effects on the system’s
finances.

15. The growth in spending for Disability Insurance has been driven
largely by an increase in the number of people receiving benefits.
In 1980, 4.7 million people received DI benefits; by 2009, there
were 9.7 million beneficiaries. See Congressional Budget Office,
Social Security Disability Insurance: Participation Trends and Their
Fiscal Implications, Issue Brief (forthcoming).

The creation of individual accounts is a frequently dis
cussed possibility that would make changes outside the
existing Social Security program. (The resources available
to an account holder at retirement would depend on how
much had been paid into such an account, most likely
through a payroll tax, and the rate of return on the
account’s assets during the account holder’s working life.)
Most proposals that would introduce individual accounts
also would involve changes to the current program, and
the interactions between the accounts and the altered
Social Security program are generally critical to such
proposals. Although in the past CBO has analyzed com
prehensive proposals that would combine the establish
ment of individual accounts with changes to various
elements of the Social Security system, such analyses are
beyond this study’s scope.16

This study also includes no options that would draw on
general government revenues for Social Security or that
would change the form of investment for the trust funds.
In addition, because this study does not consider revenue
options that would directly affect taxes other than the
Social Security payroll tax, it does not examine what
might happen if income taxes on Social Security benefits
were increased. Such changes could have sizable effects on
the system’s finances.

Many other changes could achieve various policy goals
for Social Security, although they would not produce a
substantial change in the system’s long run finances.17
Several possibilities that have received attention elsewhere
are not included in this study:

B Changes could be aimed at reducing poverty by
increasing benefits as Social Security beneficiaries get
older. Poverty tends to be greater among the very old
than among younger retirees, so directing benefit
increases to the oldest beneficiaries could reduce
poverty by more than if the same total amount of
additional benefits were distributed more broadly.
For example, benefits could be increased by fixed
amounts or by percentages tied to beneficiaries’ ages.
Alternatively, the COLA could rise as people age so
that the oldest beneficiaries would receive larger
increases.

16. For example, see Congressional Budget Office, letter to the
Honorable Paul Ryan about an analysis of the Roadmap for
America’s Future Act of 2010 (January 27, 2010).

17. Some options of this type that CBO has presented elsewhere
are not analyzed here. See Congressional Budget Office,
Budget Options, Volume 2, pp. 141–155.

http://www.cbo.gov/doc.cfm?index=10851

http://www.cbo.gov/ftpdocs/102xx/doc10294/08-06-BudgetOptions

SOCIAL SECURITY POLICY OPTIONS 11

B The treatment of spousal benefits could be changed to
reduce the disparity between the benefits due to dual
and single income couples with the same earnings. For
example, spousal benefits for couples with similar
earnings could be increased, decreased, or adjusted
after the death of a spouse. (Under current law, an eli
gible spouse of a retired or disabled worker is entitled
to a spousal benefit that is equal to 50 percent of the
worker’s benefit, but only if it is higher than the
spouse’s own earned benefit. Therefore, benefits gen
erally replace a larger portion of lifetime earnings for
couples with one earner than for couples with two
earners.)

B Benefit increases could be targeted toward parents
who had low earnings during years when they were
caring for children.

B The earnings of those state and local government
workers who now are exempt from the Social Security
payroll tax could be taxed, and coverage could be
made mandatory for all public sector employees.

Effects of the Options on the System’s Finances
As a summary measure of the consequences of each
option, CBO estimated the change in the actuarial bal
ance as a percentage of GDP. That change is estimated
using scheduled benefits because, by definition, the sys
tem is in financial balance with payable benefits, which
would be automatically reduced to eliminate any short
fall. The calculations are based on the projections
described in CBO’s 2010 The Long Term Budget Outlook,
and the effects of various options were analyzed using
CBO’s long term Social Security model.18 Analysis of the
effects of uncertainty in the projected values on the
results for each option is outside the scope of this study.19

Some options would, by themselves, eliminate most or all
of the actuarial imbalance of 0.6 percent of GDP (see

18. The results presented in this study are based on a single simulation
for each option, reflecting CBO’s long term demographic and
economic assumptions. For a description of the model, see
Congressional Budget Office, CBO’s Long Term Model: An
Overview, Background Paper (June 2009).

19. For more information, see Congressional Budget Office,
Quantifying Uncertainty in the Analysis of Long Term Social Security
Projections, Background Paper (November 2005).

Table 2 on page 33). For example, Option 2 would
increase the payroll tax by 2 percentage points over the
next two decades, and Option 12 would cut benefits for
all new recipients by 15 percent. Most individual options
could be altered to have a smaller or larger effect on the
actuarial balance by affecting the same key elements of
the system, but with different tax rates, benefit calcula
tion rates, or speed at which the policy would phase in.
Depending on the timing of the changes encompassed by
the option, eliminating the actuarial imbalance might or
might not avert exhaustion of the trust fund balance.

Most of the options presented would not eliminate the
75 year imbalance on their own; to achieve that goal, it
would be necessary to combine several. For simplicity,
however, CBO evaluated each policy in isolation. Com
bining several options might introduce changes to the
overall finances of the system that were larger or smaller
than would be produced by summing the effects of those
options, because they would interact with one another.

Some options would have their full effect immediately,
and they would change annual revenues or outlays by
roughly the same percentage each year (such as Options 6
to 9, which would increase payroll tax rates in 2012).
Others would phase in slowly, resulting in increasingly
larger changes in annual revenues or outlays (such as
Option 3, which would increase taxes gradually, or
Options 15 to 21, which would reduce benefits gradu
ally). Although most options would improve the system’s
finances, three (Options 23 to 25) would focus on
increasing benefits for people with low lifetime earnings.
Those options would increase scheduled outlays and,
taken alone, would worsen the system’s finances.

Effects of Delayed Implementation. The effects of an
option would be sensitive to the date of its implementa
tion. In particular, changing the year in which scheduled
benefits are reduced could interact with changes in demo
graphics that will occur over the next 30 years. Delaying
the start of such a reduction could help people to make
informed decisions about preparing for retirement
because they would have earlier warning about changes in
Social Security’s rules. With every year that goes by, how
ever, larger changes would be needed to create a balance
over the next 75 years between scheduled revenues and
scheduled benefits. To demonstrate the effect of delaying

CBO

http://www.cbo.gov/doc.cfm?index=11559

http://www.cbo.gov/ftpdocs/103xx/doc10328/06-26-CBOLT

http://www.cbo.gov/ftpdocs/68xx/doc6873/11-16-MonteCarlo

12 SOCIAL SECURITY POLICY OPTIONS

CBO

Figure 5.

Effect of Delaying a Payroll Tax Increase or Benefit Reduction on
Social Security’s Finances
(Trust fund ratio)

Source: Congressional Budget Office.

Notes: The trust fund ratio is the ratio of the trust fund balance (the amount in the OASDI trust funds at the beginning of a year) to one year’s
outlays (Social Security benefits and administrative costs). The trust funds are exhausted when the trust fund ratio reaches zero.
Under current law, the trust funds cannot incur negative balances. A negative balance is a projected shortfall, reflecting the trust
funds’ inability to pay scheduled benefits out of current-law revenues.

OASDI = Old-Age, Survivors, and Disability Insurance; PIA = primary insurance amount.

Option 12: Reduce All PIA Factors by 15 Percent

Option 2: Increase the Payroll Tax Rate by 2 Percentage Points Over 20 Years

1985 1990 1995 2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050 2055 2060 2065 2070 2075 2080
-10

-8

-6

-4

-2

0
2
4
6

8
Actual Projected

Option Starts in 2022

Option Starts in 2012

Current Law
1985 1990 1995 2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050 2055 2060 2065 2070 2075 2080
-10
-8
-6
-4
-2
0
2
4
6
8

Current Law
Option Starts in 2027

Option Starts in 2017

Actual Projected

the implementation of Social Security reforms, CBO
analyzed the effects of varying the start dates for two
options.

In the first case, CBO estimated the effect of boosting the
payroll tax rate by 2.0 percentage points gradually over
two decades (Option 2) but starting in 2022 instead of
2012. If the increase started in 2012, the Social Security
trust funds would not be exhausted until 2083 (see the
top panel of Figure 5). However, if the onset of payroll
tax increases were delayed by 10 years, until 2022, the

result would be quite different: The Social Security
trust funds would be exhausted by 2056. If policymakers
wanted to delay implementation by 10 years and still
achieve the same improvement in the 75 year actuarial
balance, the increase in the tax rate would need to be
more than a quarter larger: The tax rate would have to
increase by 2.6 percentage points rather than by
2.0 percentage points.

In the second instance, CBO evaluated the effect of a flat
15 percent cut in benefits for new beneficiaries

SOCIAL SECURITY POLICY OPTIONS 13

(Option 12) but implemented in 2027 rather than in
2017. Beginning this cut in 2017 would be nearly
sufficient to restore Social Security to solvency over 75
years, and the trust funds would be exhausted in 2076
(see the bottom panel of Figure 5). With a 10 year delay,
however, the trust fund exhaustion date would be 2044,
only five years later than CBO projects under current law.
If policymakers wanted to implement a benefit reduction
in 2027 and still achieve the same improvement in the
75 year actuarial balance as a 15 percent reduction in
2017, the benefit cut in 2027 would need to be one third
larger: 20 percent, rather than 15 percent.

Sustainable Solvency. Different policies can have similar
effects on the actuarial balance but different effects on
Social Security’s finances at various points in time. For
example, a policy that would implement large tax
increases or benefit cuts in 30 years could eliminate the
75 year actuarial imbalance but not prevent trust fund
exhaustion. And a policy that would immediately cut
benefits or increase taxes by a flat percentage could elimi
nate the 75 year imbalance and delay exhaustion beyond
the projection period but still allow large and growing
imbalances to remain in the 76th year and beyond.

One way to sustain solvency is to have a trust fund ratio
that is positive throughout the projection period and then
stable or growing after 75 years. Neither increasing the
payroll tax by 2.0 percentage points over two decades nor
cutting benefits by 15 percent would result in sustainable
solvency; the trust funds would be exhausted around
the end of the projection period and the trust fund ratio
would still be declining after 75 years (see Figure 5). Only
one option CBO analyzed (Option 17) would, by itself,
result in sustainable solvency.

Effects of the Options on Payroll Taxes Paid and
Benefits Received by Various Groups
Some options would affect people in all earnings groups
similarly; others could have greater effects on people with
higher or lower earnings or on members of other specific
populations. In addition, some options would have
greater effects on people born earlier or later. This section
discusses the distributional effects of various options on
initial benefits, lifetime benefits, and lifetime payroll
taxes. It assumes that scheduled benefits are paid, consis
tent with the discussion of the Social Security system’s
finances (see Table 2 on page 33). (For an analysis of the
distributional effects of the options on scheduled bene
fits, see Table 3 on page 39; for an analysis of the
distributional effects of the options on payable benefits,

see Table 4 on page 43.) Specifically, CBO examined
three measures to identify distributional effects:

B The percentage change in initial benefits for retirees
(calculations were made under the assumption that all
workers claim benefits at age 65);

B The percentage change in the present value at age 62
of lifetime benefits; and

B The percentage change in the present value at age 62
of lifetime payroll taxes.

Present values were computed using a real discount rate
of 3 percent.

An option could have different effects on initial benefits
and lifetime benefits. The initial benefits presented here
are the initial retirement benefits that would be received
by workers, assuming that all people claim retiree benefits
at age 65.20 Lifetime benefits, by contrast, include the
present value of all payments to recipients over time,
including cost of living adjustments, and they include
payments to disabled workers and to dependents and sur
vivors. Unlike the estimated initial benefits, the projected
lifetime benefits take into account the age at which each
person is actually expected to claim benefits.21

CBO examined the way changes resulting from the
various policy options would affect beneficiaries in
the low, middle, and high quintiles (the lowest, middle,
and highest fifths) of households’ lifetime earnings in
three 10 year birth cohorts: people born in the 1960s,
people born in the 1980s, and people born in the
2000s.22

20. Values are based on earnings through age 61 and are net of the
income taxes paid on benefits that are credited to the Social
Security trust funds.

21. Values are net of the income taxes paid on benefits that are
credited to the Social Security trust funds. The measure includes
benefits received by old age workers, disabled workers, old age
spouses, and old age widows. Because there are insufficient data
on benefits received by young widows and children for years
before 1984, young widows, spouses of disabled workers, and
child beneficiaries are excluded from this measure.

22. Each person who lives at least to age 45 is ranked by lifetime
household earnings. For someone who is single in all years,
lifetime earnings equals the sum of real earnings over a lifetime. In
any year a person is married, the earnings measure is a function of
his or her earnings plus those of his or her spouse (adjusted for
economies of scale in household consumption). A person’s lifetime
earnings consist of the present value of those annual amounts.

CBO

14 SOCIAL SECURITY POLICY OPTIONS

CBO

In all, therefore, the effects are identified for nine groups
of people, measured in terms of how the option would
affect the median value of each outcome; in other words,
CBO’s analysis assesses the options’ effects on the person
at the middle of the distribution of outcomes in a group.
Outcomes for half of the people in the group would be
lower and outcomes for half would be higher. In this
study, “benefits for low earners” refers to median benefits
for people in the lowest quintile of lifetime household
earnings. A “change in taxes (or benefits) for high earn
ers” refers to the percentage change caused by a policy
option, relative to current law, in the median of the pres
ent value of lifetime taxes (or benefits) at age 62 for peo
ple in the highest quintile of lifetime household earnings.

Outcomes shown in Table 3 and Table 4 are estimated
percentage changes from the outcomes that would occur
under current law. They are rounded to the nearest three
percentage points to give a sense of likely effects on bene
fits and taxes without showing numerous small differ
ences in outcomes that are not analytically meaningful.
The estimates are based on samples of people from the
relevant groups, and different characteristics of people
sampled can lead to small differences in outcomes across
groups that are not meaningful from a policy perspective.
For example, if an unusually high number of sampled
people in a group were to turn out to have a shorter than
average lifespan, then an option that would reduce
COLAs would result in smaller changes in median life
time benefits for that group than in a more representative
sample of the particular group, and the difference could
lead to somewhat misleading comparisons with other
groups.

To analyze distributional consequences of several policies
that would produce the same effect on Social Security’s
finances, CBO also analyzed eight variants of the main
options in this study, each of which would improve the
actuarial balance by one quarter, or by 0.15 percentage
points of GDP (see the Appendix).

Options with Proportionate Effects on People in
Different Earnings Categories. Several options presented
in this study would make a proportionate change in ini
tial benefits for people of a particular age, regardless of
their earnings:

B Change All PIA Factors. Among the possibilities CBO
examined are a one time flat reduction (Option 12)
and a gradual reduction over time (Option 15). Other

approaches that would phase in the reduction of
scheduled benefits over time could link reductions in
the PIA factors to the increasing longevity of the U.S.
population (Option 16) or to real growth in earnings
(Option 17).

B Increase the Full Retirement Age. An increase by one
year in the FRA would reduce monthly benefits by
between 5 percent and 8 percent, depending on the
age at which a person chose to begin receiving bene
fits. Changes could be phased in either at a constant
rate (Options 26 and 27) or according to the increase
in life expectancy (Option 28).

An option that reduced all benefits by a similar percent
age would nevertheless have different consequences for
different categories of beneficiaries. In dollar terms, the
reduction would be larger for people who receive higher
benefits. But measured as a percentage of total income,
the reduction would be greatest for people who rely most
heavily on Social Security. For example, someone who has
no income other than Social Security would find a
10 percent cut in benefits more burdensome than some
one for whom Social Security provides just one quarter of
retirement income. (In 2008, Social Security benefits
accounted for 90 percent or more of income for 21 per
cent of retired married couples who were beneficiaries.
Forty three percent of beneficiaries who were not cur
rently married received 90 percent or more of their
income from Social Security.)23

For the most part, options that would have proportionate
effects for people in all earnings categories would have
greater effects on younger people. However, instituting a
one time reduction in the PIA factors would have similar
effects on everyone born in 1955 or later.

Options with Very Different Effects on High and Low
Earners. Some options would have much greater effects
on the amount of payroll taxes levied on people or on the
benefits they are scheduled to receive, depending on their
lifetime earnings.

B Change the Maximum Amount of Earnings Subject to
Payroll Taxes. The method for setting the taxable
maximum (which usually increases to keep pace with
average nominal earnings) could be changed. The tax
able maximum could be eliminated (Option 4) or

23. See Social Security Administration, “Income Sources.”

SOCIAL SECURITY POLICY OPTIONS 15

increased so a larger percentage of covered earnings
would be subject to payroll taxes (Option 5).
Changing the taxable maximum also would change
benefits, because the amount of earnings included in
the AIME would change.

B Impose Taxes on Earnings Above the Current Taxable
Maximum. Changing the taxable maximum would
affect benefits, but an additional tax could be applied
solely to raise revenue without affecting benefits. Such
a policy could extend the existing tax rate to all earn
ings (Option 6) or to earnings up to a higher thresh
old (Option 7), or it could apply a different rate above
the taxable maximum (Options 8 and 9) with no
effect on benefits.

B Change the PIA Factors Applied to Higher Earnings.
The top two PIA factors (currently 32 percent and
15 percent) could be changed (Option 13), or the
top factor alone could be changed (Option 14). The
benefit formula could be changed so that it would
have four different PIA factors rather than three
(Options 18 and 19); that approach, commonly called
progressive price indexing, would then reduce the top
two PIA factors gradually until they reached zero.

B Increase Benefits for Low Earners. Benefits for people
with low lifetime earnings could be boosted by
increasing the existing special minimum benefit or
by creating a new minimum benefit with a different
structure (Options 23 and 24). Alternatively, benefits
for people who have many years of low earnings
could be increased by raising their standard benefit
(Option 25).

Most of those options would have greater effects on peo
ple born later. Effects of a one time reduction in one or
two of the top PIA factors would not vary much from
one birth cohort to another.

Effects of the Options on Work and Saving
Social Security affects people’s decisions about how much
to work, when to retire, and how much to save for retire
ment. Changing the rates and structure of Social Security
taxes and benefits would influence those decisions.
Because those behavioral responses are difficult to quan
tify, this study’s analysis of the effects on Social Security’s
finances generally does not incorporate changes in behav
ior that might result from implementing the various

options, such as the potential for alterations in a worker’s
lifetime earnings.24

Like all taxes on earnings, Social Security taxes reduce the
reward from work, which tends to decrease how much
people work. At the same time, Social Security taxes and
other taxes on earnings reduce take home pay for any
given amount of work, and the desire to earn a certain
amount of take home pay can lead people to work more.
The net effect of taxes on work reflects the balance of
those forces; most economists conclude that, on average,
the negative effects of lower after tax earnings for each
additional hour worked slightly outweigh the positive
effects of lower after tax income from current working
hours.25 Thus, in CBO’s estimation, increasing Social
Security tax rates would tend to decrease modestly the
hours of labor that workers supply. Increasing payroll tax
rates also would encourage workers to shift some of their
compensation to tax exempt fringe benefits. High earn
ers, who tend to have more flexibility about how to struc
ture their compensation, are particularly likely to reduce
their taxable earnings by electing to receive more of their
compensation as tax exempt fringe benefits.

Options that would result in increased tax rates only up
to a particular amount would affect all earners regardless
of whether their earnings were above or below that
amount. For example, Option 1 would increase the tax
rate but would not change the existing taxable maximum;
Option 5 would raise the taxable maximum but would
not change the tax rate. People whose earnings would be
well above the range of earnings affected by changes in
tax rates would not confront the same disincentives to
work, but they would have less income after taxes, which
might tend to slightly increase their work effort.

24. The current analysis incorporates some small changes in work
behavior in response to changes in expected lifetime benefits. See
Congressional Budget Office, Projecting Labor Force Participation
and Earnings in CBO’s Long Term Microsimulation Model,
Background Paper (October 2006).

25. For discussion, see Congressional Budget Office, The Effect of Tax
Changes on Labor Supply in CBO’s Microsimulation Tax Model,
Background Paper (April 2007), and Labor Supply and Taxes,
CBO Memorandum (January 1996). The 1996 memorandum
assumed that the spouse of a household’s primary breadwinner
would be more responsive to higher taxes on earnings than would
the primary worker in a household. In recent years, CBO has
revised downward its estimates of the responsiveness of such
spouses because of evidence that their responsiveness has declined
over time as their participation in the labor force has grown.

CBO

http://www.cbo.gov/ftpdocs/76xx/doc7676/10-27-LaborForce

http://www.cbo.gov/ftpdocs/79xx/doc7996/04-12-LaborSupply

http://www.cbo.gov/ftpdocs/33xx/doc3372/labormkts

16 SOCIAL SECURITY POLICY OPTIONS

CBO

Workers would probably consider not only the Social
Security taxes they pay but also the benefits they expect to
receive. Therefore, Option 6, which would eliminate the
taxable maximum but would not affect benefits, would
probably have a larger effect on work incentives than
would Option 4, which would eliminate the maximum
but include the additional taxable earnings in the benefit
computation.

Options that would modify the way benefits are deter
mined also would influence how long people remain in
the workforce and how much they work while they
are in the workforce. If workers expected lower Social
Security benefits, for example, they might stay in the
workforce longer to claim benefits at a later age.26 How
ever, a reduction in benefits also could mean (depending
on the formula through which benefits are reduced) that
an extra year of work would increase future benefits by a
smaller amount; that would amount to an increased tax
on earnings, which would discourage work. On net, older
workers would probably choose to work longer in
response to a reduction in benefits, leading to an increase
in the size of the labor force.27

The decision about how long someone would remain in
the workforce would be influenced differently by options
that would change benefits and by options that would
raise the full retirement age. Because many workers claim
benefits at the full retirement age, increasing that age
would probably result in beneficiaries’ claiming benefits
later than they would if an effectively identical policy
were implemented through adjustments in the benefit
formula.

Increasing the full retirement age also would create a
somewhat stronger incentive for some older workers—
particularly those in poor health—to leave the labor force
and apply for DI benefits rather than continue to
work and then claim Old Age benefits at age 62 in a

26. Before the recent recession, the increase in the full retirement age
caused a benefit cut that led many workers to delay claiming
Social Security benefits. See Jae Song and Joyce Manchester, Have
People Delayed Claiming Retirement Benefits? Responses to Changes
in Social Security Rules, Congressional Budget Office Working
Paper 2008 04 (May 2008).

27. A reduction in scheduled benefits would extend the date of trust
fund exhaustion and would result in higher payable benefits for
several years (see Figure 5 on page 12), so a reduction in scheduled
benefits might actually discourage work during that period.

decreased amount.28 (Changes in the full retirement age
would not affect the benefits of workers who qualify for
Disability Insurance.) Under current law, workers who
claimed retirement benefits at age 62 in 2033 would
receive 70 percent of their PIA (the benefit they would
have received if they had claimed benefits at their full
retirement age); if they qualified for DI benefits, however,
they would receive 100 percent of that amount. Increas
ing the full retirement age would increase the difference
between retirement and DI benefits if retirement benefits
are claimed before a worker reaches the FRA.

Social Security also affects private saving. People who
expect to receive Social Security benefits probably save
less for their retirement than they would if there were no
such program. In effect, Social Security substitutes to
some extent for retirement saving: Some workers view the
tax on their wages as a form of saving money each month
for retirement; instead of accumulating assets to draw
down when they retire, those workers are counting on
receiving benefits from the government.29 Therefore, ben
efit reductions would probably result in higher personal
saving.

To the extent that changes in Social Security increase pri
vate saving without increasing government deficits, those
changes also would increase national saving—the total
amount of saving in the economy by the government and
private sector. Over time, greater national saving would
raise the stock of capital and result in greater total wealth
and larger incomes.

Options That Would Change the
Taxation of Earnings
Payroll taxes for Social Security are proportional to earn
ings below the taxable maximum, and they are not col
lected on earnings above that amount. Currently, about
93 percent of workers have earnings below the taxable
maximum, and they pay Social Security taxes on all of
their earnings. The remaining 7 percent have some

28. Empirical evidence of the response in DI enrollment is discussed
in Mark Duggan, Perry Singleton, and Jae Song, “Aching to
Retire? The Rise in Full Retirement Age and Its Impact on the
Social Security Disability Rolls,” Journal of Public Economics,
vol. 91 (2007), pp. 1327–1350.

29. See Congressional Budget Office, Social Security and Private
Saving: A Review of the Empirical Evidence, CBO Memorandum
(July 1998).

http://www.cbo.gov/ftpdocs/90xx/doc9077/2008-04

http://www.cbo.gov/doc.cfm?index=731

SOCIAL SECURITY POLICY OPTIONS 17

earnings that are not taxed and therefore have a lower
average tax rate on their earnings.

The options in this section would increase Social Security
revenues beginning in 2012 by making changes to pay
roll taxes. Options that increased the taxable maximum
without making other changes to the system also would
increase the amount of earnings used in the computation
of benefits, so in those options, a portion of the increase
in revenues would be offset by increased outlays. (See
Table 2 on page 33 for the effects of the options on
Social Security’s finances, Table 3 on page 39 for effects
on distributional outcomes assuming that scheduled
benefits are paid, and Table 4 on page 43 for effects on
distributional outcomes assuming that only payable
benefits are paid.)

Option 1: Increase the Payroll Tax Rate by
1 Percentage Point in 2012
This option would raise the payroll tax rate for employers
and employees by 0.5 percentage points each, beginning
in 2012. The overall rate would be 13.4 percent: 6.7 per
cent paid by employers and by employees. (As with the
other options discussed in this section, the tax rate for
self employed workers would increase in line with the
combined tax rate on employers and employees.)

Social Security’s total revenues would increase by about
0.4 percentage points of GDP in 2040, or by about
7 percent, relative to current law. (Taxable earnings are
only a portion of GDP, so a 1 percentage point increase in
the payroll tax rate leads to significantly less than a 1 per
centage point increase in revenues as a share of GDP.)
This option would improve the 75 year actuarial balance
by 0.3 percentage points of GDP and would extend the
trust fund exhaustion date by 17 years, to 2056. As a
result, payable benefits would be higher for people who
receive benefits in 2039 or later (because, under current
law, total payable benefits are determined by total reve
nue during that period). The percentage increase in life
time payroll taxes paid would be similar for people in all
categories of lifetime earnings within the same birth
cohort.

Option 2: Increase the Payroll Tax Rate by
2 Percentage Points Over 20 Years
Whereas Option 1 would increase taxes by a fixed
amount in 2012 and later, this option would raise the
combined payroll tax rate gradually, by 0.1 percentage

point (0.05 percentage points each for employers and
employees) every year from 2012 to 2031. By the end
of that period, the rate would stand at 14.4 percent—
7.2 percent apiece for employers and employees—a
total of 2 percentage points higher than the current rate
of 12.4 percent.

The Social Security payroll tax rate would increase by
16 percent under this option relative to current law,
but Social Security revenues would rise a bit less because
the option would not affect income taxes on benefits.
This option would improve the 75 year actuarial balance
by 0.6 percentage points of GDP and would extend the
trust fund exhaustion date to 2083. As a result, payable
benefits would be higher for people who receive benefits
in 2039 or later.

Lifetime payroll taxes would increase by a small amount
for people born in the 1960s and by about 15 percent for
those born in the 2000s. After the option is fully phased
in, the percentage increase in lifetime payroll taxes paid
would be similar for people in all categories of lifetime
earnings.

Option 3: Increase the Payroll Tax Rate by
3 Percentage Points Over 60 Years
Under this option, the combined payroll tax rate would
increase gradually, by 0.05 percentage points (0.025 per
centage points each for employers and employees) every
year from 2012 to 2071. By that time, the rate would
stand at 15.4 percent (3 percentage points higher than
the current rate of 12.4 percent), with employers and
employees each paying 7.7 percent. This option is similar
to Option 2, except that the tax increase would be imple
mented more gradually and ultimately the tax rate would
be higher. Under both options, the tax rate would be
14.4 percent in 2051. Before that, it would be higher
under Option 2, and in later years, it would be higher
under this option.

Social Security’s total revenues would increase by about
10 percent under this option in 2040, or by 0.5 percent
age points of GDP relative to current law. After the
option was fully phased in (by 2071), revenues would
increase by 20 percent. The option would improve the
75 year actuarial balance by 0.5 percentage points of
GDP and would extend the trust fund exhaustion date by
19 years, to 2058. As a result, payable benefits would be
higher starting in 2039.

CBO

18 SOCIAL SECURITY POLICY OPTIONS

CBO

Lifetime payroll taxes would increase by a small amount
for people born in the 1960s and by about 15 percent for
the 2000s cohort. The percentage increase in lifetime
payroll taxes paid would be similar for people in all cate
gories of lifetime earnings within the same birth cohort.

Option 4:

Eliminate the Taxable Maximum

In 1983, 91 percent of all earnings from jobs covered by
the Social Security program were below the maximum
taxable amount. That percentage declined as the earnings
of workers in the highest income groups grew faster than
average earnings did. Thus, in 2009, about 83 percent of
earnings from employment covered by OASDI (corre
sponding to 93 percent of workers, as mentioned above)
was below the maximum taxable amount, now set at
$106,800. Under this option, all covered earnings would
be taxed at the current rate of 12.4 percent (6.2 percent
paid by the employer and 6.2 percent paid by the
employee) in 2012 and later. The additional taxable earn
ings would be included in benefit computations, resulting
in higher benefits for the higher earning workers who
would be subject to the additional tax.

Social Security’s total revenues would increase by 0.9 per
centage points of GDP in 2040, or by 19 percent relative
to current law, and outlays would increase by 0.3 percent
age points of GDP, with further increases in subsequent
years. This option would improve the 75 year actuarial
balance by 0.6 percentage points of GDP and extend the
trust fund exhaustion date to 2083.

This option would primarily affect taxes paid by high
earners. The increase in taxes for high earners would be
12 percent, 15 percent, and 18 percent for people born in
the 1960s, 1980s, and 2000s, respectively. The increase
would be greater for people with the very highest earn
ings. Among people born in the 1980s, lifetime taxes
would rise by at least 40 percent for people in the top
5 percent of lifetime earnings. The increase in benefits for
the highest earners would be slightly smaller than the
increase in their payroll taxes in percentage terms.30 In
dollar terms, benefits would increase by much less than
taxes because, under current law, over their lifetimes most
high earners receive much less in benefits than they pay in
taxes.

Option 5: Raise the Taxable Maximum to Cover
90 Percent of Earnings
Because the earnings of workers in the highest income
groups have grown faster than average earnings in recent

decades, the share of all earnings from jobs covered by the
Social Security program that were below the taxable max
imum has fallen from about 91 percent in 1983 to about
83 percent in 2009. This option would raise the taxable
maximum so that, beginning in 2012, 90 percent of earn
ings would be taxable; the additional amounts subject to
the payroll tax would be used in benefit calculations. The
taxable maximum in 2012 would be about $156,000
under this option, CBO estimates, an amount 38 percent
higher than the maximum of $113,700 estimated under
current law. (The current taxable maximum is $106,800.)
After 2012, the taxable maximum would increase so that
90 percent of covered earnings would continue to be sub
ject to payroll taxes.

Social Security’s total revenues would increase by about
0.4 percentage points of GDP in 2040, or by about
8 percent, relative to current law. This option would
improve the 75 year actuarial balance by 0.2 percentage
points of GDP and would extend the trust fund exhaus
tion date by 11 years, to 2050. As a result, payable bene
fits would rise, especially for those who receive benefits in
the 2040s.

This option would primarily affect high earners, whose
taxes would increase by about 6 percent for the 1960s
cohort and by approximately 15 percent for the 2000s
cohort. Benefits would be affected in the same manner
as under Option 4.

Option 6: Tax Covered Earnings Above the Taxable
Maximum; Do Not Increase Benefits
Under this option, starting in 2012, all covered earnings,
including earnings above the taxable maximum, would be
taxed at 12.4 percent (6.2 percent paid by the employer
and 6.2 percent paid by the employee). This option
would levy the same taxes as Option 4, but the taxable
maximum under current law would still be used to

30. The change in lifetime benefits is greater than that of initial
benefits for two main reasons. First, lifetime benefits are based on
the highest 35 years of earnings at any age. In contrast, initial
benefits shown in this study are calculated on the basis of the
highest 35 years of earnings through age 61 (and they omit
earnings at older ages, including any earnings after age 61 above
the taxable maximum). Second, lifetime benefits include spousal
benefits on the basis of the earnings of the household’s primary
earner (including earnings above the taxable maximum), whereas
initial benefits shown here are calculated based on the spouse’s
own earnings.

SOCIAL SECURITY POLICY OPTIONS 19

calculate benefits, and the option would therefore have no
direct effect on scheduled benefits.

Under this option, Social Security’s total revenues would
increase by about 0.9 percentage points of GDP in 2040,
or by about 18 percent relative to current law. This
option would improve the 75 year actuarial balance by
0.9 percentage points of GDP and would extend the trust
fund exhaustion date beyond the 75 year projection
period. As a result, payable benefits would be higher from
2039 onward, especially for people born later.

This option would primarily affect taxes paid by high
earners. The effects on payroll taxes would be the same as
in Option 4, but there would be no effect on scheduled
benefits.

Option 7: Tax Covered Earnings Up to $250,000;
Do Not Increase Benefits
Under this option, starting in 2012, all covered earnings
between the taxable maximum and $250,000 would be
taxed at 12.4 percent (6.2 percent paid by the employer
and 6.2 percent paid by the employee). The $250,000
threshold would cover 93 percent of earnings in 2012
and, in future years, the threshold would be indexed to
earnings. Because the taxable maximum under current
law would continue to be used to calculate benefits, the
option would have no direct effect on scheduled benefits.
This option is similar to Option 6, although the increase
in the payroll tax would be smaller, as would be the finan
cial effects.

Social Security’s total revenues would increase by about
0.5 percentage points of GDP in 2040, or by about
11 percent relative to current law. This option would
improve the 75 year actuarial balance by 0.5 percentage
points of GDP and would extend the trust fund exhaus
tion date by 38 years, to 2077. Payable benefits would be
higher in 2039 and later. There would be no increase in
scheduled benefits.

Like Option 6, this option would primarily affect taxes
paid by high earners. Their taxes would increase by
about 12 percent for the 1960s cohort and by approxi
mately 18 percent for the 2000s cohort.

Option 8: Tax All Earnings Above the Taxable
Maximum at 4 Percent; Do Not Increase Benefits
Under this option, starting in 2012, all covered earnings
above the taxable maximum would be taxed at 4 percent.

Because the current law maximum would still be used for
calculating benefits, this option would have no direct
effect on scheduled benefits. This option is similar to
Option 6, but the payroll tax rate above the taxable
maximum would be substantially smaller, as would be the
financial effects on the trust funds.

Social Security’s total revenues would increase by
about 0.3 percentage points of GDP in 2040, or by about
6 percent, relative to current law. The option would
improve the 75 year actuarial balance by 0.3 percentage
points of GDP and would extend the trust fund exhaus
tion date by 12 years, to 2051. Although there would be
no increase in scheduled benefits, higher payroll taxes
would result in higher payable benefits in 2039 and later.

This option would primarily affect taxes paid by high
earners. For high earners born in the 1960s, lifetime taxes
would increase by about 6 percent; high earners born in
the 2000s would see their taxes increase by about 9 per
cent. The increase would be greater for people with the
very highest earnings.

Option 9: Tax All Earnings Above $250,000 at
4 Percent; Do Not Increase Benefits
Like Option 8, this option would institute a 4 percent tax
on high earners, but the tax would apply only to covered
earnings above $250,000. (The Social Security payroll tax
would not apply to earnings between the taxable maxi
mum and $250,000.) That threshold would apply to less
than 1 percent of people with earnings in 2012. In future
years, the threshold would rise at the rate of average wage
growth. The current law taxable maximum would still be
used for calculating benefits, so this option would have
no direct effect on scheduled benefits.

Social Security’s total revenues would rise by approxi
mately 0.1 percentage point of GDP in 2040, or by
about 3 percent relative to current law. This option
would improve the 75 year actuarial balance by 0.1 per
centage point of GDP and would extend the trust fund
exhaustion date by 4 years, to 2043. There would be no
increase in scheduled benefits, but higher payroll taxes
would result in higher payable benefits in 2039 and later.

This option would primarily affect high earners, whose
taxes would typically increase by a small percentage. The
increase would be greatest for people with the very high
est earnings.

CBO

20 SOCIAL SECURITY POLICY OPTIONS

CBO

Options That Would Change the
Benefit Formula
The formulas for calculating earnings histories for Social
Security and translating those amounts into initial
monthly benefits could be changed in many ways. The
options in this section either would make a single change
that would affect all new beneficiaries or they would
make gradual changes that would have larger effects for
people born later rather than earlier. All options would be
implemented beginning in 2017.

Some options would affect high earners more than low
earners, for whom Social Security benefits generally
constitute a larger percentage of lifetime earnings. CBO
projects, for example, that under current law, scheduled
lifetime benefits would be 15 percent of lifetime taxable
earnings for low earners born in the 1960s and 8 percent
of lifetime taxable earnings for high earners in the same
birth cohort.

All of the options in this section would produce small
effects on revenues, in addition to their main effects on
outlays, because changes in benefits would change the
amount collected in taxes on those benefits. (See Table 2
on page 33 for the effects of the options on Social
Security finances, Table 3 on page 39 for effects on distri
butional outcomes assuming that scheduled benefits are
paid, and Table 4 on page 43 for effects on distributional
outcomes assuming that only payable benefits are paid.)

Option 10: Raise from 35 to 38 the Years of
Earnings Included in the AIME
For a person who reaches age 62 after 1990, the calcula
tion of average indexed monthly earnings under current
law incorporates the highest 35 years of indexed earnings
in which that person paid Social Security taxes. This
option would extend the period for the AIME calculation
by 3 years, phased in between 2017 and 2019. Beginning
in 2019, the calculation would take the average of the
38 highest years of indexed monthly earnings. The new
average would apply only to newly eligible retired work
ers, so there would be no effect on DI benefits.31

This option would reduce Social Security’s total outlays
by about 0.1 percentage point of GDP in 2040, or by
2 percent, relative to currently scheduled outlays. As a

31. For a more detailed analysis of this option, see Congressional
Budget Office, Budget Options, Volume 2, p. 146.

result, it would improve the 75 year actuarial balance by
0.1 percentage point of GDP; however, it would not
significantly extend the trust fund exhaustion date. (That
is, by CBO’s estimate, the option would change the
exhaustion date by two years or less.)

This option would reduce scheduled lifetime benefits by
a similar amount for all birth cohorts. It would have the
largest effect on people who worked for fewer than
38 years, because they would have additional years with
no earnings included in the calculation of their benefits.
However, the option would reduce benefits even for
workers who worked 38 or more years, because those
people would almost always have had lower average earn
ings in the additional computation years than they would
have had in the 35 years of their highest earnings. The
reduction would, on average, be larger for women than
for men, because women tend to spend more years out of
the workforce.

Option 11: Index Earnings in the
AIME Formula to Prices
Under current law, as part of the computation of a retired
worker’s average indexed monthly earnings, past earnings
are indexed to total average earnings nationwide through
the year that is two years prior to benefit eligibility, thus
incorporating rising prices and growth in real earnings.
Under this option, those earnings would be indexed to
the growth in prices only.32 As a result, initial benefits
would be lower than those calculated under current law.

This option would reduce Social Security’s outlays by
0.1 percentage point of GDP in 2040, or by 2 percent
compared with currently scheduled outlays. It would
improve the 75 year actuarial balance by 0.2 percentage
points of GDP. The effects would be small initially, so the
option would not significantly extend the trust fund
exhaustion date.

The magnitude of the reduction in benefits would
increase over time, and by 2040, the average AIME

32. Specifically, in computing the AIME, all earnings from 2016 and
earlier would be increased by the growth in the average wage index
from the earnings year to the year that is two years before the
AIME computation year. Earnings from 2017 and later would be
increased by the growth in the CPI W from the earnings year to
the year that is two years before the AIME computation year. The
bend points in the PIA formula would continue to be indexed to
nominal wage growth. Beginning in 2017, this option would
apply to newly eligible retired and disabled workers.

http://www.cbo.gov/ftpdocs/102xx/doc10294/08-06-BudgetOptions

SOCIAL SECURITY POLICY OPTIONS 21

Figure 6.

Calculating the PIA in 2010 Under the

Current Social Security System

(PIA)

Source: Congressional Budget Office.

Notes: A bend point (represented by a dot on the line) is the threshold at which a PIA factor changes.

PIA = primary insurance amount; AIME = average indexed monthly earnings.

0

1,000

2,000

3,000

4,000

5,000

6,000

0 3,000 6,000 9,000 12,000 15,000

AIME

Benefits Rise $0.90 for
Each $1 Increase in the AIME

Benefits Rise $0.32 for
Each $1 Increase in the AIME

Benefits Rise $0.15 for
Each $1 Increase in the AIME

$761

$4,586

would be 7 percent lower than it would be under current
law. After price indexing was in place for several decades,
the scheduled lifetime benefits for people born in the
2000s would be reduced by about 12 percent for low
earners and by around 6 percent for high earners. The
reduction in benefits would be smaller for high earners
because the reduction in the AIME would be multiplied
by the 90 percent replacement rate for most low earners
but multiplied by the 15 percent replacement rate for a
large fraction of earnings among high earners. Payable
lifetime benefits would not fall as much as scheduled
benefits because benefits before the exhaustion of the
trust funds (and the gap between outlays and revenues
that would have to be closed by reducing scheduled
benefits) would be lower.

Option 12: Reduce All PIA Factors by 15 Percent
Under current law, the primary insurance amount factors
used in calculating initial benefits are 90 percent (applied
to the first $761 of the AIME in 2010), 32 percent
(applied to the AIME between $761 and $4,586 in
2010), and 15 percent (applied to the AIME over $4,586
in 2010) (see Figure 6). This option would reduce the
PIA factors for newly eligible beneficiaries, including dis
abled workers, by 15 percent in 2017 (to 77 percent,
27 percent, and 13 percent), thus reducing initial benefits

by 15 percent. For example, if the change was applied in
2010, a worker with an AIME of $5,000 would have a
monthly benefit of $1,672 instead of the current system’s
benefit of $1,971.

Social Security’s total outlays would decline by 0.7 per
centage points of GDP in 2040, or by 12 percent relative
to currently scheduled outlays. This option would
improve the 75 year actuarial balance by 0.5 percentage
points of GDP and would extend the trust fund exhaus
tion date by 37 years, to 2076.

Percentage reductions in scheduled lifetime benefits for
people affected by the change would be similar. Payable
lifetime benefits would be lower than under current law
for people in older cohorts but higher for people who
were born in the 1980s and 2000s.

Option 13:

Reduce the Top Two PIA Factors by
Roughly One-Third

Starting in 2017, this option would reduce the top two
primary insurance amount factors for newly eligible
retired and disabled workers from 32 percent to 20 per
cent and from 15 percent to 10 percent. In contrast to
Option 12, the benefit reduction under this option
would be greater for people with higher earnings.

CBO

22 SOCIAL SECURITY POLICY OPTIONS

CBO

Social Security’s total outlays would decline by 1.0 per
centage point of GDP in 2040, or by 16 percent from
currently scheduled outlays. This option would improve
the 75 year actuarial balance by 0.7 percentage points of
GDP and extend the trust fund exhaustion date beyond
the 75 year projection period. As a result, payable life
time benefits would generally be higher than under
current law for people other than high earners who
receive benefits in 2039 and several decades thereafter.
However, this option is not sustainably solvent because
outlays would increase more rapidly than revenues after
implementation.

Scheduled lifetime benefits would fall by 24 percent for
high earners and by about 3 percent for low earners. Peo
ple with an AIME below the first bend point would not
be affected.

Option 14: Reduce the Top PIA Factor by One-Third
This option would implement one part of Option 13: It
would reduce the top primary insurance amount factor
for newly eligible retired and disabled workers from
15 percent to 10 percent in 2017. It would affect only
those new beneficiaries whose AIMEs were above the
second bend point (in 2010, $4,586; 29 percent of
62 year olds). For example, under this option, in 2017
that bend point would be $5,114 in 2010 dollars, and a
worker with an AIME of $6,000 would receive monthly
benefits that were $44 lower than under current law.

Social Security’s total outlays would decline by 0.1 per
centage point of GDP in 2040, or by 2 percent from cur
rently scheduled outlays. This option would improve the
75 year actuarial balance by 0.1 percentage point of GDP,
and it would not significantly extend the trust fund
exhaustion date.

This option would primarily affect high earners, whose
scheduled lifetime benefits would be reduced by approxi
mately 6 percent. Payable lifetime benefits would be
slightly lower than under current law for high earners
and slightly higher for others who receive benefits in
2039 or later.

Option 15: Reduce All PIA Factors by
0.5 Percent Annually
Beginning in 2017, this option would reduce the primary
insurance amount factors for newly eligible retired and
disabled workers by 0.5 percent annually. Specifically,
each year’s PIA factors would equal the previous year’s
factors multiplied by 0.995. By 2080, the PIA factors

would be 65 percent, 23 percent, and 11 percent, equal
to about three quarters of what they are now. In 2048,
this option would match the reduction in initial benefits
provided by Option 12 (which would cut benefits by
15 percent in 2017), but it would provide for smaller cuts
in earlier years and larger reductions in later years.

Social Security’s total outlays would decline by about
0.3 percentage points of GDP in 2040, or by about
6 percent from currently scheduled outlays. This option
would improve the 75 year actuarial balance by 0.4 per
centage points of GDP and would extend the trust fund
exhaustion date by 3 years, to 2042.

Scheduled lifetime benefits would be reduced by about
3 percent and 21 percent for people born in the 1960s
and 2000s, respectively. This option would have little
effect on lifetime payable benefits.

Option 16:

Index Initial Benefits to
Changes in Longevity

Under this option, benefits for newly eligible retired
workers would be reduced in proportion to the increase
in life expectancy at age 62; reductions would begin in
2017. For example, life expectancy at age 62 in 2040 will
be about 8 percent longer than in 2016, CBO projects, so
initial benefits would be reduced by about 8 percent in
2040. The option would not affect DI beneficiaries, but
benefits would decline for disabled beneficiaries when
they converted to OASI (conversion occurs automatically
when a beneficiary reaches the full retirement age).33

Under this option, increases in average life expectancy
would not result in higher average retirement benefits
paid over a lifetime.34 However, the computation would
be based on the average life expectancy of the entire

33. Under this option, when a disabled beneficiary reached the full
retirement age, DI benefits would be reduced by an amount
proportional to the number of years between age 22 and age 62
that the beneficiary was not entitled to receive benefits. For
example, someone who became entitled to benefits at 62 without
receiving any DI benefits would experience the full reduction,
whereas the reduction for someone who became disabled at age 42
would be half as large.

34. The reduction would not exactly equal the change in life
expectancy after 2016 because it would be implemented by
adjusting PIA factors by a ratio of the life expectancy at age 62
for the birth cohort reaching age 62 in 2013 to the life expectancy
at age 62 for the birth cohort reaching age 62 three years before
the birth cohort in question. (The computation depends on life
expectancy from three years before the year in question because
of the lag in collecting and processing mortality data.)

SOCIAL SECURITY POLICY OPTIONS 23

population. On average, low earners have shorter life
expectancies than high earners do, and some evidence
suggests that the gap is growing.35 If that trend continues,
the reduction in lifetime benefits per percentage point of
additional life expectancy would be greater for low earn
ers than for high earners (rather than equal, as it would be
if low and high earners were to have the same increase in
life expectancy).

Social Security’s total outlays would decline by 0.2 per
centage points of GDP in 2040, or by 3 percent from
currently scheduled outlays. This option would improve
the 75 year actuarial balance by 0.2 percentage points of
GDP. The early effects of the option would be too small
to significantly extend the trust fund exhaustion date.

The reduction in scheduled lifetime benefits would
increase over time, reaching about 12 percent for people
born in the 2000s. This option would have little effect on
payable lifetime benefits.

Option 17: Reduce PIA Factors to Index Initial
Benefits to Prices Rather Than Earnings
Under this option, often called “price indexing,” the bend
points in the benefit formula would be indexed to earn
ings, as under current law, but the PIA factors would be
reduced each year by measured growth in real earnings
from two years earlier. Beginning in 2017, average initial
benefits for newly eligible retirees would increase with
prices rather than with prices and real earnings. Given
CBO’s long term projections for growth in real earnings,
initial benefits would be 1.3 percent lower in the first year
than under current law, the next year they would be 2.6
percent lower, and they would decline in the same way in
each succeeding year.36 In reality, however, the incremen
tal reduction would vary from year to year, depending on
actual growth in real earnings. The reductions would be
smaller during periods of slower earnings growth and
larger when earnings grew more quickly. By 2060, sched
uled initial benefits would be 48 percent below those pro
jected under current law, CBO estimates. The percentage
reductions in initial benefits for retired workers would be
the same for all beneficiaries in a birth cohort.

35. See Congressional Budget Office, Growing Disparities in Life
Expectancy, Issue Brief (April 17, 2008).

36. CBO projects that growth in real wages will average 1.3 percent
annually (see Congressional Budget Office, CBO’s Long Term
Projections for Social Security: 2009 Update, p. 9).

This option is similar in structure to Option 15, but
instead of reducing benefits at a fixed rate, the reduction
relative to benefits scheduled under current law would
vary each year with growth in real earnings. Under cur
rent law, average real scheduled benefits grow over time,
and the ratio of initial scheduled benefits to average earn
ings (as measured by the AIME) remains roughly con
stant. Under this option, average real benefits would
remain constant, and the ratio of initial scheduled bene
fits to the AIME would decline over time from an average
of 0.44 for people retiring in 2010 to 0.29 for people
who retire in 2060.37 This option would not affect people
who collect DI benefits, but, as under Option 16, the
decline in benefits upon conversion to OASI would be
smaller for people who had received DI benefits for a
longer period.

Social Security’s total outlays would decline by 0.9 per
centage points of GDP in 2040, or by 14 percent from
currently scheduled outlays. The savings would continue
to grow. This option would improve the 75 year actuarial
balance by 1.0 percentage point of GDP and would result
in long term sustainable solvency.

For people born in the 1980s, scheduled lifetime benefits
would decline by about 30 percent; later cohorts would
face bigger reductions. Payable lifetime benefits also
would be lower than those under current law, but the
cuts would not be as large as under scheduled benefits
because payable benefits are lower to begin with.

Option 18: Lower Initial Benefits for the Top
70 Percent of Earners
This option, often called “progressive price indexing,” is
similar to Option 17. However, scheduled benefits for
people in the bottom 30 percent of lifetime average earn
ings would not change relative to current law, and the
reductions for people at the high end of the earnings dis
tribution would be greater than those for people closer to
the low end. Under this option, the PIA factors applicable
to the top 70 percent of earners would be gradually
reduced so that initial benefits for such earners would
decline over time relative to those scheduled under current
law. Beginning in 2017, initial benefits for newly eligible
Social Security beneficiaries who earned the taxable maxi
mum for 35 years—“maximum earners”—would increase
from year to year with prices (as in Option 17) rather
than with prices and real earnings. For beneficiaries

37. See Congressional Budget Office, Budget Options, Volume 2,
p. 143.

CBO

http://www.cbo.gov/ftpdocs/91xx/doc9104/04-17-LifeExpectancy_Brief

http://www.cbo.gov/ftpdocs/102xx/doc10294/08-06-BudgetOptions

http://www.cbo.gov/ftpdocs/104xx/doc10457/08-07-SocialSecurity_Update

24 SOCIAL SECURITY POLICY OPTIONS

CBO

whose lifetime earnings were between those two groups—
above the 30th percentile but below the taxable maxi
mum—average initial benefits would increase more rap
idly than prices but more slowly than earnings.38 This
option would not affect people who collect DI benefits,
but, as under Option 16, the decline in benefits upon
conversion to OASI would be smaller for people who had
received DI benefits for a longer time.

This option would be implemented by adding a third
bend point to the PIA formula in 2017, initially set at the
30th percentile (encompassing the lowest 30 percent) of
earners. In 2040, the new bend point would be at about
$2,560, between the first bend point at $1,130 and the
highest bend point at $6,830 (see Figure 7). Between the
first bend point and the new one, the PIA factor would
remain at 32 percent. The PIA factors in the next two
brackets initially would be 32 percent and 15 percent,
but they would be reduced annually—at the rate needed
to keep a maximum earner’s initial benefits growing with
prices.

The top two PIA factors would fall to zero by about
2080, CBO projects, when monthly benefits for a worker
with earnings at the new bend point (which would have
increased at the rate of earnings growth) would be about
$2,560, roughly the same as the benefits paid to maxi
mum earners (which would have increased at the rate of
prices). Thereafter, scheduled initial benefits for all newly
retired beneficiaries would increase with earnings, but
benefits would effectively be capped at the amount
received by workers with earnings at the new bend point.

Social Security’s total outlays would decline by 0.4 per
centage points of GDP in 2040, or by 7 percent from
currently scheduled outlays. This option would improve
the 75 year actuarial balance by 0.5 percentage points of
GDP but would extend the trust fund exhaustion date
only by five years, to 2044, because it would be phased in
slowly.

Lifetime scheduled benefits for low earners would be
essentially unchanged; those for high earners born in the
1980s would decline by approximately 30 percent; later
cohorts would face bigger reductions. Lifetime payable
benefits would be similar to those paid under current law

38. Ibid.

for people in the middle of the earnings distribution but
reduced for high earners and increased for low earners.

Option 19: Lower Initial Benefits for the
Top 50 Percent of Earners
This option differs from Option 18 in that benefits
would remain as scheduled under current law for benefi
ciaries in the bottom 50 percent of career average earn
ings, rather than for those in the bottom 30 percent.
Under this option, the PIA factors applicable to the top
50 percent of earners would be gradually reduced so that
initial benefits would decline over time relative to those
scheduled under current law. The change would be
achieved by adding a bend point to the PIA formula
between the first and second bend points. In 2017, the
new bend point would be set initially at the 50th percen
tile of the lifetime earnings distribution, which CBO
estimates would be 74 percent of the way between the
original first and second bend points.

The top two factors would fall to zero in 2057, CBO
estimates. The top two factors reach zero earlier than in
Option 18 because the new second bend point occurs at a
higher level of earnings. At that time, benefits for a
worker with earnings at the new bend point (which
would have increased at the rate of earnings) would equal
the benefits received by maximum earners (which would
have increased at the rate of prices). Thereafter, scheduled
initial benefits for retired beneficiaries would be indexed
to earnings, but those benefits would effectively be
capped at the new bend point. This option would not
affect people collecting DI benefits, but their benefits
would decline upon conversion to OASI. That reduction
would be smaller for people who had received DI benefits
for a longer period, as under Option 16.

Social Security’s total outlays would decline by 0.4 per
centage points of GDP in 2040, or by 6 percent from
currently scheduled outlays. This option would improve
the 75 year actuarial balance by 0.4 percentage points of
GDP and would extend the trust fund exhaustion date by
four years, to 2043. By 2060, the reduction in outlays
would be about 40 percent of what it would be with
indexing to prices (as in Option 17) and about
80 percent of the amount it would be with progressive
price indexing for the top 70 percent of earners (as in
Option 18).

SOCIAL SECURITY POLICY OPTIONS 25

Figure 7.

Calculating Initial Benefits with

Progressive Price Indexing

(PIA in 2010 dollars)

Source: Congressional Budget Office.

Notes: Progressive price indexing is the subject of Option 18: Lower Initial Benefits for the Top 70 Percent of Earners. In 2010, the bend
points (represented by the dots on the lines) under current law are $761 and $4,586. In 2040, the bend points (in 2010 dollars) are
$1,130 and $6,830 under current law and $1,130, $2,560, and $6,830 with progressive price indexing. In 2080, the bend points (in
2010 dollars) are $1,890 and

$11,380

under current law and $1,890, $4,270, and $11,380 with progressive price indexing. A bend
point is the threshold at which a PIA factor changes.

PIA = primary insurance amount; AIME = average indexed monthly earnings.
0
1,000
2,000
3,000
4,000
5,000
6,000
0 3,000 6,000 9,000 12,000 15,000

AIME in 2010 Dollars

0
1,000
2,000
3,000
4,000
5,000
6,000
0 3,000 6,000 9,000 12,000 15,000
AIME in 2010 Dollars

2040

2080
Current Social Security System
Current Social Security System
Progressive Price Indexing
Progressive Price Indexing

$1,130

$1,130

$6,830

$6,830

$2,560

$1,890

$1,890

$11,380

$4,270

$11,380
CBO

26 SOCIAL SECURITY POLICY OPTIONS

CBO

Scheduled lifetime benefits for middle earners born in the
2000s would decrease by about 9 percent; those for high
earners would decline by around 36 percent. Payable
benefits would be reduced for high earners but increased
somewhat for low earners who receive benefits in 2039
or later.

Option 20: Index the Bend Points in the
PIA Formula to Prices
Under this option, beginning in 2017, the bend points in
the formula that determines the primary insurance
amount would be indexed to prices rather than to average
earnings, as they are under current law. (Because this
option would change the bend points, it differs signifi
cantly from Options 17, 18, and 19, which would mod
ify the PIA factors.) This option would apply to newly
eligible retired and disabled workers. (As under current
law, workers’ earnings would still be indexed to nominal
earnings in the computation of the AIME.)

CBO projects that annual growth in real earnings will
average 1.3 percent, so the bend points would increase
1.3 percent more slowly under this option than under
current law. By 2040, the bend points would be almost
30 percent lower than they would be under current law
(see Figure 8). By 2080, they would be almost 60 percent
lower. The bend points would have remained the same in
real terms (about $820 and

$4,960

in 2010 dollars) from
2017 to 2080, whereas the bend points under current law
would have increased markedly, to about $1,890 and
about $11,380 (in 2010 dollars). In 2080, the PIA under
current law would be about $1,700 for a worker with an
AIME of $2,000; under this option, the PIA would be
$1,100, or 35 percent less.

Social Security’s total outlays would decline by 0.4 per
centage points of GDP in 2040, or by 7 percent from
currently scheduled outlays. This option would improve
the 75 year actuarial balance by 0.5 percentage points of
GDP, but it would extend the trust fund exhaustion date
only by five years, to 2044, because it would be phased in
slowly.

Lifetime scheduled benefits for low earners born in the
2000s would decrease by about 18 percent; for high
earners they would decrease by approximately 24 percent.
Payable lifetime benefits would be about the same as
under current law.

Option 21: Index Earnings in the AIME and
Bend Points in the PIA Formula to Prices
This option would combine Options 11 and 20 by
switching from wage indexing to price indexing for com
puting average indexed monthly earnings and for calcu
lating the bend points in the formula for the primary
insurance amount. Beginning in 2017, the option would
apply to newly eligible retired and disabled workers.

Social Security’s total outlays would decline by 0.5 per
centage points of GDP in 2040, or by 8 percent, from
currently scheduled outlays. This option would improve
the 75 year actuarial balance by 0.6 percentage points of
GDP. Even though it would result in a positive 75 year
actuarial balance, the benefit reductions would be rela
tively small in the first few decades. This option therefore
would extend the trust fund exhaustion date only by
seven years, to 2046.

Lifetime scheduled benefits for low earners born in the
2000s would be reduced by around 27 percent; for high
earners, those benefits would decline by about 30 percent.
Lifetime scheduled benefits for people born earlier also
would decline but by a smaller proportion. Payable bene
fits generally would be lower than those under current
law.

Option 22: Replace the Current PIA Formula with a
New Two-Part Formula
Beginning in 2017, this option would introduce a new
formula for calculating the primary insurance amount for
newly eligible retired workers. The PIA would equal the
sum of two amounts: The first would provide each
worker with a benefit based on the number of years of
work, and the second would provide each worker with
additional benefits proportional to the worker’s average
indexed monthly earnings.

The first part of the new formula would provide a fixed
amount of benefits for every quarter of coverage accumu
lated by a worker, regardless of earnings. As under current
law, a retired worker would need at least 40 quarters of
coverage to receive benefits. A newly eligible retired
worker in 2017 would receive a monthly benefit of $6 for
each quarter up to 160 quarters of coverage accumulated.
The amount would increase over time at the rate of aver
age wage growth. The second part of the formula would
provide additional benefits proportional to earnings, cal
culated as 15 percent of the AIME.

SOCIAL SECURITY POLICY OPTIONS 27

Figure 8.

Calculating Initial Benefits with Indexing of Bend Points to Prices
(PIA in 2010 dollars)

Source: Congressional Budget Office.

Notes: Price indexing of the bend points (represented by the dots on the lines) is the subject of two options, Option 20: Index the Bend Points
in the PIA Formula to Prices and Option 21: Index Earnings in the AIME and Bend Points in the PIA Formula to Prices. In 2010, the
bend points under current law are $761 and $4,586. In 2040, the bend points (in 2010 dollars) are $1,130 and $6,830 under current
law and $820 and $4,960 with price indexing. In 2080, the bend points (in 2010 dollars) are $1,900 and $11,380 under current law
and $820 and $4,960 with price indexing. A bend point is the threshold at which a PIA factor changes.

PIA = primary insurance amount; AIME = average indexed monthly earnings.
0
1,000
2,000
3,000
4,000
5,000
6,000
0 3,000 6,000 9,000 12,000 15,000
AIME in 2010 Dollars
2040
Current Social Security System

Bend Points Indexed to Prices

$1,130

$820

$6,830

$4,960

0
1,000
2,000
3,000
4,000
5,000
6,000
0 3,000 6,000 9,000 12,000 15,000
AIME in 2010 Dollars
2080
Current Social Security System
Bend Points Indexed to Prices
$820
$1,890
$11,380
$4,960
CBO

28 SOCIAL SECURITY POLICY OPTIONS

CBO

Under this option, in 2017, a worker with 160 quarters
of coverage and an AIME of $5,000 would have a PIA of
$1,710. Of that amount, $960 would be attributable to
the 160 quarters of coverage and $750 would be 15 per
cent of the AIME. Under current law, in 2017, that
worker would have a PIA of $2,150 (in 2017 dollars).

Social Security’s total outlays would decline by 5 percent
in 2040 under this option, or by 0.3 percentage points of
GDP relative to current law. The option would improve
the 75 year actuarial balance by 0.2 percentage points of
GDP and would extend the trust fund exhaustion date by
six years, to 2045.

Scheduled lifetime benefits would decrease by roughly
9 percent for high earners. Scheduled lifetime benefits for
low earners would decrease for people born in the 1960s
but increase somewhat for people born later (because ear
lier cohorts of low earners have fewer years of work).
Payable lifetime benefits would be slightly higher for low
earners in later birth cohorts.

Options That Would Increase
Benefits for Low Earners
One goal of Social Security is to ensure an adequate
income for beneficiaries. In 2008, 8 percent of all Social
Security beneficiaries over the age of 65 were considered
poor. Only 3 percent of married beneficiaries over 65
were poor, but 14 percent of beneficiaries who were not
currently married and 16 percent of never married bene
ficiaries in the same age group were poor. One reason for
the higher poverty rates among those groups is that bene
ficiaries who never married (or, if divorced, who had been
married for less than 10 years) are not eligible to receive
auxiliary benefits for widows, widowers, or divorcees. A
minimum benefit for workers could help provide a larger
income for those beneficiaries and others. The options in
this section would increase worker benefits for some peo
ple who worked and contributed to Social Security for
many years yet had low average annual earnings and thus
would receive low Social Security benefits under current
law. (The options would not affect the benefits received
by married people with low earnings if their own worker
benefits remained less than half of their spouse’s benefits.)

Because the options in this group are based on earnings
during a working lifetime, they would not necessarily

benefit people who have low income during retirement.
In addition, the options would not distinguish between
workers who had low annual earnings because they
earned low hourly wages and workers who had higher
hourly wages but worked for only part of the year. (See
Table 2 on page 33 for the effects of the options on Social
Security’s finances, Table 3 on page 39 for effects on dis
tributional outcomes assuming that scheduled benefits
are paid, and Table 4 on page 43 for effects on distribu
tional outcomes assuming that only payable benefits are
paid.)

Option 23: Modify the Special Minimum Benefit and
Index It to Growth in Earnings
Current law includes a special minimum benefit, which
currently can be as much as $763 a month, about
85 percent of the federal poverty guideline for an individ
ual.39 Beneficiaries receive the larger of the standard bene
fit or the special minimum benefit. The special minimum
benefit was created to increase payments to people who
had low earnings over a long working lifetime. However,
the benefit is indexed to prices, whereas regular Social
Security benefits are indexed to earnings—and because
earnings have grown faster than prices, regular Social
Security benefits have increased in real terms, and few
beneficiaries now qualify for the special minimum bene
fit. After 2010, benefits computed under the standard
formula are projected to be greater than the special mini
mum benefit for all new beneficiaries.

This option would restructure the primary insurance
amount used in calculating the special minimum benefit,
basing it on the number of years of qualifying work but
not varying it otherwise with earnings. The effect would
be to increase the benefit as the number of years of work
(above the 10 needed to qualify for Social Security bene
fits) increased, up to 30 years. For example, in 2012, the
monthly benefit for a single person age 65 or older in that
year who has worked for 30 years or more would be
about $1,170, which CBO projects would equal 125 per
cent of the poverty guideline. For someone with fewer
than 30 years of work, the special minimum benefit
would be reduced proportionately, down to zero for

39. See Kelly A. Olsen and Don Hoffmeyer, “Social Security’s Special
Minimum Benefit,” Social Security Bulletin, vol. 64, no. 2 (2001/
2002).

SOCIAL SECURITY POLICY OPTIONS 29

people with 10 or fewer years of work.40 In 2013 and
later, the dollar thresholds would increase at the same rate
as average earnings.

Social Security’s total outlays under this option would
increase by 0.2 percentage points of GDP in 2040, or by
4 percent above currently scheduled outlays. This option
would worsen the 75 year actuarial balance by 0.2 per
centage points of GDP, and the trust funds would be
exhausted two years earlier, in 2037. Payable benefits
would generally be higher for the affected low earners,
but high earners would be affected by the earlier trust
fund exhaustion date and by the larger gap between
outlays and revenues thereafter.

Under this option, scheduled lifetime benefits for low
earners born in the 2000s would increase by about
30 percent. In 2040, about 50 percent of new OASI ben
eficiaries and about 35 percent of new DI beneficiaries
would have higher initial benefits; about 45 percent of
the group would be women.

Option 24:

Introduce a New Poverty-Related
Minimum Benefit

This option would introduce a new benefit for workers
who have relatively low earnings over a long period. For
someone with 20 years of earnings, the minimum PIA
would typically be 80 percent of the poverty guideline for
a single person age 65 or older in 2016. (In most of the
country, the 2009 poverty guideline for a single person
was $10,830.) With 40 years of earnings, the amount
would be 120 percent of the poverty guideline.41 Those
PIAs would amount to $720 and $1,080, respectively, in
2016 (in 2010 dollars). Beneficiaries would receive the
higher of the regular benefit or the new minimum
benefit. This option’s minimum benefits would be higher
than those under Option 23 for workers with

40. Under this option, a year of coverage for the special minimum
benefit is defined as a year in which a worker earns four quarters
of coverage. Years of coverage would accumulate after 10 years of
coverage but not increase beyond 30 years. That is, someone who
worked 30 or more years would be credited with benefits based on
20 years of coverage. The additional PIA per year of coverage
above 10 years would be 1/20 of the benefit for a 30 year worker,
or $58.50 in 2012. So, for example, the minimum monthly
benefit for a worker with 15 years of coverage would be about
$293 (5 × $58.50).

11 to 24 years of earnings and lower for those with 25 or
more years of earnings; neither option would provide a
minimum benefit to workers with 10 or fewer years of
earnings.

Social Security’s total outlays would increase by less than
0.05 percentage points of GDP in 2040, or by less than
1 percent from currently scheduled outlays. This option
would worsen the 75 year actuarial balance by less than
0.05 percentage points of GDP and would not signifi
cantly change the trust fund exhaustion date.

Scheduled and payable lifetime benefits for low earners
would be roughly 6 percent higher than under current
law. In 2040, about 13 percent of new OASI beneficiaries
and about 18 percent of new DI beneficiaries would
receive higher initial benefits than under current law;
about 55 percent of those beneficiaries would be women.

Option 25: Enhance Low-Earners’ Benefits on the
Basis of Years Worked
Under this option, beginning in 2012, benefits would
increase for workers who have both low lifetime average
earnings and at least 20 years of covered earnings. This
option would raise the standard benefit for qualified
workers by a specified percentage that would depend on
the number of years worked and a worker’s AIME. The
largest benefit increase would be 40 percent for someone
with 35 or more years in the workforce and an AIME at
or below the AIME of someone who had worked full
time and earned the minimum wage for 30 years. The
benefit increase would be smaller for people with fewer
years of work or higher AIMEs, and there would be no
increase for people whose AIME was above that of a
worker who had worked for 35 years or who always
earned an amount equal to or greater than the average

41. To qualify for the new benefit, a beneficiary would need to have
worked at least 10 years. Specifically, the minimum PIA would be
2 percent of the poverty guideline for each quarter of coverage
above 40 (10 years of earnings) and up to 80 quarters of coverage,
and 0.5 percent of the poverty guideline for quarters of coverage
above 80 and up to 160. (For disabled workers, fewer quarters
would be required because of their shortened careers.) This new
minimum benefit would be phased in from 2012 to 2016.
Beginning in 2016, the effective poverty guidelines would increase
with average wages.

CBO

30 SOCIAL SECURITY POLICY OPTIONS

CBO

wage index (AWI, the average amount of total earnings in
the United States in a year).42

Social Security’s total outlays in 2040 would increase
by 0.4 percentage points of GDP, or by 7 percent from
currently scheduled outlays. This option would worsen
the 75-year actuarial balance by 0.3 percentage points of
GDP, and the trust funds would be exhausted in 2034,
five years earlier than CBO anticipates otherwise. Payable
benefits for high earners would be somewhat lower than
under current law because of the earlier exhaustion of the
trust funds and the larger gap between outlays and
revenues thereafter.

Under this policy, scheduled lifetime benefits for low
earners born in the 2000s would increase by about
24 percent. For people in the same cohort and in
the middle of the household earnings distribution,
scheduled lifetime benefits would increase by 12 percent;
they would receive larger benefits because many of them
would have an AIME that was less than that of a worker
who worked for 35 years and always earned an amount
equal to the AWI. Under this option, about 57 percent of
new OASI beneficiaries and about 65 percent of DI ben-
eficiaries would receive an increase in their initial benefits
in 2040; about 53 percent of those beneficiaries would be
women.

42. The standard benefit would be multiplied by 1 + (40 percent ×
AIME factor × coverage factor). The two factors range from 0 to
1, so this option would increase benefits by as much as 40 percent.
The AIME factor would be 1 for workers with an AIME equal to
or less than the AIME of a full-time worker who earned the
minimum wage for 30 years. It would be zero for workers with an
AIME greater than the AIME of a worker who worked for
35 years, always earning an amount equal to the AWI. For workers
with earnings between those amounts the factor would be set
proportionately: AIME factor for a given worker = (AIME of
35-year AWI worker – AIME of given worker)/(AIME of 35-year
AWI worker – AIME of minimum wage worker). The coverage
factor would give a larger increase to workers with more quarters
of coverage. For most retired workers the factor would be 1 if the
worker had at least 35 years in covered employment. It would be
zero if the worker had 20 years or less in covered employment. For
workers whose employment was between 20 and 35 years, the
formula would be as follows: Coverage factor = the minimum of
1 or 1 – {[(3.5 × elapsed years) – quarters of coverage]/(1.5 ×
elapsed years)}. “Elapsed years” would be set to 40 for retired
workers or equal to the number of years from age 22 to the age of
entitlement for disabled workers (see Congressional Budget
Office, Budget Options, Volume 2, p. 154).

Options That Would Raise the
Full Retirement Age
People who turn 65 today will, on average, collect Social
Security benefits for significantly longer than retirees did
in the past because the average life span in the United
States has lengthened considerably. In 1940, life expec-
tancy at age 65 was 11.9 years for men and 13.4 years for
women. The Social Security trustees project that life
expectancy has increased by more than 5 years for 65-
year-olds today, to 17.0 years for men and 19.4 years for
women, and that those figures will increase to 18.7 years
and 20.8 years by 2035. Therefore, a commitment to
provide people with a specific monthly benefit for the rest
of their lives would be more costly if made to those who
will be 65 in 2035 than to 65-year-olds today.

Increasing the full retirement age is, in most ways, equiv-
alent to cutting initial benefits. In particular, for people
who claim benefits at any given age, a higher FRA results
in lower benefits. (The options presented here would not
change the early eligibility age, which under current law
is fixed at 62 for retired workers.) Depending on the age
at which the worker claims his or her benefits, a one-year
increase in the FRA is equivalent to a reduction in a
retired worker’s monthly benefit of between 5 percent
and 8 percent.43 For example, under current law, benefits
would be reduced by 30 percent for someone with a FRA
of 67 (that is, someone born in 1960 or later) who
claimed benefits at age 62. The reduction would be
35 percent for the same worker if the FRA was set at 68.

The options in this section would increase the full retire-
ment age to different ages at different speeds. (See Table 2
on page 33 for the effects of the options on Social Secu-
rity’s finances, Table 3 on page 39 for effects on distribu-
tional outcomes assuming that scheduled benefits are
paid, and Table 4 on page 43 for effects on distributional
outcomes assuming that only payable benefits are paid.)

43. When a worker claims benefits before the full retirement age,
benefits are reduced by 5/9 of 1 percent for each month, or
6-2/3 percent per year, before full retirement age, up to
36 months. If the number of months exceeds 36, then the
benefit is further reduced by 5/12 of 1 percent per month, or
5 percent per year. People who claim benefits after reaching
their FRA generally receive a delayed-retirement credit, which is
8 percent for those born in 1943 and later. No additional credit
is given after a person turns 70.

http://www.cbo.gov/ftpdocs/102xx/doc10294/08-06-BudgetOptions

SOCIAL SECURITY POLICY OPTIONS 31

Option 26: Raise the FRA to 68
This option would continue to increase the full retire
ment age after it reaches 67 in 2022 under current law.
Specifically, the FRA would rise by an additional two
months per birth year for another six years, reaching 68
for workers born in 1966, who will turn 62 in 2028.

Social Security’s total outlays would decline by 0.2 per
centage points of GDP in 2040, or by 3 percent from
currently scheduled outlays. This option would improve
the 75 year actuarial balance by 0.1 percentage point of
GDP and would not significantly extend the trust fund
exhaustion date.

After this option was fully phased in, scheduled lifetime
benefits for people born in the 1980s and 2000s would
be reduced by about 6 percent relative to current law.
Payable benefits would not change significantly.

Option 27: Raise the FRA to 70
Like Option 26, this option would continue to increase
the full retirement age after it reaches 67 in 2022 under
current law, but this option would ultimately make the
age of full retirement later than would Option 26. Under
this option, the FRA would rise by an additional two
months per birth year for another 18 years, reaching age
70 for workers who were born in 1978 and who will turn
62 in 2040.44 That change, relative to Social Security’s
original FRA of 65, would roughly match the increase in
life expectancy that has occurred since 1940.

Under this option, Social Security’s total outlays would
decline by 0.4 percentage points of GDP in 2040, or by
6 percent from currently scheduled outlays. This option
would improve the 75 year actuarial balance by 0.3 per
centage points of GDP and would not significantly
extend the trust fund exhaustion date.

After this option was fully phased in, scheduled lifetime
benefits for people born in the 1980s and 2000s would
be reduced by about 15 percent relative to current law.
Payable benefits would decline by smaller percentages.

Option 28: Index the FRA to Changes in Longevity
This option would maintain a constant ratio of projected
years of benefit receipt to years of work—that is, the ratio

44. CBO has examined the effects of increasing the FRA to 70 for
people born in 1971 and then by 1 month every second year. See
Congressional Budget Office, Budget Options, Volume 2, p. 145.

of life expectancy at the full retirement age to the number
of years from age 21 to the full retirement age would be
held constant.45 Under current law, the FRA will be 67
starting in 2022. Under Options 26 and 27, the FRA
would increase at a rate of two months per birth year.
Under this option, the increase would be more gradual:
The FRA would rise by approximately half a month per
birth year, and the rate of increase would vary depending
on the actual rate of mortality improvement. Under this
approach, CBO estimates, the FRA for people born in
1979 would be 68, and the FRA for people born in 2003
would be 69. The FRA would reach 70 for people born
in 2026; that group would become eligible for retirement
benefits in 2088, which is beyond the 75 year projection
period for this study.

Social Security’s total outlays in 2040 would decline by
0.1 percentage point of GDP, or by just over 2 percent,
relative to currently scheduled outlays. This option would
improve the 75 year actuarial balance by 0.2 percentage
points of GDP and would not significantly extend the
trust fund exhaustion date.

Under this policy, scheduled lifetime benefits for people
born in the 2000s would be reduced by roughly 12 per
cent. Payable benefits would be reduced by a smaller
amount.

Options That Would Reduce
Cost-of-Living Adjustments
Current law requires that the benefits paid to existing
beneficiaries generally rise each year with the application
of a cost of living adjustment. At the end of each year,
the Social Security Administration adjusts each benefi
ciary’s PIA by an amount that is equal to any increase in
the consumer price index for urban wage earners and
clerical workers from the third calendar quarter of the
prior year to the third calendar quarter of the current
year. (When prices decline, the COLA is set at zero, as
occurred in 2010.)

Many analysts believe that the CPI W overstates increases
in the cost of living because it does not fully account for
the fact that consumers generally adjust their spending
patterns as some prices change relative to others. Another

45. Because of the delay in the availability of mortality data, this
option would be linked to life expectancy data from three years
before the affected year.

CBO

http://www.cbo.gov/ftpdocs/102xx/doc10294/08-06-BudgetOptions

32 SOCIAL SECURITY POLICY OPTIONS

CBO

consideration, however, is that the cost of living could
grow faster for elderly people than for the rest of the pop
ulation. Inflation as measured by the CPI E, an experi
mental version of the CPI that reflects the purchasing
patterns of older people, has been 0.3 percentage points
higher than the CPI W over the past three decades.

Because most proposals to reduce the size of COLAs
would result in an annually compounding reduction dur
ing the course of a beneficiary’s receipt of benefits, the
reduction in scheduled annual benefits would grow larger
over time relative to current law. The difference in annual
benefits would be most pronounced for beneficiaries who
would receive benefits under a modified COLA option
for long periods, such as very old retirees (who are more
likely than younger retirees to be poor) and people who
begin receiving DI benefits at an early age.

A change in COLAs also would affect initial benefits
claimed after the early eligibility age (currently set at 62),
because COLAs are applied to the age 62 benefit every
year, even for people who wait until after age 62 to
claim benefits. Thus, a change in COLAs has a small
effect on initial benefits at age 65. (See Table 2 on
page 33 for the effects of the options on Social Security’s
finances, Table 3 on page 39 for effects on distributional
outcomes assuming that scheduled benefits are paid, and
Table 4 on page 43 for effects on distributional outcomes
assuming that only payable benefits are paid.)

Option 29: Reduce COLAs by 0.5 Percentage Points
This option would reduce the annual cost of living
adjustment by 0.5 percentage points. The reduction
would begin in 2012.

Social Security’s total outlays would decline by 0.4 per
centage points of GDP in 2040, or by 7 percent from
currently scheduled outlays. This option would improve
the 75 year actuarial balance by 0.3 percentage points of

GDP and would extend the trust fund exhaustion date by
nine years, to 2048.

Scheduled lifetime benefits would be reduced by about
6 percent for people born in the 1950s or later and would
be smaller for older birth cohorts. For example, lifetime
benefits would be reduced only slightly for people who
turn 90 in 2012. Payable benefits would be slightly
higher for people who will collect larger portions of their
benefits in 2039 and later.

Option 30:

Base COLAs on the Chained CPI-U

Beginning in 2012, this option would link Social Security
cost of living adjustments to another measure of infla
tion—the chained CPI U (consumer price index for all
urban consumers)—which takes into account that con
sumers generally adjust their spending patterns as some
prices change relative to others. CBO projects that the
chained CPI U will increase, on average, by 0.3 percent
age points more slowly per year than will the CPI W.46

Social Security’s total outlays would decline by 0.2 per
centage points of GDP in 2040, or by 4 percent, from
currently scheduled outlays. This option would improve
the 75 year actuarial balance by 0.2 percentage points of
GDP and would extend the trust fund exhaustion date by
four years, to 2043.

Compared with those scheduled under current law, life
time benefits for people in all earnings categories would
be reduced by about 3 percent. Payable lifetime benefits
would not change significantly.

46. The estimate of the effect of this option is based on CBO’s
projection of the difference between growth in the CPI W and
the chained CPI U, but the actual difference (and thus the effect
of the option) would vary from year to year and could average
more or less than 0.3 percent. For additional information, see
Congressional Budget Office, Using a Different Measure of
Inflation for Indexing Federal Programs and the Tax Code, Issue
Brief (February 24, 2010), and Budget Options, Volume 2, p. 147.

http://www.cbo.gov/ftpdocs/112xx/doc11256/CPI_brief

http://www.cbo.gov/ftpdocs/102xx/doc10294/08-06-BudgetOptions

SOCIAL SECURITY POLICY OPTIONS 33

Table 2.

Changes to Social Security’s Finances Under Various Options with Scheduled Benefits
(Percentage of GDP)

Continued

2020 2040 2060 2080 Annual Finances

Revenues and Outlaysb

Revenues 4.9 4.9 4.9 5.0 5.2 14.4

Outlays 5.2 6.2 6.0 6.3 5.8 16.0

Balance -0.3 -1.3 -1.1 -1.3 -0.6 -1.6

Change the Taxation of Earnings Change in Annual Balancec

1 Revenues 0.4 0.4 0.3 0.3 0.3 1.0

Outlaysd * * * * * *

Balance 0.4 0.4 0.4 0.4 0.3 1.0

2 Revenues 0.3 0.7 0.7 0.7 0.5 1.6

Outlaysd * * * * * *

Balance 0.3 0.7 0.7 0.8 0.6 1.6

3 Revenues 0.2 0.5 0.8 1.0 0.5 1.5

Outlaysd * * * * * *

Balance 0.2 0.5 0.9 1.1 0.5 1.4

4 Revenues 0.8 0.9 0.9 0.9 0.9 n.a.

Outlays * 0.3 0.5 0.5 0.3 n.a.

Balance 0.8 0.6 0.4 0.4 0.6 n.a.

5 Revenues 0.3 0.4 0.4 0.4 0.4 n.a.

Outlays * 0.1 0.2 0.2 0.1 n.a.

Balance 0.3 0.3 0.2 0.2 0.2 n.a.

Raise the Taxable
Maximum to Cover 90% of

Earningse

75-Year

Taxable

Percentage-Point Change from

Current Lawa

Increase the Payroll Tax
Rate by 1 Percentage Point
in 2012

Current Lawa
Percentage of

GDP

Present Value as a

Payroll

Eliminate the Taxable

Maximume

Increase the Payroll Tax
Rate by 2 Percentage
Points Over 20 Years

Increase the Payroll Tax
Rate by 3 Percentage
Points Over 60 years

CBO

34 SOCIAL SECURITY POLICY OPTIONS

CBO

Table 2. Continued

Changes to Social Security’s Finances Under Various Options with Scheduled Benefits
(Percentage of GDP)
Continued
2020 2040 2060 2080 Annual Finances

Change the Taxation of Earnings (Continued) Change in Annual Balancec

6 Revenues 0.8 0.9 0.9 0.9 0.8 n.a.

Outlaysd * * * * * n.a.

Balance 0.9 0.9 0.9 0.9 0.9 n.a.

7 Revenues 0.5 0.5 0.5 0.5 0.5 n.a.

Outlaysd * * * * * n.a.

Balance 0.5 0.6 0.6 0.6 0.5 n.a.

8 Revenues 0.3 0.3 0.3 0.3 0.3 n.a.

Outlaysd * * * * * n.a.

Balance 0.3 0.3 0.3 0.3 0.3 n.a.

9 Revenues 0.1 0.1 0.1 0.1 0.1 n.a.

Outlaysd * * * * * n.a.

Balance 0.1 0.1 0.1 0.1 0.1 n.a.

Change the Benefit Formula

10 Revenues * * * * * *

Outlays * -0.1 -0.1 -0.1 -0.1 -0.2

Balance * 0.1 0.1 0.1 0.1 0.2

11 Revenues * * * * * *

Outlays * -0.1 -0.4 -0.6 -0.2 -0.5

Balance * 0.1 0.4 0.5 0.2 0.5

Tax Covered Earnings Up to
$250,000; Do Not Increase

Benefitse

Tax All Earnings Above
$250,000 at 4%; Do Not

Increase Benefitse

Raise from 35 to 38 the
Years of Earnings Included
in the AIME

Tax Covered Earnings
Above the Taxable
Maximum; Do Not Increase

Benefitse
Percentage of

GDP Payroll

Percentage-Point Change from Current Lawa (Continued)

Tax All Earnings Above the
Taxable Maximum at 4%;

Do Not Increase Benefitse

Taxable
Present Value as a

Index Earnings in the AIME
Formula to Prices

SOCIAL SECURITY POLICY OPTIONS 35

Table 2. Continued
Changes to Social Security’s Finances Under Various Options with Scheduled Benefits
(Percentage of GDP)
Continued
2020 2040 2060 2080 Annual Finances

Change the Benefit Formula (Continued) Change in Annual Balancec

12 Revenues * -0.1 -0.1 -0.1 * -0.1

Outlays -0.2 -0.7 -0.9 -0.9 -0.6 -1.6

Balance 0.2 0.7 0.8 0.8 0.5 1.5

13 Revenues * -0.1 -0.1 -0.1 -0.1 -0.2

Outlays -0.2 -1.0 -1.1 -1.2 -0.8 -2.1

Balance 0.2 0.9 1.0 1.1 0.7 2.0

14 Revenues * * * * * *

Outlays * -0.1 -0.1 -0.1 -0.1 -0.2
Balance * 0.1 0.1 0.1 0.1 0.2

15 Revenues * * -0.1 -0.1 * -0.1

Outlays * -0.3 -0.8 -1.3 -0.5 -1.3

Balance * 0.3 0.8 1.2 0.4 1.2

16 Revenues * * * -0.1 * -0.1

Outlays * -0.2 -0.4 -0.7 -0.2 -0.6

Balance * 0.2 0.4 0.6 0.2 0.6

17 Revenues * -0.1 -0.2 -0.3 -0.1 -0.2

Outlays * -0.9 -1.9 -2.9 -1.1 -2.9

Balance * 0.8 1.7 2.6 1.0 2.7

Percentage of

Payroll
Taxable

GDP

75-Year
Present Value as a

Percentage-Point Change from Current Lawa (Continued)

Reduce the Top PIA Factor
by One-Third

Reduce All PIA Factors by
0.5% Annually

Reduce All PIA Factors
by 15%

Reduce the Top Two
PIA Factors by Roughly
One-Third

Reduce PIA Factors to
Index Initial Benefits to
Prices Rather Than
Earnings

Index Initial Benefits to
Changes in Longevity
CBO

36 SOCIAL SECURITY POLICY OPTIONS

CBO
Table 2. Continued
Changes to Social Security’s Finances Under Various Options with Scheduled Benefits
(Percentage of GDP)
Continued
2020 2040 2060 2080 Annual Finances
Change the Benefit Formula (Continued) Change in Annual Balancec

18 Revenues * * -0.1 -0.1 -0.1 -0.1

Outlays * -0.4 -1.0 -1.5 -0.5 -1.5

Balance * 0.4 0.9 1.4 0.5 1.4

19 Revenues * * -0.1 -0.1 0 -0.1

Outlays * -0.4 -0.8 -1.1 -0.4 -1.2

Balance * 0.3 0.7 1.0 0.4 1.1

20 Revenues * * -0.1 -0.1 * -0.1

Outlays * -0.4 -0.9 -1.5 -0.5 -1.5

Balance * 0.4 0.9 1.4 0.5 1.3

21 Revenues * * -0.1 -0.2 -0.1 -0.1

Outlays * -0.5 -1.2 -1.9 -0.7 -1.9

Balance * 0.5 1.1 1.7 0.6 1.7

22 Revenues * * * * * *

Outlays -0.1 -0.3 -0.2 -0.3 -0.2 -0.6

Balance 0.1 0.3 0.2 0.2 0.2 0.5

Increase Benefits for Low Earners

23 Revenues * * * 0.1 * 0.1

Outlays 0.1 0.2 0.5 0.5 0.3 0.7

Balance -0.1 -0.2 -0.4 -0.5 -0.2 -0.7

Earnings

Lower Initial Benefits for
the Top 70% of Earners

75-Year
Present Value as a

Percentage of
Taxable

GDP Payroll
Percentage-Point Change from Current Lawa (Continued)

Index Earnings in the AIME
and Bend Points in the PIA
Formula to Prices

Modify the Special
Minimum Benefit and
Index It to Growth in

Replace the Current
PIA Formula with a New
Two-Part Formula

Index the Bend Points in
the PIA Formula to Prices

Lower Initial Benefits for
the Top 50% of Earners

SOCIAL SECURITY POLICY OPTIONS 37

Table 2. Continued
Changes to Social Security’s Finances Under Various Options with Scheduled Benefits
(Percentage of GDP)
Continued
2020 2040 2060 2080 Annual Finances

Increase Benefits for Low Earners (Continued) Change in Annual Balancec

24 Revenues * * * * * *

Outlays * * 0.1 * * 0.1

Balance * * * * * -0.1

25 Revenues * * * * * 0.1

Outlays 0.2 0.4 0.5 0.5 0.4 1.0

Balance -0.2 -0.4 -0.4 -0.4 -0.3 -0.9

Raise the Full Retirement Age

26 Revenues 0 * * * * *

Outlays 0 -0.2 -0.3 -0.3 -0.2 -0.4

Balance 0 0.2 0.2 0.2 0.1 0.4

27 Revenues 0 * -0.1 -0.1 * -0.1

Raise the FRA to 70f Outlays 0 -0.4 -0.7 -0.7 -0.4 -1.0

Balance 0 0.3 0.6 0.6 0.3 0.9

28 Revenues 0 * * -0.1 * -0.1

Outlays 0 -0.1 -0.4 -0.6 -0.2 -0.5

Balance 0 0.1 0.3 0.5 0.2 0.5

GDP Payroll
Percentage-Point Change from Current Lawa (Continued)

Raise the FRA to 68f

Index the FRA to Changes

in Longevityf

75-Year
Present Value as a
Percentage of
Taxable

Introduce a New Poverty-
Related Minimum Benefit

Enhance Low-Earners’
Benefits on the Basis of
Years Worked

CBO

38 SOCIAL SECURITY POLICY OPTIONS

CBO
Table 2. Continued
Changes to Social Security’s Finances Under Various Options with Scheduled Benefits
(Percentage of GDP)
Source: Congressional Budget Office.

Notes: Scheduled benefits are full benefits as calculated under current law, regardless of the amounts available in the Social Security trust
funds.

The 75-year period is 2010 through 2084. Revenues consist of payroll taxes and income taxes on benefits (but not interest credited to
the trust funds) in the specified year. Outlays consist of Social Security benefits and administrative costs. The balance is the surplus or
deficit, which is the difference between revenues and outlays. Details of specific options are contained in the text.

GDP = gross domestic product; AIME = average indexed monthly earnings; PIA= primary insurance amount;
FRA = full retirement age; COLA =cost-of-living adjustment; CPI-U = consumer price index for all urban consumers;
* = between -0.05 and 0.05 percentage points, but not exactly zero; 0 = exactly zero, with no rounding; n.a. = not applicable.

a. “Current law” refers to current Social Security provisions for calculating benefits and payroll taxes. See Congressional Budget Office,
The Long-Term Budget Outlook (June 2010).

b. The line graph shows projected revenues (lower line) and outlays (upper line) as a percentage of GDP over the period from 2020 to 2080.
The range is from 3.5 percent to 7.0 percent of GDP.

c. The area graphs depict the change in the annual trust fund balance over the period from 2020 to 2080. The range is from -1.0 percent to
3.0 percent of GDP.

d. For options that would increase payroll taxes but not benefits, the ratio of outlays to GDP would be slightly reduced because GDP would
increase slightly. Although CBO’s model generally keeps GDP growth steady, adjustments in response to reduced budget deficits occur
with a lag, allowing small variations.

e. Under this option, the size of the tax base would change, so the changes in the 75-year present values of revenues, outlays, and the
actuarial balance are more clearly represented as a percentage of GDP.

f. Because this option would take effect after the FRA reaches age 67 under current law in 2022, no system finance changes are reported for
2020.

2020 2040 2060 2080 Annual Finances

Reduce Cost-of-Living Adjustments Change in Annual Balancec

29 Revenues * * * -0.1 * -0.1

Outlays -0.2 -0.4 -0.4 -0.4 -0.3 -0.8

Balance 0.2 0.4 0.4 0.4 0.3 0.8

30 Revenues * * * * * *

Outlays -0.1 -0.2 -0.2 -0.3 -0.2 -0.5

Balance 0.1 0.2 0.2 0.2 0.2 0.5

Reduce COLAs by
0.5 Percentage Points

Base COLAs on the Chained
CPI-U

75-Year
Present Value as a
Percentage of
Taxable
GDP Payroll
Percentage-Point Change from Current Lawa (Continued)

http://www.cbo.gov/doc.cfm?index=10457

http://www.cbo.gov/ftpdocs/115xx/doc11559/

SOCIAL SECURITY POLICY OPTIONS 39

Table 3.

Changes to Social Security’s Scheduled Benefits and Payroll Taxes for
Different Groups Under Various Options

Continued

Lifetime
Household
Earnings
Quintilea 1960 1980 2000 1960 1980 2000 1960 1980 2000

Low 11 14 18 110 130 180 90 110 140
Middle 20 24 32 250 300 420 320 370 490

High 31 38 50 400 500 680 670 780 1,070

Change the Taxation of Earnings

1 Low 0 0 0 0 0 0 3 6 6
Middle 0 0 0 0 0 0 3 6 9

High 0 0 0 0 0 0 3 6 6

2 Low 0 0 0 0 0 0 3 6 15
Middle 0 0 0 0 0 0 3 9 15

High 0 0 0 0 0 0 3 12 15

3 Low 0 0 0 0 0 0 * 3 12
Middle 0 0 0 0 0 0 * 6 15

High 0 0 0 0 0 0 3 6 15

4 Low * * * * * * * * *
Middle * * * * * * * * *

High 3 6 9 9 15 15 12 15 18

5 Low * * * * * * * * *
Middle * * * * * * * * *

High 3 6 6 3 9 9 6 12 15

6 Low 0 0 0 0 0 0 * * *
Middle 0 0 0 0 0 0 * * *

High 0 0 0 0 0 0 12 15 18

7 Low 0 0 0 0 0 0 * * *
Middle 0 0 0 0 0 0 * * 3

High 0 0 0 0 0 0 12 15 18

8 Low 0 0 0 0 0 0 * * *
Middle 0 0 0 0 0 0 * * *

High 0 0 0 0 0 0 6 6 9

Median Initial Benefits

for Retired Workers by Benefits by 10-Year Payroll Taxes by

Median Lifetime Median Lifetime

10-Year Birth Cohortb

Birth Cohortc 10-Year Birth Cohortc

Increase the Payroll Tax Rate by
1 Percentage Point in 2012

Eliminate the Taxable Maximum

Current Lawd (Thousands of 2010 dollars)

Increase the Payroll Tax Rate by
2 Percentage Points Over 20 Years

Increase the Payroll Tax Rate by
3 Percentage Points Over 60 Years

Percentage Change from Current Lawd

Raise the Taxable Maximum to
Cover 90% of Earnings

Tax Covered Earnings Above the
Taxable Maximum; Do Not Increase
Benefits

Tax Covered Earnings Up to
$250,000; Do Not Increase Benefits

Tax All Earnings Above the Taxable
Maximum at 4%; Do Not Increase
Benefits

CBO

40 SOCIAL SECURITY POLICY OPTIONS

CBO

Table 3. Continued

Changes to Social Security’s Scheduled Benefits and Payroll Taxes for
Different Groups Under Various Options
Continued
Lifetime
Household
Earnings
Quintilea 1960 1980 2000 1960 1980 2000 1960 1980 2000

Change the Taxation of Earnings (Continued)

9 Low 0 0 0 0 0 0 * * *
Middle 0 0 0 0 0 0 * * *

High 0 0 0 0 0 0 3 3 3

Change the Benefit Formula

10 Low -3 -3 -3 -3 -3 -3 0 0 0
Middle -3 -3 -3 -3 -3 -3 0 0 0

High -3 -3 -3 -3 -3 -3 0 0 0

11 Low * -6 -12 * -6 -12 0 0 0
Middle * -9 -12 * -9 -12 0 0 0

High * -6 -6 * -3 -6 0 0 0

12 Low -15 -15 -15 -12 -12 -12 0 0 0
Middle -15 -15 -15 -12 -15 -15 0 0 0

High -15 -15 -15 -15 -12 -15 0 0 0

13 Low -3 -3 -3 -6 -3 -3 0 0 0
Middle -18 -18 -18 -18 -18 -18 0 0 0

High -24 -24 -24 -24 -24 -24 0 0 0

14 Low * * * * * * 0 0 0
Middle * * * * * * 0 0 0

High -3 -3 -3 -6 -6 -6 0 0 0

15 Low -6 -15 -21 -3 -12 -18 0 0 0
Middle -6 -15 -21 -3 -15 -21 0 0 0

High -6 -15 -21 -6 -12 -21 0 0 0

16 Low -3 -9 -12 -3 -6 -12 0 0 0
Middle -3 -9 -12 -3 -9 -12 0 0 0

High -3 -9 -12 -3 -9 -12 0 0 0

17 Low -15 -33 -48 -12 -27 -39 0 0 0
Middle -15 -33 -48 -12 -30 -45 0 0 0

High -15 -33 -48 -12 -33 -48 0 0 0

Reduce the Top PIA Factor by
One-Third

Reduce All PIA Factors by 0.5%
Annually

Index Earnings in the AIME
Formula to Prices

Reduce All PIA Factors by 15%

Tax All Earnings Above $250,000 at
4%; Do Not Increase Benefits

Raise from 35 to 38 the Years of
Earnings Included in the AIME

Reduce the Top Two PIA Factors by
Roughly One-Third

Median Initial Benefits
for Retired Workers by

Index Initial Benefits to Changes in
Longevity

Reduce PIA Factors to Index Initial
Benefits to Prices Rather Than
Earnings

Median Lifetime Median Lifetime

Birth Cohortc 10-Year Birth Cohortc
Benefits by 10-Year Payroll Taxes by

Percentage Change from Current Lawd (Continued)

10-Year Birth Cohortb

SOCIAL SECURITY POLICY OPTIONS 41

Table 3. Continued
Changes to Social Security’s Scheduled Benefits and Payroll Taxes for
Different Groups Under Various Options
Continued
Lifetime
Household
Earnings
Quintilea 1960 1980 2000 1960 1980 2000 1960 1980 2000

Change the Benefit Formula (Continued)

18 Low * * * * * * 0 0 0
Middle -6 -15 -21 -6 -15 -21 0 0 0

High -12 -30 -45 -12 -30 -45 0 0 0

19 Low * * * * * * 0 0 0
Middle * * -3 -3 -6 -9 0 0 0

High -12 -27 -36 -12 -30 -36 0 0 0

20 Low -9 -21 -27 -6 -12 -18 0 0 0
Middle -6 -12 -21 -3 -12 -21 0 0 0

High -9 -18 -27 -6 -15 -24 0 0 0

21 Low -9 -24 -36 -6 -18 -27 0 0 0
Middle -6 -21 -30 -6 -18 -30 0 0 0

High -9 -24 -36 -9 -21 -30 0 0 0

22 Low -15 -6 -6 -6 3 3 0 0 0
Middle -15 -15 -15 -9 -6 -9 0 0 0

High -21 -21 -21 -9 -9 -12 0 0 0

Increase Benefits for Low Earners

23 Low * 15 24 6 24 30 0 0 0
Middle * * * 6 9 9 0 0 0

High * * * * * * 0 0 0

24 Low 3 6 6 3 6 6 0 0 0
Middle * * * * * * 0 0 0

High * * * * * * 0 0 0

25 Low 15 21 24 18 21 24 0 0 0
Middle 9 15 15 9 12 12 0 0 0

High * * * * * * 0 0 0
Raise the Full Retirement Age

26 Low -6 -9 -9 -3 -6 -6 0 0 0
Raise the FRA to 68 Middle -6 -9 -9 -3 -6 -6 0 0 0

High -6 -9 -9 -3 -6 -6 0 0 0

Index the Bend Points in the PIA
Formula to Prices

Lower Initial Benefits for the
Top 50% of Earners

Index Earnings in the AIME and
Bend Points in the PIA Formula to
Prices

Replace the Current PIA Formula
with a New Two-Part Formula

Introduce a New Poverty-Related
Minimum Benefit

Enhance Low-Earners’ Benefits on
the Basis of Years Worked

Modify the Special Minimum
Benefit and Index It to Growth in

Lower Initial Benefits for the
Top 70% of Earners

10-Year Birth Cohortb Birth Cohortc 10-Year Birth Cohortc

Median Initial Benefits Median Lifetime Median Lifetime

Percentage Change from Current Lawd (Continued)

for Retired Workers by Benefits by 10-Year Payroll Taxes by
CBO

42 SOCIAL SECURITY POLICY OPTIONS

CBO
Table 3. Continued
Changes to Social Security’s Scheduled Benefits and Payroll Taxes for
Different Groups Under Various Options
Source: Congressional Budget Office.

Notes: Scheduled benefits are full benefits as calculated under current law, regardless of the amounts available in the Social Security trust
funds. Percentage changes are rounded to 3 percentage points to give a sense of the likely effects on benefits and payroll taxes
without showing numerous small differences in outcomes that are not analytically meaningful. Median values are within a group; half
of the people in each group (defined by lifetime household earnings category and birth cohort) would have a lower value and half
would have a higher value. Details of specific options are contained in the text.

AIME = average indexed monthly earnings; PIA = primary insurance amount; FRA = full retirement age; COLA = cost-of-living
adjustment; CPI-U = consumer price index for all urban consumers; * = between -1.5 percent and 1.5 percent, but not exactly zero;
0 = exactly zero, with no rounding.

a. The lowest fifth, middle fifth, and highest fifth of people ranked by lifetime household earnings, within a 10-year birth cohort.

b. Assumes that all workers claim benefits at age 65.

c. Lifetime benefits and lifetime taxes are present values discounted to age 62.

d. “Current law” refers to current Social Security provisions for calculating benefits and payroll taxes. See Congressional Budget Office,
The Long-Term Budget Outlook (June 2010).

Lifetime
Household
Earnings
Quintilea 1960 1980 2000 1960 1980 2000 1960 1980 2000

Raise the Full Retirement Age (Continued)

27 Low -6 -18 -18 -3 -15 -15 0 0 0
Raise the FRA to 70 Middle -6 -18 -18 -3 -15 -15 0 0 0

High -6 -18 -18 -6 -15 -15 0 0 0

28 Low -3 -9 -15 * -6 -9 0 0 0
Middle -3 -9 -15 -3 -9 -12 0 0 0

High -3 -9 -15 -3 -6 -12 0 0 0

Reduce Cost-of-Living Adjustments

29 Low * * * -6 -6 -6 0 0 0
Middle * * * -6 -6 -6 0 0 0

High * * * -6 -6 -6 0 0 0

30 Low * * * -3 -3 -3 0 0 0
Base COLAs on the Chained CPI-U Middle * * * -3 -3 -3 0 0 0

High * * * -3 -3 -3 0 0 0

Index the FRA to Changes in
Longevity

Reduce COLAs by 0.5 Percentage
Points

Percentage Change from Current Lawd (Continued)
10-Year Birth Cohortb Birth Cohortc 10-Year Birth Cohortc

Median Initial Benefits
for Retired Workers by Benefits by 10-Year Payroll Taxes by

Median Lifetime Median Lifetime

http://www.cbo.gov/doc.cfm?index=10457

http://www.cbo.gov/ftpdocs/115xx/doc11559/

SOCIAL SECURITY POLICY OPTIONS 43

Table 4.

Changes to Social Security’s Payable Benefits and Payroll Taxes for
Different Groups Under Various Options

Continued
Lifetime
Household
Earnings
Quintilea 1960 1980 2000 1960 1980 2000 1960 1980 2000

Low 11 11 14 100 110 150 90 110 140
Middle 20 20 25 230 250 320 320 370 490

High 30 31 40 370 410 530 660 790 1,070

Change the Taxation of Earnings

1 Low * 21 9 3 12 9 3 6 9
Middle * 21 12 9 15 9 3 6 9

High * 21 9 9 12 9 3 6 6

2 Low * 21 27 3 15 21 3 6 15
Middle * 21 27 6 21 24 3 9 15

High * 21 24 9 21 21 3 12 15

3 Low * 21 21 6 15 21 * 3 12
Middle * 21 24 9 21 24 * 6 15

High * 21 21 9 18 21 3 6 15

4 Low * 21 27 3 15 18 * * *
Eliminate the Taxable Maximum Middle * 21 27 6 18 21 * * *

High 3 30 36 21 39 42 12 15 18

5 Low * 15 6 3 6 6 * * *
Middle * 15 9 9 9 9 * * *

High 3 21 12 12 15 15 6 12 15

6 Low * 21 27 3 18 24 * * *
Middle * 21 27 6 21 27 * * *

High * 21 24 9 21 27 12 15 18

7 Low * 21 27 3 18 18 * * *
Middle * 21 27 9 21 21 * * 3

High * 21 27 9 21 18 12 15 18

8 Low * 18 9 6 9 9 * * *
Middle * 18 9 9 9 9 * * *

High * 18 9 9 9 6 6 6 9

Current Lawd (Thousands of 2010 dollars)

Percentage Change from Current Lawd

Median Initial Benefits Median Lifetime Median Lifetime
for Retired Workers by Benefits by Payroll Taxes by

10-Year Birth Cohortb 10-Year Birth Cohortc 10-Year Birth Cohortc

Raise the Taxable Maximum to Cover
90% of Earnings

Increase the Payroll Tax Rate by
1 Percentage Point in 2012

Tax All Earnings Above the Taxable
Maximum at 4%; Do Not Increase
Benefits

Tax Covered Earnings Up to
$250,000; Do Not Increase Benefits

Increase the Payroll Tax Rate by
2 Percentage Points Over 20 Years

Increase the Payroll Tax Rate by
3 Percentage Points Over 60 Years

Tax Covered Earnings Above the
Taxable Maximum; Do Not Increase
Benefits
CBO

44 SOCIAL SECURITY POLICY OPTIONS

CBO

Table 4. Continued

Changes to Social Security’s Payable Benefits and Payroll Taxes for
Different Groups Under Various Options
Continued
Lifetime
Household
Earnings
Quintilea 1960 1980 2000 1960 1980 2000 1960 1980 2000
Change the Taxation of Earnings (Continued)

9 Low * 3 3 3 3 3 * * *
Middle * 3 3 3 3 3 * * *

High * 3 3 3 3 3 3 3 3

Change the Benefit Formula

10 Low -3 -3 * -3 -3 * 0 0 0
Middle -3 -3 -3 * * * 0 0 0

High -3 * * * * * 0 0 0

11 Low * -3 -6 * -3 -3 0 0 0
Middle * -6 -6 * -3 -3 0 0 0

High * -3 * * * 3 0 0 0

12 Low -15 3 9 -9 3 6 0 0 0
Reduce All PIA Factors by 15% Middle -15 3 9 -6 6 6 0 0 0

High -15 3 9 -6 6 3 0 0 0

13 Low -3 18 24 * 15 21 0 0 0
Reduce the Top Two PIA Factors by Middle -18 * 3 -12 3 6 0 0 0
Roughly One-Third High -24 -9 -6 -18 -6 -3 0 0 0

14 Low * * 3 * 3 3 0 0 0
Middle * 3 3 3 3 3 0 0 0

High -3 -3 -3 -3 -3 -3 0 0 0

15 Low -6 -6 -6 * * * 0 0 0
Middle -6 -6 -6 * * * 0 0 0

High -6 -6 -6 * * * 0 0 0

16 Low -3 -6 -6 * -3 -3 0 0 0
Middle -3 -6 -6 * -3 * 0 0 0

High -3 -6 -6 * -3 -3 0 0 0

17 Low -15 -18 -33 -9 -15 -27 0 0 0
Middle -15 -18 -33 -6 -15 -30 0 0 0

High -15 -18 -33 -6 -18 -33 0 0 0

10-Year Birth Cohortb 10-Year Birth Cohortc 10-Year Birth Cohortc
Percentage Change from Current Lawd (Continued)

Median Initial Benefits
for Retired Workers by Benefits by Payroll Taxes by

Median Lifetime Median Lifetime

Tax All Earnings Above $250,000 at
4%; Do Not Increase Benefits

Raise from 35 to 38 the Years of
Earnings Included in the AIME
Reduce All PIA Factors by 0.5%
Annually
Index Initial Benefits to Changes in
Longevity

Index Earnings in the AIME Formula
to Prices

Reduce the Top PIA Factor by
One-Third

Reduce PIA Factors to Index Initial
Benefits to Prices Rather Than
Earnings

SOCIAL SECURITY POLICY OPTIONS 45

Table 4. Continued
Changes to Social Security’s Payable Benefits and Payroll Taxes for
Different Groups Under Various Options
Continued
Lifetime
Household
Earnings
Quintilea 1960 1980 2000 1960 1980 2000 1960 1980 2000

Change the Benefit Formula (Continued)
18 Low * 12 24 3 12 27 0 0 0

Middle -6 -3 -3 * * * 0 0 0
High -12 -24 -33 -6 -18 -30 0 0 0

19 Low * 9 18 3 12 18 0 0 0
Middle * 6 15 3 6 9 0 0 0

High -12 -24 -24 -9 -21 -24 0 0 0

20 Low -9 -12 -12 * -3 * 0 0 0
Middle -3 * * * * * 0 0 0

High -9 -12 -12 * -3 -3 0 0 0

21 Low -9 -12 -21 -3 -3 -9 0 0 0
Middle -6 -6 -9 3 -3 -9 0 0 0

High -9 -12 -18 * -6 -9 0 0 0

22 Low -15 * -3 -3 6 9 0 0 0
Replace the Current PIA Formula Middle -15 -9 -12 -3 -3 -3 0 0 0
with a New Two-Part Formula High -21 -18 -21 -6 -9 -9 0 0 0

Increase Benefits for Low Earners

23 Low * 9 12 3 15 21 0 0 0
Middle * -6 -6 * 3 3 0 0 0

High * -6 -9 -3 -6 -9 0 0 0

24 Low 3 6 6 3 6 6 0 0 0
Middle * * * * * * 0 0 0
High * * * * * * 0 0 0

25 Low 15 12 12 9 12 12 0 0 0
Middle 9 6 6 * 6 3 0 0 0

High * -6 -9 -9 -6 -9 0 0 0

Raise the Full Retirement Age

26 Low -6 -6 -3 * -3 * 0 0 0
Raise the FRA to 68 Middle -6 -6 -3 -3 * * 0 0 0

High -6 -6 -3 * -3 -3 0 0 0

10-Year Birth Cohortb 10-Year Birth Cohortc 10-Year Birth Cohortc
Percentage Change from Current Lawd (Continued)
Median Initial Benefits
for Retired Workers by Benefits by Payroll Taxes by
Median Lifetime Median Lifetime

Modify the Special Minimum Benefit
and Index It to Growth in Earnings

Index the Bend Points in the PIA
Formula to Prices
Lower Initial Benefits for the
Top 70% of Earners
Lower Initial Benefits for the
Top 50% of Earners
Index Earnings in the AIME and
Bend Points in the PIA Formula to
Prices
Introduce a New Poverty-Related
Minimum Benefit
Enhance Low-Earners’ Benefits on
the Basis of Years Worked
CBO

46 SOCIAL SECURITY POLICY OPTIONS

CBO
Table 4. Continued
Changes to Social Security’s Payable Benefits and Payroll Taxes for
Different Groups Under Various Options
Source: Congressional Budget Office.

Notes: Payable benefits are benefits as calculated under current law, reduced as necessary to make outlays equal the Social Security system’s
revenues. Percentage changes are rounded to 3 percentage points to give a sense of the likely effects on benefits and payroll taxes
without showing numerous small differences in outcomes that are not analytically meaningful. Median values are within a group; half
of the people in each group (defined by lifetime household earnings category and birth cohort) would have a lower value and half
would have a higher value. Details of specific options are contained in the text.

AIME = average indexed monthly earnings; PIA = primary insurance amount; FRA = full retirement age; COLA = cost-of-living
adjustment; CPI-U = consumer price index for all urban consumers; * = between -1.5 percent and 1.5 percent, but not exactly zero;
0 = exactly zero, with no rounding.
a. The lowest fifth, middle fifth, and highest fifth of people ranked by lifetime household earnings, within a 10-year birth cohort.
b. Assumes that all workers claim benefits at age 65.
c. Lifetime benefits and lifetime taxes are present values discounted to age 62.

d. “Current law” refers to current Social Security provisions for calculating benefits and payroll taxes. See Congressional Budget Office,
The Long-Term Budget Outlook (June 2010).

Lifetime
Household
Earnings
Quintilea 1960 1980 2000 1960 1980 2000 1960 1980 2000
Raise the Full Retirement Age (Continued)

27 Low -6 -12 -9 * -6 -3 0 0 0
Raise the FRA to 70 Middle -6 -12 -9 * -6 -6 0 0 0

High -6 -12 -9 * -9 -6 0 0 0

28 Low -3 -6 -9 * -3 -3 0 0 0
Middle -3 -6 -9 * -3 -3 0 0 0

High -3 -6 -9 * -3 -6 0 0 0

Reduce Cost-of-Living Adjustments

29 Low -3 12 6 * 3 3 0 0 0
Middle * 12 6 * 3 * 0 0 0

High * 12 6 3 * * 0 0 0

30 Low * 3 3 * * 3 0 0 0
Middle * 3 6 * * * 0 0 0

High * 3 3 * * * 0 0 0

Percentage Change from Current Lawd (Continued)
10-Year Birth Cohortb 10-Year Birth Cohortc 10-Year Birth Cohortc
Median Initial Benefits
for Retired Workers by Benefits by Payroll Taxes by
Median Lifetime Median Lifetime

Reduce COLAs by 0.5 Percentage
Points

Base COLAs on the Chained CPI-U
Index the FRA to Changes in
Longevity

http://www.cbo.gov/doc.cfm?index=10457

http://www.cbo.gov/ftpdocs/115xx/doc11559/

Appendix:
Distributional Effects of Options with

Similar Effects on the System’s Finances

The 30 policy options discussed in the main portion
of this study would have a variety of effects on the Social
Security system’s finances. The distributional trade offs
become clearer, however, if the options are compared
while their overall effects on the system’s finances are
held constant. Therefore, in another exercise, the
Congressional Budget Office compared the distributional
effects of 8 additional policy options it derived from the
original 30 with the objective of producing a single effect
on the actuarial balance—each would reduce the 75 year
actuarial deficit, relative to current law, by about one
quarter, or by 0.15 percent of gross domestic product (see
Table A 1 on page 49).

B Option 1a: Increase the payroll tax rate by 0.42 per
centage points in 2012 (Option 1 calls for an increase
of 1 percentage point in the payroll tax).

B Option 5a: Raise the taxable maximum to cover
87 percent of earnings (Option 5 would raise the
taxable maximum to cover 90 percent of earnings).

B Option 7a: Apply the payroll tax to covered earnings
between the taxable maximum and $136,000 with no
additional benefits (Option 7 would cap earnings
subject to tax at $250,000).

B Option 8a: Apply a 1.9 percent tax to all covered earn
ings above the taxable maximum with no additional
benefits (Option 8 would apply a 4.0 percent tax).

B Option 12a: Reduce all of the primary insurance
amount (PIA) factors by 4 percent (Option 12 would
reduce them by 15 percent).

B Option 14a: Reduce the top PIA factor from 15 per
cent to 4 percent (Option 14 would reduce the top
factor from 15 percent to 10 percent).

B Option 26a: Increase the full retirement age to
68 years and 1 month (Option 26 would increase
the age to 68 years).

B Option 29a: Reduce cost of living adjustments
(COLAs) by 0.25 percentage points (Option 29
would reduce them by 0.5 percentage points).

The first four options listed here (Options 1a, 5a, 7a, and
8a) would primarily affect payroll taxes. (The increase in
the taxable maximum also results in higher benefits.) In
general, increasing taxes in 2012 would have less of an
effect on workers born before the 1980s than on workers
born in the 1980s and later because most of the younger
groups’ earnings would be subject to higher taxes. Raising
the payroll tax rate (Option 1a) would increase lifetime
taxes by a similar proportion for all workers with earnings
below the taxable maximum (in 2010, $106,800). The
other three payroll options, by contrast, would raise life
time taxes mainly for high earners, and the effects of the
tax on all earnings above the taxable maximum would be
more concentrated among the very highest earners.

The remaining four options (Options 12a, 14a, 26a, and
29a) would affect benefits but not tax rates. Reducing all
of the PIA factors in various ways or reducing COLAs
would have proportionately similar effects on lifetime
benefits, regardless of when people were born or how
much they earned. The lower PIA factors would reduce
initial benefits; lower COLAs would diminish benefits
over time and therefore cause greater reductions in

CBO

48 SOCIAL SECURITY POLICY OPTIONS

CBO

benefits for people who receive benefits for long periods.
In addition, changes in COLAs would reduce benefits for
existing beneficiaries; changes to the benefit formula
would affect future beneficiaries only. An increase in the
full retirement age, phased in over time, would lead to

a greater reduction in benefits for people born later. By
contrast, reducing just the top PIA factor would reduce
benefits only for high earners and would result in similar
reductions for everyone who was born after 1954 and has
sufficiently high earnings.

APPENDIX SOCIAL SECURITY POLICY OPTIONS 49

Table A-1.

Changes to Social Security’s Scheduled Benefits and Payroll Taxes for
Different Groups Under Various Options That Have Similar Effects on the
System’s Finances

Continued

Lifetime
Household
Earnings
Quintilea 1960 1980 2000 1960 1980 2000

Low 110 130 180 90 110 140
Middle 250 300 420 320 370 490
High 400 500 680 670 780 1,070

Change Taxation of Earnings

1a Low 0 0 0 * 3 3
Middle 0 0 0 * 3 3
High 0 0 0 * 3 3

5a Low * * * * * *
Middle * * * * * *
High 3 3 6 3 9 9

7a Low 0 0 0 * * *
Middle 0 0 0 * * *
High 0 0 0 3 6 6

8a Low 0 0 0 * * *
Middle 0 0 0 * * *
High 0 0 0 3 3 3

Tax All Earnings Above the Taxable
Maximum at 1.9%; Do Not Increase
Benefits

Increase the Payroll Tax Rate by 0.42
Percentage Points in 2012

Raise the Taxable Maximum to Cover
87% of Earnings

Tax Covered Earnings Up to $136,000;
Do Not Increase Benefits

10-Year Birth Cohortb
Median Lifetime Payroll Taxes byMedian Lifetime Benefits by

10-Year Birth Cohortb

Current Lawc (Thousands of 2010 dollars)

Percentage Change from Current Lawc

CBO

50 SOCIAL SECURITY POLICY OPTIONS

CBO

Table A-1. Continued

Changes to Social Security’s Scheduled Benefits and Payroll Taxes for
Different Groups Under Various Options That Have Similar Effects on the
System’s Finances
Source: Congressional Budget Office.
Notes: Scheduled benefits are full benefits as calculated under current law, regardless of the amounts available in the Social Security trust
funds. Percentage changes are rounded to 3 percentage points to give a sense of the likely effects on benefits and payroll taxes
without showing numerous small differences in outcomes that are not analytically meaningful. Median values are within a group; half
of the people in each group (defined by lifetime household earnings category and birth cohort) would have a lower value and half
would have a higher value. Details of specific options are contained in the text.

PIA = primary insurance amount; FRA = full retirement age; COLA = cost-of-living adjustment; * = between -1.5 percent and
1.5 percent, but not exactly zero; 0 = exactly zero, with no rounding.

a. The lowest fifth, middle fifth, and highest fifth of people ranked by lifetime household earnings, within a 10-year birth cohort.

b. Lifetime benefits and lifetime taxes are present values discounted to age 62.

c. “Current law” refers to current Social Security provisions for calculating benefits and payroll taxes. See Congressional Budget Office,
The Long-Term Budget Outlook (June 2010).

Lifetime
Household
Earnings
Quintilea 1960 1980 2000 1960 1980 2000
Change the Benefit Formula

12a Low -3 -3 -3 0 0 0
Middle -3 -3 -3 0 0 0
High -3 -3 -3 0 0 0

14a Low * * * 0 0 0
Middle * * * 0 0 0
High -12 -9 -12 0 0 0

26a Low -3 -6 -6 0 0 0
Middle -3 -6 -6 0 0 0
High -3 -6 -6 0 0 0

29a Low -3 -3 -3 0 0 0
Middle -3 -3 -3 0 0 0
High -3 -3 -3 0 0 0

10-Year Birth Cohortb 10-Year Birth Cohortb

Reduce the Top PIA Factor from 15%
to 4%

Raise the FRA to 68 and 1 month

Reduce COLAs by 0.25 Percentage
Points

Reduce All PIA Factors by 4.0%

Percentage Change from Current Lawc (Continued)

Median Lifetime Benefits by Median Lifetime Payroll Taxes by

http://www.cbo.gov/doc.cfm?index=10457

http://www.cbo.gov/ftpdocs/115xx/doc11559/

Glossary

actuarial balance: The present value over a specified
period of the stream of projected trust fund revenues plus
the trust funds’ initial balance minus the stream of
projected outlays minus the value of a year’s worth of
benefits as a reserve at the end of the period, expressed as
a percentage of the present value of gross domestic
product or taxable payroll over the same period.

average indexed monthly earnings (AIME): For retired
workers who attain age 62 after 1990, the AIME is
calculated on the basis of the highest 35 years of earnings
on which someone paid Social Security taxes (up to the
taxable maximum, which is $106,800 in 2010). Earnings
before age 60 are indexed to compensate both for
inflation and for real (inflation adjusted) growth in
earnings; earnings after age 59 enter the computations at
actual amounts. (For disabled workers, earnings in the
two years before the initial benefit computation enter at
their actual amounts, and earlier earnings are indexed.)
Dividing the total earnings (after indexing) by 420
(35 years multiplied by 12 months) yields the AIME for
retired workers. For disabled workers, total earnings are
divided by a number of months that is linked to the age
at which benefits begin.

average wage index (AWI): The average amount of total
wages in the United States in a year, including earnings in
employment not covered by Social Security. Several
automatic adjustments in Social Security are based on the
AWI.

baby boom generation: People born between 1946 and
1964.

bend point: An element in the formula for calculating
initial benefits—namely, the threshold at which a PIA
(primary insurance amount) factor changes. Under
current law, there are two: in 2010, $761 and $4,568.

Bend points change each year to keep pace with changes
in the average earnings of the workforce as a whole.
Therefore, bend points occur at approximately the same
place in the distribution of average indexed monthly
earnings each year and average initial benefits rise at a
pace that matches the increase in average earnings over
time.

birth cohort: A group of people born during a given
period. This analysis places people into 10 year birth
cohorts: The 1960s birth cohort consists of people born
between 1960 and 1969, the 1980s cohort includes those
born between 1980 and 1989, and so on.

cost of living adjustment (COLA): An annual increase
in benefits tied to the increase in the cost of living. Under
current law, the COLA for Social Security benefits is
equal to the percentage increase in the consumer price
index for urban wage earners and clerical workers from
the third calendar quarter of the prior year to the third
calendar quarter of the current year, provided that there is
an increase.

covered earnings: Total earnings (from wages and
self employment income) for employment covered by
Social Security.

Disability Insurance Trust Fund: One of two Social
Security trust funds, it is used to finance the activities of
the Disability Insurance (DI) program. See trust funds.

full retirement age (FRA): Also called the normal
retirement age. The age at which a person becomes
entitled to unreduced retirement benefits (benefits that
are equal to the primary insurance amount). Currently
age 66, the FRA is being increased gradually so it will be
67 for people born in 1960 or later.

CBO

52 SOCIAL SECURITY POLICY OPTIONS

CBO

gross domestic product (GDP): The total market value
of goods and services produced domestically in a given
period.

initial benefits for retired workers: Benefits that would
be received by workers eligible to claim Old Age
Insurance benefits at age 62 who have not yet claimed any
other benefit. In this study, benefits are computed
assuming that all workers claim benefits at age 65 and are
based only on earnings through age 61. Values are net of
income taxes paid on benefits and credited to the Social
Security trust funds.

initial replacement rate: The initial benefit as a
percentage of average career annual earnings.

lifetime benefits: In this study, the present value at age 62
of benefits received by a person over a lifetime, net of
income taxes paid on those benefits and credited to the
Social Security trust funds. Lifetime benefits include
payments received by people in the following beneficiary
classes: old age workers, disabled workers, old age
spouses, and old age widows. Because there are
insufficient historical data on benefits for young widows
and children for years before 1984, young widows,
spouses of disabled workers, and child beneficiaries are
not considered in this study.

lifetime household earnings: For someone who is single
in all years, the sum of real (inflation adjusted) earnings
over a lifetime. In any year a person is married, the
earnings measure is a function of his or her earnings plus
those of his or her spouse (adjusted for economies of scale
in household consumption). A person’s lifetime earnings
consist of the present value of those annual amounts. In
this study, for the initial benefits for retired workers and
for the present value of lifetime benefits and payroll taxes,
the values shown are changes in benefits or taxes for
people in the lowest, middle, and highest one fifth
(quintile) of lifetime household earnings.

lifetime payroll taxes: The present value at age 62 of Old
Age, Survivors, and Disability Insurance taxes paid by the
employer and the employee over a lifetime; under current
law, the tax is 12.4 percent of taxable earnings. This
measure includes taxes paid by people who receive
benefits from the Old Age and Survivors Insurance

program (except young widows, spouses of disabled
workers, and child beneficiaries).

Old Age and Survivors Insurance Trust Fund: One of
two Social Security trust funds, it is used to finance the
activities of the Old Age and Survivors Insurance (OASI)
program. See trust funds.

payable benefits: Benefits as calculated under current law,
reduced as necessary to make outlays equal the Social
Security system’s revenues. Upon exhaustion of the Social
Security trust funds, the Social Security Administration
would reduce all scheduled benefits—those paid to
existing beneficiaries and to new beneficiaries—by the
percentage necessary to make the program’s total annual
outlays equal its total available revenues. That percentage
would vary each year.

present value: A single amount that expresses a flow of
current and future revenues or outlays in terms of an
equivalent lump sum received or spent at one time,
calculated by discounting future cash flows using a given
interest rate. (The Congressional Budget Office projects a
long term interest rate of 3 percent above price growth,
which is used as the discount rate.)

primary insurance amount (PIA): The monthly amount
payable to a worker who begins receiving Social Security
retirement benefits at the age at which he or she is eligible
for full benefits, or the amount payable to a disabled
worker who has never received a retirement benefit
reduced for age. For workers who turn 62 or become
disabled in 2010, for all of their dependents, and for
dependents of workers who die in 2010, the PIA formula
is 90 percent of the first $761 of the average indexed
monthly earnings (AIME), plus 32 percent of the AIME
between $761 and $4,586, plus 15 percent of the AIME
above $4,586.

primary insurance amount (PIA) factor: The percentage
of the average indexed monthly earnings replaced in the
PIA formula. Under current law, the PIA factors are
90 percent below the first bend point, 32 percent
between the two bend points, and 15 percent above the
second bend point.

quarter of coverage: The basic unit of measurement for
determining insured status. In 2010, a worker receives

GLOSSARY SOCIAL SECURITY POLICY OPTIONS 53

one quarter of coverage (up to a total of four quarters in
the year) for each $1,120 of annual covered earnings. The
basic amount of earnings required for a quarter of
coverage is increased annually at the same rate as the
increase in the average wage index.

replacement rate: The percentage of a worker’s past
average earnings that his or her benefits represent.

scheduled benefits: Full benefits as calculated under
current law, regardless of the amounts available in the
Social Security trust funds.

sustainable solvency: A condition under which positive
trust fund ratios are maintained throughout the 75 year
projection period and ratios are stable or rising at the end
of the period.

taxable maximum: The maximum covered earnings
upon which the Social Security payroll tax is levied each
year. In 2010, that amount is $106,800. The taxable
maximum amount is increased annually at the same rate
as the increase in the average wage index.

taxable payroll: The total earnings (from wages and self
employment income) for employment covered by Social
Security that is below the applicable annual taxable
maximum.

trust funds: Accounts to which Social Security taxes are
credited. Interest on the balances in the funds also is
credited to the trust funds, and Social Security benefits
and administrative expenses are drawn from the funds.
Social Security has two trust funds: the Old Age and
Survivors Insurance (OASI) Trust Fund and the
Disability Insurance (DI) Trust Fund. They are often
treated collectively as the OASDI trust funds.

trust fund balance: The amount credited to the Social
Security trust funds; the balance determines the
program’s current spending authority. The balance is held
in the form of special Treasury securities, and the cash
that generates the balance is used by the Treasury for
other operations and activities of the government.

trust fund exhaustion: The combined OASDI trust
funds are exhausted if outlays in any given year are greater
than the balance (including interest credited to the trust
funds) at the beginning of the year.

trust fund ratio: The balance in the Social Security trust
funds at the beginning of the year, divided by projected
outlays in that year.

trust fund solvency: The combined OASDI trust funds
are said to be solvent if the actuarial balance, calculated
over 75 years, is greater than zero.

CBO

Notes
Preface
Contents
Summary
Social Security Policy Options
Introduction
An Overview of Social Security
Social Security Projections
Assessing Options for Changing Social Security
Key Elements of Social Security
Scope of the Options
Effects of the Options on the System’s Finances
Effects of the Options on Payroll Taxes Paid and Benefits Received by Various Groups
Effects of the Options on Work and Saving
Options That Would Change the Taxation of Earnings
Options That Would Change the Benefit Formula
Options That Would Increase Benefits for Low Earners
Options That Would Raise the Full Retirement Age
Options That Would Reduce Cost-of-Living Adjustments
• Appendix: Distributional Effects of Options with Similar Effects on the System’s Finances

Glossary
Tables
1. Social Security’s Revenues and Outlays Under Current Law with Scheduled Benefits
2. Changes to Social Security’s Finances Under Various Options with Scheduled Benefits
3. Changes to Social Security’s Scheduled Benefits and Payroll Taxes for Different Groups Under Various Options
4. Changes to Social Security’s Payable Benefits and Payroll Taxes for Different Groups Under Various Options
A-1. Changes to Social Security’s Scheduled Benefits and Payroll Taxes for Different Groups Under Various Options That Have Similar Effects on the System’s Finances
Figures
1. Effects of the Policy Options on the OASDI Trust Fund Actuarial Balance
1. Distribution of Social Security Beneficiaries, by Type of Benefits Received, 2010
2. U.S. Population Age 65 or Older as a Percentage of the Population Ages 20 to 64, 1962 to 2080
3. Social Security’s Revenues and Outlays with Scheduled and Payable Benefits
4. Projected Outlays for Social Security Under Current Law and with a Gradual Reduction in Benefits Starting in 2012
5. Effect of Delaying a Payroll Tax Increase or Benefit Reduction on Social Security’s Finances
6. Calculating the PIA in 2010 Under the Current Social Security System
7. Calculating Initial Benefits with Progressive Price Indexing
8. Calculating Initial Benefits with Indexing of Bend Points to Prices
Options
Option 1: Increase the Payroll Tax Rate by 1 Percentage Point in 2012
Option 2: Increase the Payroll Tax Rate by 2 Percentage Points Over 20 Years
Option 3: Increase the Payroll Tax Rate by 3 Percentage Points Over 60 Years
Option 4: Eliminate the Taxable Maximum
Option 5: Raise the Taxable Maximum to Cover 90 Percent of Earnings
Option 6: Tax Covered Earnings Above the Taxable Maximum; Do Not Increase Benefits
Option 7: Tax Covered Earnings Up to $250,000; Do Not Increase Benefits
Option 8: Tax All Earnings Above the Taxable Maximum at 4 Percent; Do Not Increase Benefits
Option 9: Tax All Earnings Above $250,000 at 4 Percent; Do Not Increase Benefits
Option 10: Raise from 35 to 38 the Years of Earnings Included in the AIME
Option 11: Index Earnings in the AIME Formula to Prices
Option 12: Reduce All PIA Factors by 15 Percent
Option 13: Reduce the Top Two PIA Factors by Roughly One-Third
Option 14: Reduce the Top PIA Factor by One-Third
Option 15: Reduce All PIA Factors by 0.5 Percent Annually
Option 16: Index Initial Benefits to Changes in Longevity
Option 17: Reduce PIA Factors to Index Initial Benefits to Prices Rather Than Earnings
Option 18: Lower Initial Benefits for the Top 70 Percent of Earners
Option 19: Lower Initial Benefits for the Top 50 Percent of Earners
Option 20: Index the Bend Points in the PIA Formula to Prices
Option 21: Index Earnings in the AIME and Bend Points in the PIA Formula to Prices
Option 22: Replace the Current PIA Formula with a New Two-Part Formula
Option 23: Modify the Special Minimum Benefit and Index It to Growth in Earnings
Option 24: Introduce a New Poverty-Related Minimum Benefit
Option 25: Enhance Low-Earners’ Benefits on the Basis of Years Worked
Option 26: Raise the FRA to 68
Option 27: Raise the FRA to 70
Option 28: Index the FRA to Changes in Longevity
Option 29: Reduce COLAs by 0.5 Percentage Points
Option 30: Base COLAs on the Chained CPI-U

<< /ASCII85EncodePages false /AllowTransparency false /AutoPositionEPSFiles true /AutoRotatePages /All /Binding /Left /CalGrayProfile (Gray Gamma 2.2) /CalRGBProfile (sRGB IEC61966-2.1) /CalCMYKProfile (U.S. Web Coated \050SWOP\051 v2) /sRGBProfile (sRGB IEC61966-2.1) /CannotEmbedFontPolicy /Warning /CompatibilityLevel 1.6 /CompressObjects /Tags /CompressPages true /ConvertImagesToIndexed true /PassThroughJPEGImages true /CreateJobTicket false /DefaultRenderingIntent /Default /DetectBlends true /DetectCurves 0.1000 /ColorConversionStrategy /LeaveColorUnchanged /DoThumbnails false /EmbedAllFonts true /EmbedOpenType false /ParseICCProfilesInComments true /EmbedJobOptions true /DSCReportingLevel 0 /EmitDSCWarnings false /EndPage -1 /ImageMemory 1048576 /LockDistillerParams false /MaxSubsetPct 100 /Optimize true /OPM 1 /ParseDSCComments true /ParseDSCCommentsForDocInfo true /PreserveCopyPage true /PreserveDICMYKValues true /PreserveEPSInfo false /PreserveFlatness true /PreserveHalftoneInfo false /PreserveOPIComments false /PreserveOverprintSettings true /StartPage 1 /SubsetFonts true /TransferFunctionInfo /Apply /UCRandBGInfo /Remove /UsePrologue true /ColorSettingsFile () /AlwaysEmbed [ true ] /NeverEmbed [ true /Arial-Black /Arial-BlackItalic /Arial-BoldItalicMT /Arial-BoldMT /Arial-ItalicMT /ArialMT /ArialNarrow /ArialNarrow-Bold /ArialNarrow-BoldItalic /ArialNarrow-Italic /ArialUnicodeMS /CenturyGothic /CenturyGothic-Bold /CenturyGothic-BoldItalic /CenturyGothic-Italic /CourierNewPS-BoldItalicMT /CourierNewPS-BoldMT /CourierNewPS-ItalicMT /CourierNewPSMT /Georgia /Georgia-Bold /Georgia-BoldItalic /Georgia-Italic /Impact /LucidaConsole /Tahoma /Tahoma-Bold /TimesNewRomanMT-ExtraBold /TimesNewRomanPS-BoldItalicMT /TimesNewRomanPS-BoldMT /TimesNewRomanPS-ItalicMT /TimesNewRomanPSMT /Trebuchet-BoldItalic /TrebuchetMS /TrebuchetMS-Bold /TrebuchetMS-Italic /Verdana /Verdana-Bold /Verdana-BoldItalic /Verdana-Italic ] /AntiAliasColorImages false /CropColorImages true /ColorImageMinResolution 150 /ColorImageMinResolutionPolicy /OK /DownsampleColorImages true /ColorImageDownsampleType /Bicubic /ColorImageResolution 150 /ColorImageDepth -1 /ColorImageMinDownsampleDepth 1 /ColorImageDownsampleThreshold 1.50000 /EncodeColorImages true /ColorImageFilter /DCTEncode /AutoFilterColorImages true /ColorImageAutoFilterStrategy /JPEG /ColorACSImageDict << /QFactor 0.76 /HSamples [2 1 1 2] /VSamples [2 1 1 2] >>
/ColorImageDict << /QFactor 0.76 /HSamples [2 1 1 2] /VSamples [2 1 1 2] >>
/JPEG2000ColorACSImageDict << /TileWidth 256 /TileHeight 256 /Quality 15 >>
/JPEG2000ColorImageDict << /TileWidth 256 /TileHeight 256 /Quality 15 >>
/AntiAliasGrayImages false
/CropGrayImages true
/GrayImageMinResolution 150
/GrayImageMinResolutionPolicy /OK
/DownsampleGrayImages true
/GrayImageDownsampleType /Bicubic
/GrayImageResolution 150
/GrayImageDepth -1
/GrayImageMinDownsampleDepth 2
/GrayImageDownsampleThreshold 1.50000
/EncodeGrayImages true
/GrayImageFilter /DCTEncode
/AutoFilterGrayImages true
/GrayImageAutoFilterStrategy /JPEG
/GrayACSImageDict << /QFactor 0.76 /HSamples [2 1 1 2] /VSamples [2 1 1 2] >>
/GrayImageDict << /QFactor 0.76 /HSamples [2 1 1 2] /VSamples [2 1 1 2] >>
/JPEG2000GrayACSImageDict << /TileWidth 256 /TileHeight 256 /Quality 15 >>
/JPEG2000GrayImageDict << /TileWidth 256 /TileHeight 256 /Quality 15 >>
/AntiAliasMonoImages false
/CropMonoImages true
/MonoImageMinResolution 1200
/MonoImageMinResolutionPolicy /OK
/DownsampleMonoImages true
/MonoImageDownsampleType /Bicubic
/MonoImageResolution 300
/MonoImageDepth -1
/MonoImageDownsampleThreshold 1.50000
/EncodeMonoImages true
/MonoImageFilter /CCITTFaxEncode
/MonoImageDict << /K -1 >>
/AllowPSXObjects true
/CheckCompliance [
/None
]
/PDFX1aCheck false
/PDFX3Check false
/PDFXCompliantPDFOnly false
/PDFXNoTrimBoxError true
/PDFXTrimBoxToMediaBoxOffset [
0.00000
0.00000
0.00000
0.00000
]
/PDFXSetBleedBoxToMediaBox true
/PDFXBleedBoxToTrimBoxOffset [
0.00000
0.00000
0.00000
0.00000
]
/PDFXOutputIntentProfile (None)
/PDFXOutputConditionIdentifier ()
/PDFXOutputCondition ()
/PDFXRegistryName ()
/PDFXTrapped /False
/CreateJDFFile false
/Description << /ARA
/CHS
/CHT
/CZE
/DAN
/DEU
/ESP
/FRA
/GRE
/HEB
/HRV (Za stvaranje Adobe PDF dokumenata pogodnih za pouzdani prikaz i ispis poslovnih dokumenata koristite ove postavke. Stvoreni PDF dokumenti mogu se otvoriti Acrobat i Adobe Reader 5.0 i kasnijim verzijama.)
/HUN
/ITA (Utilizzare queste impostazioni per creare documenti Adobe PDF adatti per visualizzare e stampare documenti aziendali in modo affidabile. I documenti PDF creati possono essere aperti con Acrobat e Adobe Reader 5.0 e versioni successive.)
/JPN
/KOR
/NLD (Gebruik deze instellingen om Adobe PDF-documenten te maken waarmee zakelijke documenten betrouwbaar kunnen worden weergegeven en afgedrukt. De gemaakte PDF-documenten kunnen worden geopend met Acrobat en Adobe Reader 5.0 en hoger.)
/NOR
/POL
/PTB
/RUM
/RUS
/SLV
/SUO
/SVE
/TUR
/ENU (Use these settings to create Adobe PDF documents suitable for reliable viewing and printing of business documents. Created PDF documents can be opened with Acrobat and Adobe Reader 5.0 and later.)
>>
>> setdistillerparams
<< /HWResolution [600 600] /PageSize [612.000 792.000] >> setpagedevice