Topic: Financial Management

Discuss statesmanship as it relates to financial management in the public administration context.  Discuss the challenges that statesmen would face in this area and the statecraft needed to lead successfully. Integrate Biblical principles into your initial personal thread and your replies to peers.

 

You will take part in 3 Discussion Board Forums in which you will post a thread presenting your own opinion on the assigned topic, writing 400–500 words. Then, you will post replies of 200–250 words each to 2 or more classmates’ threads. Each reply must be unique and must integrate ideas (and citations) from the required reading. Merely posting the same reply in 2 places is not sufficient and may be treated as a form of academic misconduct. The original thread must incorporate ideas and citations from all of the required readings and presentations for that module/week. It must also address statecraft as part of a meaningful discussion of effective statesmanship. The reply posts must also integrate ideas and citations from the required readings and presentations for the module/week. Use only peer reviewed sources (no dictionaries, encyclopedias, websites, or Wikipedia).  All cites and references should conform to APA Style Manual (current edition).

Integrate Biblical principles within your discussion for both your personal thread and in all replies to peers.

Copyright© Cengage Learning. All rights reserved.

CHAPTER 5

Budgeting and Financial Management

 LEARNING OBJECTIVES

1. Understanding the role of the budget as an instrument of fiscal policy and public policy

2. Exploring where resources come from and how they are expended

3. Understanding the steps in the budgeting process

4. Learning about the various types of budgets

5. Understanding the importance of financial management in the public sector

 SUMMARY OVERVIEW

This chapter provides an overview of public budgeting and financial management. While

budgeting and financial management in public organizations have a great deal in common with

those activities in the private sector, there are important differences that come from the need for

public agencies to account to elected officials and citizens for their activities, including raising

revenue and spending money. In modern society, the way in which government at all levels

spends money affects the overall economy, meaning that the budget is an instrument of fiscal

policy. More importantly, the budget is a primary expression of government priorities, meaning

that the budget documents the decisions that governments make concerning public policies.

Chapter 5 examines the budget both as an instrument of fiscal policy and of public policy. The

budget as a device for government accountability is emphasized throughout the chapter.

The authors first address the budget as an instrument of fiscal policy, which they define as being

concerned with the impact of government taxation and spending on the economy. This section

covers the ways in which government spending can affect the economy along with theories

behind various approaches to fiscal policy. The discussion moves to the budget as an instrument

of public policy, noting that while the overall pattern of government spending defined by a budget

has an effect on the overall economy, the individual entries in a budget represent the measure of

support, or lack thereof, for specific policy choices. This discussion includes an exploration of

government revenue sources and the processes of choosing how to spend the government’s

money. This section also provides a review of the budget policies of recent presidential

administrations and offers insight into revenue and spending issues at the state and local level.

After examining the fiscal and public policy dimensions of public budgeting, the chapter turns to

a review of the budget as a tool of management. The budget process is critical to public

administrators as it establishes the level of funding for an agency and its programs. Thus, an

administrator should be familiar with the steps involved that enable an agency to spend money,

including budget formulation, approval, execution, and auditing. This is followed by a look at

budgeting strategies—line-item budgeting, performance budgeting, program budgeting, and

outcome-based budgeting—along with a discussion about the political nature of the budgeting

process. This section also addresses a variety of budgetary strategies and strategies for program

development.

58 Chapter 5: Budgeting and Financial Management

Copyright © Cengage Learning. All rights reserved.

The chapter then moves to an exploration of various aspects of the financial management of

public-sector resources. Included in this are capital budgeting, which addresses outlays for

infrastructure, facilities, and other items intended for long-term use; debt management, which is

concerned with various methods of borrowing funds; risk management, which is the ways in

which public organizations anticipate and cope with the variety of risks to which they are subject;

and purchasing, which involves considerations about how governmental organizations buy goods

and services. The final section of the chapter covers accounting and related information systems

in the public sector. Because keeping track of public revenues and expenditures has become an

enormously complex undertaking, an understanding of the requirements and practices involved in

government accounting is important to public managers. As the authors note, knowing the

technical side of the budget process—being able to follow the budget process and clearly

understand preparation, administration, and review—is extremely helpful as managers attempt to

influence the operations of their organizations.

 CHAPTER OUTLINE

I. THE BUDGET AS AN INSTRUMENT OF FISCAL POLICY

II. THE BUDGET AS AN INSTRUMENT OF PUBLIC POLICY

A. Where the Money Comes From
1. Individual Income Tax
2. Corporation Income Tax
3. Payroll Taxes
4. Sales and Excise Taxes
5. Property Taxes
6. Other Revenue Sources

B. Where the Money Goes

C. From Deficits to Surplus and Back
1. The Bush Tax Plan
2. Obama and Economic Recovery

 Exploring Concepts: ISSUES IN BUDGETING

D. State and Local Expenditures

III. THE BUDGET AS A MANAGERIAL TOOL

A. Budget Formulation

 Take Action: FINANCIAL MANAGEMENT: THE PROGRAM

MANAGER’S ROLE

B. Budget Approval

C. Budget Execution

D. Audit Phase

IV. APPROACHES TO PUBLIC BUDGETING

A. The Line-Item Budget

B. The Performance Budget

C. Program Budgeting

Chapter 5: Budgeting and Financial Management 59

Copyright © Cengage Learning. All rights reserved.

D. Outcome-Based Budgeting

E. Budgetary Strategies and Political Games

F. Strategies for Program Development

V. ASPECTS OF FINANCIAL MANAGEMENT

A. Capital Budgeting

B. Debt Management

C. Risk Management

D. Purchasing

VI. ACCOUNTING AND RELATED INFORMATION SYSTEMS

A. Government Accounting

B. Computer-Based Information Systems

VII. SUMMARY AND ACTION IMPLICATIONS

 KEY TERMS

Accounting The process of identifying, measuring, and communicating economic information to

permit informed judgment and decision making.

Allotments Amounts that agencies are authorized to spend within a given period.

Apportionment Process by which funds are allocated to agencies for specific portions of the

year.

Authorizing legislation Legislative action that permits establishment or continuation of a

particular program or agency.

Bond Promise to repay a certain amount (principal) at a certain time (maturity date) at a

particular rate of interest.

Budget padding Proposing a higher budget than is actually needed.

Business cycle Periods of economic growth featuring inflation and high employment followed

by periods of recession or depression and unemployment.

Capital expenditures Spending for items that will be used over a period of several years.

Continuing resolution Resolution permitting the government to continue operating until an

appropriations measure is passed.

Debt capacity Value of a city’s resources combined with the ability of the government to draw

on them to provide payment.

Deferral Decision by the president to withhold expenditure of funds for a brief period.

Discretionary spending That portion of the budget still open to changes by the president and

Congress.

Entitlement programs Programs that provide a specified set of benefits to those who meet

certain eligibility requirements.

Excise tax Tax applied to the sale of specific commodities.

60 Chapter 5: Budgeting and Financial Management

Copyright © Cengage Learning. All rights reserved.

Fiduciary funds Funds used when government must hold assets for individuals or when

government holds resources to be transmitted to another organization.

Fiscal policy Public policy concerned with the impact of government taxation and spending on

the economy.

Fiscal year (FY) Government’s basic accounting period.

General fund Fund that handles “unrestricted” funds of government.

Impoundment Withholding of funds authorized and appropriated by law.

Item veto Allows a governor to veto specific items in an appropriations bill.

Line-item budget Budget format for listing categories of expenditures along with amounts

allocated to each.

Performance auditing Analysis and evaluation of the effective performance of agencies in

carrying out their objectives.

Performance budget Budget format organized around programs or activities; includes various

performance measurements that indicate the relationship between work actually done and

its cost.

Planning-programming-budgeting system (PPBS) Effort to connect planning, systems

analysis, and budgeting in a single exercise.

Preaudit Review in advance of an actual expenditure.

Progressive tax One that taxes those with higher incomes at a higher rate.

Proportional tax One that taxes everyone at the same rate.

Proprietary funds Used to account for government activities that more closely resemble private

business.

Reconciliation bill Legislative action that attempts to reconcile individual actions in taxes,

authorizations, or appropriations with the totals.

Regressive tax One that taxes those with lower incomes at a proportionally higher rate than

those with higher incomes.

Rescission Presidential decision to permanently withhold funds.

Risk management Ways that public organizations anticipate and cope with risks.

Supplemental appropriation Bill passed during the fiscal year, adding new money to an

agency’s budget for the same fiscal year.

Chapter 5: Budgeting and Financial Management 61

Copyright © Cengage Learning. All rights reserved.

 WEB LINKS

The following are links to general information on budgeting organizations and financial

management:

Governmental Accounting Standards Board: (http://www.gasb.org/).

National Association of State Budget Officers: (http://www.nasbo.org/).

American Association for Budget and Program Analysis: (http://www.aabpa.org/).

Association for Budgeting & Financial Management: (http://www.abfm.org/).

Government Finance Officers Association: (http://www.gfoa.org/).

The following are links to information about the budget of the United States and how to get

to the budgets of other governments:

Department of the Treasury: (http://www.treasury.gov).

U.S. Government Printing Office:

(http://www.gpo.gov/fdsys/browse/collectionGPO.action?collectionCode=BUDGET).

National Debt Clock: (http://brillig.com/debt_clock/).

Office of Management and Budget: (http://www.whitehouse.gov/omb).

Congressional Budget Office: (http://www.cbo.gov/).

General Accountability Office: (http://www.gao.gov/).

http://www.gasb.org/

http://www.nasbo.org/

http://www.aabpa.org/

http://www.abfm.org/

http://www.gfoa.org/

http://www.treasury.gov/

http://www.gpo.gov/fdsys/browse/collectionGPO.action?collectionCode=BUDGET

http://brillig.com/debt_clock/

http://www.whitehouse.gov/omb

http://www.cbo.gov/

http://www.gao.gov/

https://doi.org/

American Review of Public Administration
2017, Vol. 47(5

)

550 –57

3

© The Author(s) 2016
Reprints and permissions:

sagepub.com/journalsPermissions.nav
DOI: 10.1177/0275074016629008

journals.sagepub.com/home/arp

Articl

e

Introduction

E-government facilitates interaction between the govern-
ments and the public, providing information and equipping
citizens to take an active role in public affairs (Flyverbom,
2015). Strategic plans for the implementation of e-govern-
ment are intended to generalize the use of new technologies
in the field of public administration and to create an environ-
ment enabling citizens to communicate their views or com-
plaints regarding public issues to influence the development
or implementation of public policy (Birchall, 2015

)

.

The importance attributed to e-government regarding
improved accountability and information transparency in the
framework of New Public Management (NPM) reforms has
motivated various studies seeking to identify factors that
determine a greater level of disclosure of public financial
information, using two communication channels, hard copy
(paper-based documents; Evans & Patton, 1987) and through
the use of information and communication technologies
(ICT), fundamentally the Internet (Caba, Rodríguez, &
López, 2008).

Nevertheless, despite the best endeavors of previous
research, there exists considerable heterogeneity in the
results, and conclusive evidence has yet to be obtained
regarding the influence of the above-mentioned factors.

Possibly, inconsistency in study design could account for the
uneven results (Pomeroy & Thornton, 2008). Whatever the
reason, there is as yet insufficient empirical evidence of the
validity and generality of these results to clearly establish the
influence of the above-mentioned factors, in quantitative
terms, on the disclosure of public financial information in
both modes of information disclosure.

This article addresses this issue by means of an objective
analysis of two aspects: (a) the communication channel used
to disclose public financial information and (b) the influence
that studies’ characteristics could have had on the conclu-
sions obtained (i.e., the moderating effects) when the public
financial disclosure was analyzed in those studies. This anal-
ysis was conducted by testing the statistical validity of the
empirical results of 51 articles (see Table 1) in a meta-
analysis to determine the underlying causes of the variations
and contradictions identified.

1University of Granada, Granada, Spain

Corresponding Author:
Laura Alcaide Muñoz, Faculty of Business and Administration, University
of Granada, Campus La Cartuja s/n, Granada 18071, Spain.
Email: lauraam@ugr.es

Transparency in Governments: A Meta-Analytic
Review of Incentives for Digital Versus
Hard-Copy Public Financial Disclosures

Laura Alcaide Muñoz1, Manuel Pedro Rodríguez Bolívar1,
and Antonio Manuel López Hernández1

Abstract
Prior research has indicated that information transparency in governments depends on institutional and environmental factors.
Nonetheless, previous studies show heterogeneity in the results, and the academic researchers cannot make consistent
conclusions. It makes it difficult to know the behavior of governments regarding their information policies. Therefore,
making use of meta-analysis techniques, we integrate the empirical results reported by studies to determine the factors
favoring the disclosure of public financial information via two modes of information disclosure—online versus hard-copy format.
Several moderating effects—administrative culture, accounting regime, impact of measure used on determining variables, and level of
government—have been considered and analyzed for their influence on the degree of correlation between the determinants
and the disclosure of public financial information in both modes of information disclosure. Our study does not only show
that the variables analyzed are positively associated with the disclosure of public financial information, but also that this
depends on the context in which the research is conducted. The administrative style and the level of government are the
main moderating effects that influence the results of analyzed studies.

Keywords
public administration (generally), budgeting/financial management, e-government, information and communication technology

https://us.sagepub.com/en-us/journals-permissions

https://journals.sagepub.com/home/arp

mailto:lauraam@ugr.es

http://doi.org/10.1177/0275074016629008

http://crossmark.crossref.org/dialog/?doi=10.1177%2F0275074016629008&domain=pdf&date_stamp=2016-02-12

551

T
a
b

le
1

.

M

ai
n

C
ha

ra
ct

er
is

ti
cs

o
f

th

e

St

ud
ie

s

A

na
ly

ze
d.

A
ut

ho
rs

/s
tu

di
es

C

o
un

tr
ie

s
O

nl
in

e
vs

.

h
ar

d

co

py
Le

ve
l o

f
go

ve
rn

m
en

t
Fi

na
nc

ia
l c

o
nd

it
io

n

(s

ig
n;

s
ig

ni
fic

an
ce

)

In
te

rg

o
ve

rn
m

en
ta

l
tr

an
sf

er
s

(s
ig
n;
s
ig
ni
fic
an
ce
)

P

o
lit

ic
al

co

m
pe

ti
ti

o
n

(s
ig
n;
s
ig
ni
fic
an
ce

)
Si

ze

(s
ig

n;
s

ig
ni

fic
an

ce
)

M
un

ic
ip

al
w

ea

lt
h

(s
ig
n;
s
ig
ni
fic
an
ce
)

B
ab

er
(

19
83

)
U

ni
te

d
St

at
es

H
ar

d
co

py
St

at
e

+
; N

o
ns

ig
ni
fic
an

t
—

+
; N
o
ns

ig
n.

/s
ig

n.
+

; S
ig

ni
fic

an
t

—
Ev

an
s

an
d

Pa
tt

o
n

(1
98

3)
U

ni
te
d
St
at
es
H
ar
d
co

py
Lo

ca
l

M
ix

ed
r

es
ul

t

s
—

—
+

; N
o
ns

ig
n.
/s
ig

n.
—

In
gr

am
(

19
84

)
U
ni
te
d
St
at
es
H
ar
d
co
py
St
at
e
+
; N
o
ns
ig
ni
fic
an

t
+

; N
o
ns
ig
ni
fic
an
t
+
; N
o
ns
ig
ni
fic
an
t
+
; N
o
ns
ig
ni
fic
an
t
+
; S
ig
ni
fic
an
t
B
ab

er
a

nd
S

en
(

19
84
)
U
ni
te
d
St
at
es
H
ar
d
co
py
St
at
e

−
; N

o
ns
ig
ni
fic
an
t
—
+
; N
o
ns
ig
n.
/s
ig
n.
—

—
R

o
bb

in
s

an
d

A
us

ti
n

(1
98

6)
U

ni
te
d
St
at
es
H
ar
d
co
py
Lo
ca
l

+
; S

ig
ni
fic
an

t
−

; N
o
ns
ig
ni
fic
an
t
—
+
; S
ig
ni
fic
an
t
−
; N
o
ns
ig
ni
fic
an

t
In

gr
am

a
nd

D
eJ

o
ng

(
19

87
)

U
ni

te
d

St
at

es
H

ar
d

c

o
py

St
at

e
+

; S
ig
ni
fic
an
t
+
; S
ig
ni
fic
an
t
—
+
; S
ig
ni
fic
an
t
—

Ev
an

s
an

d
Pa

tt
o
n

(1
98

7)
U

ni
te
d
St
at
es
H
ar
d
co
py
Lo
ca
l
+
; S
ig
ni
fic
an
t
—
M
ix
ed
r
es
ul

ts
M

ix
ed

r
es

ul
ts

—
G

ir

o
ux

(
19

89
)

U
ni
te
d
St
at
es
H
ar
d

co
py

L

o
ca

l
—

—
−

; N
o
ns
ig
ni
fic
an
t
—
M
ix
ed
r
es
ul

ts
D

w
ye

r

an

d
W

ils
o
n

(1
98

9)
U

ni
te
d
St
at
es
H
ar
d
co
py
Lo
ca
l
−
; N
o
ns
ig
n.
/s
ig
n.
—
—
+
; N
o
ns
ig
ni
fic
an
t
—

B
an

ke
r,

B
un

ch
, a

nd
S

tr
au

ss
(

19
89

)
U
ni
te
d
St
at
es
H
ar
d
co
py
Lo
ca
l
+
; S
ig
ni
fic
an
t
+
; N
o
ns
ig
ni
fic
an
t
—

—
M

ix
ed
r
es
ul
ts

C
o
pl

ey
(

19
91

)
U
ni
te
d
St
at
es
H
ar
d
co
py
Lo
ca
l
+
; S
ig
ni
fic
an
t
−
; S
ig
ni
fic
an
t
—
+
; S
ig
ni
fic
an
t

—
C

ar
pe

nt
er

(
19

91
)

U
ni
te
d
St
at
es
H
ar
d
co
py
St
at
e
+
; S
ig
ni
fic
an
t
—
−
; N
o
ns
ig
ni
fic
an
t
+
; S
ig
ni
fic
an
t

—
Fe

ro
z

an
d

W
ils

o
n

(1
99

2)
U

ni
te
d
St
at
es
H
ar
d
co
py
Lo
ca
l

—
—

—
+
; N
o
ns
ig
ni
fic
an
t
—

C
he

ng
(

19
92

)
U
ni
te
d
St
at
es
H
ar
d
co
py
St
at
e
−
; N
o
ns
ig
ni
fic
an
t
−
; N
o
ns
ig
ni
fic
an

t
M

ix
ed
r
es
ul
ts
+
; S
ig
ni
fic
an
t
+
; N
o
ns
ig
ni
fic
an

t
G

ir
o
ux
a
nd

D
ei

s
(1

99
3)

U
ni
te
d
St
at
es
H
ar
d
co
py

Lo
ca

l
M

ix
ed
r
es
ul
ts
−
; N
o
ns
ig
n.

/S
ig

n.
—
—
—

Li
m

a
nd

M
c

K

in
no

n
(1

99
3)
U
ni
te
d
St
at
es
H
ar
d
co
py
Lo
ca
l
—
—
—
+
; S
ig
ni
fic
an
t
—

A
lle

n
an

d
Sa

nd
er

(
19

94
)

U
ni
te
d
St
at
es
H
ar
d
co
py
Lo
ca
l
—
—
—
+
; N
o
ns
ig
ni
fic
an
t
—

C
hr

is
ti
ae

ns
(

19
99

)
N

et
he

rl
an

ds
H

ar
d
co
py
Lo
ca

l
+

; N
o
ns
ig
ni
fic
an
t
—
—
+
; S
ig
ni
fic
an
t
+
; N
o
ns
ig
ni
fic
an
t
M

cL
el

la
nd

a
nd

G
ir

o
ux

(
20

00
)

U
ni
te
d
St
at

es
O

nl
in

e
Lo

ca
l
—
+
; S
ig
ni
fic
an
t
—
+
; S
ig
ni
fic
an
t

—
T

ay
lo

r
an

d

R

o
sa

ir
(

20
00

)
A

us
tr

al
ia

H
ar
d
co
py
St
at
e
—
—
—
M
ix
ed
r
es
ul
ts

R

ya
n,

S
ta

nl
ey

, a
nd

N
el

so
n

(2
00

2)
U
ni
te
d
St
at
es
H
ar
d
co
py
Lo
ca
l
—
—
—
+
; N
o
ns
ig
n.
/s
ig
n.
—
G
ir
o
ux
a
nd

M
cL

el
la

nd
(

20
03

)
U
ni
te
d
St
at
es
H
ar
d
co
py
Lo
ca
l
+
; N
o
ns
ig
n.
/s
ig

n.
M

ix
ed
r
es
ul
ts
—
M
ix
ed
r
es
ul
ts
+
; N
o
ns
ig
ni
fic
an
t
G

o
re

(
20

04
)

U
ni
te
d
St
at
es
H
ar
d
co
py
Lo
ca
l
+
; S
ig
ni
fic
an
t
—
—
+
; S
ig
ni
fic
an
t
—

G
o
re

, S
ac

hs
, a

nd
T

rz
ci

nk
a

(2
00

4)
U

ni
te
d
St
at
es
H
ar
d
co
py
Lo
ca
l
+
; S
ig
ni
fic
an
t
—
—
+
; S
ig
ni
fic
an
t

—
Sm

it
h

(2
00
4)
U
ni
te
d
St
at
es
H
ar
d
co
py
Lo
ca
l
M
ix
ed
r
es
ul
ts
—
M
ix
ed
r
es
ul

ts
+

; N
o
ns
ig
ni
fic
an
t
M
ix
ed
r
es
ul
ts

La
sw

ad
, F

is
he

r,
a

nd
O

ye
le

re
(

20
05

)
N

ew
Z

ea
la

nd
H

ar
d
co
py

/o
nl

in
e

Lo
ca
l
+
; N
o
ns
ig
ni
fic
an
t
—
−
; N
o
ns
ig
ni
fic
an
t
+
; N
o
ns
ig
ni
fic
an
t
+
; N
o
ns
ig
ni
fic
an
t
M

al
o
ne

(
20

06
)

U
ni
te
d
St
at
es
H
ar
d
co
py
Lo
ca
l
M
ix
ed
r
es
ul
ts
—
—
+
; S
ig
ni
fic
an
t
+
; S
ig
ni
fic
an
t

Ju
st

ic
e,

M
el

it
sk

i,
an

d
Sm

it
h
(2
00
6)
U
ni
te
d
St
at
es

O
nl

in
e
St
at

e/
Lo

ca
l
—
—
—
+
; S
ig
ni
fic
an
t

—
St

yl
es

a
nd

T
en

ny
so

n
(2

00
7)

U
ni
te
d
St
at
es
O
nl
in
e
Lo
ca
l
+
; S
ig
ni
fic
an
t
—
—
+
; S
ig
ni
fic
an
t
+
; S
ig
ni
fic
an
t
M

ill
s,

C
ar

sl
aw

, a
nd

M
as

o
n
(2
00
7)
U
ni
te
d
St
at
es
H
ar
d
co
py
Lo
ca
l
—
+
; S
ig
ni
fic
an
t
—
—

—
da

C

o
st

a

C

ar
va

lh
o
, J

o
rg

e,
a

nd
F

er
na

nd
ez

(
20

07
)

Po
rt

ug
al

H
ar
d
co
py
Lo
ca
l
M
ix
ed
r
es
ul
ts
+
; S
ig
ni
fic
an
t
—
M
ix
ed
r
es
ul
ts

—
Pi

na
, T

o
rr

es
, a

nd
R

o
yo

(
20
07
)

EU
O

nl
in
e
Lo
ca
l
—
—
—
+
; S
ig
ni
fic
an
t
—
St

ec
co

lin
i,

A
ne

ss
i-
Pe

ss
in

a,
a

nd
N

as
i (

20
08

)
It

al
y

H
ar
d
co
py
Lo
ca
l
—
—
—
+
; N
o
ns
ig
ni
fic
an
t
—

G
an

dí
a

an
d

A
rc

hi
do

na
(

20
08

)
Sp

ai
n

O
nl
in
e
Lo
ca
l
—
—
+
; S
ig
ni
fic
an
t
+
; S
ig
ni
fic
an
t
+
; S
ig
ni
fic
an
t

C
ab

a,
R

o
dr

íg
ue

z,
a

nd
L

ó
pe

z
(2

00
8)

Sp
ai

n
O

nl
in
e
Lo
ca
l
+
; S
ig
ni
fic
an
t
−
; N
o
ns
ig
ni
fic
an
t
+
; N
o
ns
ig
ni
fic
an
t
−
; N
o
ns
ig
ni
fic
an
t
—

C
ár

ca
ba

a
nd

G
ar

cí
a

(2
00

8)
Sp

ai
n
O
nl
in
e
Lo
ca
l
M
ix
ed
r
es
ul
ts
—
+
; S
ig
ni
fic
an
t
+
; S
ig
ni
fic
an
t
—

Se
rr

an
o
, R

ue
da

, a
nd

P
o
rt

ill

o
(

20
09

)
Sp
ai
n
O
nl
in
e
Lo
ca
l
+
; N
o
ns
ig
ni
fic
an
t
—
+
; S
ig
ni
fic
an
t
+
; S
ig
ni
fic
an
t
+
; S
ig
ni
fic
an

t
Pi

na
, T
o
rr
es
, a
nd
R
o
yo
(
20

09
)

EU
O
nl
in
e
Lo
ca
l
—
—
—
+
; S
ig
ni
fic
an
t
—
Pi
na
, T
o
rr
es
, a
nd
R
o
yo
(
20

10
)

EU
O
nl
in
e
Lo
ca
l
—
—
—
+
; S
ig
ni
fic
an
t
—
C

ár
ca

ba
a

nd
G

ar
cí

a
(2

01
0)

Sp
ai
n
O
nl
in
e
Lo
ca
l
M
ix
ed
r
es
ul
ts
—
+
; S
ig
ni
fic
an
t
+
; S
ig
ni
fic
an
t
—
G

ui
lla

m
ó
n,

B
as

ti
da

, a
nd

B
en

it
o
(

2

0
11

)
Sp
ai
n
O
nl
in
e
Lo
ca
l
+
; N
o
ns
ig
ni
fic
an
t
+
; S
ig
ni
fic
an
t
—
+
; S
ig
ni
fic
an
t
M
ix
ed
r
es
ul
ts
M

ar
ta

ni
a

nd
L

es
ti
an

i (
20

12
)

In
do

ne
si

a
O

nl
in
e
Lo
ca
l
—
+
; N
o
ns
ig
ni
fic
an
t
—
+
; S
ig
ni
fic
an
t
+
; S
ig
ni
fic
an
t

A
lb

al
at

e
(2

01
3)

Sp
ai
n
O
nl
in
e
Lo
ca
l
+
; N
o
ns
ig
ni
fic
an
t
—
—
+

:S
ig

ni
fic
an
t
—
C

aa
m

añ
o
-A

le
gr

e,
L

ag
o
-P

eñ
as

, R
ey

es
-S

an
ti
as

, a
nd

S
an

ti
ag

o
–

B
o
ub

et
a

(2
01

3)
Sp

ai
n
O
nl
in
e
Lo
ca
l
+
; N
o
ns
ig
ni
fic
an
t
—
+
; S
ig
ni
fic
an
t
+
; N
o
ns
ig
ni
fic
an

t
— (c

on
tin

ue

d)

552

A
ut
ho
rs
/s
tu
di
es
C
o
un
tr
ie
s
O
nl
in
e
vs
. h
ar
d
co
py
Le
ve
l o
f
go
ve
rn
m
en
t
Fi
na
nc
ia
l c
o
nd
it
io

n
(s

ig
n;
s
ig
ni
fic
an
ce
)
In
te

rg
o
ve

rn
m
en
ta
l
tr
an
sf
er
s

(s
ig
n;
s
ig
ni
fic
an
ce
)
Po
lit
ic
al

co
m
pe

ti

ti
o
n

(s
ig
n;
s
ig
ni
fic
an
ce
)
Si
ze

(s
ig
n;
s
ig
ni
fic
an
ce
)
M
un
ic
ip
al
w
ea
lt
h
(s
ig
n;
s
ig
ni
fic
an
ce
)

R
io

s,
B

en
it
o
, a

nd
B

as
ti
da

(
20

13
)

In
te

rn
at

io
na

l
O

nl
in

e
St

at
e
+
; N
o
ns
ig
ni
fic
an
t
—
−
; N
o
ns
ig
ni
fic
an
t
+
; S
ig
ni
fic
an
t

—
W

eh
ne

r
an

d
R

en
zi

o
(

20
13

)
In

te
rn

at
io

na
l

O
nl
in
e
St
at

e
—

—
+
; S
ig
ni
fic
an
t
—
—
C
ab

a
Pé

re
z,

R
o
dr

íg
ue

z
B

o
lív

ar
, a

nd
L
ó
pe

z
H

er
ná

nd
ez
(
20

14
)

O
EC

D
c

o
un
tr
ie
s
O
nl
in
e
St
at
e
−
; N
o
ns
ig
n.
/S
ig
n.
—
+
; S
ig
ni
fic
an
t
+

/−
; N

o
ns
ig
ni
fic
an
t
+
; N
o
ns
ig
ni
fic
an

t
K

ee

ra

su
nt

o
np

o
ng

, D
un

st
an

, a
nd

K
ha

nn
a

(2
01

5)
N

ew
Z
ea
la

nd
O

nl
in
e
Lo
ca
l
—
—
—
+
; S
ig
ni
fic
an
t
+
; N
o
ns
ig
ni
fic
an

t
D

a
C
o
st

a
B
ai

rr
al

, C

o
ut

in
ho

e
S

ilv
a,

a
nd

D

o
s

Sa
nt

o
s

A
lv

es
(

20
15

)
B

ra
zi

l
O
nl
in
e
St
at
e
—
+
/−
; N
o
ns
ig
ni
fic
an
t
—
−
; N
o
ns
ig
ni
fic
an
t
—

Lo
w

at
ch

ar
in

a
nd

M
en

ifi
el

d
(2

01
5)

U
ni
te
d
St
at
es
O
nl
in
e
Lo
ca
l
—
—
—
+
; S
ig
ni
fic
an
t
+
; N
o
ns
ig
ni
fic
an
t
G

ar
ci

a-
T

ab
uy

o
, S

áe
z-

M
ar

tí
n,

a
nd

C
ab

a-
Pé

re
z

(2
01

5)
C

en
tr

al
A

m
er

ic
a

O
nl
in
e
Lo
ca
l
—
—
—
+
; N
o
ns
ig
ni
fic
an
t
—

N
ot

e.
T

he
s

tu
di

es
c

o
nt

ri
bu

te
m

o
re

t
ha

n
o
ne

o
bs

er
va

ti
o
n

t

o
t

he
s

a

m
pl

e
be

ca
us

e
th

ey
in

cl
ud

e
di

ffe
re

nt
e

st
im

at
io

ns
w

it
h

di
ffe

re
nt

d
at

a
se

ts
, d

iff
er

en
t

ex
pl

an
at

o
ry

v
ar

ia
bl

es
, o

r
di

ffe
re

nt
m

o
de

ls
. “

+
; N
o
ns
ig
ni
fic
an

t”
=

t
he

v
ar
ia
bl

e

o
ffe

rs

a

po

si

ti
ve

r
el

at
io

ns
hi

p
an

d
is

s
ta

ti
st

ic
al

ly
n

o
ns
ig
ni
fic
an

t.
“

+
; N
o
ns
ig
n.
/S
ig

n.
”

=
t

he
v

ar
ia

bl
e

o
ffe

rs
a

p
o
si

ti
ve
s
ta
ti
st
ic
al

ly
s

ig
ni
fic
an

t
re

la

ti
o
ns

hi
p

an
d
a
po

si
ti
ve

s
ta
ti
st
ic
al
ly
n
o
ns
ig
ni
fic
an
t
re
la
ti
o
ns

hi
p.

“
+

; S
ig
ni
fic

an
t”

=
t

he
v
ar
ia
bl
e
o
ffe
rs
a

po
si
ti
ve
r
el
at
io
ns
hi
p
an
d
is
s
ta
ti
st
ic
al
ly
s
ig
ni
fic
an
t.
“
−
; N
o
ns
ig
n.
/S
ig
n.
”
=
t
he
v
ar
ia
bl
e
o
ffe
rs
a

n
eg

at
iv

e
st

at
is

ti
ca

lly
s

ig
ni
fic
an
t
re
la
ti
o
ns
hi
p
an
d

a

ne

ga
ti
ve

s
ta
ti
st
ic
al
ly
n
o
ns
ig
ni
fic
an
t
re
la
ti
o
ns
hi
p.

“
−

; N
o
ns
ig
ni
fic
an
t”
=
t
he
v
ar
ia
bl

e
o
ffe

rs
a

ne
ga
ti
ve
r
el
at
io
ns
hi
p
an
d
is
s
ta
ti
st
ic
al
ly
n
o
ns
ig
ni
fic
an
t.
“

−
; S

ig
ni
fic
an
t”
=
t
he
v
ar
ia
bl
e
o
ffe
rs
a
n
eg
at
iv

e
re

la
ti
o
ns
hi
p
an
d

is
s

ta
ti
st

ic
al
ly
s
ig
ni
fic
an
t.
“
M
ix
ed
r
es
ul

ts
”

=
t
he
v
ar
ia
bl
e
o
ffe
rs
a

ll
o
pt

io
ns

w
he

n
di

ffe
re

nt
s

tu
di

es
a

re
a

na
ly

ze
d,

t
ha

t
is

, t
he

v
ar
ia
bl

e
ha

s
a

po
si

ti
ve

, s
ta

ti
st
ic
al
ly
s
ig
ni
fic
an

t
an

d
no

ns
ig

ni
fic

an
t

re
la

ti
o
ns

hi
p,

a
nd

it
h

as
a

n
eg
at
iv
e
st
at
is
ti
ca
lly
s
ig
ni
fic
an
t
an
d
no
ns
ig
ni
fic
an
t
re
la
ti
o
ns
hi
p.

E
U

=
E

ur
o
pe

an
U

ni
o
n;

O
EC

D
=

O
rg

an
is

at
io

n
fo

r
Ec

o
no

m
ic

C
o
-O

pe
ra

ti
o
n
an
d

D
ev

el
o
pm

en
t.

T
a
b
le
1

.
(c

o
n

ti
n

u
e
d

)

Alcaide Muñoz et al. 553

In short, the use of the meta-analysis technique in this
article seeks to investigate the effectiveness of previously
identified factors for public financial disclosure via hard-
copy format and compares it with the effectiveness of these
factors through the Internet, regardless of the conditions
under which the prior studies were carried out. The meta-
analysis was conducted in a sample of individual quantita-
tive studies. We incorporate different models and variables to
explain the relationships between different levels of informa-
tion disclosure and incentives of financial reporting, consid-
ering the two different communication channels used for
information disclosure (online via the Internet or hard copy).

Financial Transparency in
Governmental Entities and Forms of
Information Disclosure: Hard Copy
Versus Online

This study addresses two scenarios for the disclosure of
financial information: via the Internet or as hard copy (paper-
based documentation). Many studies focusing on the latter
question have analyzed the factors underlying the decision
by the U.S. Government to comply with generally accepted
accounting principles (GAAP) in preparing its financial
reports (Baber & Sen, 1984; Evans & Patton, 1987) or con-
solidated financial statements (Cheng, 1992; Giroux, 1989).

In contrast, recent studies of public-sector financial infor-
mation disclosure via the Internet have examined administra-
tive reforms conducted under the NPM framework and
concluded that the adoption and implementation of ICTs
tends to favor innovation within public organizations,
improving their transparency and accountability (Caba et al.,
2008; Cárcaba & García, 2008). Indeed, Internet-based
reporting has produced a shift from “push” to “pull” com-
munications in relations between government and citizens
(Mergel, 2013). The former “push” paradigm was character-
ized by the provision of batch-processed, producer-driven

information, whereas in the “pull” system, information is
tailored and citizen-driven (Mergel, 2013). These changes
are consistent with those taking place elsewhere under NPM
frameworks, with general-purpose information giving way
to made-to-measure data (see Figure 1).

The Internet has the potential to revolutionize information
disclosure due to its global reach, versatility, interactive capac-
ity, and speed (Jensen & Xiao, 2001), and the web is currently
perceived as providing the best available platform for informa-
tion stewardship and disclosure for both financial and nonfi-
nancial information (see Figure 1). Indeed, as opposed to
hard-copy format, the Internet is a good channel of communi-
cation to make information accessible to a great number of
users at a low cost as well as to centralize users’ demands for
information, allowing them to satisfy their need for information
when they must make financial decisions. The Internet can dis-
close continuous financial data and improve the flexibility of
information provided by financial statements by introducing
figures or elements that make the information more attractive
and comprehensible for users (see Figure 1). So, the disclosure
of financial information through the Internet could help public
administrators be publicly accountable and promote dialogue
regarding the use of public financial resources.

Nonetheless, despite the importance of online reporting
for improving transparency and accountability in the area of
public-sector financial information (see Figure 1), to date,
the volume of information disclosed online appears to be
about the same as that provided in hard-copy format
(Rodríguez Bolívar, Caba, & López, 2015). Indeed, as in the
private sector (Ziek, 2009), little progress seems to have
been made in online reporting, and online public financial
information consists mainly of the display of hard-copy
annual reports in an electronic format (Rodríguez Bolívar,
Caba, & López, 2007). Thus, the level of technology and the
disclosure environment are associated with the presentation
format, but not with information content.

Thus, despite the potential benefits of online reporting,
doubts have been expressed about whether it is designed or

Figure 1. Trends in governmental financial information reporting.
Source. Adapted from Institute of Chartered Accountants in England and Wales (ICAEW; 1998).

554 American Review of Public Administration 47(5)

used to effectively disseminate information. For example, as
investigation has found in the private sector (Unerman &
Bennett, 2004), online information may not be accessible to
some of the user groups most strongly affected by govern-
ment activity due to the lack of technology, infrastructure,
education, or language skills needed to use the government’s
official websites. In this regard, it has been observed that the
level of Internet pervasiveness is a major determinant of
Internet-based information disclosure; where general levels
of Internet use are higher, users will expect the government
to provide more information online (Caba et al., 2008;
Gandía & Archidona, 2008).

As researchers have found in the area of private business
(Debreceny, Gray, & Rahman, 2002), the environment
regarding the disclosure of financial information may favor
such disclosure or may hinder it. Thus, Rodríguez Bolívar
et al. (2015) and Pina et al. (2010) concluded that differences
in cultural variables could produce differing patterns of
accounting and would show varying levels of disclosure
from one country to another. In addition, expectations of
online reporting in developed countries may not be identical
to those in developing countries because of the “digital
divide” arising from populations’ ability to access the
Internet.

Finally, the Internet could generate higher quality finan-
cial reporting if better technical use were made of the poten-
tial to disclose information online. Thus, the technical use of
the Internet could be a significant factor to improve hard-
copy disclosure (Litan & Wilson, 2000). In this regard,
empirical research on the private business sector (Beattie &
Pratt, 2003) has indicated that various user groups, such as
auditors, demand Internet-based disclosure because of the
additional information that can be provided electronically,
the value of navigation and search aids, and the portability of
information in different formats.

According to the Inter-Agency Standing Committee
(IASC; 1999) and the Business Reporting Research Project
(BRRP; 2002), websites differ considerably from traditional
media. With this in mind and together with the above consid-
erations, we sought to inquire whether the results obtained
by individual studies might be influenced by the modes of
disclosure through which their information was accessed.
Empirical evidence of this heterogeneity can be observed
when the variables analyzed are income per capita (see the
conclusions of Christiaens, 1999 and Giroux & McLelland,
2003 in contrast to those of Caba et al., 2008 and Gandía &
Archidona, 2008) or the transfers received from other gov-
ernments (see Ingram & DeJong, 1987 and Robbins &
Austin, 1986, on one hand, and Laswad, Fisher, & Oyelere,
2005 and Caba et al., 2008, on the other).

In many cases, earlier studies have provided little theo-
retical underpinning for their analysis, and problems can
arise when interpreting and integrating the findings. In the
present article, the mode of disclosure is the main attribute to
analyze regarding governments’ incentives for financial

reporting, and we have tested this assumption using meta-
analysis techniques.

Meta-Analysis Techniques to Identify the
Incentives for Financial Transparency

Incentives for Financial Transparency in
Information Disclosure by Governments

Since the mid-1970s, many studies have analyzed the factors
conducive to more and better disclosure of public financial
information. These studies have mainly been based on agency
theory, neoinstitutional theory, and legitimacy theory
(Carpenter & Feroz, 2001; Feroz, Carpenter, & Cheng, 2007).
The key idea underpinning agency theory is that differences
exist between the interest of policymakers and public manag-
ers, on one hand, and the interests of ordinary citizens, on the
other. These differences force public representatives to be
held accountable for their actions and to demonstrate that
they have acted according to their responsibilities (Thompson,
1998). This theory has been influential in public-sector
reforms based on the NPM tenets and promotes the use of
government financial statements as instruments for external
users (citizens) to inform themselves of the actions taken by
policymakers and public managers (Mack & Ryan, 2006).

In contrast, neoinstitutional theory has been applied to
explain the adoption of innovations in management account-
ing (Carpenter & Feroz, 2001). According to this theory,
organizations seek to adopt structures and practices that are
considered legitimate and socially acceptable, thus produc-
ing homogeneous practices and structures (Feroz et al., 2007;
Powell & DiMaggio, 1991). Information disclosure becomes
a symbol of trust and modernity, constituting a trend adopted
by governments with the aim to project an image of good
governance and transparency (Hoffman, 1999).

Finally, legitimacy theory (Suchman, 1995) argues that
the legitimacy of the organization’s actions is influenced by
dissemination of information to stakeholders (Archel,
Husillos, Larrinaga, & Spence, 2009). In particular, the
greater the chance of negative impacts of a public policy, the
greater the need to try to influence the process through infor-
mation disclosure (Patten, 1992).

Regarding the determining factors of public-sector report-
ing, there are different incentives to disclose public financial
information, such as internal-external, political-economic,
managerial, and social factors. In this article, we group these
factors into two main categories: (a) institutional—financial
condition, intergovernmental transfers, and political compe-
tition, and (b) environmental—organization size and munici-
pal wealth (see Figure 2). Institutional factors are linked to
the assumptions of the agency and neoinstitutional theories
because they are related to the need of governments in per-
forming structures and practices with the aim of demonstrat-
ing that they are accomplishing their duty of transparency
and accountability on the use of the public financial resources.

Alcaide Muñoz et al. 555

The implementation of ICTs is being promoted by interna-
tional organizations as fundamental pillars of the innovations
in the public-sector management driven to improve transpar-
ency and trust on the use of public finances. By contrast,
environmental factors are strongly linked to the assumptions
proposed by agency and legitimacy theories, because they
are centered on the pressures and demands for financial
information made on government agencies by citizens. Also,
public managers could use the disclosure of public informa-
tion with the aim of influencing the negative opinions of the
public-sector management in the municipality.

This difference between institutional and environmental
factors is relevant because it enables us to determine whether
information transparency and accountability policies are
applied by public organizations through the greater disclo-
sure of public financial information that depends mainly on
an organizational decision or whether, on the contrary, the
environment affecting these government organizations has
greater significance. These factors will be analyzed in both
modes of information disclosure examined in this article.

Financial condition. One of the variables most commonly ana-
lyzed in this field is the government’s financial condition,
which is positively related to the motivation of public man-
agers to provide information transparency (Baber, 1983;
Ingram, 1984). This variable is included in our analysis of
financial information disclosure because it is an integral
component of the financial credibility of governmental agen-
cies vis-à-vis external agents (Baber, 1983; Ingram, 1984), as
well as the government’s capacity to meet its payment com-
mitments (Giroux & Deis, 1993). Taking into account the
agency theory approach, public managers must respond to
greater demands for information disclosure to minimize con-
flicts of interest (Baber & Sen, 1986; Gore, 2004; Zimmer-
man, 1977).

Previous research has defined financial condition as the
capacity of governments to finance the provision of public
services and programs, and it is measured in relation to the

level of public borrowing (Caba et al., 2008; Guillamón,
Bastida, & Benito, 2013). In fact, higher levels of debt may
represent a burden of future interest costs and principal
repayments that would reduce a governmental agency’s abil-
ity to meet demand for services and thus increase the fiscal
pressure on future generations of taxpayers (Rodríguez
Bolívar, Navarro Galera, Alcaide Muñoz, & López Subirés,
2016). For this reason, the evaluation of debt burden has
become an integral component of public managers’ responsi-
bility (Styles & Tennyson, 2007), and investors and creditors
are more interested in determining the ability of the govern-
ment to repay its debt (Daniels & Daniels, 1991). Besides,
given asymmetric information whereby investors are unable
to fully determine the default risk characteristics of issuers,
public managers could use increased disclosure to reduce the
cost of capital (Gore, 2004). Therefore, in accordance with
agency theory (Zimmerman, 1977), to generate positive sig-
nals about their performance with the aim of attracting finan-
cial funds from investors and creditors, public managers are
encouraged to disclose public financial information as a
mechanism to allow their actions to be monitored (Cárcaba
& García, 2008).

In addition, several authors argue that one cause for a lack
of trust in government is that citizens are not often provided
with enough factual documentation about government pro-
cesses and performance (Cook, Jacobs, & Kim, 2010).
Perhaps, it is produced because when government perfor-
mance is fully revealed, no state is fully successful in the
eyes of its citizens (Schick, 2003). Nonetheless, giving citi-
zens the possibility to monitor policymaking and scrutinize
its results will enhance the legitimacy of the institutional
structures—neoinstitutional theory approach (Curtin &
Meijer, 2006). To achieve this aim, previous studies have
highlighted that citizens are likely to want more accountabil-
ity from government and more information about where their
money is spent (Ipsos MORI, 2010).

This demand for information has intensified as a result of
the recent global financial crises in which many governments

Figure 2. Incentives for the disclosure of public financial information and moderating effects of empirical investigations.
Note. ME = moderating effects.

556 American Review of Public Administration 47(5)

have increased public expenditure to stimulate the economy
(Rodríguez Bolívar et al., 2016). They want reports to con-
tain information that would help them develop their own
evaluations of financial position (Hary & Antonio, 1999),
which could improve the credibility and integrity of public
finance and contribute to effective management of public
resources—neoinstitutional theory approach. In this sense,
Brusca and Montesinos (2006) affirmed that citizens could
control public management at election time, as the informa-
tion can reflect the results of public policies and, conse-
quently, serve as a vehicle for communicating the economic
effect of political management.

However, no conclusive evidence of this relationship has
been reported. The results show that if debt levels rise, politi-
cians make use of diverse channels of information disclosure
as a way to demonstrate a government’s ability to meet its
obligations to creditors. Thus, the Internet provides an effec-
tive mode of disseminating information and enables creditors
to readily monitor government activities (Debreceny et al.,
2002). In contrast, however, other studies have found no sig-
nificant association between these two variables (Guillamon,
Bastida, & Benito, 2011; Serrano, Rueda, & Portillo, 2009).
Accordingly, the following hypothesis is proposed:

Hypothesis 1 (H1): There is a positive relationship
between financial condition and the disclosure of public
financial information

Intergovernmental transfers. Intergovernmental transfers is
another variable that previous research has considered a pos-
sible determinant of financial information disclosure, as
these transfers are subject to control by the central or supra-
national institutions that disburse the funds (Copley, 1991;
Ingram, 1984; Ingram & DeJong, 1987). According to previ-
ous research in the field of the agency theory, the receipt of
such funds by central or supranational governments requires
the recipient to disclose information to account for the use of
the funds received, seeking to demonstrate that public man-
agers have acted according to their responsibilities (Ingram
& DeJong, 1987). Thus, governments in need of this funding
are obliged to provide high quality public financial reports
because their fund providers demand to see how the trans-
ferred funds are used, thus achieving more and better
accountability (Ingram, 1984; Ingram & DeJong, 1987).

With respect to the hard-copy disclosure of financial
statements, Giroux and Deis (1993) found a negative rela-
tionship with intergovernmental funding, which suggests
that cities receiving intergovernmental transfers have less
incentive to disclose financial information. However, Copley
(1991) concluded that resources obtained from other public
bodies must be accountable to the providing authority, to
demonstrate that the assigned funds have been used in accor-
dance with the objectives of the program for which they
were granted. In the case of online information disclosure, a
variety of results have been reported; Caba et al. (2008) do
not consider intergovernmental revenues to be a significant

variable regarding the disclosure of financial information,
but in this respect, Guillamon et al. (2011) found a positive
and statistically significant relationship. Accordingly, the
following hypothesis is proposed:

Hypothesis 2 (H2): There is a positive relationship
between transfers and funds received from other public
organizations and the disclosure of public financial
information.

Political competition. Political competition is a variable that
has been widely analyzed with respect to the disclosure of
financial information. After taking office, politicians often
ignore their preelection promises, giving priority to their
own interests rather than the public good, fully aware of the
difficulty for citizens to exercise imminent, effective control
(Cárcaba & García, 2008). However, the existence of politi-
cal rivals presenting strong opposition to the party in power
can increase the long-term costs of this opportunistic behav-
ior (Baber, 1983; Evans & Patton, 1987), and the public
managers are forced to justify their actions. Opposition par-
ties in this situation will call on the ruling political party to
exercise responsible management and will advise the public
of any divergences from its election program (Giroux, 1989;
Serrano et al., 2009). Therefore, the existence of alternative
officeholders and the corresponding risk of defeat in future
elections will reduce the deviation from the interests of vot-
ers and other political agents (Zimmerman, 1977). Accord-
ingly, the governing party will closely observe the degree of
implementation of its preelection promises to remain in
office for several terms and the greater the political competi-
tion, the more incentive it will have to do so (Baber, 1983;
Caba et al., 2008; Gandía & Archidona, 2008). In conse-
quence, taking into account the legitimacy theory approach,
the government has an interest in voluntarily demonstrating
its commitment to efficient management and will make use
of different modes of information disclosure (Cárcaba &
García, 2010; Laswad et al., 2005).

Nevertheless, previous studies of the hard-copy disclo-
sure of financial statements and its relationship to political
competition have reported conflicting evidence (Carpenter,
1991); some have obtained mixed results (Cheng, 1992;
Smith, 2004), and another found a relationship with a non-
significant negative sign (Giroux, 1989). However, all the
evidence points in a single direction in the studies that have
examined online information transparency. All are agreed
that political competition within governments encourages
public managers to be more responsive to the needs of the
electorate, who then offer a greater amount of information
(Caamaño-Alegre, Lago-Peñas, Reyes-Santias, & Santiago-
Boubeta, 2013; Cárcaba & García, 2008, 2010). Accordingly,
the following hypothesis is proposed:

Hypothesis 3 (H3): There is a positive relationship
between political competition and the disclosure of public
financial information.

Alcaide Muñoz et al. 557

Size. From the standpoint of agency theory (Zimmerman,
1977), large governments have a high degree of information
asymmetry, and thus high agency costs and a great conflict of
interest with citizens. Thus, to reduce agency costs, govern-
ments should disclose more information (Baber, 1983; Serrano
et al., 2009). In addition, large cities must provide services to a
greater number of citizens, and have more resources at their dis-
posal to do so. Accordingly, citizens will demand a greater vol-
ume of information to discover how well the government fulfills
its duties (Giroux & McLelland, 2003).

From the empirical point of view, although Christiaens
(1999) and Ingram and Robbins (1988) reported significant
positive relationships between disclosure levels in hard-copy
format and the size of government, other authors (Robbins &
Austin, 1986) found no significant association. In any case
and in general, empirical evidence shows that large munici-
palities are more likely to disclose financial information in
hard copy and to facilitate public access to this information
(Copley, 1991; Giroux & McLelland, 2003; Gore, 2004). For
the online disclosure of financial information, the empirical
evidence is overwhelming; most empirical studies conclude
that cities with large populations make greater use of the
Internet than smaller cities to provide financial information
(Gandía & Archidona, 2008; Guillamón et al., 2011). In this
respect, Moon and Norris (2005) argued that governments of
large cities are more likely to adopt e-government, because
they are under great pressure to find the most effective and
efficient way of providing services and information.
Therefore, the following hypothesis is proposed:

Hypothesis 4 (H4): There is a positive relationship
between the size of public organizations and the disclo-
sure of public financial information.

Municipal wealth. Finally, decisions regarding information
disclosure may be affected by circumstances external to the
local government, such as income per capita. It has been
shown that with increasing local income, the population
expects better service and more information to confirm that
their taxes are being spent effectively (Cheng, 1992; Giroux,
1989; Giroux & McLelland, 2003; Ingram, 1984). Under this
milieu, legitimacy theory could support the interest of public
managers and policymakers to disclose financial statements
with the aim of confirming citizens the management of pub-
lic finances and of influencing citizens’ opinions regarding
the negative impacts of public policies.

In the case of the hard-copy disclosure of financial infor-
mation, opinions are divided (Evans & Patton, 1987; Smith,
2004). Thus, Giroux and McLelland (2003) reported a signifi-
cant relationship with municipal income levels for a study
conducted in 1996 but not for one conducted in 1983; Ingram
(1984) found this variable was significantly related to munici-
pal wealth, but the relationship was nonsignificant for Robbins
and Austin (1986) and Giroux (1989). In contrast, the empiri-
cal evidence reported in studies of online information

disclosure leads us to conclude that in cities where income per
capita is high, residents are motivated to demand information
(Caamaño-Alegre et al., 2013; Serrano et al., 2009; Styles &
Tennyson, 2007). In this regard, too, Ho (2002) argued that
cities with a low income per capita are not likely to adopt a
progressive web design, due to low demand for online ser-
vices. Therefore, the following hypothesis is proposed:

Hypothesis 5 (H5): There is a positive relationship
between municipal income levels and the disclosure of
public financial information.

Moderating Effects Identified in Previous
Research on Financial Transparency

The conditions under which empirical investigations are
implemented, for example sample size, country analyzed,
year of publication, theoretical basis, and so on, must be
taken into account, because these moderating effects may
prove inconsistent results (see Table 1). Therefore, a method-
ology is needed to identify the statistically significant factors
that are independent of the underlying considerations of the
studies in which they were examined, and determine possible
incentives to the disclosure of public financial information
that influence public managers in their design of information
transparency strategies. Thus, designing public policies to
take such incentives into account will enable public manag-
ers to comply with their duty to be accountable.

Accounting for our analysis of the literature review, we
conclude that there are five essential modulating effects
under which different investigations have been implemented
and that could have influenced the results. The first is the one
analyzed in section “Meta-Analysis Techniques to Identify
the Incentives for Financial Transparency,” namely, the
hypothesis that the modes employed for the disclosure of
financial information (hard copy versus digital) may have
influenced the findings of previous empirical studies (ME

1
—

see Figure 2). In addition to this moderating effect, another
four were also considered: the administrative culture of the
countries from which we obtained the study data, the
accounting regimes adopted by the countries studied, the
measurement unit utilized to quantify determinants analyzed,
and the level of government constituting the research focus.

Administrative culture. The tenets of NPM have been imple-
mented worldwide, but in widely varying forms (Pina et al.,
2010). These differences are due to divergent concepts of
state and of the doctrine of separation that underlies adminis-
trative thinking. Thus, implementation of NPM models con-
forms to common patterns in countries that share certain
cultural values and are in accord with the national context,
which influences the role played by the state and its relation-
ship with its citizens. This, in turn, contributes to the impor-
tance granted to transparency and a government’s duty to be
publicly accountable (Kickert, 1997).

558 American Review of Public Administration 47(5)

Prior research has highlighted differences in govern-
ments’ degree of involvement with the citizenry, the empha-
sis on the implementation of NPM reforms, and the
bureaucratic structures and legal systems as well as differ-
ences in administrative culture that affect the evolution of
e-government (Rodríguez Domínguez, García, & Gallego,
2011), both to provide online public services (Torres, Pina, &
Acerete, 2006) and to publicly disclose financial information
(Pina et al., 2009, 2010; Rodríguez Bolívar et al., 2015).

Based on Hofstede (2001), administrative cultures are
considered as patterns of behavior acquired and transmitted
by human groups in the public sector; such patterns under-
line cultural values, beliefs, and principles, resulting in
shared understandings, and legitimize the function of their
acts. Indeed, national culture is an important driver for the
way people think and act in any given society (Hofstede,
2001) and also for the e-government diffusion (Zhao, Shen,
& Collier, 2014). In addition, studies on transparency over-
look the effect of cultural differences between countries on
how transparency is viewed and related to citizen attitudes
(Bouckaert & van de Walle, 2003). The administrative cul-
ture affects government transparency (Rodríguez Bolívar
et al., 2015), and it has been demonstrated that government
transparency fits less in countries with national cultures that
possess higher power distances and long-term orientation
(Grimmelikhuijsen, Porumbescu, Hong, & Im, 2013). In
fact, in contexts of high power distance, relations between
leaders and citizens can be viewed as paternalistic in the
sense that citizens are said to perceive themselves as largely
dependent upon (and by extension vulnerable to) their gov-
ernment (Park & Shin, 2005). Therefore, we would expect a
meta-analysis to reveal discrepancies in the disclosure of
public financial information between different countries, in
accordance with the corresponding administrative culture
(ME

2
—see Figure 2).

Four major styles of management culture in governmental
entities have been distinguished: Anglo-Saxon, Germanic,
Southern European, and Nordic countries, the latter is
regarded as a mixed form of the Anglo-Saxon and Germanic
administrative cultures (Kickert, 1997). The term Anglo-
Saxon is applied to countries such as the United Kingdom,
the United States, Australia, and New Zealand. South
American countries considered include Argentina, Colombia,
Peru, and Paraguay. The term Continental European is used
to refer to Mediterranean countries such as Italy, Portugal,
and Spain, and the Nordic category also includes Denmark
and the Netherlands (Pina et al., 2010).

In the present study, after analyzing all the articles in the
sample (see Table 1), two main groups were differentiated: (a)
articles focusing on the Anglo-Saxon style of administrative
culture and (b) all other articles. This differentiation makes
sense if we consider that sample studies in our research come
mainly from the Anglo-Saxon countries (31 studies) and from
Continental European countries (10 studies). In addition, three
sample studies come from other different cultures, and five of

them analyze the governments of the European Union (EU)
and others worldwide. These last studies are not considered in
our analysis of the moderating effect of administrative culture,
as they address diverse administrative cultures together.

Accounting regime. Although, we have accounted for the differ-
ent practices about disclosing public financial information and
have emphasized the implementation of NPM reforms with
the moderating effect of administrative culture, we think that
countries with similar cultures do exist, but have different
accounting requirements; for example, the United States has
adopted the Governmental Accounting Standard Board
(GASB), and Australia has implemented the Australian
Accounting Standard Board (AASB)—both of them have an
Anglo-Saxon administrative culture.

In previous research, empirical evidence shows these con-
tradictions. In this sense, Gore (2004) analyzed the hard-
copy disclosure of public financial information in the United
States, and the results showed that large municipalities are
likely to disclose financial information by hard copy and to
facilitate public access to this information. However, Taylor
and Rosair (2000) did not find clear evidence about the influ-
ence of organizational size on disclosure of public financial
information in Australian local governments. However,
Styles and Tennyson (2007) affirmed that policymakers were
motivated to disclose public financial information to demon-
strate to creditors in the United States of local governments’
ability to meet their obligations; however, Laswad et al.
(2005) found no significant association between the financial
condition and disclosure of public financial information in
New Zealand’s local governments.

Taking into account these contradictory results, we con-
sidered the accounting regime was a moderating effect, as
legal requirement and accounting practices could influence
hard copy versus digital financial reporting. In this sense,
after analyzing all the articles in the sample, two main groups
were differentiated: (a) articles that analyzed U.S. govern-
ments with the GASB accounting regime and (b) articles that
analyzed countries with their own accounting system. In this
regard, Keerasuntonpong, Dunstan, and Khanna (2015) has
presented the exception, analyzing New Zealand local gov-
ernments, which have adopted International Public Sector
Accounting Standards (IPSAS) at the time of research
(ME

3
—see Figure 2).

Impact of measure used on determining variables. In the litera-
ture, the use of different measurement to define the variables
can produce conflicting findings (Pomeroy & Thornton,
2008). In this regard, studies have highlighted the problems
of multidimensionality of accounting and reporting choice—
an aspect not easily described using specific quantitative
indices—as well as problems regarding the measurement of
data (Carpenter, 1991; Cheng, 1992). Thus, and in agreement
with Carpenter (1991) and Serrano et al. (2009), we found
measurement of the variable “political competition” led to

Alcaide Muñoz et al. 559

heterogeneous results regarding its association with financial
information disclosure. While Baber (1983) and Ingram
(1984) suggested that the effect of parliamentary political
competition on disclosure of financial information was posi-
tive, the obtained results were not statistically significant
when political competition was measured in terms of lobby
groups (Baber & Sen, 1984). In addition, greater divergence
of results is obtained when the incentive being analyzed is a
financial condition expressed as municipal indebtedness
(Baber & Sen, 1984; ME

4
—see Figure 2).

Level of government. Another of the moderating effects that
could produce the heterogeneity observed among different
studies is the level of government under analysis (statewide
or municipal). At the local level, citizens are increasingly
interested in being informed of the actions taken by munici-
pal governments, as these actions and reforms have direct
impact on their daily lives (Piotrowski & Van Ryzin, 2007).
Accordingly, a study of the situation in the United States
reported widespread interest in having access to information
to understand the realities of municipal management
(Thomas & Streib, 2003).

In addition, local governments offer a large number of
public-sector services, which must often be adapted to citi-
zens’ needs, to make public-sector services more efficient
and effective (Zafra, López, & Hernández, 2009). In addi-
tion, these governments undertake management reforms to
introduce technological changes (Connolly, Bannister, &
Keamey, 2010). Therefore, assuming the relations between
government and citizens can be of different degrees of inten-
sity (Edmiston, 2003), the empirical results obtained and the
relations observed among the variables analyzed may vary
accordingly (ME

5
—see Figure 2).

In brief, Figure 2 shows a scheme of the theoretical founda-
tion of the hypotheses contained in previous research by sort-
ing determinants according to institutional factors and
environmental factors. In this sense for our study, we tested the
influence of these determining factors on the disclosure of pub-
lic financial information by proposing five hypotheses.
Nonetheless, to account for variability within the results and in
findings of previous research regarding the influence of these
factors, in this study, we analyze whether they are affected by
the five considered moderating effects.

Sample Selection

In this study, we first conducted a comprehensive analysis
of all publications listed in the categories of Public
Administration, Information Science, Computer Science
and Information Systems, and Business and Finance,
indexed by the Institute for Scientific Information for the
period 1980 to 2015. To do this, we examined the title, the
abstract, the keywords, and the introduction of each article
to ensure that the stated research goals were relevant to our
investigation. In the few cases in which application of these

discrimination criteria was not decisive, we read the entire
article.

A systematic search was then made of the ABI/INFORM,
EJS Ebsco, Blackwell, Scopus, Emerald/Insight, SpringerLink,
ScienceDirect, Social Science Research Network (SSRN),
Econlit, Ecopapers, and Business Source Premier databases,
using descriptors and keywords, such as “public financial
reporting practices,” “voluntary disclosure,” “government
accounting,” “local government,” and “accounting disclosure.”
Then, to avoid the non-inclusion of relevant research, even
after carrying out these two steps, we selected the most signifi-
cant articles identified in the first and second searches and per-
formed an exhaustive analysis of the references they cited.

To be included in our sample, studies had to provide the
information needed to perform the meta-analysis technique,
namely the Pearson or Spearman (r) correlations for the
relations between the different variables and the disclosure
of public financial or public economic-financial informa-
tion. When r statistics were not reported, but other statistics
were transformable into r statistics, we applied formulas
given by Lipsey and Wilson (2001).

Some of the articles extracted did not provide the statisti-
cal information needed for the meta-analysis technique, and
so the authors in question (Cárcaba & García, 2008, 2010;
Pina et al., 2009, 2010) were contacted and requested to sup-
ply the missing information. Thus, all 51 of the articles ini-
tially identified were included in the sample for final analysis.
It should be noted that the meta-analysis technique does not
require a large number of studies to produce useful results
(Lipsey & Wilson, 2001), and so our review of previous
research highlighted some meta-analytical studies based on a
smaller sample than our own (Bel, Fageda, & Warner, 2010;
García & Sánchez, 2010; Pomeroy & Thornton, 2008).

Research Method

According to Hunter and Schmidt (2004), meta-analysis is a
robust statistical technique that is used to accumulate and
integrate results obtained from previous statistical analyses
to draw overall conclusions. This technique enables the
researcher to achieve clear, coherent conclusions, systemati-
cally extracted from previous research and highlighting
points common to all that would be difficult to identify by
descriptive analysis alone (Rosenthal, 1979). Hence, through
the use of appropriate statistical procedures, we can identify
the moderator variables explaining the heterogeneity of ear-
lier results (Stanley, 2005). In this regard, the moderator vari-
ables are factors inherent to the study design, such as sample
size, data of publication, country in which the study was car-
ried out, and so on, which may influence the findings of these
individual studies and, therefore, explain variation in the
results (Stanley, 2005).

We used the meta-analytic technique developed by Hunter
and Schmidt (2004), calculating, first, three magnitudes: (a)
the weighted mean correlation coefficient, ρ = ΣN

i
r

i
/ ΣN

i
;

560 American Review of Public Administration 47(5)

(b) the total observed variance, Sr
2 = ΣNi (ri − ρ)

2 / ΣN
i
; and

(c) the sampling error variance, Sr
2 = (1 − ρ2)2 k / ΣN

i
, where

N
i
is the number of observations in each sample, r is the

effect size for the sample, and i and k are the numbers of
effect sizes. Larger sample sizes are given more weight to
reduce sampling error, which declines as the sample size
increases (Hunter & Schmidt, 2004). In this sense, the mean
correlation is the weighted average of each of the correla-
tions provided in the studies. In our study, it is showing the
relationship between the variable analyzed and the disclo-
sure of public financial information. In addition, the total
observed variance and the sampling error variance are two
intermediate magnitudes that are used to evaluate whether
empirical correlations are homogeneous.

The correlations were examined taking into account the
content of the variables and not the names used for them, as
there are similar variables that frequently appear under dif-
ferent denominations, such as the size of public organiza-
tions for which the population size can be used (Gore, 2004;
Ingram, 1984) or the level of assets and revenue (Serrano
et al., 2009) or the financial condition of public organiza-
tions, which are valued using diverse measures (Laswad
et al., 2005; Serrano et al., 2009). In the same way, consider-
ing the measurement units for the different variables, we
used the definitions provided by the authors as a reference. In
addition, the dependent variable—disclosure of public infor-
mation—is measured through the list of items as performed
by researchers in prior studies (Gandía & Archidona, 2008;
Pina et al., 2010).

To evaluate whether empirical correlations are homoge-
neous, two tests were used: (a) the 75% rule, according to
which if 75% of the observed variance across studies can be
explained by sampling errors: (100) Se

2 / Sr
2 ≥ 75, there is no

true variance in the studies, and thus, the association is non-
modulated and homogeneous (i.e., if this statistic is less than
75%, it would mean that the association is not homogeneous
and variability does exist, which is influenced by other factors
such as moderating effects); and (b) the Q statistic: (k) ( Sr

2 ) /
( Se

2 ) = (N Sr
2 ) / (1 − ρ2)2, where k is the number of effect

sizes included in the analysis, Sr
2 is the total observed vari-

ance, Se
2 is the sampling error variance, N is the total sample

size of the effect sizes, and ρ is the mean correlation coeffi-
cient. This statistical function has a chi-square distribution
with (k −1) degrees of freedom, and a significant Q would
indicate rejection of the null hypothesis of homogeneity. In
other words, if this statistic were statistically significant, it
would mean that the association is not homogeneous, so we
have to search the moderating effects that cause variability
within the results.

The hypothesis of homogeneity is rejected in many cases,
and so to limit the Type I error rate, we utilized a random
effect model (Hunter & Schmidt, 1990), using ( / )Sr k

2 as
the standard error to create a 95% confidence interval (CI)
around the mean effect size and to assess the significance of
the null hypothesis, H

0
: ρ = 0.

In our analysis, no correction is made for statistical arte-
facts other than sampling error, such as range restriction and
measurement unreliability, because this information was not
provided in the analyzed studies. This is a characteristic limi-
tation of research studies in the financial accounting field,
which, however, are not usually subject to the problems of
reliability that affect other areas, such as psychology or busi-
ness management (García & Sánchez, 2010). Moreover,
some researchers do not believe this type of correction
should be applied (Rosenthal, 1979).

However, we do analyze the existence of possible publi-
cation bias. The meta-analysis may be subject to the implicit
existence of these biases, and editors and reviewers may be
attracted by the publication of studies showing a significant
relationship between the variable of interest. This outcome is
more likely to occur in large sample studies, which would
distort any analysis based on the empirical observation of the
variables (Stanley, 2005). The lack of nonsignificant pub-
lished studies has been termed the “file-drawer problem”
(Rosenthal, 1979), and publication bias would tend to over-
estimate the number of significant results on a given topic.

To analyze publication bias in our sample selection and to
assess the stability of the results, we followed Rosenthal’s
(1979) technique to determine the fail-safe number N. A fail-
safe N indicates the number of nonsignificant, unpublished
(or missing) studies that would need to be added to a meta-
analysis to reduce an overall statistically significant observed
result to nonsignificance (Rosenberg, 2005). If this number
is large relative to the number of observed studies, one can
feel fairly confident in the summary conclusions. Thus, the
fail-safe N can indicate stability of the relationship.

We also wish to determine whether the time factor influ-
ences the moderating effects considered in this empirical
study. Therefore, the studies in the sample are divided into
those published between 1983 and 1999 (when all the articles
in question analyzed the disclosure of economic and financial
information in hard-copy format) and studies published
between 2000 and 2015. In the latter case, we differentiated
between studies considering the disclosure of financial infor-
mation in hard-copy format versus empirical studies that ana-
lyzed the online disclosure of this information.

Analysis of Results

Institutional Factors

Financial condition. As can be seen in Table 2, the question of
financial condition has been widely considered in this field;
thus, a total of 111 observations present an average correla-
tion of .087 and a CI = [.073, .103], which indicates a posi-
tive and statistically significant relation, confirmed by z
value (z = 11.04, ρ < .001), to the level of disclosure of finan- cial information in the public sector irrespective of the mode of information disclosure. Moreover, as reflected in the N (fail-safe) value, the stability of the data is confirmed; thus,

Alcaide Muñoz et al. 561

publication bias does not exist. Therefore, these results con-
firm the relationship expressed in H1 (see Figure 2). How-
ever, the explanatory power of the error variance (59.76%) is
limited, and it means that the relationship is nonhomoge-
neous and influenced by the existence of moderating
effects—chi-square statistic (Q = 185.736, ρ < .001). Thus, we have to search which of the considered moderating effects could modulate the analyzed relationship (see Figure 2).

Table 2 also shows that this relationship is greater when
financial disclosure is made in hard-copy format than when
it is provided online (ρ = .098 vs. ρ = .040); in both cases, the
relationship is positive and statistically significant. Therefore,
when the financial condition is healthy, that is, the debt is
reduced and its costs do not besiege the public organization,
the public managers are interested in informing citizens
about the public good management and accountability.
However, the level of variability has increased, although the
heterogeneity has not disappeared as shown by the chi-
square statistic (Q = 134.662, ρ < .05). In other words, we cannot affirm that the modes of dissemination (ME

2
) are a

moderating effect in the relationship between financial con-
dition and the disclosure of public financial information.

These differences are even more apparent in Table 3. In
this sense, the influence is stronger in countries with an
Anglo-Saxon administrative culture (ρ = .100 vs. ρ = .067)
than in non-Anglo-Saxon countries, and a comparable result
is obtained for online information disclosure (ρ = .104 vs. ρ =
.032). In both cases, the heterogeneity disappears (chi-square
statistic is nonsignificant). Accordingly, the administrative
culture modulates the relationship between financial condi-
tion and the disclosure of public financial information (ME

2
).

In addition, we can see that the use of different accounting
requirements also influences the analyzed relationship,

although the heterogeneity has not disappeared. So, as the
degree of variability persists and the statistical tools do not
show clear evidences of this influence, we could affirm that the
accounting regimes do not show evidence of modulation in the
analyzed relationships (ME

3
).

Taking into account the measurement unit used to quan-
tify the financial condition, we can see that the total debt
ratio and leverage offers the most statistically significant
positive association, followed closely by financial viability,
when hard-copy information disclosure was analyzed. In this
case, all of them show statistically significant positive asso-
ciation, and the variability within the results has disappeared.
In the case of online information disclosure, the financial
condition does not have much influence. This relationship is
not significant in the case of measured leverage, but the het-
erogeneity has disappeared. Therefore, we could affirm that
the measure used to quantify the financial condition modu-
lates the analyzed relationship (ME

4
).

Furthermore, in the case of hard-copy information disclo-
sure, financial condition is an incentive that exerts a stronger
influence on state governments (ρ = .116 vs. ρ = .096) than
on local ones. In the former case, the variability within the
data has disappeared (chi-square statistic is nonsignificant),
so the level of government analyzed (ME

4
) in the studies

could have influenced the relationship of the financial condi-
tion. Moreover, in local governments, this incentive exerts a
stronger influence in countries with an Anglo-Saxon admin-
istrative culture (ρ = .099 vs. ρ = .067) than in those with
other types of administrative cultures.

With respect to online financial information, our review
cannot be as detailed as in the case of hard-copy information
disclosure because there is no representative sample found in
studies examining this question for countries with an

Table 2. Meta-Analytic Data on the Determinant Factors of the Disclosure of Public Financial Information.

Independent variable K M correlation (ρ) % S
2

e/ S
2

r

Confidence interval (95%)

χ
2

k-1
Fail-safe NMinimum Maxim

um

Financial condition 111 .087*** 59.76 .071 .103 185.736*** 6,23

6

*

Hard copy 89 .098† 66.09 .080 .115 134.662**
Online 22 .040*** 50.52 .002 .077 43.547**
Intergovernmental transfers 47 .115*** 37.01 .105 .124 126.982*** 12,504*
Hard copy 36 .118*** 32.60 .108 .127 110.426***
Online 11 .010 100 −.047 .066 2.598
Political competition 68 .091*** 21.64 .069 .113 314.259*** 2,174*
Hard copy 44 .034† 58.54 .007 .061 75.166**
Online 24 .201*** 12.09 .163 .239 198.541***
Size 116 .166*** 19.09 .152 .180 607.701*** 28,900*
Hard copy 75 .154*** 18.51 .136 .173 405.247***
Online 41 .184*** 20.74 .162 .207 197.700***
Municipal wealth 46 .087*** 67.27 .062 .112 68.380** 1,019*
Hard copy 28 .094*** 65.55 .053 .134 42.713***
Online 18 .083*** 70.73 .052 .115 25.449***

†p < .1. *p < .05. **p < .01. ***p < .001.

562

T
a
b

le
3

.
M

et
a-

A
na

ly
ti
c

D
at

a
o
n

th
e

R
el

at
io

n
B

et
w

ee
n

A
na

ly
ze

d
V

ar
ia

bl
es

a
nd
t
he

D
is

cl
o
su

re
o

f
Pu

bl
ic

F
in

an
ci

al
I
nf

o
rm

at
io

n
(O

nl
in
e
vs

. H
ar

d
C

o
py
).

In
de

pe
nd

en
t

va
ri

ab
le

H
ar
d
co

py
O

nl
in
e
K

M

co
rr

el
at

io
n

(ρ

)
%

S

2 e
/

S2 r

C
o
nf

id
en

ce
in

te
rv

al

(9
5%

)

χ2
k-

1
K

M

co
rr
el
at
io
n

(ρ
)

%
S

2 e
/

S2 r

C
o
nf
id
en
ce
in
te
rv
al

(9
5%
)
χ2
k-

1
M

in
im

um
M

ax
im

um
M
in
im
um
M
ax
im
um

Fi
na

nc
ia

l c
o
nd

it
io

n

A
dm

in
is

tr
at

iv
e

cu

lt
ur

ea

A
ng

lo
-S

ax
o
n

83
.1

00

**

*
67

.8
1

0.
08

2
0.

11
9

12
2.

40
6*

*
5

.1
04

**
*

10
0

0.
00

6
0.

20
1

1.
13

7

N

o
n-

A
ng
lo
-S
ax
o
n

6
.0

67

†

53
.3

0
0.

00
6

0.
13

0
11

.2
56

†
14

.0
32

34
.7

1
−

0.
01

1
0.

07
5

40
.3

34
**

*

A
cc

o
un

ti
ng

r
eg

im
es

G
A

SB
83

.1
00

**
*

67
.8

1
0.

08
2

0.
11

0
12

2.
40

6*
*

—
—
—
—
—
—

O
th

er
s

6
.0

67
†

53
.3
0
0.
00
6
0.
13
0
11
.2
56

†
19

.0
40

5

9
.4

4
−

0.
13
0
0.

21
1

31
.9

65
**

M

ea
su

re
u

se
d

Fi
na
nc
ia

l v
ia

bi
lit

y
14

.0
87

**
*

6

6
.2

8
0.

04
0

0.
13

3
21

.1
22

†
—

—
—
—
—

—

R
at
io

t

o
ta

l d
eb

t
69

.1
00
**
*

7

4
.0

6
0.

08
0

0.
11

9
93

.1
65

†
15

.0
43

39
.4

7
−

0.
00

4
0.

09
0

3

8
.0

07
**

Le
ve

ra
ge

4
.1

75
**

*
10

0
0.

08
8

0.
26

1
1.

18
0

6
.0

14
10

0
−

0.
05

1
0.

07
9

1.
50

4

Le
ve

l o
f
go

ve
rn
m
en

tb

St
at

e
go

ve
rn
m
en

t
24

.1
16

**
*
10
0
0.
05
8
0.

17
4

9.
28

0
—

—
—
—
—
—

Lo

ca
l g

o
ve
rn
m

en
t

6

5
.0

96
**

*
51

.8
5

0.
07

7
0.

11
4

12
5.

36
3*

*

*
17

.0
53

*

*
50

.1
1

0.
01

3
0.

09
3

33
.9

24
**

Lo
ca
l g
o
ve
rn
m

en
tc

A
ng
lo
-S
ax
o
n

59
.0

99
**

*
52

.1
1
0.
07

9
0.

11
8

11
3.

22
2*

*

*
—

—
—
—
—
—

N
o
n-
A
ng
lo
-S
ax
o
n
6
.0
67
†
53
.3
0
0.

00
4

0.
13
0
11
.2
56
†
—
—
—
—
—

—
In

te
rg

o
ve
rn
m
en
ta

l

t
ra

ns
fe

rs

A
dm
in
is
tr
at
iv
e

cu
lt
ur

e

A

ng
lo

-S
ax

o
n

3

2
.1

18
**

*
29

.6
0

0.
10

8
0.

12
8

10
8.

10
0*

**
—
—
—
—
—
—

N
o
n-
A
ng
lo
-S
ax
o
n
4
.1

02
**

*
10
0
0.

02
9

0.
17

6
2.

14
7

1

1
.0

1

0
10

0
−

0.
04

7
0.

06
6

2.
59

8

A
cc
o
un
ti
ng
r
eg
im
es

G
A

SB
32

.1
18

**
*

29
.6

0
0.

10
8

0.
12

8
10

8.
10

0*
**

—
—
—
—
—
—

O
th
er
s
4
.1
02
**
*
10
0
0.
02
9
0.
17
6
2.
14
7

1

0
.0

02
10

0
−
0.
05
9
0.

06
3

0.
30

3

M
ea

su
re

u
se

d

C

ur
re

nt
a

nd
c

ap

it
al

t
ra
ns
fe

rs
5

.1
27

**
*

91
.4

3
0.

11
7

0.
13

7
5.

46
9

4
.0

10
10

0
−

0.
09

5
0.

11
4

0.
67

6

R
at
io

in
te

rg
o
ve
rn
m
en
ta

l r
ec

ei
pt

s
31

.0
29

*
44

.9
2

−
0.

00
3

0.
06

1
69

.0
17

**
*

7
.0

10
10
0
−
0.
05
8
0.

07
7

1.
89

9

Le
ve
l o
f
go
ve
rn
m
en

t

St

at
e

go
ve

rn
m
en
t
6
.0

72
*

10
0
0.
01
6
0.

1

8
7

2.
00

7
6

−
.0

02
10
0
−
0.
07
7
0.

07
2

1.
36

7

Lo
ca
l g
o
ve
rn
m
en
t

3

0
.1

18
**

*
27

.8
5
0.
10
8
0.
12
8

10
7.

73
8*

**
5

.0
26

10
0
−
0.

06
0

0.
11

2
1.

17
5

Lo
ca
l g
o
ve
rn
m
en
t

A
ng
lo
-S
ax
o
n

2

6
.1

18
**

*
24

.6
5

0.
10
9
0.
12
8

10
5.

47
4*

**
—
—
—
—
—

—
Po

lit
ic

al
c

o
m

pe
ti

ti
o
n

A

dm
in

is
tr

at
iv

e
cu

lt
ur

ed
,e

N
o
n-

A
ng
lo
-S
ax
o
n
—
—
—
—
—
—

1

1
.1

43
**

*
10
0
0.

09
8

0.
18

8
5.

46
9

A

cc
o
un

ti
ng
r
eg
im
es

G
A

SB
44

.0
34

†
58

.5
4

0.
00
7
0.

06
1

7

5
.1

66
**

—
—
—
—
—
—

O
th
er
s
—
—
—
—
—
—

22
.1

96
**

*
11

.3
1

0.
15

8
0.

23
4

19
4.

47
1*

**

M
ea
su
re
u
se
d

R
at
io

c
an

di
da

te
s/

av
ai

la
bl

e
po

si
ti
o
n

—
—
—
—
—
—

18
.2

91
**

*
9.

18
0.

24
2

0.
22

9
19

6.
11

7*
**

Pe
rc

en
ta

ge
s

o
f
vi

ct

o
ri

es
10

.0
75

*
58

.4
3

−
0.

00
2

0.
15

3
17

.1
15

*
—
—
—
—
—
—

D

ic
ho

to
m

o
us

v
ar
ia
bl

e
21

−
.0

04
10

0
−

0.
03

5
0.

02
7

18
.2

63
2

.1
92

**
95

.2
8

0.
07
0
0.

31
5

2.
09

9

In

di
ce

s
o
f
po

lit
ic
al
c
o
m

pe
ti
ti
o
n

5
.2

48
**

*
10
0
0.

13
2

0.
36

4
1.

28
1

4
.1

37
**

*
10
0
0.

08
5

0.
18
8
0.

85
4

%
s

ea
ts

h
el

d
by

m
in

o
ri

ty
p

ar
ty

6
.2

19
**

*
10
0
0.

11
3

0.
32

5
3.

18
5

—
—
—
—
—
—

Le

ve
l o
f
go
ve
rn
m
en
t

St
at
e
go
ve
rn
m
en
t

21
.2

19
**
*
10
0
0.

15
9

0.
27

8
16

.7
46

9
.4

70
**

*
5.

13
0.

42
4

0.
51

7
17

5.
35

1*
**

Lo
ca

l g
o
ve

rn
m

en
td

2

3
−

.0
09

10
0
−
0.

03
9

0.
02

1
17

.6
68

15
.1

16
**

*
91

.5
8

0.
07
3
0.
15
9

16
.3

79

(c
on

tin
ue

d)

563

In
de
pe
nd
en
t
va
ri
ab
le
H
ar
d
co
py
O
nl
in
e
K
M

co
rr
el
at
io
n

(ρ
)
%

S
2 e
/

S2 r
C
o
nf
id
en
ce
in
te
rv
al

(9
5%
)
χ2
k-
1
K
M

co
rr
el
at
io
n
(ρ
)
%
S
2 e
/
S2 r

C
o
nf
id
en
ce
in
te
rv
al

(9
5%
)
χ2
k-
1
M
in
im
um
M
ax
im
um
M
in
im
um
M
ax
im
um

Si
ze

A
dm
in
is
tr
at
iv
e
cu
lt
ur
ee

A
ng
lo
-S
ax
o
n

6

9
.1

82
**

*
22

.6
1

0.
16

3
0.
20
1

30
5.

21
4*

**
11

.1
00
**
*

5

3
.5

3
0.

06
7

0.
13

3
20

.5
50

*

N
o
n-
A
ng
lo
-S
ax
o
n

6
−

.0
96

f
23

.0
2

−
0.

15
5

−
0.

03
7

26
.0

59
**

*
14

.3
01

**
*

36
.3

4
0.

26
4

0.
33

9
38

.5
20

**
*

A
cc
o
un
ti
ng
r
eg
im
es

G
A

SB
67

.1
82

**
*

22
.0

0
0.

16
3

0.
20

0
30

4.
53

7*
**
6
.0

83
**

*
21

.1
1
0.
04
8
0.
11
8

28
.4

28
**

*

IP

SA
S

—
—
—
—
—
—

4
.2

99
**
*
10
0
0.

20
4

0.
29

5
0.

24
6

O
th
er
s

8
−

.0
37

**
*

17
.8

9
−

0.
08
9
0.

01
6

44
.7

23
**

*
28

.2
64

**
*

25
.3

6
0.

23
5

0.
29

2
11

0.
41

5*
**

M
ea
su
re
u
se
d

R
eg

is
te

re
d

in
ha

bi
ta

nt
s

2

0
.2

94
**

*
23

.0
2

0.
25

6
0.

33
2

86
.8

65
**
*
10

.2
86

**
*

38
.9

8
0.

24
4

0.
32

7
25

.6
54

**

Po

pu
la

ti
o
n

lo
ga

ri
th

m
43

.1
30

**
*

31
.5

4
0.

10
9

0.
15

1
13

6.
35

0*
**
18
.2
91
**

*
35

.5
9

0.
25
2
0.

32
9

50
.5

78
**

*

T

o
ta

l r
ev

en
ue

6
.4

59
**
*
10
0
0.

38
2

0.
53

6
3.

51
9

3
.2

10
**

*
75

.2
9

0.
09
0
0.

33
0

3.
98

5

T
o
ta

l a
ss

et
s

3
.5

30
**

*
10
0
0.

44
6

0.
61

3
1.

61
1

2
.1

86
*

50
.1

5
0.

02
6

0.
34

5
3.

98
8*

N
um

be
r

o

f
pu

bl
ic

e
m

pl
o
ye

es
—

—
—
—
—

—
7

.1
17

**
*
10
0
0.
04
7
0.

18
6

3.
90

8

Le
ve
l o
f
go
ve
rn
m
en
t

St
at
e
go
ve
rn
m
en
t

17
.3

41
**

*
28

.1
6

0.
29
6
0.

38
6

60
.3

76
**

*
—
—
—
—
—
—

Lo
ca
l g
o
ve
rn
m
en
t

58
.1

25
**

*
20

.1
4

0.
10

5
2.

11
5

28
7.

93
3*

**
28

.1
94

**
*

15
.4

1
0.

17
1

0.
21

8
18

1.
74

6*
**

Lo
ca
l g
o
ve
rn
m
en
tc

A
ng
lo
-S
ax
o
n

52
.1

53
**

*
25

.5
8
0.
13
2
0.
17
4

20
3.

26
2*

**
11
.1
00
**
*

25
.5

6
0.
06
7
0.
13

3
43

.0
29
**
*

N
o
n-
A
ng
lo
-S
ax
o
n
6
−
.0
96
f
23
.0
2
−
0.
15
5
−
0.
03
7
26
.0
59
**

*
13

.3
07

**
*

34
.6

4
0.

26
9

0.
34

6
37

.5
25

**
*
M
un
ic
ip
al
w

ea
lt
h

A
dm
in
is
tr
at
iv
e
cu
lt
ur
e

A
ng
lo
-S
ax
o
n

27
.0

98
**

*
63

.5
8
0.
05
7
0.

13
9

42
.4

63
**

*
11

.0
69

**
*
10
0
0.
03
5
0.

10
3

7.
07

1

N
o
n-
A
ng
lo
-S
ax
o
n
—
—
—
—
—
—
5
.1

58
**

*
53

.0
4

0.
07
6
0.

24
1

9.
42

6*

A
cc
o
un
ti
ng
r
eg
im
es

G
A

SB
27

.0
98

**
*

63
.5

8
0.

05
7

0.
13

9
42

.4
63

**
*
5
.0

60
**

*
68

.7
0

0.
02
4
0.

09
6

7.
27

8*
**

IP
SA

S
—

—
—
—
—

—
5

.1
13

**
*
10
0
0.
01
2
0.

21
4

0.
17

2

O

th
er

s
—
—
—
—
—

—
6

.1
69

**
*

55
.7

2
0.

09
1

0.
24

7
10

.7
67

†

M
ea
su
re
u
se
d

Pe
r

ca
pi

ta
l c

it
iz

en
s’

in
co

m
e

18
.1

05
**

*
67

.8
6

0.
05
7
0.

15
2

26
.5

25
†

8
.0

35
**

*
42

.1
1
0.
00
0
0.

05
2

18
.9

97
**

*

O

w
n

re
ve

nu
e

pe
r

ca
pi

ta
8

.

1
24

**
*

77
.2

6
0.

04
2

0.
20

6
10

.3
55

10
.1

24
**

*
77

.2
6

0.
04
2
0.

20
6

10
.3

55

Le
ve
l o
f
go
ve
rn
m
en
t

St
at
e
go
ve
rn
m
en
t

9
.2

70
**
*
10
0
0.

18
2

0.
35

7
2.

43
8

—
—
—
—
—
—

Lo
ca
l g
o
ve
rn
m
en
t

19
.0

52
*

81
.4

0
0.

00
7

0.
09

8
23

.3
42

16
.0

81
**

*
66

.9
5

0.
05
0
0.

11
2

23
.8

97
**
*

Lo
ca
l g
o
ve
rn
m
en
t

A
ng
lo
-S
ax
o
n

18
.0

55
*

77
.0

7
0.

00
9

0.
10

1
23

.3
54

11
.0

69
**

*
10
0
0.

03
5

0.
10

3
7.

07
1

N
o
n-
A
ng
lo
-S
ax
o
n
—
—
—
—
—
—
5
.1
58
**
*
53
.0
4
0.
07
6
0.
24
1
9.
42
6*
N
ot

e.
G

A
SB

=
G

o
ve
rn
m
en
ta

l A
cc

o
un
ti
ng
S
ta

nd
ar

d
B

o
ar

d;
I
PS

A
S

=
I
nt

er
na

ti
o
na

l P
ub

lic
S

ec
to

r
A

cc
o
un
ti
ng
S
ta
nd
ar

ds
; E

U
=

E
co

no
m

ic
U

ni
o
n;
O
EC
D
=
O
rg
an
is
at
io
n
fo
r
Ec
o
no
m
ic
C
o
-O
pe
ra
ti
o
n
an
d
D
ev
el
o
pm
en
t.

a T
he

s
tu

dy
o

f
R

io
s

et
 a

l.
(2

01
3)

w
as

n

o
t

in
cl

ud
ed

w
he

n
w

e
an

al
yz

ed
“

A
dm
in
is
tr
at
iv
e

C
ul

tu
re

,”
b

ec
au

se
t

hi
s

ar
ti
cl

e
an
al
yz

ed
t

he
b

ud
ge

t
tr

an
sp

ar
en

cy
o

f
st

at
e

pu
bl

ic
a

dm
in
is
tr
at
io

n
ar

o
un

d
th

e
w

o
rl

d,
a

nd
t

he
re

a
re

d
iff

er
en

t
st

yl
es
o
f
pu
bl
ic
a
dm
in
is
tr
at
io

n
in

t
hi

s
sa

m
pl

e.
b R

io
s
et
 a
l.
(2
01
3)
w
as
t
he

o
nl

y
st

ud
y

ab
o
ut

b
ud

ge
t

tr
an

sp
ar

en
cy

o
f
st

at
e
pu
bl
ic
a
dm
in
is
tr
at
io

n.
c T

he
o

nl
in
e
st
ud
ie

s
ab

o
ut

f
in

an
ci
al
c
o
nd
it
io
n
an

d
si

ze
a

na
ly

ze
d

lo
ca

l g
o
ve
rn
m

en
ts

, a
nd

s
o
t

he
r

es
ul

ts
a

re
t

he
s

am
e

as
in

t
he

c
as

e
o
f
“L

o
ca

l A
dm

in
is
tr
at

io
n.

”
d T

he
h

ar
d-

co
py
s
tu
di
es

a
bo

ut
p

o
lit
ic
al

c
o
m

pe
ti
ti
o
n

an
al

yz
ed

lo
ca

l g
o
ve
rn
m
en
ts

in
A

ng
lo
-S
ax
o
n

“A
dm

in
is
tr
at
iv
e
C
ul
tu
re

,”
a

nd
s

o
t
he
r
es
ul
ts
a
re
t
he
s
am
e
as
in
t
he
c
as
e
o
f
“L
o
ca
l A
dm
in
is
tr
at
io
n.

”
e P

in
a

et
 a
l.
(2

00
7,

2
00

9
, 2

01
0)

; C
ab

a
Pé
re
z
et
 a
l.
(2

01
4)

; R
io

s
et

 a
l.

(2
01

3)
; a

nd
W

eh
ne
r
an
d
R
en
zi
o
(
20
13

)
ha

ve
n

o
t

be
en

in
cl

ud
ed
w
he
n
w
e
an
al
yz
ed
“
A
dm
in
is
tr
at
iv
e
C
ul
tu
re

”
an

d

“A
cc

o
un
ti
ng

R
eg

im
es
,”
b
ec
au
se
t

he
y

an
al
yz
ed
t
he
t
ra

ns
pa

re
nc

y
in

E
U

c
o
un

tr
ie

s,
O

EC
D

c
o
un
tr
ie

s,
a

nd
f
ro

m
t

he
in

te
rn
at
io

na
l p

er
sp

ec
ti
ve

.
f T

he
m

o
de

ra
ti
ng

e
ffe

ct
o

f
“A

dm
in
is
tr
at
iv

e
C

ul
tu

re
”

w
he

n
it
is

a
na

ly
ze

d
in

t
he

d
is

cl
o
su
re
o
f
pu
bl
ic
f
in
an
ci

al
in

fo
rm

at
io
n
in
h
ar
d
co

py
is

s
ig
ni
fic
an
t

at
t

he
1

2%
le

ve
l.

† p
<

.1
. *

p
<

.0
5.

*
*p

< .0

1.
*

**
p

< .0

01
.

T
a
b
le
3
.
(c
o
n
ti
n
u
e
d
)

564 American Review of Public Administration 47(5)

Anglo-Saxon administrative culture and for state governments.
However, the data available seem to indicate that public man-
agers in local governments are motivated to disclose informa-
tion in hard-copy format, rather than online.

Upon analyzing the data and taking into account the year
of publication of each study (see Table 4), we see that in
more recent years, financial condition has come to exert a
strong influence on public managers to disseminate informa-
tion in hard-copy format, especially in countries with an
Anglo-Saxon administrative culture. However, no such
change in influence is apparent with respect to the online
disclosure of public information, although here, too, the
influence is strong in countries with an Anglo-Saxon admin-
istrative culture. Nevertheless, managers in local govern-
ments are more likely to represent the financial condition of
their organizations in hard-copy format, rather than online (ρ
= .149 vs. ρ = .040).

Intergovernmental transfers. The funds and grants received by
governments is a factor that has been studied in detail by
many researchers; we found an average correlation of .115
and a CI = [.105, .124] (see Table 2). Although some authors
conclude that the receipt of funds from other public agencies
may discourage public managers from disclosing economic
and financial information (Giroux & Deis, 1993), our results
indicate that public managers are motivated to disclose
information when public administrators receive funds and
grants irrespective of the mode of information disclosure. In
addition, the file-drawer test shows that this association is
robust and not influenced by publication bias, which con-
firms H2. However, as the previous variable showed, the
limited explanatory power of the error variance (37.01%)
suggests a high degree of heterogeneity (Q = 115.609; ρ < .01) and the existence of factors moderating the relationship in question (see Figure 2).

Table 2 shows that although intergovernmental funds are
an incentive for public managers when economic and finan-
cial information is published as hard copy, this is not the
same when the mode of disclosure is online. In this sense,
the heterogeneity has disappeared (chi-square statistic is
nonsignificant); so, with such empirical evidence, we could
affirm that the environment in which financial information is
disclosed does exert a moderating effect and influences the
relationship between transfers and funds received and the
disclosure of public financial information (ME

1
).

Table 3 shows that the receipt of funds from another pub-
lic body is an incentive that influences public managers to
disclose information in hard-copy format, and its influence is
greater in countries with an Anglo-Saxon administrative cul-
ture (ρ = .118 vs. ρ = .102). The results also indicate that
public managers of municipal governments disclose more
financial information in hard copy to justify the use of
received funds (ρ = .118 vs. ρ = .072). In both cases, the vari-
ability within the data has disappeared (chi-square statistic is
nonsignificant). Therefore, with this empirical evidence, we

can affirm that the administrative culture and the level of
government are moderating effects and have influence in the
analyzed relationship (ME

2
and ME

5
).

In contrast, taking into account the accounting regimes,
the results are very similar to those of the administrative cul-
ture variable. Results show that the public managers of coun-
tries that have their own accounting systems have little
motivation to disclose financial information online.
However, the results are very different when the information
is released as hard copy. Public managers are more inclined
to disclose managerial information in countries using
GASB’s accounting regime. So, we could affirm that the
countries’ accounting requirements and practices is a moder-
ating effect on the analyzed relationship (ME

3
).

Regarding the measurement unit used, the association
between grants received and the disclosure of public finan-
cial information is stronger when quantified through current
and capital transfers rather than through ratio intergovern-
mental receipts. Also, we can observe that the variability
within the data has disappeared (chi-square statistic is non-
significant); so, the measure used to quantify the intergov-
ernmental transfers modulates the analyzed relationship
(ME

4
).

In Table 4, we observe that the receipt of funds from other
government agencies and the justification for their use is an
incentive to disclose financial information in hard-copy for-
mat and that this accountability requirement has increased
over time; hence, the relationship is stronger in recently pub-
lished articles than articles published between 1983 and
1999 (ρ = .126 vs. ρ = .009). Few studies have been pub-
lished regarding the disclosure of information through the
Internet, but the results available in this respect seem to show
that there is no statistically significant relationship.

Political competition. Regarding political competition (see
Table 2), the average correlation presented is .091, CI =
[.069, .113], although this positive association is statistically
significant and stable (fail-safe, as shown by the N value),
and so there is a positive relationship between political com-
petition and the disclosure of public financial information
irrespective of the mode of information disclosure (H3
should be accepted). However, the observed variance
(21.64%) reveals a high degree of heterogeneity in the stud-
ies analyzed (Q = 314.259; ρ < .001). Therefore, we believe it is necessary to search for the moderating effects that this variability may have produced within the reported findings.

In this sense, Table 2 shows that the public managers have
more motivation when financial information is disclosed
online (ρ = .201 vs. ρ = .034), than when it is provided in
hard-copy format; in both cases, the relationship is statisti-
cally significant. In addition, for the particular case of online
disclosure of financial information, the heterogeneity has not
disappeared. So, with these empirical tests, we could not
affirm that the format chosen for information disclosure has
a moderating effect on the reported results (ME

1
).

565

T
a
b

le
4

.
M
et
a-
A
na
ly
ti
c
D
at
a
o
n
th
e
R
el
at
io
n
B
et
w
ee
n
A
na
ly
ze
d
V
ar
ia
bl
es
a
nd
t
he
D
is
cl
o
su
re
o
f
Pu
bl
ic
F
in
an
ci
al
I
nf
o
rm
at
io

n
(P

ub
lis

he
d

St
ud

ie
s

19
83

-1
99

9
vs

.
20

00
-2

01
5)
.
In
de
pe
nd
en
t
va
ri
ab
le

Pu
bl

is
he

d
st

ud
ie

s

19
83
-1
99

9
(H

ar
d
co
py
)
Pu
bl
is
he
d
st
ud
ie

s
20

00
-2

01
5

H
ar
d
co
py
O
nl
in
e
K
M

co
rr
el
at
io
n
(ρ
)
%
S
2 e
/
S2 r
C
o
nf
id
en

ce

in
te

rv
al

(
95

%
)

χ2
k-
1
K
M

co
rr
el
at
io
n

(ρ
)
%
S
2 e
/
S2 r
C
o
nf
id
en
ce
in
te
rv
al

(9
5%
)
χ2
k-
1
K
M
ea

n

co
rr
el
at
io
n

(ρ
)
%
S
2 e
/
S2 r
C
o
nf
id
en
ce
in
te
rv
al

(9
5%
)
χ2
k-
1
M
in
im
um
M
ax
im
um
M
in
im
um
M
ax
im
um
M
in
im
um
M
ax
im
um
Fi
na
nc
ia
l c
o
nd
it
io

n
63

.0
84

**
*

80
.7

4
.0

64
.1

04
78

.0
27

26
.1

49
**

*
53

.4
4

0.
11
1
0.
18
6

48
.6

51
*

22
.0

40
**

*
50

.5
2

.0
02

.0
77

43
.5

47
**

A
dm
in
is
tr
at
iv
e
cu
lt
ur
ea

A
ng
lo
-S
ax
o
n

62
.0

84
**

*
78

.9
4

.0
64

.1
04

78
.5

38
21

.1
91

**
*

73
.6

2
0.
14
7

0.
23

5
28

.5
26

5
.1

04
*

10
0

.0
07

.2
00

1.
15

1

N
o
n-
A
ng
lo
-S
ax
o
n
—
—
—
—
—
—
5
.0

64
44

.3
9

−
0.
00
2
0.
13
0
11

.2
63

*
12

.0
42

†
37

.9
1

−
.0

01
.0

86
31

.6
56

**
*

A
cc
o
un
ti
ng
r
eg
im
es

G
A

SB
62

.0
84
**
*

78
.9

4
.0
64
.1
04
78

.5
38

21
.1

91
**

*
73

.6
2

0.
14

7
0.
23
5

28
.5

26
—

—
—
—
—
—

O
th
er
s
—
—
—
—
—
—
5
.0
64

44
.6

5
−

0.
00
2
0.

13
0

11
.1

99
**

*
19

.0
40

59
.4

4
−
.1
30

.2
11

31
.9
65
**

M
ea
su
re
u
se
d

Fi
na
nc
ia
l v
ia
bi
lit

y
7

.0
29

41
.9

8
−
.0
40

.0
99

16
.6

76
*

7
.1

36
**

*
10
0
0.

07
4

0.
19

9
2.

30
2

—
—
—
—
—
—

R
at

io
t

o
ta
l d
eb

t
52

.0
82

**
*
10
0

.0
60

.1
03

49
.8

70
17

.2
12

**
*

69
.0

0
0.
16
3
0.
26
1
24

.6
39

*
15

.0
43
39
.4
7
−

.0
04

.0
90

38
.0

07
**

Le
ve
ra
ge
4
.2

21
**

*
10
0
.1

37
.3

04
0.

16
4

—
—
—
—
—
—
6
.0
14
10
0
−

.0
51

.0
79

1.
50
4

Le
ve
l o
f
go
ve
rn
m
en
t

St
at
e
go
ve
rn
m
en
t
11
.0

96
10

0
.0
11
.1

82
3.

02
6
—
—
—
—
—
—
—
—
—
—
—
—

Lo
ca
l g
o
ve
rn
m
en
t

52
.0

83
**
*
68
.9
4

.0
63

.1
03

74
.4

23
*

26
.1
49
**
*
53
.4
4
0.
11
1
0.
18
6
48
.6
51
*

17
.0

53
**
49
.8
4
.0

13
.0

93
34

.1
00

**
In

te
rg
o
ve
rn
m
en
ta

l t
ra

ns
fe

r
23

.0
09

37
.4

3
−

.0
28

.0
47

61
.4

46
**

*
13

.1
26

**
*
10
0
0.
11
6
0.

13
6

11
.4

74
11

.0
10

10
0
−
.0

47
.0

66
2.

59
8

A
dm
in
is
tr
at
iv
e
cu
lt
ur
e

A
ng
lo
-S
ax
o
n

23
.0

09
37

.4
3
−
.0

28
.0

47
61

.4
46

**
*
9
.1

26
**

*
10
0
0.

11
6

0.
13

6
8.

88
7

—
—
—
—
—
—

N
o
n-
A
ng
lo
-S
ax
o
n
—
—
—
—
—
—
4
.1
02
**
10
0
0.
03
0
0.
17
5

2.
16

0
10
.0
02
10
0
−
.0
59
.0

63
2.

07
9

A
cc
o
un
ti
ng
r
eg
im
es

G
A

SB
23

.0
09
37
.4
3
−
.0
28
.0
47
61
.4
46
**

*
9

.1
26
**
*
10
0
0.
11
6
0.
13
6

8.
88

7
—

—
—
—
—
—

O
th
er
s
—
—
—
—
—

—
4

.1
02

**
10

0
0.

03
0

0.
17
5
2.

16
0

10
.0

02
10
0
−

.0
59

.0
63
0.
30
3

M
ea
su
re
u
se
d

C
ur

re
nt
a
nd

c
ap

it
al
t
ra
ns
fe

rs
—

—
—
—
—

—
9

.1
26
**
*
10
0
0.
11
6
0.
13
6
8.
88
5

R
at

io
in

te
rg
o
ve
rn
m
en
ta
l r
ec
ei
pt

s
23

.0
09
37
.4
3
−
.0
28
.0
47
61
.4
46
**

*
4

.1
02
**
10
0
0.
03
0
0.
17
5
2.
16
0
11
.0
10
10
0
−
.0
47

.0
66

2.
59
8

Le
ve
l o
f
go
ve
rn
m
en
t

St
at
e
go
ve
rn
m
en
t
6
.0

72
10

0
−
.0
42

.1
86

2.
04

8
—

—
—
—
—
—
—
—
—
—
—
—

Lo
ca
l g
o
ve
rn
m
en
t
17
.0

02
29

.0
0

−
.0
38
.0

42
58

.6
18

**
*

13
.1

26
**
*
10
0
0.
11
6
0.
13

6
11

.4
74

10
.0
02
10
0
−
.0
59
.0
63

2.
07

9
Po

lit
ic
al
c
o
m
pe
ti
ti
o
n

40
.0

30
†

55
.1

1
.0
01
.0

58
72

.5
81

**
*
4
.0

89
a

10
0
−
0.
00
6
0.
18
3
0.

04
8

24
.2

01
**

*
12

.0
9

.1
63

.2
39

19
8.

54
1*

**

A
dm
in
is
tr
at
iv
e
cu
lt
ur
ea

A
ng
lo
-S
ax
o
n
40
.0
30
†
55
.1
1
.0
01
.0
58
72
.5
81
**
*
4
.0
89
a
10
0
−
0.
00
6
0.
18
3
0.
04
8
—
—
—
—
—
—

N
o
n-
A
ng
lo
-S
ax
o
n
—
—
—
—
—
—
—
—
—
—
—
—
15
.1
16
**

*
90

.9
1

.0
74

.1
58

16
.4

99

A
cc
o
un
ti
ng
r
eg
im
es

G
A

SB
40

.0
30

†
55

.1
1

.0
01

.0
58

72
.5

81
**
*
4

.0
89

a
10

0
−
0.
00
6
0.

18
3

0.
04
8
—
—
—
—
—
—

O
th
er
s
—
—
—
—
—
—
—
—
—
—
—

—
22

.1
96

**
*

11
.3

1
.1

58
.2

34
19

4.
47

1*
**

M
ea
su
re
u
se
d

R
at

io
c

an
di

da
te

s/
av

ai
la

bl
e
po
si
ti
o
n
—
—
—
—
—
—
—
—
—
—
—
—
18
.2
91
**
*
9.
18
.2

42
.2

29
19

6.
11
7*
**

Pe
rc
en
ta
ge
s
o
f
vi

ct
o
ri

es
10
.0
75
*
58
.4
3
−
.0

02
.1

53
17

.1
15
*
—
—
—
—
—
—
—
—
—
—
—
—

D
ic
ho
to
m
o
us
v
ar
ia
bl

e
17

−
.0

15
10

0
−

.0
48

.0
18

12
.3

93
4

.0
89
*
10
0
−
0.
00
5
0.
18
3

1.
79

2
2

.1
92
**
95
.2
8

.0
70

.3
15

2.
09
9

In
di
ce
s
o
f
po
lit
ic
al
c
o
m
pe
ti
ti
o
n
5
.2
48
**
*
10
0
.1

32
.3

64
1.

28
1
—
—
—
—
—
—
4
.1
37
**
*
10
0
.0

85
.1

88
0.

85
4

%
s
ea
ts
h
el
d
by
m
in
o
ri
ty
p
ar
ty
6
.2
19
**
*
10
0
.1

13
.3

25
3.

18
5
—
—
—
—
—
—
—
—
—
—
—
—
(c
on
tin
ue
d)

566

In
de
pe
nd
en
t
va
ri
ab
le
Pu
bl
is
he
d
st
ud
ie
s

19
83
-1
99
9
(H
ar
d
co
py
)
Pu
bl
is
he
d
st
ud
ie
s
20
00
-2
01
5
H
ar
d
co
py
O
nl
in
e
K
M

co
rr
el
at
io
n
(ρ
)
%
S
2 e
/
S2 r
C
o
nf
id
en
ce

in
te
rv
al
(
95
%
)
χ2
k-
1
K
M

co
rr
el
at
io
n

(ρ
)
%
S
2 e
/
S2 r
C
o
nf
id
en
ce
in
te
rv
al

(9
5%
)
χ2
k-
1
K
M
ea
n

co
rr
el
at
io
n

(ρ
)
%
S
2 e
/
S2 r
C
o
nf
id
en
ce
in
te
rv
al

(9
5%
)
χ2
k-
1
M
in
im
um
M
ax
im
um
M
in
im
um
M
ax
im
um
M
in
im
um
M
ax
im
um

Le
ve
l o
f
go
ve
rn
m
en
t

St
at
e
go
ve
rn
m
en
t
21
.2
19
**
*
10
0
.1

62
.2

75
17

.1
02
—
—
—
—
—
—
9
.4
70
**
*
5.

13
.4

24
.5

17
17

5.
35
1*
**

Lo
ca
l g
o
ve
rn
m
en
t

19
−

.0
20

10
0
−
.0
52
.0

12
1.

07
2
4
.0
89
a
10
0
−
0.
00
6
0.
18
3
0.
04
8
15
.1
16
**
*
90
.9
1
.0
74
.1
58
16
.4

99
Si

ze
48

.1
64

**
*
18
.0
1
.1

43
.1

86
26

6.
45

1*
**

29
.1

24
**
*
21
.5
2
0.
08
9
0.
16
0

13
4.

72
8*

**
41

.1
84

**
*

20
.7

4
.1
62
.2

07
19

7.
70

0*
**

A
dm
in
is
tr
at
iv
e
cu
lt
ur
ea

A
ng
lo
-S
ax
o
n

45
.1

66
**
*
17

.0
1

.1
45

.1
87

26
4.

57
2*

**
24

.2
39
**
*
80
.7
9
0.

20
0

0.
27

8
29

.7
08

11
.1
00
**
*
25

.5
6

.0
67

.1
33

43
.0

29
**

*

N
o
n-
A
ng
lo
-S
ax
o
n
—
—
—
—
—
—
5
−

.1
09

**
*

20
.3

2
−

0.
10
7
−
0.
10

3
24

.6
05

**
*

14
.3

01
**

*
36

.3
4

.2
64

.3
39

38
.5

20
**

*

A
cc
o
un
ti
ng
r
eg
im
es

G
A

SB
45

.1
66

**
*
17
.0
1
.1
45
.1

87
26

4.
57

2*
**

22
.2

55
**

*
3.

42
0.

21
2

0.
29

7
64

3.
71

6*
**
6
.0
83
**
*
21
.1
1
.0
48
.1
18
28
.4
28
**
*

IP
SA
S
—
—
—
—
—
—
—
—
—
—
—
—
4
.2
99
**
*
10
0
.2

04
.2

95
0.

24
6

O
th
er
s
—
—
—
—
—
—
7
−
.0
42
**
*

15
.7

0
−
0.
09
6
0.

01
3

44
.5

73
**

*
28
.2
64
**
*
25
.3
6
.2

35
.2

92
11

0.
41
5*
**

M
ea
su
re
u
se
d

R
eg
is
te
re
d
in
ha
bi
ta
nt
s
17
.3

12
**

*
26

.4
0

.2
71

.3
53

64
.3

89
**

*
3

.2
20

**
*
15
.4
6
0.
06
0

0.
38

0
19

.4
09

**
*

10
.2

86
**

*
38

.9
8

.2
44

.3
27

25
.6

54
**

Po
pu

la
ti
o
n

lo
ga
ri
th

m
16

.2
16

**
*
10
0

.1
67

.2
64

8.
51

6
27

.1
13
**
*

23
.4

6
0.

09
4

0.
13
2
11

5.
09

1*
**
18
.2
91
**
*
35
.5
9

.2
52

.3
29

50
.5
78
**
*

T
o
ta
l r
ev
en
ue
3
.2
10
**
*
75
.2
9
.0
90

.3
30

3.
98
5

T
o
ta
l a
ss
et
s
3
.5
30
**
*
10

0
.4

46
.6

13
1.

61
1
—
—
—
—
—
—
2
.1
86
*
50
.1
5
.0

26
.3

45
3.

98
8*

N
um
be
r

o
f
pu

bl
ic
e
m
pl
o
ye
es
—
—
—
—
—
—
7
.1
17
**
*
10
0
0.
04
7
0.
18
6
3.
90
8
7
.1
17
**
*
10
0
.0
47
.1
86
3.
90
8

Le
ve
l o
f
go
ve
rn
m
en
t

St
at
e
go
ve
rn
m
en
t

14
.4

18
**
*
26
.1
6

.3
66

.4
71

53
.5

19
**
*
—
—
—
—
—

—
12

.1
18
**
*
10
0
.0
60

.1
75

9.
68

2

Lo
ca
l g
o
ve
rn
m
en
t

34
.1

42
**

*
19

.2
5

.1
21

.1
64

17
6.

60
4*

**
25

.1
14

**
*
19
.0
2
0.
07
5
0.
15

3
13

1.
43

**
*
29
.1
98
**
*
15

.5
3

.1
75

.2
22

18
6.

75
3*

**
M
un
ic
ip
al
w
ea
lt
h

21
.0

79
**

*
54

.7
3

.0
33

.1
26

38
.3

70
**
7
.1
35
**
*
10
0
0.
05
7
0.
21

4
3.

23
4
18
.0
83
**

*
70

.7
3

.0
52

.1
15

25
.4

49
**
*

A
dm
in
is
tr
at
iv
e
cu
lt
ur
e

A
ng
lo
-S
ax
o
n

20
.0

84
**
*
52
.8
6

.0
36

.1
32

37
.8

33
**

7
.1
35
**
*
10
0
0.
05
7
0.
21
4
3.
23
4
11
.0
69
**
*
10
0
.0

35
.1

03
7.

07
1

N
o
n-
A
ng
lo
-S
ax
o
n
—
—
—
—
—
—
—
—
—
—
—
—
5
.1
58
**
*
53
.0
4

.0
76

.2
41

9.
42
6*

A
cc
o
un
ti
ng
r
eg
im
es

G
A

SB
20

.0
84
**
*

52
.8

6
.0

36
.1

32
37

.8
33

**
7

.1
35

**
*
10
0
0.
05
7
0.
21
4

3.
23

4
5

.0
60
**
*

68
.7

0
.0

24
.0

96
7.

27
8*

**

IP
SA
S
—
—
—
—
—
—
—
—
—
—
—
—
5
.1

13
**

*
10
0
.0

12
.2

14
0.

17
2

O
th
er
s
—
—
—
—
—
—
—
—
—
—
—
—
6
.1
69
**

*
55

.7
2

.0
91

.2
47

10
.7

67
†

M
ea
su
re
u
se
d

Pe
r
ca
pi
ta
l c
it
iz
en
s’
in
co
m
e

12
.0

88
**

*
53

.6
9

.0
28

.1
48

22
.3

50
**

6
.1
34
**
*
10
0
0.

05
6

0.
21

3
3.

26
2

8
.0
35
**
*
42
.1
1

.0
00

.0
52
18
.9
97
**
*

O
w
n
re
ve
nu
e
pe
r
ca
pi
ta
8

.1
24

**
*
77
.2
6
.0
42
.2

06
10

.3
55
—
—
—
—
—
—
10
.1
24
**
*
77
.2
6
.0
42

.2
06

10
.3
55

Le
ve
l o
f
go
ve
rn
m
en
t

St
at
e
go
ve
rn
m
en
t
9
.2
70
**
*
10
0
.0

29
.5

10
2.

48
9

—
—
—
—
—
—
—
—
—
—
—
—

Lo
ca
l g
o
ve
rn
m
en
t
12
.0

14
83

.8
4

−
.0

04
.0

33
14

.3
13

7
.1
35
**
*
10
0
0.
05
7
0.
21
4
3.
23
4
16
.0
81
**
*
66
.9
5

.0
50

.1
12

23
.8
97
**
*
N
ot
e.
G
A
SB
=
G
o
ve
rn
m
en
ta
l A
cc
o
un
ti
ng
S
ta
nd
ar

d
B
o
ar

d;
IP

SA
S

=
In

te
rn
at
io

na
l P

ub
lic

S
ec

to
r

A
cc
o
un
ti
ng
S
ta
nd
ar
ds
; E
U
=
E
co
no
m
ic
U
ni
o
n;
O
EC
D
=
O
rg
an
is
at
io
n
fo
r
Ec
o
no
m
ic
C
o
-O
pe
ra
ti
o
n
an
d
D
ev
el
o
pm
en
t.

a P
in

a
et

 a
l.
(2
00

7,
2

00
9,

2
01

0)
; C

ab
a

Pé
re

z
et

 a
l.
(2
01

4)
; R

io
s
et
 a
l.
(2
01
3)

; a
nd

W
eh

ne
r

an
d

R
en

zi
o
(

20
13

)
w

er
e

no
t

in
cl
ud
ed
w
he
n
w
e
an
al
yz
ed
“
A
dm
in
is
tr
at
iv
e
C
ul
tu
re
,”
b
ec
au
se
t
he
y
an
al
yz
ed
t
he
t
ra
ns
pa
re
nc
y
in
E
U
c
o
un
tr
ie

s,

O
EC
D
c
o
un
tr
ie
s,
a
nd
f
ro
m
t
he
in
te
rn
at
io
na
l p
er
sp
ec
ti
ve

.
† p

< .1

. *
p

< .0

5.
*

*p
<

.0
1.

*
**

p
<

.0
01

.
T
a
b
le
4
.
(c
o
n
ti
n
u
e
d
)

Alcaide Muñoz et al. 567

The results described in Table 3 show that political com-
petition is an incentive for state governments to disclose
financial information in hard copy and online; in both cases,
the relationship observed is positive and statistically signifi-
cant. However, this relationship is stronger when the infor-
mation is disclosed via Internet than via hard-copy format (ρ
= .470 vs. ρ = .219). In summary, these empirical results
show that the administrative culture and the level of govern-
ment both constitute moderating effects on the results
reported in the studies of this question (ME

2
and ME
5
).

In the case of accounting regimes, there are no studies
regarding the disclosure through hard-copy format with their
own accounting regimes and studies about the disclosure
through the Internet with GASB’s accounting regimes,
although the results available in this respect seem to show
conclusive evidence that the accounting requirements is a
moderating effect.

However, when we focus on the measurement unit used,
we can see that the indices of political competition and per-
centage of seats held by the minority party offer the most
statistically significant positive association, followed dis-
tantly by percentages of victories, when hard-copy informa-
tion is analyzed. When the online information disclosure is
analyzed, the ratio of candidates and dichotomous variable
offer the most statistically significant positive associations,
followed closely by indices of political competition. In most
cases, the variability within the data has disappeared (chi-
square statistic is nonsignificant), so we affirm that the mea-
sure used to quantify the political competition modulates the
analyzed relationship (ME

4
).

In Table 4, we see that over time, political competition has
come to exert a great influence on public managers to dis-
close information, in any format, although this influence is
strongest with respect to online disclosure.

Environmental Factors

Size. The variable “size of the government” presents an
average correlation of .166 and CI of [.152, .180], and the
relation is positive and statistically significant (z = 9.53, ρ < .001). Moreover, the N fail-safe value reflects the stability of the relationship (see Table 2). These findings lead us to affirm that there is a positive relation between the size of government and the disclosure of public financial informa- tion irrespective of the mode of information disclosure (H4). However, the significance of the chi-square statistic (Q = 607.701, ρ < .001) suggests other factors might modulate this association, thus, we have to search for moderating effects that influence the analyzed relationship (see Figure 2).

The above table also shows that the relationship is stron-
ger when information is disclosed via the Internet than when
it is provided as hard copy (ρ = .184 vs. ρ = .154; ρ = .001).
However, the data variability persists (chi-square statistic is
significant), so we cannot conclude that the channel used for
disclosing financial information affects the results (ME

1
).

Table 3 shows that when the information is provided in
hard-copy format, the size of the government affects posi-
tively to the disclosure of financial information in countries
that act under an Anglo-Saxon administrative culture (ρ =
.182 vs. ρ = .096). However, in the case of online information
disclosure, the size of the population addressed by govern-
ment action positively influences information transparency;
this ratio is higher among governments where the administra-
tive culture is non-Anglo-Saxon (ρ = .307 vs. ρ = .100). In
contrast, when we focus on the level of government, we see
that state administrators are more motivated to disclose infor-
mation as hard copy, while local government officials seem to
be more motivated to disclose online. Due to the persistence
of variability (chi-square statistic is significant), the mode of
information disclosure and the level of government cannot be
considered moderating effects (ME

2
and ME
5
).

Taking into account the accounting regimes (see Table 3),
we can see that the relationship is strongest when the infor-
mation is disclosed via the Internet and, especially, when the
country has adopted the IPSAS, followed closely by the case
in which the country has adopted its own accounting stan-
dard (ρ = .299 vs. ρ = .264). Also, when the information is
disclosed in hard copy, a negative significant relationship
exists in cases of the country adopting its own accounting
standards. In this sense, we can see that the heterogeneity has
disappeared, so the accounting regimes adopted by the coun-
tries influence the analyzed relationship (ME

3
).

Regarding the measurement used to quantify the size of
government, we can see that the relationship is stronger
when this variable is quantified by the total revenue and the
total assets than when this variable is quantified by the regis-
tered inhabitants and the population logarithm, and, in the
case of hard-copy information, is statistically significant and
positive in all cases. Meanwhile, in the case of online infor-
mation, this relationship is stronger when the academic
researcher used registered inhabitants and a population loga-
rithm to quantify the size of government than when this vari-
able is quantified by the total revenue, the total assets, and
the number of public employees, which is statistically sig-
nificant and positive in all cases. Also, Table 3 shows that
there is no variability within the data, so we can affirm that
the measurement modulates the analyzed relationship (ME

4
).

In Table 4, we can see that in governments where the
Anglo-Saxon administrative culture prevails, a greater popu-
lation size encourages information disclosure, and this influ-
ence has increased over time. In addition, we confirm that in
countries with a non-Anglo-Saxon administrative culture,
this relation is negative and statistically significant for the
hard-copy provision of information, but is positive and sta-
tistically significant for online disclosure.

Municipal wealth. The “municipal wealth” variable pres-
ents a positive and statistically significant relationship (z =
8.13, ρ < .001, as shown by the CI of [.062, .112]), and there- fore we can affirm that there is a positive relation between

568 American Review of Public Administration 47(5)

municipal wealth and the disclosure of public financial
information irrespective of the mode of information disclo-
sure (H5—Table 2). Furthermore, the fail-safe test confirms
the stability of this relationship. Thus, we conclude that the
economic status of the population has a positive and statisti-
cally significant influence (ρ = .087, ρ < .001), moderating the disclosure of economic and financial information in the public sector, although the Q value (68.360; ρ < .05) suggests the existence of other factors that could modulate this asso- ciation—see Figure 2.

The results presented in Table 2 show that municipal
wealth is an incentive that exerts a stronger influence in the
case of hard-copy information disclosure (r = .094 vs. r =
.083) than in online format. In this case, the explanatory
power of the error variance has no increase and the level of
heterogeneity is the same, and so we cannot affirm that the
mode employed for the disclosure of public financial infor-
mation is not a moderating effect, because it has no influence
on the data (ME

1
).

Table 3 shows that municipal wealth is a determinant fac-
tor in government information disclosure in non-Anglo-
Saxon administrative cultures, and that it exerts a stronger
influence with respect to online information disclosure (ρ =
.158 vs. ρ = .069). In turn, public managers in countries with
an Anglo-Saxon administrative culture are more motivated to
disclose information in hard-copy format (ρ = .098 vs. r =
.069). Moreover, this variable has a stronger influence in state
administrations with respect to the provision of information
as hard copy (ρ = .270 vs. ρ = .052) than in local governments,
although the latter are more motivated to disclose information
online (ρ = .081). In these cases, the variability on the data has
disappeared, so the format in which the information is pro-
vided and the level of government are both moderating effects
on the findings reported (ME

2
and ME
5
).

However, the accounting regime adopted by the country
shows a stronger relationship when the information is dis-
closed by the Internet, and this relationship is greater when
the country adopted its own accounting requirement, fol-
lowed closely by the countries that adopted the IPSAS. In this
case, the heterogeneity has disappeared (chi-square statistic is
nonsignificant), so the accounting regimes adopted by the
countries have influence on the analyzed relationship (ME

3
).

When we focus on the measurement used to quantify the
municipal wealth, the relationship is stronger when this vari-
able is measured by countries’ own revenue per capita than
when it is measured by per capita citizens’ income irrespec-
tive of whether the information is disclosed via the Internet
or in hard-copy format. In this sense, the variability has dis-
appeared (chi-square statistic is nonsignificant), so we can
affirm that the measurement modulates the analyzed rela-
tionship (ME

4
).

In Table 4, it can be seen that municipal wealth has had a
stronger influence in more recent studies, and especially so
regarding the online disclosure of information. Thus, gov-
ernments with an Anglo-Saxon administrative culture have

been shown to be more motivated than non-Anglo-Saxon
governments to disclose hard-copy information according to
more recent studies and are also more strongly motivated
when the disclosure is made online.

Discussion and Conclusion

With respect to financial condition, public managers are moti-
vated to disclose and justify their management role by report-
ing financial information and to reduce the cost of debt to
increase resources available for other programs (Zimmerman,
1977); that is, they will have an interest in disclosing informa-
tion voluntarily to reduce information asymmetries and, thus,
the risk perceived by external agents, which in turn will
reduce the cost of borrowing (Gore, 2004). This result sup-
ports the view that agency theory is relevant for explaining
disclosure of public-sector financial reporting, because gov-
ernment financial statements are used by external users to
inform themselves of the financial condition of public agen-
cies (Mack & Ryan, 2006).

Moreover, recent studies have shown that public managers
face increasing demands from citizens for financial informa-
tion, especially in view of rapid changes taking place in the
technological innovations that have led to alternative mecha-
nisms of accountability (Schillemans, Van Twist, &
Vanhommerig, 2013). Taking into account the assumptions of
neoinstitutional theory and legitimacy theory, policymakers
and public managers have adopted innovations in management
and accounting and have implemented the new technologies
that have favored the transparency and accountability, as a
result of the global tendencies in adopting homogeneous prac-
tices and structures. Also, they try to inform about management
of governmental organizations and their financial health in
order to the possible influence in citizenry’s opinion.

In our study, this influence is exerted more strongly in
countries where the administrative culture is Anglo-Saxon,
and this factor motivates public managers to disclose finan-
cial information both online and as hard copy. This associa-
tion is modulated by the administrative culture, a finding that
is consistent with previous studies (Rodríguez Bolívar et al.,
2015) and is according to the height of the level of informa-
tion disclosure by public bodies where Anglo-Saxon tradition
prevails; in common law countries, the government relies
heavily on markets as sources of capital (La Porte, Demchak,
& De Jong, 2002), and so there is a highly developed tradition
of public financial accountability and transparency. Therefore,
public managers, especially in Anglo-Saxon countries, should
establish financial transparency policies to reduce the pres-
sure to which they are exposed.

In addition, in countries where governmental accounting
standards have been professionalized, the direction and extent
of changes to the standards will depend on the number of orga-
nizations involved and the strength of support for the financial
standard bodies within these organizations (Cheng, 1992).
These financial standard bodies have a notable influence on

Alcaide Muñoz et al. 569

the public-sector accounting reform implemented to improve
public financial accountability and transparency (Cheng,
1992). In this regard, according to the ideas of neoinstitutional
theory, public-sector organizations have had to face external
pressures and adopt these standards and practices issued by
these financial standard bodies that are considered legitimate
and socially acceptable.

Furthermore, the results obtained show that financial con-
dition has a more positive influence on the public managers
of state rather than local governments. This evidence leads us
to believe that either the contextual and environmental fac-
tors or the pressure exerted by stakeholders in state office-
holders provides greater accountability when debt levels are
high and prompts these officials to release financial state-
ments. Indeed, according to the neoinstitutional theory, orga-
nizations must respond to external pressure or innovative
trends, and, as information disclosure is a symbol of trust and
modernity, this constitutes a trend that governments should
adopt in response to those factors (Hoffman, 1999). For this
reason, state government administrators are more motivated
to minimize government debt and the costs arising from it, as
this will reduce the pressure on them to disclose information
and, at the same time, reduce the tax burden on the popula-
tion—a situation that would probably be favorably viewed at
election time (Laswad et al., 2005) and maintain the coun-
try’s credit rating (legitimacy theory approach).

Also, our results indicate that, over time, the govern-
ment’s financial condition is increasingly a factor that influ-
ences public managers to disclose financial information to
defend the policies adopted and show that the government’s
situation is financially sustainable (Rodríguez et al., 2014).
This result may be a consequence of the economic and finan-
cial crisis that has severely impacted governments and soci-
ety in recent years. In fact, from the point of view of the
legitimacy theory, public managers could disclose public
information with the aim of showing the management and
financial situation of governments, seeking to change the
possible negative impact of management cuts. These find-
ings underline the need to create transparency policies and
portals to communicate financial information effectively.

With respect to intergovernmental transfers, meta-ana-
lytic data show that intergovernmental grants have a great
influence on public financial information disclosed by local
authorities, who are obliged to account for transfers received
from state or supranational administrations, usually in hard-
copy format. In fact, these entities are particularly motivated
to disclose public economic and financial information and to
demonstrate that the funds received have been used in accor-
dance with the requirements of the program for which they
were assigned (Copley, 1991), thus corroborating the effi-
ciency of their management and the good use of resources
(Robbins & Austin, 1986). In brief, the principal–agent rela-
tionship discussed in agency theory is also seen here; local
government agents need to justify the use of financial
resources to demonstrate they have acted according to their

responsibilities. Moreover, this pressure has increased over
time, perhaps as a result of the greater control exercised over
public finances in response to the present economic and
financial crises.

In addition, public-sector financial transparency and
accountability are greater when the government has no abso-
lute majority, and opposition political parties demand the full
disclosure of public financial information to exercise their
role of financial control and present effective opposition.
Moreover, highly competitive political environments are
conducive to the adoption of communication strategies via
websites and technological reforms and to the provision of
high levels of information transparency (Tolbert, Mossberger,
& McNeal, 2008), thus allowing the continuous monitoring
of governance, which in turn can benefit from the use of new
technologies, as shown by our findings.

Currently, this is especially relevant at the central govern-
ment level, perhaps due to economic crisis and fiscal distress
experienced by central governments in many countries,
which put politicians under more pressure to disclose infor-
mation. According to the neoinstitutional theory, politicians
in central government play a leading role in this question,
which constitutes models for municipal politicians regarding
transparent management and the performance of their finan-
cial duties (Rodríguez Domínguez et al., 2011).

Regarding the size of the government and taking into
account the pronouncements of agency theory, public man-
agers are pressured by the population to disclose information
on their performance (Moon & Norris, 2005) and to accept a
high level of control (Evans & Patton, 1987; Serrano et al.,
2009). In addition, the preparation and disclosure of finan-
cial information is costly (Herian, Hamm, Tomkins, & Pytlik,
2012) and represents a fixed cost for the organization. In
comparison to small- and medium-sized cities, larger ones
can afford to invest financial and technical resources that are
necessary to prepare and disclose financial information and,
consequently, are more likely to provide such information.
Therefore, expenditure by large governments can and should
be devoted to adopting needed measures for offering appro-
priate financial transparency.

Finally, the results obtained confirm that municipal wealth
is a variable that is positively associated with the level of dis-
closure of public information, and so municipalities with a
higher per capita income disclose a greater volume of public
financial information (Giroux & McLelland, 2003; Ingram,
1984). In addition, previous investigations have shown that
wealthier governments have greater access to this informa-
tion through the implementation of new technologies; thus,
there is a relationship between per capita income and access
to new technologies (Ho, 2002). Hence, the associations
found in our meta-analytic study are strongest with respect to
information disclosed online. Therefore, access to the Internet
and the public financial information disclosed via this mode
depend to a large extent on municipal wealth and on the per
capita income of the population (Serrano et al., 2009). This

570 American Review of Public Administration 47(5)

result could be explained by the legitimacy theory, since pub-
lic managers seek to transmit the perception that their actions
meet stakeholders’ expectations through the dissemination of
information (Archel et al., 2009), and they, therefore, improve
citizens’ access to information and acquire legitimacy before
a well-informed and educated citizenry.

In regard to the findings made in this study, future research
could usefully identify the main stakeholders who influence
information disclosure and make use of surveys to directly
examine the need for public-sector financial information.
Nevertheless, the present article represents a first step toward
achieving a framework to design the complex econometric
model previously mentioned, and our findings show that the
five variables analyzed should be included in any future
econometric model in this area.

All these considerations lead us to inquire whether it is
possible to construct a single model of incentives for the dis-
closure of public financial information, one that is valid for
all environments. Nonetheless, it appears that the disclosure
of state information consistently disappoints: There is never
enough of it (Fenster, 2015). Open Data projects could help
to solve this problem, especially for the disclosure of public
financial information. In this regard, the English Open Data
experiment has especially targeted financial information and
would be a useful place to start to examine behavioral and
institutional factors that are important to the research ques-
tions proposed in this article. Therefore, future studies should
further investigate such aspects by adding new variables and
seeking effects that might modulate the relations we have
examined and subsequently develop a framework enabling
public managers and citizens to obtain a useful instrument
that will facilitate government accountability.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect
to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for
the research, authorship, and/or publication of this article: This
research was carried out with financial support from the Regional
Government of Andalusia (Spain), Department of Innovation, Science
and Enterprise with Research Project (number PII-SEJ-7700).

References

Albalate del Sol, D. (2013). The institutional, economic and social
determinants of local government transparency. Journal of
Economic Policy Reform, 16(1), 90-107.

Allen, A., & Sanders, G. D. (1994). Financial disclosure in US
municipalities: Has the Governmental Accounting Standars
Board made a difference? Financial Accountability &
Management, 10(3), 0267-4424.

Archel, P., Husillos, J., Larrinaga, C., & Spence, C. (2009).
Social disclosure, legitimacy theory and the role of the state.
Accounting, Auditing & Accountability Journal, 22, 1284-1307.

Baber, W. R. (1983). Towards understanding the role of auditing
in the public sector. Journal of Accounting & Economics, 5,
213-227.

Baber, W. R., & Sen, P. K. (1984). The role of generally accepted
reporting methods in the public sector: An empirical test.
Journal of Accounting and Public Policy, 3, 91-106.

Baber, W. R., & Sen, P. K. (1986). The political process and the
use of debt financing by state governments. Public Choice, 48,
201-215.

Banker, R. D., Bunch, B. S., & Strauss, R. P. (1989). Factors influ-
encing school district financial reporting practices. Research in
Governmental and Nonprofit Accounting, 5, 27-56.

Beattie, V., & Pratt, K. (2003). Issues concerning web-based busi-
ness reporting: An analysis of the views of interested parties.
The British Accounting Review, 35, 155-187.

Bel, G., Fageda, X., & Warner, M. (2010). A meta-regression
analysis of solid waste and water services. Journal of Policy
Analysis and Management, 29, 553-577.

Birchall, C. (2015). “Data.gov-in-a-box”: Delimiting transparency.
European Journal of Social Theory, 18, 185-202.

Bouckaert, G., & van de Walle, S. (2003). Comparing measures of
citizen trust and user satisfaction as indicators of “good gover-
nance”: Difficulties in linking trust and satisfaction indicators.
International Review of Administrative Sciences, 69, 329-343.

Brusca, I., & Montesinos, V. (2006). Are citizens significant
users of government financial information? Public Money &
Management, 26, 205-209.

Business Reporting Research Project. (2002). Electronic distribu-
tion of business reporting information (Steering Committee
Report Series). Financial Accounting Standards Board.
Retrieved from http://www.fasb.org/brrp/brrp1.shtml

Caamaño-Alegre, J., Lago-Peñas, S., Reyes-Santias, F., & Santiago-
Boubeta, A. (2013). Budget transparency in local governments:
An empirical analysis. Local Government Studies, 39, 182-207.

Caba, C., Rodríguez, M. P., & López, A. M. (2008). e-Government
process and incentives for online public financial information.
Online Information Review, 32, 379-400.

Caba Pérez, M. C., Rodríguez Bolívar, M. P., & López Hernández,
A. M. (2014). The determinants of government financial res-
ports online. Transylvanian Review of Administrative Sciences,
42, 5-31.

Cárcaba, A., & García, J. (2008). Factors behind the disclosure of
accounting information online on behalf of local governments.
Revista Española de Financiación y Contabilidad, 137, 63-84.

Cárcaba, A., & García, J. (2010). Determinants of online report-
ing of accounting information by Spanish local government
authorities. Local Government Studies, 36, 679-695.

Carpenter, V. L. (1991). The influence of political competition on
the decision to adopt GAAP. Journal of Accounting and Public
Policy, 10, 105-134.

Carpenter, V. L., & Feroz, E. H. (2001). Institutional theory and
accounting rule choice: An analysis of four US state govern-
ments’ decisions to adopt generally accepted accounting prin-
ciples. Accounting, Organizations and Society, 26, 565-596.

Cheng, R. H. (1992). An empirical analysis of theories on factors
influencing state government accounting disclosure. Journal of
Accounting and Public Policy, 11, 1-42.

Christiaens, J. (1999). Financial accounting reforms in Flemish
municipalities: An empirical investigation. Financial
Accountability & Management, 15, 21-40.

Alcaide Muñoz et al. 571

Connolly, R., Bannister, F., & Keamey, A. (2010). Government
website service quality: A study of the Irish Revenue online ser-
vice. European Journal of Information Systems, 19, 649-667.

Cook, F. L., Jacobs, L. R., & Kim, D. (2010). Trusting what you
know: Information, knowledge, and confidence in social secu-
rity. The Journal of Politics, 72, 397-412.

Copley, P. A. (1991). The association between municipal disclosure
practices and audit quality. Journal of Accounting and Public
Policy, 10, 245-266.

Curtin, D., & Meijer, A. (2006). Does transparency strengthen
legitimacy? A critical analysis of European Union policy docu-
ment. Information Policy. Available at SSRN: http://ssrn.com/
abstract=1434862 or http://dx.doi.org/10.2139/ssrn.1434862

Da Costa Bairral, M. A.l, Coutinho e Silva, A. H., & Dos Santos
Alves, F. J. (2015). Transparência no setor público: uma
análise dos relatórios de gestao anuais de entidades públicas
federais no ano de 2010. Revista de Administraçao Pública,
49(3), 643-675.

da Costa Carvalho, J. B., Camões, P. J., Jorge, S. M., & Fernandes,
M. J. (2007). Conformity and diversity of accounting and
financial reporting practices in Portuguese Local Government.
Canadian Journal of Administrative Sciences, 24, 2-14.

Daniels, J. D., & Daniels, C. E. (1991). Municipal financial reports:
What users want. Journal of Accounting and Public Policy, 10,
15-38.

Debreceny, R., Gray, G. L., & Rahman, A. (2002). The determi-
nants of Internet financial reporting. Journal of Accounting and
Public Policy, 21, 371-394.

Dwyer, P. D., & Wilson, E. R. (1989). An empirical investigation
of factor81s affecting the timeliness of reporting by munici-
palities. Journal of Accounting and Public Policy, 8(1), 29-55.

Edmiston, K. D. (2003). State and local e-government. The American
Review of Public Administration, 33, 20-45.

Evans, J. H., III, & Patton, J. M. (1983). An economic analysis
of participation in the municipal finance officers’ association
certificate of conformance program. Journal of Accounting &
Economics, 5, 151-175.

Evans, J. H., III, & Patton, J. M. (1987). Signaling and monitoring
in public-sector accounting. Journal of Accounting Research,
25(Suppl.), 130-158.

Fenster, M. (2015). Transparency in search of a theory. European
Journal of Social Theory, 18, 150-167.

Feroz, E. H., Carpenter, V. L., & Cheng, R. H. (2007). Toward
an empirical institutional theory. Corporate Ownership and
Control, 4(4), 30-46.

Feroz, E. H., & Wilson, E. R. (1992). Market segmentation and
the association between municipal financial disclosure and Net
Interest Costs. The Accounting Review, 67(3), 480-495.

Flyverbom, M. (2015). Sunlight in cyberspace? On transparency
as a form of ordering. European Journal of Social Theory, 18,
168-184.

Gandía, J. L., & Archidona, M. C. (2008). Determinants of web
site information by Spanish city councils. Online Information
Review, 32, 35-57.

García, E., & Sánchez, J. P. (2010). The association of board indepen-
dence and ownership concentration with voluntary disclosure: A
meta-analysis. European Accounting Review, 19, 603-627.

Garcia-Tabuyo, M., Sáez-Martín, A., & Caba-Pérez, M. C. (2015).
Mandatory versus voluntary disclosures: Drivers of proac-
tive information provision by local governments in Central
America. Information Development, 1-17.

Giroux, G. (1989). Political interest and government accounting dis-
closure. Journal of Accounting and Public Policy, 8, 199-217.

Giroux, G., & Deis, D. (1993). Investor interest and government
accounting disclosure. Accounting, Auditing & Accountability
Journal, 63, 63-78.

Giroux, G., & McLelland, A. J. (2003). Governance structures and
accounting at large municipalities. Journal of Accounting and
Public Policy, 22, 203-230.

Gore, A. K. (2004). The effects of GAAP regulation and bond
market interaction on local government disclosure. Journal of
Accounting and Public Policy, 23, 23-52.

Gore, A. K., Sachs, K., & Trzcinka, C. (2004). Financial disclosure and
bond insurance. Journal of Law and Economics, XLVII, 275-306.

Grimmelikhuijsen, S. G., Porumbescu, G., Hong, B., & Im, T.
(2013). The effect of transparency in trust in government: A
cross-national comparative experiment. Public Administration
Review, 73, 575-586.

Guillamón, M. D., Bastida, F., & Benito, B. (2011). The deter-
minants of local government’s financial transparency. Local
Government Studies, 37, 391-406.

Guillamón, M. D., Bastida, F., & Benito, B. (2013). The electoral
budget cycle on municipal police expenditure. European
Journal of Law Economics, 36, 447-469.

Hary, L. E., & Antonio, J. F. (1999). What users want in govern-
ment financial reports. Journal of Accounting, 170, 91-98.

Herian, M. N., Hamm, J. A., Tomkins, A. J., & Pytlik, L. M. (2012).
Public participation, procedural fairness, and evaluations of
local governance: The moderating role of uncertainty. Journal
of Public Administration Research and Theory, 22, 815-840.

Ho, A. (2002). Reinventing local governments and the e-govern-
ment initiative. Public Administration Review, 62, 434-444.

Hoffman, A. J. (1999). Linking organizational and field-level
analyzes: The diffusion of corporate environmental practice.
Organization Environment, 14, 133-156.

Hofstede, G. (2001). Culture’s consequences: Comparing values,
behaviors, institutions, and organizations across nations (2nd
ed.). Thousand Oaks, CA: Sage.

Hunter, J. E., & Schmidt, F. L. (1990). Dichotomization of conti-
nous variables: The implications for meta-analysis. Journal of
Applied Psychology, 75(3), 334-349.

Hunter, J. E., & Schmidt, F. L. (2004). Methods of meta-analysis:
Correcting error and bias in research findings (2nd ed.).
Thousand Oaks, CA: Sage.

Ingram, R. W. (1984). Economic incentives and the choice of
state government accounting practices. Journal of Accounting
Research, 22, 126-144.

Ingram, R. W., & DeJong, D. V. (1987). The effect of regulation on
local government disclosure practices. Journal of Accounting
and Public Policy, 6, 245-270.

Ingram, R. W., & Robbins, W. A. (1988, April). Financial report-
ing practices of local governments: An overview. Government
Finance Review, 17-21.

Institute of Chartered Accountants in England and Wales (ICAEW)
(1998). The 21st Century Annual Report. ICAEW, London

Inter-Agency Standing Committee. (1999). Business report-
ing on the Internet: A report prepared for the International
Accounting Standards Committee (Prepared by A. Lymer, R.
Debreceny, G. L. Gray, & A. Rahman). London, England:
International Accounting Standards Committee.

Ipsos MORI. (2010). What do people want, need and expect from
public services? London, England: 2020 Public Services Trust.

572 American Review of Public Administration 47(5)

Jensen, R., & Xiao, J. Z. (2001). Customized financial reporting,
networked databases and distributed file sharing. Accounting
Horizons, 15, 202-222.

Justice, J. B., Melitski, J., & Smith, D. L. (2006). E-Government
as an isntrument of Fiscal Accountability and Responsiveness:
Do the best practitioners employ the best practices? American
Review of Public Administration, 36(3), 301-322.

Keerasuntonpong, P., Dunstan, K., & Khanna, B. (2015). Factors
influencing disclosures of statements of services performance
of New Zealand local authorities. Pacific Accounting Review,
27, 304-328.

Kickert, W. J. M. (1997). Public governance in the Netherlands:
An alternative to Anglo-American managerialism. Public
Administration, 75, 731-752.

La Porte, T. M. C., Demchak, C. C., & De Jong, M. (2002).
Democracy and bureaucracy in the age of the web. Empirical
Findings and Theoretical Speculations. Administration &
Society, 34(4), 411-446.

Laswad, F., Fisher, R., & Oyelere, P. (2005). Determinants of vol-
untary Internet financial reporting by local government author-
ities. Journal of Accounting and Public Policy, 24, 101-121.

Lim, S., & McKinnon, J. (1993). Voluntary disclosure by NSW
statutory authorities: The influence of political visibility.
Journal of Accounting and Public Policy, 12, 189-216.

Lipsey, M. W., & Wilson, D. B. (2001). Applied Social
Research Methods, Vol. 49: Practical meta-
analysis. Thousand Oaks, CA: Sage.

Litan, E. R., & Wilson, P. (2000, May 24). Corporate disclosure
in the Internet age. The Financial Times. Available at SSRN:
http://ssrn.com/abstract=235274

Lowatcharin, G., & Menifield, C. E. (2015). Determinants of
internet-enabled transparency at the local level: A study of
Midwestern County Web Sites. State and Local Government
Review, 1-14.

Mack, J., & Ryan, C. (2006). Reflections on the theoretical underpin-
ning of the general-purpose financial reports of Australian gov-
ernment departments. Accounting, Auditing & Accountability
Journal, 19, 592-612.

Malone, D. (2006). An exploration of municipal financfial disclo-
sure and certain dimensions of political culture. Academy of
Accounting and Financial Studies Journal, 10(1), 7-23.

Martani, D., & Lestiani, A. (2012). Disclosure in Local Government
Financial Statements: the Case of Indonesia. Global Review of
Accounting and Finance, 3 (1), 67-84.

McLelland, A. J., & Giroux, G. (2000). An empirical analysis
of auditor report timing by large municipalities. Journal of
Accounting and Public Policy, 19, 263-281.

Mergel, I. (2013). Social media adoption and resulting tactics in the
U.S. Federal Government. Government Information Quarterly,
30, 123-130.

Mills, J. R., Carslaw, C., & Mason, R. (2007). Audit timeliness of
School Distritc audits. Journal of Public Budgeting, Accounting
& Financial Management, 19(3), 290-316.

Moon, D. F., & Norris, M. J. (2005). Advancing e-government at
the grassroots: Tortoise or hare? Public Administration Review,
65, 64-75.

Park, C. M., & Shin, D. C. (2005). Social capital and democratic
citizenship: The case of South Korea. Japanese Journal of
Political Science, 6, 63-85.

Patten, D. (1992). Intra-industry environmental disclosures in
response to the Alaskan oil spill: A note on legitimacy theory.
Accounting, Organizations and Society, 17, 471-475.

Pina, V., Torres, L., & Royo, S. (2007). Are ICTs improving trans-
parency and accountability in the EU regional and local gov-
ernments? An empirical study. Public Administration, 85(2),
449-472.

Pina, V., Torres, L., & Royo, S. (2009). E-government evolution
in EU local government: A comparative perspective. Online
Information Review, 33, 1137-1168.

Pina, V., Torres, L., & Royo, S. (2010). Is e-government promoting
convergence towards more accountable local governments?
International Public Management Journal, 13, 350-380.

Piotrowski, S. J., & Van Ryzin, G. G. (2007). Citizen attitudes
toward transparency in local government. American Review of
Public Administration, 37, 306-323.

Pomeroy, B., & Thornton, D. B. (2008). Meta-analysis and the
accounting literature: The case of audit committee indepen-
dence and financial reporting quality. European Accounting
Review, 17, 305-330.

Powell, W. W., & DiMaggio, P. J. (Eds.). (1991). The new institu-
tionalism in organizational analysis. Chicago, IL: University
of Chicago Press.

Rios, A. M., Benito, B., & Bastida, F. (2013). Determinants of
central government budget disclosure: An international com-
parative analysis. Journal of Comparative Policy Analysis:
Research and Practice, 15(3), 235-254.

Robbins, W. A., & Austin, K. R. (1986). Disclosure quality in gov-
ernmental financial reports: An assessment of the appropriate-
ness of a compound measure. Journal of Accounting Research,
24, 412-421.

Rodríguez Bolívar, M. P., Caba, C., & López, A. M. (2007).
E-government and public financial reporting: The case of
Spanish regional government. American Review of Public
Administration, 37, 142-177.

Rodríguez Bolívar, M. P., Caba, C., & López, A. M. (2015). Online
budget transparency in OECD member countries and admin-
istrative culture. Administration and Society, 47, 943-982.
doi:10.1177/0095399713509238

Rodríguez Bolívar, M. P., Navarro Galera, A., Alcaide Muñoz,
L., & López Subire, M. D. (2014). Factors influencing local
government financial sustainability: An empirical study. Lex-
Localis – Journal of Local Self-Government, 12(1), 31-54.

Rodríguez Bolívar, M. P., Navarro Galera, A., Alcaide Muñoz,
L., & López Subirés, M. D. (2016). Risk factors and drivers
of financial sustainability in local government: An empirical
study. Local Government Studies, 46, 29-51.

Rodríguez Domínguez, L., García, I., & Gallego, I. (2011).
Determining factors of e-government development: A world-
wide national approach. International Public Management
Journal, 14, 218-248.

Rosenberg, M. S. (2005). The file-drawer problem revisited: A
general weighted method for calculating fail-safe numbers in
meta-analysis. Evolution, 59, 464-468.

Rosenthal, R. (1979). The “file drawer problem” and tolerance for
null results. Psychology Bulletin, 86, 638-641.

Ryan, C., Stanley, T., & Nelson, M. (2002). Accountability disclo-
sures by Queensland Local Government Councils: 1997-1999.
Financial Accountability & Management, 18(3), 0267-4424.

Alcaide Muñoz et al. 573

Schick, A. (2003). The performing state: Reflection on an idea
whose time has come but whose implementation has not.
OECD Journal on Budgeting, 3, 70-103.

Schillemans, T., Van Twist, M., & Vanhommerig, I. (2013).
Innovation in accountability: Learning through interactive,
dynamic, and citizen-initiated forms of accountability. Public
Performance & Management Review, 36, 407-435.

Serrano, C., Rueda, M., & Portillo, P. (2009). Factors influencing
e-disclosure in local public administration. Environment and
Planning C: Government and Policy, 27, 355-378.

Smith, K. A. (2004). Voluntarily reporting performance measures
to the public: A test of accounting reports from U.S. cities.
International Public Management Journal, 7, 19-48.

Stanley, T. D. (2005). Beyond publication bias. Journal of Economic
Surveys, 19, 209-3445.

Steccolini, I., Anessi-Pessina, E., & Nasi, G. (2008). Accounting
reforms: Determinants of local government choices. Financial
Accountability & Management, 24(3), 0267-4424.

Styles, A. K., & Tennyson, M. (2007). The accessibility of finan-
cial reporting of U.S. municipalities on the Internet. Journal
of Public Budgeting, Accounting & Financial Management,
19(1), 56-92.

Suchman, M. C. (1995). Managing legitimacy: Strategic and insti-
tutional approaches. The Academy Management Review, 20,
571-610.

Taylor, D. W., & Rosair, M. (2000). The effect of participating par-
ties, the public and size on government departments’ account-
ability disclosure in annual reports. Accounting, Accountability
& Performance, 6(1), 77-97.

Thomas, J. C., & Streib, G. (2003). The new face of government:
Citizen-initiated contacts in the era of e-government. Journal
of Public Administration Research and Theory, 13, 83-102.

Thompson, F. (1998). Public economics and public administration.
In J. Rabin, W. B. Hildreth, & G. J. Miller (Eds.), Handbook of
public administration (2nd ed., pp. 995-1063). New York, NY:
Marcel Dekker.

Tolbert, C. J., Mossberger, K., & McNeal, R. (2008). Institutions,
policy and e-government in the American States. Public
Administration, 68, 549-563.

Torres, L., Pina, V., & Acerete, B. (2006). E-governance develop-
ment in European Union cities: Reshaping government’s rela-
tionship with citizens. Governance, 19, 277-302.

Unerman, J., & Bennett, M. (2004). Increased stakeholder dialogue
and the Internet: Towards greater corporate accountability or
reinforcing capitalist hegemony? Accounting Organizations
and Society, 29, 685-707.

Wehner, J., & De Renzio, P. (2013). Citizens, legislators, and exec-
utive disclosure: The political determinant of fiscal transpar-
ency. World Development, 41, 96-108.

Zafra, J. L., López, A. M., & Hernández, A. (2009). Developing an
alert system for local governments in financial crisis. Public
Money & Management, 29, 175-181.

Zhao, F., Shen, K. N., & Collier, A. (2014). Effects of national cul-
ture on e-government diffusion—A global study of 55 coun-
tries. Information & Management, 51, 1005-1016.

Ziek, P. (2009). Making sense of CSR communication. Corporate
Social Responsibility and Environmental Management, 16,
137-145.

Zimmerman, J. L. (1977). The municipal accounting maze: An anal-
ysis of political incentives. Journal of Accounting Research,
15(Suppl.), 107-144

Author Biographies

Laura Alcaide Muñoz is an associate professor in Accounting at
the University of Granada. She is interested in how E-government
has favoured the process of reform and modernization of Public
Administrations, giving rise to greater accessibility to public infor-
mation and services, and information transparency.

Manuel Pedro Rodríguez Bolívar is a professor in Accounting at
the University of Granada. He has authored numerous articles in
international journals, and is a member of the Editorial Board of
Government Information Quarterly.

Antonio M. López Hernández is a professor of Accounting at the
University of Granada. He teaches public sector management and
control. His research interests are center of management systems
and financial information in federal and local government.

Niels Opstrup is assistant professor in

the Department of Political Science and

Public Management at the University of

Southern Denmark. His research focuses

on the management and organization of

public organizations and, specifi cally, on the

governance of universities and management

of academics. He has also investigated

developments in the backgrounds and

career paths of top civil servants.

E-mail: nop@sam.sdu.dk

Anders R. Villadsen is associ-

ate professor at Aarhus University,

Denmark. His research interests include

public management and diversity in

public organizations. He has published

in journals such as Journal of Public

Administrative Research and Theory,

Public Administrative Review, and

International Public Management

Journal.

E-mail: avilladsen@badm.au.dk

The Right Mix? Gender Diversity in Top Management Teams and Financial Performance 29

1

Public Administration Review,

Vol. 75, Iss. 2, pp. 291–301. © 2014 by

The American Society for Public Administration.

DOI: 10.1111/puar.12310.

Niels Opstrup
University of Southern Denmark, Denmark

Anders R. Villadsen
Aarhus University, Denmark

Abstract: Recent research has illustrated that demographic diversity infl uences the outcomes of public sector organiza-
tions. Most studies have focused on workforce diversity; by comparison, little is known about how managerial diversity
aff ects organizational outcomes. Th is article focuses on gender diversity in the top management teams of public organi-
zations and its relationship to fi nancial performance. Th eory suggests that management diversity can be a positive asset
for organizations, allowing for the use of more diverse knowledge and human skill sets. Results of this study, however,
suggest that organizations may only be able to leverage these advantages if they have a supporting management
structure. In a longitudinal study of top management teams in Danish municipalities, the authors fi nd that gender
diversity in top management teams is associated with higher fi nancial performance, but only in municipalities with a
management structure that supports cross-functional team work. Th ese results are interpreted in light of existing theory,
and implications are suggested.

Practitioner Points
• Little is known about how the composition of the top management team in public organizations matters for

organizational outcomes.
• Greater gender diversity has the potential to lead to superior organizational outcomes but does not automati-

cally do so.
• To reap the benefi ts of diverse top management teams, the organizational structure must facilitate behavioral

integration between members of the top management team.
• Gender diversity in the top management team is related to better fi nancial outcomes when organizational

structures promote integration and discretion of the top management team.
• Gender diversity in top management positions has no eff ect in traditional hierarchical organizations.

contracting (Hefetz and Warner 2004, 2012),
and performance-information use (Moynihan and
Ingraham 2004; Moynihan and Pandey 2010). Th is
substantial line of research has provided valuable
insights about how managers can aff ect the outcomes
of public organizations. Much less research has been
devoted to the composition of management teams and
the dynamics among individuals occupying manage-
rial roles in organizations. Th is is a critical gap in our
understanding of public management. Top manage-
ment is composed of individuals who apply their
knowledge, perspectives, and worldviews to contribute
to decision making as well as the overall direction of
organizations (Hambrick, Cho, and Chen 1996).

In this article, we explore how gender diversity in
top management teams (TMTs) aff ects fi nancial
performance in public organizations. Research
focusing on private fi rms indicates that managerial
gender diversity is related to positive performance
outcomes (Auh and Menguc 2006; Carter, Simkins,

Th e Right Mix? Gender Diversity in Top Management
Teams and Financial Performance

Recent research has explored how characteris-tics such as tenure (Juenke 2005), race (Pitts 2005), experience (Villadsen 2012), and
gender (Meier, O’Toole Jr, and Goerdel 2006) infl u-
ence the actions of public managers. Th is research,
however, generally fails to allow for the fact that
management is rarely an individual endeavor; while a
single person may top the hierarchy and serve as the
public face of an organization, actual top management
is often better described as a team eff ort (Finkelstein
and Hambrick 1990). In this article, we move the
focus from the individual manager to the top manage-
ment team and examine how its composition aff ects
organizational outcomes.

A considerable amount of research has investigated
how various management practices aff ect public
organizations. Th ese practices include leadership
(Fernandez 2005; Van Wart 2013), managerial net-
working (Meier and O’Toole 2003), organizational
strategies (Andrews, Boyne, and Walker 2006),

292 Public Administration Review • March | April 2015

steering of public spending (Serritzlew 2005; Wildavsky and
Caiden 2004).

More specifi cally, we study the relationship between TMT gender
diversity and fi nancial performance in a longitudinal analysis of
91 Danish municipalities. Our results indicate the importance of
organizational structure for accruing the benefi ts of TMT diversity.
Only in municipalities where the management team is organized
according to an executive board model do higher levels of gender
diversity correlate with better fi nancial results (measured in terms of
operational results and outperforming the budget in the course of
the year). No such eff ect is found for municipalities using a tradi-
tional specialized organization of management.

With this study, we make three important contributions. First, we
expand the very small number of studies that call attention to the
diff erences between male and female public managers. While the
Weberian public manager is a genderless bureaucrat, in the real
world, men and women bring diff erent backgrounds, experiences,
and expectations to the executive fl oors of ministries, agencies, and
city halls. Th is study points clearly to the importance of increasing
our understanding of how gender diversity aff ects top management
decision making and outcomes. Second, we address a call for more
research exploring the importance of diversity for organizational
outcome (Pitts and Wise 2010). Most diversity research in public
organizations has studied workforce diversity using the theory of
representative bureaucracy (Kingsley 1944; Park 2013; Roch and
Pitts 2012). We add to this knowledge by showing that top manage-
ment diversity can have a predictable, signifi cant impact on fi nancial
performance. Th ird, we shed light on one of the mechanisms that
connects gender diversity to positive outcomes. We suggest that

in order to accrue the benefi ts of diversity,
structures must be put in place to facilitate
behavioral integration of the team. In this
way, our study helps clarify one of the reasons
why previous research on diversity yielded
mixed results.

Next, we lay out our arguments for a positive
relationship between TMT gender diversity
and fi nancial performance and set forth our
hypotheses. Following this, we describe the

research setting and methods used. After reporting the results of our
analyses, we discuss our results with regard to existing research and
managerial relevance.

Gender Diversity in Top Management Teams
A top management team is the “dominant coalition” responsible
for setting an organization’s direction (Cyert and March 1963).
Th e TMT is charged with identifying opportunities and problems,
gathering and interpreting information, and making decisions that
propel the organization forward (Mintzberg 1979). In this process,
TMT members must draw on individual as well as collective human
and social capital. Each individual brings a diff erent background
and perspectives to the table. However, decisions are also shaped in
cooperation. While there is a relatively limited number of stud-
ies on management team composition and functioning in the
public sector, a substantial line of research has investigated TMT
and board composition in the private sector and its relationship to

and Simpson 2003; Dwyer, Richard, and Chadwick 2003; Erhardt,
Werbel, and Shrader 2003). However, little research has yet explored
the outcomes of TMT gender composition in public sector organi-
zations (Pitts and Wise 2010), and to our knowledge, no studies
have focused on how diversity may aff ect fi nancial performance.
We respond to an increasing interest in diversity in public organiza-
tions (Park 2013; Pitts 2005; Pitts and Wise 2010) and draw on
recent theories about team diversity (for reviews, see Jackson, Joshi,
and Erhardt 2003; Van Knippenberg and Schippers 2007) in order
to explore how the gender composition of TMTs aff ects fi nancial
outcomes in public organizations.

In this article, we argue that TMT gender diversity is a positive
asset for public sector organizations, for three reasons. First, draw-
ing on the theory of representative bureaucracy, gender diversity
is likely to generate a better understanding of the organization’s
environment because a collective of diverse team members can
relate to many diff erent aspects of increasingly heterogeneous
organizational surroundings (Keiser et al. 2002; Park 2013).
Second, diversity may inspire more creative and innovative output
because it enables organizations to draw on a wider range of
social and human capital as well as diff erent cognitive templates
(Østergaard, Timmermans, and Kristinsson 2011). Finally, diversity
may provide more eff ective problem solving because diverse teams
evaluate more alternatives and explore more potential consequences
(Carter, Simkins, and Simpson 2003). Further, we draw on theories
of group decision-making processes to suggest that male- dominated
TMTs may be excessively risk seeking compared with those
with a more even gender balance. Based on this, we predict that
fi nancial performance is likely to be better in organizations with a
gender-integrated TMT.

We propose that the benefi ts of diversity are
contingent on organizational factors. Gender
diversity in the top management team is
likely to be related to superior perform-
ance specifi cally when management struc-
tures facilitate behavioral integration and
cross-gender

TMT work.

We suggest that
the mixed results of previous research were
caused by disregarding the role of organiza-
tional structure as an important mediating
factor. Th is is consistent with research pointing to the impor-
tance of the structural context for achieving active representation
(Keiser et al. 2002).

Previous studies of the outcomes of diversity and representation
in public organizations have predominantly taken education as
their point of departure, relying on various measures of student
performance (Pitts and Wise 2010). In this article, we focus on
the fi nancial performance of multipurpose public organizations.
Of course, we recognize that “making money” is not the prime
objective of public sector organizations. However, sound fi nancial
management is a fundamental prerequisite for eff ective service
delivery, particularly in the post–New Public Management period,
when many organizations, agencies, and units have gained greater
fi nancial autonomy. Th e ability to control fi nances eff ectively is
an important factor underlying the work public organizations
carry out. Defi cits and budget overruns may impede the eff ective

Gender diversity in the top
management team is likely to be
related to superior performance
specifi cally when management
structures facilitate behavioral
integration and cross-gender

TMT work.

The Right Mix? Gender Diversity in Top Management Teams and Financial Performance 293

diverse and complex environments, which may make female leaders
an asset (Kent and Moss 1994). Second, women in top management
are likely to be highly skilled, as they typically have to overcome
signifi cant challenges to break the glass ceiling and get promoted to
higher ranks. (Th is point is particularly salient in the context of the
present study, as only around 20 percent of municipality top man-
agers in Denmark are women.) Th is means that not only are women
who attain TMT positions likely to be skilled and knowledgeable,
but also they are adept at working the politics of the organiza-
tion. Th ird, compared with men, women’s cognitive style tends to
emphasize harmony (Krishnan and Park 2005). Th is gender diff er-
ence may lead to superior team dynamics when men and women
are brought together in continuous interaction, where the distinct
strengths of each gender can be leveraged. Further, it can be argued
that the degree of gender diversity in top management constitutes a
signal to lower hierarchical levels. Female middle managers become
more motivated if they see women in the TMT and, consequently,
see a viable promotion path (Dezsö and Ross 2012), and female
employees in general are likely to feel more accepted and to feel that
diff erent gender roles are accepted in the organization (Guy and
Newman 2004).

Th e foregoing arguments suggest that there may be positive eff ects
of gender-diverse TMTs. Th is has generally been supported in
studies of private sector fi rms (for a review, see Jackson, Joshi, and
Erhardt 2003). Krishnan and Park (2005) fi nd that TMT gender
diversity has a positive eff ect on organizational performance in a
study of Fortune 1000 companies. In a study of 127 large U.S.
companies, Erhardt, Werbel, and Shrader (2003) report a posi-
tive association between board diversity and fi nancial performance
indicators. As mentioned, relatively few studies have explored
outcomes of diversity in public organizations (Pitts and Wise 2010).
Still, representative bureaucracy theory would predict that a repre-
sentative TMT will be more attuned to the demands and needs of
citizens and, consequently, likely to make better fi nancial decisions
(Park 2013).

In this article, we focus specifi cally on how TMT gender diversity is
related to the fi nancial performance of public sector organizations.
Assessing the performance of public sector organizations is notori-
ously diffi cult because such organizations often serve multiple objec-
tives and must balance the interests of multiple stakeholders. Unlike
research on private sector performance, where fi nancial indicators
are the sine qua non, research in public management has tended to
focus on service-related performance measures such as student test
scores (Andersen and Serritzlew 2007; Meier and O’Toole 2001),
composite measures (Andrews, Boyne, and Walker 2006), or self-
reported assessments. Diversity research is no exception: as already
mentioned, relatively little diversity research is outcome focused, and
the few studies that do exist have largely focused on the educational
sector and student outcomes (e.g., Johansen 2007; Pitts 2005).

To further understand how TMT dynam-
ics may aff ect fi nancial outcomes, it is worth
considering how group composition may
aff ect risky behavior in groups. Research in
economic psychology indicates diff erences in
men’s and women’s fi nancial decision mak-
ing. Using experimental methods, Powell

organizational performance (Finkelstein and Hambrick 1990; Miller
and Triana 2009; Zahra and Pearce 1989). Pitts argues that diversity
is a strength when “[t]he ‘whole’ of the group’s eff ort will be greater
than the sum of the parts” (2005, 618).

Given that management may be stereotypically thought of as a male
task, an important challenge for research has been to investigate
how the gender composition of top management matters for the
outcomes of organizations. In a comprehensive review of private
sector research on diversity, Jackson, Joshi, and Erhardt (2003)
fi nd gender to be the most widely studied dimension of diversity,
ahead of age and functional background. Research has long been
interested in gender segregation in public administration as well
(Guy and Newman 2004; Stivers 1995). However, in a systematic
review of studies of workforce diversity in 12 core public adminis-
tration journals, Pitts and Wise (2010) fi nd that only a few studies
have explored the outcomes of diversity. Of these, even fewer have
focused on gender. Th is is a critical gap in our understanding of
public management.

Th e central premise underlying the hypothesis that gender diver-
sity matters to top management teams is that male and female
managers diff er, not only in general but specifi cally in terms of
the way they perform in their (management) jobs. Research has
provided numerous studies of how leadership styles diff er between
men and women. Generally, studies fi nd that women engage in
leadership behavior that is more participatory, collaborative, and
democratic than men’s (Dezsö and Ross 2012; Eagly and Carli
2003). While there are, to our knowledge, no studies focusing on
diversity in the top management teams of public sector organi-
zations, a small amount of research has illuminated the conse-
quences of gender diff erences among public managers. In a study
of Texas school districts, Meier, O’Toole, and Goerdel (2006)
fi nd gender diff erences in the performance eff ects of managerial
networking. Johansen’s (2007) study of the same empirical setting
suggests that prospector strategies may be especially eff ective when
carried out by women managers. In a study of state agency heads
in the United States, Jacobsen, Palus, and Bowling (2010) fi nd
that women managers devote less time to internal management
and engage in less external networking behavior than their male
counterparts.

Top Management Gender Diversity and Financial
Performance
Building on the assumption that male and female managers exhibit
distinct characteristics, there are further reasons to believe that gen-
der diversity is related to organizational outcomes. Next we develop
the relationship between TMT gender diversity and fi nancial
outcomes by drawing on the general diversity literature as well as
dedicated fi nancial literature.

Krishnan and Park (2005), drawing on social
identity theory, highlight three arguments
why TMTs may benefi t from gender integra-
tion after a history of favoring men. First,
women are more likely than men to be per-
ceived as leaders in environments that involve
a signifi cant degree of social

interaction.

Public sector organizations exist in highly

Women are more likely than
men to be perceived as leaders
in environments that involve
a signifi cant degree of social

interaction.

294 Public Administration Review • March | April 2015

TMTs must have a structure allowing integration, which, in turn,
can facilitate interaction and cultivate mutual exchange of knowl-
edge and opinions. We expect that organizations will be better
able to reap the benefi ts of gender diversity when they adopt an
organizing approach that allows for higher degrees of integration.
Conversely, if a TMT’s responsibilities are disintegrated and distrib-
uted across sectorial areas, diversity will be much less important,
if at all. Th e benefi t of bringing together members from diff erent
groups (in this case, diff erent genders) will simply not be realized
because managers will maintain a strong focus on their distinct area
of responsibility with little benefi t accrued from cooperation.

Besides integration, team discretion may also aff ect the outcomes
of diversity. If the management structure is hierarchical, with a
dominating top executive, gender integration is likely to matter less
than if the management structure emphasizes teamwork and shared

responsibilities. In the latter situation, in
which management discretion resides at the
TMT level rather than the individual level,
group dynamics matter more for decision
quality and outcomes. Management discre-
tion has been shown to positively aff ect

organizational effi ciency in local governments (Garrone, Grilli, and
Rousseau 2013).

Th is reasoning can also been interpreted in light of representative
bureaucracy research. Studies have distinguished between passive
and active forms of representation (Park 2013; Wilkins and Keiser
2006). Passive representation may increase the legitimacy of public
agencies (Riccucci, Van Ryzin, and Lavena 2014), but active repre-
sentation may create more substantial outcomes. Such more active
representation, however, may depend on structural conditions of
the organization. As argued by Keiser and colleagues, “Within the
organization, a […] necessary condition for representative bureauc-
racy is that the bureaucrats in question have discretion to infl uence
outcomes” (2002, 558). With regard to overall fi nancial outcomes,
discretion is greater when management structures promote real team
work and shared responsibilities.

We predict that the organization of management may signifi cantly
aff ect the organizational outcomes of TMT gender diversity. When
management is structured around team integration and team discre-
tion, the benefi t is likely to be higher.

Hypothesis 2: Th e positive relationship between TMT
gender diversity and fi nancial performance is moderated by
management structure, such that the relationship is stronger
when TMT integration and discretion are higher.

Research Setting
Danish municipalities (local governments) provide an excellent
setting for studying the impact of gender diversity in top manage-
ment teams. Th e 98 municipalities are highly comparable, semi-
autonomous units having the same tasks and subject to the same
regulations (Christensen 1998; Greve 2006). Local government is
responsible for the larger part of public service, including primary
schooling, child care, elder care, local infrastructure, and social
security benefi ts. Th e municipalities levy their own taxes and have
considerable autonomy in service provision.1 Government rests with

and Ansic (1997) fi nd that women are less risk seeking than men
and that they adopt diff erent strategies in their fi nancial decision
making. A similar point is made by Francis et al. (2009) in a study
of chief fi nancial offi cers (CFOs) showing that female CFOs are
more conservative than male CFOs. Th ese individual-level gender
diff erences aggregate through group composition to create system-
atic diff erences in organizational-level outcomes. Group behaviors
conform to the assumptions, values, and norms that are shared by
the majority of the group’s members (Cartwright and Zander 1960).
Th is means that if the prevailing view held by the majority favors
a more risky decision, the minority group members are likely to
conform. Stoner (1968), for instance, fi nds that groups give more
weight to risk-related attitudes shared by more members. If men and
women tend to hold diff erent attitudes toward risk, this research
suggests that TMT gender segregation may lead to fi nancial deci-
sions that are either too risky or too safe depending on the majority
gender of the group. A gender-balanced team
is less susceptible to the weaknesses of such
groupthink. In public organizations looking
to fi nd the right balance between being fi nan-
cially progressive and conservative, a balance
of male and female TMT members is likely to
be positive.

In total, we base our hypothesis on two lines of argument: fi rst, on
general research claiming that gender diversity is positively related to
organizational outcomes, and second, on specifi c research pointing
to gender diff erences in fi nancial decision making.

Hypothesis 1: Higher TMT gender diversity is positively
related to fi nancial performance.

The Moderating Role of Management Structure
Th e arguments outlined earlier imply that for a group to obtain the
positive outcomes of diversity, its members must be able to interact
productively and constructively and feed off each other’s strengths.
Previous research has pointed out how structures can facilitate posi-
tive outcomes from representation (Meier and Bohte 2001). Th e
expectation expressed in hypothesis 1 may therefore be contingent
on the organization of the top management team. To release posi-
tive TMT dynamics, the TMT must work in a way that facilitates
such interaction. Specifi cally, we propose that team integration and
team discretion positively moderate the relationship between gender
diversity and fi nancial performance.

Upper-echelon theory has argued that the outcome of a TMT’s
work depends on its organization and internal dynamics. Hambrick
points out that TMTs “often consist of semiautonomous ‘barons,’
each engaging in bilateral relations with the CEO but having little
to do with each other and hardly constituting a team” (2007, 336).
In light of this, he argues that TMTs with more behavioral integra-
tion are likely to perform better. Behavioral integration describes the
extent of a team’s collaboration, knowledge exchange, and emphasis
on joint decision making (Hambrick 1994). TMTs that are more
behaviorally integrated have been shown to be able to embrace
ambidexterity to a larger degree and also exhibit higher performance
(Lubatkin et al. 2006). Another study fi nds higher TMT behavioral
integration to be positively related to decision quality (Carmeli and
Schaubroeck 2006). For diversity to aff ect organizational outcomes,

A gender-balanced team is less
susceptible to the weaknesses of

such groupthink.

The Right Mix? Gender Diversity in Top Management Teams and Financial Performance 295

However, the full eff ect of an executive board is presumably only
apparent in the pure version of the executive board model, in which
the managing directors are formally totally independent of sector
interests (Bækgaard 2011, 1068). We explore these diff erences in the
organization of management to assess whether the eff ect of manage-
ment diversity is contingent on the structure of management.

Methods
Data and Analysis
To investigate the impact of gender diversity on fi nancial performance
in the TMTs of Danish municipalities, we analyze yearly data for the
period 2008 to 2012. Th e starting point for the analysis is given by
a major structural reform (Strukturreformen) implemented in 2007,
which reduced the number of municipalities from 271 to 98 and
changed their responsibilities and fi nancing (see Vrangbæk 2010).

2

Th e data are compiled from register data and offi cial statistics.
Th e composition of municipality top management teams and the
municipalities’ administrative structure are coded based on the
offi cial local government handbook (Mostrup Kommunal HÃ¥ndbog),
which is issued annually and contains, among other information,
the names of the local government CEOs and the other manag-
ing directors in each of the 98 municipalities. According to this
handbook, the size of the municipality TMTs varied between 1 and
12 members in the analyzed period. In order to make a measure of
diversity meaningful, we only include observations with more than
two registered municipal managers, eliminating seven municipali-
ties. Th erefore, only 91 municipalities feature in the analysis. Data
on fi nancial performance is obtained from statements made by the
Danish Evaluation Institute for Local Government.3 Furthermore, a
number of control variables are integrated from “the yearly munici-
pal key indicators” (De Kommunale Nøgletal) published by the
Ministry of Economy and the Interior.4

In our analysis, we apply generalized least squares (GLS) regres-
sion in order to deal with the potential problems of heteroscedas-
ticity attributable to the longitudinal data design (Frees 2004).
Furthermore, because each municipality appears multiple times
in the analysis, observations may not be independent. Th erefore,
we cluster the analysis on the observations (municipalities) and
calculate robust standard errors. When analyzing panel data, there
is always a choice between random- and fi xed-eff ects specifi cations.
Various reasons justify using a random-eff ects specifi cation for
this study. First, there is relatively little variation in our independ-
ent variable across the years, as changes only happen when there
is a replacement in the TMT, rendering within-observation vari-
ation modest and fi xed eff ects unfeasible. Further, we conducted
a Hausman test, which clearly rejected fi xed eff ects as superior to
random eff ects (p < .43). Th erefore, we use random-eff ects estima- tions in all models.

To ensure that multicollinearity is not problematic, we calculated
variance infl ation factors (VIFs), which did not reach any levels of
concern. Th e mean VIF score is 1.67, and the highest value is 2.45.

To assess whether results are sensitive to autocorrelation, we ran
additional models with an AR1 autocorrelation structure. We
obtained substantially very similar results (available from the authors
upon request).

the elected municipal council, and executive power is shared among
a compulsory fi nancial committee, various standing committees,
and the mayor. Th e mayor chairs the council and the fi nancial com-
mittee meetings and is formally the “daily leader of the municipal
administration” (Mouritzen and Svara 2002). Most mayors come
from one of the two biggest political parties, the Social Democrats
or the Liberals (Berg and Kjær 2005). Th e other members of the
council are “part-time politicians” who usually also hold a day job.

While municipal councilors are formally in charge of providing
direction and overall decision making, actual operations are left to
the administrative organization. Top management consists of career
bureaucrats who are promoted from the ranks of civil servants. In
practice, municipal administrative top management teams consist
of a chief executive offi cer (CEO or kommunaldirektøren) and a
number of managing directors. Th e formal hierarchical relationship
within the TMT is somewhat ambiguous because of the sharing
of executive power among the fi nancial committee, the standing
committees, and the mayor (Ejersbo, Hansen, and Mouritzen 1998,
109). However, the CEO is usually regarded as the person in charge
of the administrative part of the municipal organization (Bækgaard
2011, 1066).

Two basic yet distinct forms of organization of the top management
are used in Danish municipalities (Bækgaard 2011). Traditionally,
most Danish municipalities have organized their administration
according to a sector-based model, in which the municipality’s
tasks are divided among a number of administrative departments,
each responsible for a defi ned service area (which typically mirrors
the responsibilities of a specifi c standing committee). Each depart-
ment has a department head. However, since the beginning of the
1990s, many municipalities have reorganized (Villadsen 2012).
Emphasizing the need for coordination at the municipal apex and
drawing inspiration from private fi rms, some municipalities have
organized the top management as executive boards with cross-
departmental responsibilities. Th e main diff erences between the
two types of models lie in the roles and the focus of the managing
directors. In the sector-based model, the department heads’ primary
responsibility is the running of their department. Th is type of
organizing is based on traditional bureaucratic thinking, in which
each top manager is responsible and accountable for a particular line
of work. Th e top managers of each department together form a top
management group, but responsibilities are relatively clearly divided.
Consequently, little TMT integration is needed, and overall mana-
gerial discretion is likely to rest with the top executive.

In the executive board model, the managing directors attend to
intraorganizational coordination, strategic decision making, and the
economic steering of the municipality. While TMTs organized in this
way have a CEO who is ultimately in charge, they are a collective in
which decision making is a shared task across the various functional
areas of the organization. Th is type of organizing is more time-con-
suming, but it allows the team to benefi t from members’ collective
human and social capital and provides a higher degree of functional
integration among diff erent lines of work. TMTs organized this way
will be more integrated and have more team-based discretion.

Some municipalities apply a mixed model, with an executive and
one or more of the managing directors also being department heads.

296 Public Administration Review • March | April 2015

each municipality’s TMT for each year. Next, we used the Blau
index of dissimilarity to create the variable gender diversity. The Blau
index is calculated by the formula 1 – ∑pi2, where p equals the
proportion of each group of interest (i.e., male and female) and i
represents the number of groups (in this case, two). For gender
diversity, the index ranges from 0 to 0.5. A value of 0 represents
complete homogeneity (thus, a TMT consisting of either all male or
all female directors); a value of 0.5 represents maximum gender
diversity (an equal number of men and woman in the TMT). Both
of these extreme values are present in our data set. In the period
under analysis, the municipalities’ average index score was 0.257,
with a standard deviation of 0.196. In order to better refl ect the
TMT’s impact on the municipality’s fi nancial performance, the
independent variable is lagged by one year.

Moderating variable. The municipalities’ administrative structure in
the given year is determined based on information from the local
government handbook. Inspired by Bækgaard (2011), we distinguish
between sector-based models, mixed models, and executive board
models. A municipality is coded as having a sector-based model when
the local government handbook lists a number of administrative
departments with defi ned service areas (e.g., Department of Children
and Youth, Department of Elderly and Disabled, etc.). When
managing directors are presented as heading a defi ned service area but
no administrative departments are listed explicitly, the municipality is
registered as having a mixed model. Finally, when managing directors
are presented without any associated area of responsibility and no
administrative departments are listed, the municipality is coded as
using an executive board model. This is a crude measure that does not
necessarily capture the actual daily functioning of each TMT, but it
does refl ect how the municipalities choose to present themselves in
the annual local government handbook.

Because we are particularly interested in the use of executive board
models, we create a dummy variable where executive board models
are coded as 1 and the other two board types are coded as 0. Next,
we create an interaction term between the measure of TMT gender
diversity and the dummy variable for executive board models in
order to test its moderating eff ect. In the analyzed period, 17.4
percent of the municipalities were registered as using an executive
board model. By comparison, Bækgaard (2011, 1069–70) fi nds,
based on short interviews with the CEOs, that 16.0 percent used an
executive board model in 2008.

Control variables. In order to better isolate the impact of gender
diversity in top management teams on the municipalities’ fi nancial
performance, we include a number of control variables in the analy-
sis. First, it is likely that structural characteristics both infl uence
the tendency to employ woman in top management positions and
impact the fi nancial performance of the municipality. To account
for this, we control for the municipalities’ population size, propor-
tion of inhabitants living in an urban area (urbanization), tax base
(a measure of municipal wealth), socioeconomic conditions (using
an indicator calculated by the Ministry of the Interior), and long-
term debts. All variables are time varying and lagged by a year to
better refl ect causality. Second, we include a dummy variable indi-
cating whether the municipality resulted from the 2007 structural
reform, which led to an “oversupply” of managing directors in the
newly merged municipalities, potentially aff ecting the composition

Measurement
Dependent variables. We operationalize fi nancial performance
using two measures to account for the fact that “good fi nancial
performance” may mean different things in different organizations.
First, we use accounted operating result because it refl ects the
organization’s ability to generate suffi cient income and control
expenditures. Second, we use budget overrun because it refl ects an
organization’s ability to follow a fi nancial plan laid out by elected
politicians. Each of these variables is further specifi ed next.

Th e fi rst dependent variable is the accounted operating result in the
given year. Th is is a widely used indicator of the fi nancial perform-
ance of Danish municipalities (Hansen, Houlberg, and Pedersen
2014). Th e operating result is measured as the municipality’s
accounted operating revenues (tax revenue and block grants from
the national government) per citizen minus its accounted operating
expenses and net interest payments per citizen—all in thousands
of Danish kroner. As stated earlier, these fi gures are obtained from
yearly statements made by the Danish Evaluation Institute for
Local Government. Between 2008 and 2012, the average accounted
operating result of the municipalities that we analyzed was a
1,527 Danish kroner surplus per citizen (approximately US$284).
However, there were major variations between “municipality years”;
the standard deviation is 1,406 Danish kroner per citizen, refl ecting
that both defi cits and surpluses are common.

Th e second dependent variable, a municipality’s budget overrun,
measures the diff erence between the accounted operating result and
the budgeted operating result in a given year, both per citizen in
thousands of Danish kroner. It is calculated according to the follow-
ing formula for municipality m in year t:

Budget Overrun

m,t

= Budgeted Operating Result
m,t

– Accounted
Operating Result

m,t

Positive values indicate that a municipality has failed to reach its
budgeted fi nancial results (i.e., an overrun). Negative values indicate
that the realized result is better than budgeted (i.e., a budget “under-
run”). Such a measure has previously been used to examine Danish
municipalities’ fi nancial performance (Serritslew 2005). Again, these
fi gures are obtained from the yearly statements made by the Danish
Evaluation Institute for Local Government. In the period under
analysis, the municipalities’ average budget overrun was a modest
negative 84 Danish kroner per citizen, refl ecting that, on average,
municipalities did a little better than budgeted. Th e standard devia-
tion is 1,117 Danish kroner per citizen, indicating that the modest
average overrun covers substantial variation across municipalities.
Some did far better than the budget; others did far worse.

It can be argued that budget overrun is not symmetrical around zero,
as doing better than budgeted is not necessarily positive and may imply
that inadequate service levels were provided. To address this, we con-
ducted an alternative analysis with only budget overruns (i.e., realized
results worse than budgeted). We found substantially similar results.

Independent variable. On the basis of the listings in the local
government handbook, we registered the names of all CEOs and
managing directors in each of the 98 municipalities for each year of
the analysis. From this, we then coded the gender composition of

The Right Mix? Gender Diversity in Top Management Teams and Financial Performance 297

As earlier, the municipality’s administrative structure is found to
moderate the relationship between TMT gender diversity and
fi nancial performance, lending further support to hypothesis 2. In
municipalities applying an executive board model, more diverse
TMTs produce smaller budget overruns (or even budget underruns,

of the TMT. Moreover, being merged has also been found to aff ect
the municipalities’ fi nancial management (Hansen, Houlberg, and
Pedersen 2014). Th ird, we control for political factors by includ-
ing a dummy variable for the party of the mayor. Liberal mayors
are used as the reference category in relation to Social Democratic
mayors and mayors from other parties. We also control for potential
eff ects of municipal council elections by including a dummy vari-
able for the year 2009. Furthermore, we include a measure for the
gender diversity on the council. We also control for the extent to
which the municipality uses private contractors for service delivery.
Finally, we control for the size of the TMT.

Results
Table 1 reports the descriptive statistics and correlations among
variables. Th e impact of top management team gender diversity on
the municipalities’ operational results is presented in table 2. Th ree
models are reported: in the fi rst model, only the control variables are
entered; in model 2, the variable measuring TMT gender diversity is
added; and in the third model, the interaction term between TMT
gender diversity and the use of an executive board model is included.

Hypothesis 1 proposed that higher TMT gender diversity is related
to better fi nancial performance. Th is hypothesis is not supported
with regard to the municipalities’ operating results. Model 2
indicates no signifi cant relationship between TMT gender diversity
and operating results. However, the degree of diversity of the TMT
is found to have a positive impact on operating results when the
municipality’s top management is organized as an executive board.
In model 3, the interaction term between TMT gender diversity and
the use of an executive board model has a signifi cant positive coef-
fi cient. Th is lends support to hypothesis 2, which proposed that the
relationship between TMT gender diversity and fi nancial perform-
ance is moderated by the organization’s administrative structure.

In table 3, the impact of TMT gender diversity on the municipality’s
budget overrun is presented. Following the same steps as earlier, three
models are reported. Model 2 indicates no signifi cant relationships
between TMT gender diversity and the size of municipal budget
overrun. Th us, hypothesis 1 is not supported in this regard either.

Table 1 Correlations and Descriptive Statistics

Mean SD 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Operating resulta 1.48 1.47 1
Budget overruna –0.10 1.14 –0.71 1
TMT gender diversity 0.26 0.19 0.10 –0.05 1
Population 53,565 56,911 0.13 –0.04 0.05 1
Population density 551 1,347 0.17 –0.07 0.12 0.39 1
Tax base 152,595 28,536 0.18 –0.19 0.21 –0.03 0.34 1
Socioeconomic index 0.94 0.25 0.13 0.01 0.07 0.19 0.20 –0.34 1
Debt 15,622 9,200 0.11 –0.06 0.06 –0.17 –0.01 0.06 0.03 1
Contracting 24.00 4.07 0.08 –0.17 0.14 0.01 0.07 0.21 0.04 0.14 1
Reformed 0.66 0.46 0.13 –0.13 0.15 0.09 0.55 0.38 0.30 –0.18 0.19 1
Social Democrat mayor 0.47 0.49 0.06 0.03 –0.02 0.14 0.04 –0.20 0.32 0.01 –0.12 0.09 1
“Other” mayor 0.19 0.39 0.01 –0.01 0.13 0.16 0.19 0.19 –0.13 0.05 0.05 0.19 –0.46 1
Elected board gender

diversity 0.40 0.08 0.18 –0.22 0.21 –0.04 0.29 0.45 –0.04 –0.01 0.17 0.31 0.04 0.13 1
TMT size 4.88 1.82 –0.06 0.15 0.09 0.29 0.09 –0.09 0.10 –0.04 –0.27 –0.16 0.17 –0.05 –0.07 1
Executive board 0.17 0.38 –0.06 –0.10 –0.09 –0.13 –0.11 0.05 –0.11 –0.08 0.10 0.04 –0.07 –0.01 –0.01 0.40 1
Election year 0.20 0.40 0.02 –0.14 0.01 0.00 –0.00 –0.03 –0.00 –0.01 0.01 0.01 –0.01 –0.01 0.09 0.03 0.00 1

aIn thousands of Danish kroner.
Note: All correlations > |0.1| are signifi cant with p < .05.

Table 2 GLS Regressions: Dependent Variable, Operating Result

Model 1 Model 2 Model 3

Population 0.00*** 0.00*** 0.00***
(0.00) (0.00) (0.00)

Population density 0.00 0.00 0.0

0

(0.00) (0.00) (0.00)

Tax base 0.00 0.00 0.00
(0.00) (0.00) (0.00)

Socioeconomic index 0.95* 0.93* 1.04*
(0.40) (0.40) (0.41)

Debt 0.00* 0.00 0.00*
(0.00) (0.00) (0.00)

Contracting 0.03 0.02 0.02
(0.02) (0.02) (0.02)

Merged –0.01 –0.05 –0.07
(0.30) (0.31) (0.30)

Social Democrat mayor 0.17 0.17 0.17
(0.19) (0.19) (0.19)

“Other” mayor –0.15 –0.17 –0.19
(0.25) (0.25) (0.25)

Elected board gender diversity 2.53 2.42 2.29
(1.43) (1.46) (1.46)

TMT size –0.23** –0.23** –0.24***
(0.07) (0.07) (0.07)

Executive board –0.81*** –0.76** –1.27***
(0.25) (0.24) (0.32)

Election year –0.06 –0.06 –0.05
(0.12) (0.12) (0.12)

TMT gender diversity 0.52 0.01
(0.44) (0.46)

TMT gender diversity * Executive board 2.61*
(1.08)

Constant –2.08 –2.00 –1.58
(1.21) (1.22) (1.21)

Number of groups 91 91 91
Number of observations 442 442 442
Wald chi-square 128.15*** 136.45*** 143.76***
R2 0.14 0.14 0.15

Note: Robust standard errors in parentheses.
***p < .001; **p < .01; *p < .05; two-sided tests.

298 Public Administration Review • March | April 2015

by an almost fl at line. Turning to municipalities using an execu-
tive board model, we see that the line is decreasing indicating that
diversity’s eff ect on budget overrun decreases as diversity increases.
Together, the graphs in fi gure 1 underline how the eff ects of
diversity are contingent on management structure. In both panels,
the eff ect of diversity does not aff ect fi nancial performance when
traditional structures are used, as illustrated by the almost fl at lines
in panels 1 and 2.

To further scrutinize the importance of gender diversity for diff er-
ent management structures and assess the substantial eff ect sizes,
in table 4, the interactions are further illustrated. Based on our
regressions we have calculated the predicted diff erences in operating
results and budget overruns for municipalities with high versus low
levels of TMT

gender diversity.

Turning fi rst to the operating results, we observe a diff erence of
1,050 Danish kroner per citizen between those municipalities with
a high level of executive board diversity and those with a low level
(the eff ect illustrated by the rising line in fi gure 1, panel 1). Th is is

where the result is better than budgeted). In model 3, the interac-
tion term between TMT gender diversity and the executive board
model has a signifi cant negative coeffi cient.

Figure 1 provides a graphical depiction of the impact of gender
diversity on fi nancial performance. Th e graphs show marginal eff ects
for diff erent values of diversity. One line depicts municipalities with
an executive board model and the other municipalities without
such a management structure. In panel 1, we see how the eff ect of
TMT diversity on the operational result is contingent on manage-
ment structure. Overall, municipalities applying an executive board
model produce poorer operational results compared with munici-
palities with other administrative structures (cf. models 1 and 2).
For low levels of diversity, the traditional management model has a
higher eff ect on operational result than the board of director model.
However, a more gender-diverse TMT makes the latter munici-
palities “catch up” with the former ones as the eff ect of diversity
increases for municipalities with the executive board model when
diversity is higher. For high levels of diversity, the two models have a
similar eff ect on operating results.

In fi gure 1, panel 2, we off er a visual presentation of the marginal
eff ects of the results investigating budget overrun. Th e eff ect of
diversity on overrun is near constant for municipalities using tradi-
tional administrative structures for all levels of diversity indicated

Table 3 GLS Regressions: Dependent Variable, Budget Overrun

Model 1 Model 2 Model 3

Population 0.00 0.00 –0.00*
(0.00) (0.00) (0.00)

Population density 0.00 0.00 0.00
(0.00) (0.00) (0.00)

Tax base –0.00* –0.00* 0.00
(0.00) (0.00) (0.00)

Socioeconomic index –0.27 –0.27 –0.37
(0.30) (0.30) (0.31)

Debt 0.00 0.00 0.00
(0.00) (0.00) (0.00)

Contracting –0.03 –0.03 –0.03
(0.02) (0.02) (0.02)

Merged 0.00 0.01 0.04
(0.20) (0.20) (0.19)

Social Democrat mayor 0.09 0.09 0.1
(0.16) (0.16) (0.16)

“Other” mayor 0.30 0.30 0.32
(0.23) (0.23) (0.22)

Elected board gender diversity –3.17** –3.16** –2.92*
(1.21) (1.22) (1.24)

TMT size 0.12* 0.12* 0.13*
(0.05) (0.05) (0.05)

Executive board 0.38 0.38 0.85***
(0.21) (0.20) (0.25)

Election year 0.14 0.14 0.13
(0.11) (0.11) (0.11)

TMT gender diversity –0.08 0.37
(0.39) (0.40)

TMT gender diversity * Executive board –2.45**
(0.78)

Constant 2.55*** 2.56*** 2.11**
(0.74) (0.77) (0.81)

Number of groups 91 91 91
Number of observations 442 442 442
Wald chi-square 72.08*** 72.54*** 78.64***
R2 0.09 0.09 0.12

Note: Robust standard errors in parentheses.
***p < .001; **p < .01; *p < .05; two-sided tests.

Figure 1 Interaction between TMT Gender Diversity and
Organization Structure

0

.5

1

1.5

2

2.5

L
in

e
a
r

P
re

d
ic

tio
n

0 .1 .2 .3 .4 .5

Diversity Index

Other management structure Executive board model

Predictive Margins with 95% Confidence Intervals

-1

-.5

0
.5
1
L
in
e
a
r
P
re
d
ic
tio
n
0 .1 .2 .3 .4 .5
Diversity Index
Other management structure Executive board model
Predictive Margins with 95% Confidence Intervals

Panel 1 Dependent Variable: Operational Result

Panel 2 Dependent Variable: Budget Overrun

The Right Mix? Gender Diversity in Top Management Teams and Financial Performance 299

Discussion
Top management team gender diversity presents a paradox for
organizations. On the one hand, it may spark creativity and the
development and use of diverse knowledge and perspectives on
new ideas. On the other hand, diversity is likely to hamper eff ec-
tive team work because of potential between-group diff erences in
values, perceptions, and cognitive styles. Th is study has advanced
the understanding of TMT diversity into the domain of public
sector organizations. It provides evidence that not only individual
managerial attributes but also the constellation of a management
team aff ect core organizational outcomes. Focusing on gender, this
study has suggested that organizations are likely to accrue posi-
tive outcomes from gender diversity in management, but only if
the management structure facilitates the eff ective use of individual
diff erences. Th e results of our analysis of the TMTs of more than
90 Danish municipalities indicate that municipalities with gender-
diverse management groups obtain better fi nancial performance—
but only if they use the executive board management model, in
which the top management group shares responsibilities broadly.
Conversely, our results indicate that the impact of gender diversity is
neutral when top managers are specialized and have clearly divided
task responsibilities. Th ese results shed light on the mixed results
of previous research into team diversity, highlighting how diff erent
structures may, to varying degrees, facilitate the productive use of

diverse human resources. Th e present analysis
only deals with gender diversity, but whether
the management structure facilitates behav-
ioral integration is potentially important for
how TMT diversity along various dimensions
aff ects organizational outcomes.

As pointed out by Pitts and Wise (2010), we
urgently need more knowledge about the out-

comes of diversity. Th is study has contributed to this agenda with an
explicit focus on management. Gender has previously been estab-
lished as one relevant dimension of diversity, with research fi nding
notable diff erences between male and female public managers. Th e
results of this study further establish the relevance of gender, show-
ing that given the right structure, a gender-integrated management
team can outperform a gender-homogenous TMT.

As with any other study, there are limitations that should be noted.
Focusing on gender diversity is, as argued earlier, an important task
for research. However, it may be that other management charac-
teristics amplify or dampen the eff ects of gender. Data limitations
inhibited the inclusion of other relevant dimensions in the present
study. Future studies should look into the eff ects of other characteris-
tics and how diff erent dimensions of diversity interact. For instance,
research on private sector organizations has suggested that diversity in
education, experience, and age are all relevant for understanding the
operational dynamics in top management teams (e.g., Knight et al.
1999; Kochan et al. 2003; Van Knippenberg and Schippers 2007).

Th is study has focused on the association between TMT diversity
and fi nancial performance. While the analyses have allowed us
to speculate about diversity’s impact on group processes, future
studies should investigate underlying mechanisms in fi ner detail.
Th is would ideally involve laboratory of fi eld experiments that can
identify causal dynamics. Such studies could follow the work of

a substantial result. Th e average municipality size is around 57,000
citizens. Th is indicates that with an executive board management
structure gender diversity is a distinct asset. When other manage-
ment models are used, we observe no signifi cant diff erence in
operating results between municipalities with high and low levels of
TMT diversity (as indicated by the fl at line in fi gure 1, panel 1).

Focusing next on budget overruns, we see a similar picture. When
the executive board model is in use, there is a diff erence of –830
Danish kroner between high- and low-diversity municipalities. Th e
negative number indicates that low-diversity municipalities run
larger budget overruns. Again, no signifi cant diff erence is found
when other management models are in place.
Together, the parts of the table clearly indicate
the relationship between TMT gender diver-
sity and fi nancial results. For both operating
results and budget overrun, the relationship is
contingent on the management model. When
the executive board model is in use, signifi –
cant and substantial diff erences exist between
high and low levels of gender diversity.

In the next section, we discuss the results and their implications.
While the control variables are not of explicit interest here, they
do yield two relevant results. First, it should be noted that greater
gender diversity among elected politicians is negatively related to
budget overruns. It is beyond the scope of the present article to
speculate about the relationship between the political and admin-
istrative systems, but this provides further support for the general
relevance of gender to fi nancial decision making. Th e data do not
show a positive impact of elected board diversity on fi nancial per-
formance with regard to operating results, however. An explana-
tion for this could be that elected politicians are more attentive to
the budget than the actual fi nancial accounts. Th e second notable
result is that TMT size is negatively related to the municipalities’
fi nancial performance; larger TMTs, on average, produce poorer
operational results and bigger budget overruns. Th ese results could
suggest that behavioral integration is impeded when the number
of TMT members increases. Although we were not explicitly
interested in TMT size, to further control for a possible moderat-
ing eff ect of TMT size on the hypothesized relationship between
TMT gender diversity and fi nancial performance, we included
an interaction term for TMT gender diversity and TMT size to
the analysis. However, the original eff ects remained unchanged,
and no additional eff ects of TMT size for either the municipali-
ties’ operational results or budget overruns were found (results not
reported).

When the executive board
model is in use, signifi cant and

substantial diff erences exist
between high and low levels of

gender diversity.

Table 4 Predicted Differences in Operating Result and Budget Overrun between
Municipalities with High/Low Values of Gender Diversity

Operating Result in DKK
per Citizen: Difference
between “High” Diversity
and “Low” Diversity

Budget Overrun in DKK
per Citizen: Difference
between “High” Diversity
and “Low” Diversity

Executive board model 1,050** –830**

Other model 10 150

Notes: High diversity is defi ned as mean +1 standard deviation; low diversity is
defi ned as mean –1 standard deviation.
**p < .01.

300 Public Administration Review • March | April 2015

Berg, Rikke, and Ulrik Kjær. 2005. Den danske borgmester [Th e Danish Mayor].
Odense: Syddansk Universitetsforlag.

Blom-Hans en, Jens, Marius Ibsen, Th orkil Juul, and Poul Erik Mouritzen. 2012.
Fra sogn til velfærdsproducent. Kommunestyret gennem fi re årtier [From Parish to
Welfare Producer: Municipalities through Four Decades]. Odense: Syddansk
Universitetsforlag.

Carmeli, Abraham, and John Schaubroeck. 2006. Top Management Team Behavioral
Integration, Decision Quality, and Organizational Decline. Leadership Quarterly
17(5): 441–53.

Carter, Dav id A., Betty J. Simkins, and W. Gary Simpson. 2003. Corporate
Governance, Board Diversity, and Firm Value. Financial Review 38(1): 33–53.

Cartwright, Dorwin, and Alvin Zander, eds. 1960. Group Dynamics: Research and
Th eory. 2nd ed. Evanston, IL: Row, Peterson.

Christensen , Jørgen Grønnegård. 1998. Det kommunale mikrokosmos [Th e
Municipal Micro-Cosmos]. Politica 30(3): 339–48.

Cyert, Richard M., and James G. March. 1963. A Behavioral Th eory of the Firm.
Englewood Cliff s, NJ: Prentice Hall.

Dezsö, Cristia n L., and David Gaddis Ross. 2012. Does Female Representation
in Top Management Improve Firm Performance? A Panel Data Investigation.
Strategic Management Journal 33(9): 1072–89.

Dwyer, Sean, Or lando C. Richard, and Ken Chadwick. 2003. Gender Diversity in
Management and Firm Performance: Th e Infl uence of Growth Orientation and
Organizational Culture. Journal of Business Research 56(12): 1009–19.

Eagly, Alice H. , and Linda L. Carli. 2003. Th e Female Leadership Advantage: An
Evaluation of the Evidence. Leadership Quarterly 14(6): 807–34.

Ejersbo, Niels, Morten Balle Hansen, and Poul Erik Mouritzen. 1998. Th e Danish Local
Government CEO: From Town Clerk to City Manager. In Th e Anonymous Leader:
Appointed CEOs in Western Local Government, edited by Kurt Klaudi Klausen and
Annick Magnier, 97–112. Odense: University Press of Southern Denmark.

Erhardt, Niclas L., James D. Werbel, and Charles B. Shrader. 2003. Board of
Director Diversity and Firm Financial Performance. Corporate Governance: An
International Review 11(2): 102–11.

Fernandez, Serg io. 2005. Developing and Testing an Integrative Framework of Public
Sector Leadership: Evidence from the Public Education Arena. Journal of Public
Administration Research and Th eory 15(2): 197–217.

Finkelstein, Sy dney, and Donald C. Hambrick. 1990. Top-Management-Team
Tenure and Organizational Outcomes: Th e Moderating Role of Managerial
Discretion. Administrative Science Quarterly 35(3): 484–503.

Francis, Bill, Iftekhar Hasan, Jong Chool Park, and Qiang Wu. 2009. Gender
Diff erences in Financial Reporting Decision-Making: Evidence from Accounting
Conservatism. Paper presented at the Annual Meeting of the Financial
Management Association, Reno, NV, October 21–14.

Frees, Edward W. 20 04. Longitudinal and Panel Data: Analysis and Applications in the
Social Sciences. New York: Cambridge University Press.

Garrone, Paola, Luc a Grilli, and Xavier Rousseau. 2013. Management Discretion and
Political Interference in Municipal Enterprises: Evidence from Italian Utilities.
Local Government Studies 39(4): 514–40.

Greve, Carsten. 200 6. Public Management Reform in Denmark. Public Management
Review 8(1): 161–69.

Guy, Mary Ellen, an d Meredith A. Newman. 2004. Women’s Jobs, Men’s Jobs: Sex
Segregation and Emotional Labor. Public Administration Review 64(3): 289–98.

Hambrick, Donald C. 1994. Top Management Groups: A Conceptual Integration
and Reconsideration of the “Team” Label. Research in Organizational Behavior 6:
171–214.

———. 2007. Upper Echelons Th eory: An Update. Academy of Management Review
32(2): 334–43.

Hambrick, Donald C. , Th eresa Seung Cho, and Ming-Jer Chen. 1996. Th e Infl uence
of Top Management Team Heterogeneity on Firms’ Competitive Moves.
Administrative Science Quarterly 41(4): 659–84.

TMTs in detail and outline why and how human diversity aff ects
organizational outcomes.

Further, in this study, we explored team dynamics but with the
organization as the eff ective unit of analysis. We urge future research
to consider teams as units of analysis. While there is a long tradition
of team-based research in the general organizational behavior litera-
ture (Van Knippenberg and Schippers 2007), this type of research
is rare in public administration. Consequently, we know too little
about team dynamics in public sector organizations. Exploring team
diversity would be a well-suited starting point for such research.

Finally, an avenue for future research relates not to the outcome of
TMT diversity but to antecedents. With the demographic charac-
teristics of managers being important for organizational outcomes,
we should increase our understanding of the rationales organizations
use in hiring managers with diverse characteristics. Factors such as
current TMT composition, administrative structure and strategy,
and environmental demands could be expected to relate to the
degree of TMT diversity pursued by public organizations.

Gender diversity has the potential to lead to superior organizational
outcomes, but it does not automatically do so. Th e present analysis
implies that to reap the benefi ts of diverse top management teams,
the management structure must facilitate behavioral integration
between TMT members. However, more knowledge of other con-
tingencies and pitfalls is needed.

Notes
1. Besides their tax revenue, municipalities are subsidized by block grants from the

national government (Greve 2006).
2. Th e basis of the current system of local government in Denmark was created

in the 1970 municipal reform (Kommunalreformen), in which 1,388 so-called
parish municipals (sognekommuner) and market towns (købstæder) were organized
into 275 municipalities. Th is structure existed until 2007, when a new structural
reform (Strukturreformen) reduced the number of municipalities to 98. Th irty-
three of the existing municipalities continued unchanged, whereas the remaining
65 were new entities based on (more or less) voluntary amalgamations. Th e fi ve
municipalities on the island of Bornholm “jumped the gun” on the structural
reform and merged into one municipality in 2003 (Blom-Hansen et al. 2012).

3. See http://krevi.dk/noegletal [in Danish]. Since June 1, 2012, the Danish
Evaluation Institute for Local Government has been a part of the National
Institute for Local and Regional Analysis and Research, an independent institu-
tion under the Ministry of Economy and the Interior.

4. See http://www.noegletal.dk/ [in Danish].

References
Ande rsen, Simon Calmar, and Søren Serritzlew. 2007. Th e Unintended Eff ects of

Private School Competition. Journal of Public Administration Research and Th eory
17(2): 335–56.

Andre ws, Rhys, George A. Boyne, and Richard M. Walker. 2006. Strategy Content
and Organizational Performance: An Empirical Analysis. Public Administration
Review 66(1): 52–63.

Auh, Seigyoung, and Bulent Menguc. 2006. Diversity at the Executive Suite:
A Resource-Based Approach to the Customer Orientation–Organizational
Performance Relationship. Journal of Business Research 59(5): 564–72.

Bækga ard, Martin. 2011. Th e Impact of Formal Organizational Structure on
Politico-Administrative Interaction: Evidence from a Natural Experiment. Public
Administration 89(3): 1063–80.

The Right Mix? Gender Diversity in Top Management Teams and Financial Performance 301

Mouritzen, Poul Erik , and James H. Svara. 2002. Leadership at the Apex: Politicians
and Administrators in Western Local Governments. Pittsburgh, PA: University of
Pittsburgh Press.

Moynihan, Donald P., and Patricia Wallace Ingraham. 2004. Integrative Leadership
in the Public Sector: A Model of Performance-Information Use. Administration
& Society 36(4): 427–53.

Moynihan, Donald P., and Sanjay K. Pandey. 2010. Th e Big Question for
Performance Management: Why Do Managers Use Performance Information?
Journal of Public Administration Research and Th eory 20(4): 849–66.

Østergaard, Christian R., Bram Timmermans, and Kari Kristinsson. 2011. Does a
Diff erent View Create Something New? Th e Eff ect of Employee Diversity on
Innovation. Research Policy 40(3): 500–509.

Park, Sanghee. 2013. Does Gender Matter? Th e Eff ect of Gender Representation of
Public Bureaucracy on Governmental Performance. American Review of Public
Administration 43(2): 221–42.

Pitts, David W. 2005. Diversity, Representation, and Performance: Evidence about
Race and Ethnicity in Public Organizations. Journal of Public Administration
Research and Th eory 15(4): 615–31.

Pitts, David W., and Lois Recascino Wise. 2010. Workforce Diversity in the New
Millennium: Prospects for Research. Review of Public Personnel Administration
30(1): 44–69.

Powell, Melanie, and David Ansic. 1997. Gender Diff erences in Risk Behaviour
in Financial Decision-Making: An Experimental Analysis. Journal of Economic
Psychology 18(6): 605–28.

Riccucci, Norma M., G regg G. Van Ryzin, and Cecilia F. Lavena. 2014.
Representative Bureaucracy in Policing: Does It Increase Perceived Legitimacy?
Journal of Public Administration Research and Th eory. Published electronically on
March 6. doi:10.1093/jopart/muu006.

Roch, Christine H., a nd David W. Pitts. 2012. Diff ering Eff ects of Representative
Bureaucracy in Charter Schools and Traditional Public Schools. American Review
of Public Administration 42(3): 282–302.

Serritzlew, Søren. 20 05. Breaking Budgets: An Empirical Examination of
Danish Municipalities. Financial Accountability and Management 21(4):
413–35.

Stivers, Camilla. 1995 . Settlement Women and Bureau Men: Constructing a
Usable Past for Public Administration. Public Administration Review 55(6):
522–29.

Stoner, James A. F. 19 68. Risky and Cautious Shifts in Group Decisions: Th e
Infl uence of Widely Held Values. Journal of Experimental Social Psychology 4(4):
442–59.

Van Knippenberg, Daan, and Michaela C. Schippers. 2007. Work Group Diversity.
Annual Review of Psychology 58: 515–41.

Van Wart, Montgomery. 2013. Lessons from Leadership Th eory and the
Contemporary Challenges of Leaders. Public Administration Review 73(4):
553–65.

Villadsen, Anders R. 2 012. New Executives from Inside or Outside? Th e Eff ect of
Executive Replacement on Organizational Change. Public Administration Review
72(5): 731–40.

Vrangbæk, Karsten. 201 0. Structural Reform in Denmark, 2007–09: Central Reform
Processes in a Decentralised Environment. Local Government Studies 36(2):
205–21.

Wildavsky, Aaron B, and Naomi Caiden. 2004. Th e New Politics of the Budgetary
Process. New York: Pearson/Longman.

Wilkins, Vicky M., and Lael R Keiser. 2006. Linking Passive and Active
Representation by Gender: Th e Case of Child Support Agencies. Journal of
Public Administration Research and Th eory 16(1): 87–102.

Zahra, Shaker A., and J ohn A. Pearce. 1989. Boards of Directors and Corporate
Financial Performance: A Review and Integrative Model. Journal of Management
15(2): 291–334.

Hansen, Sune Wellin g, Kurt Houlberg, and Lene Holm Pedersen. 2014. Do
Municipal Mergers Improve Fiscal Outcomes? Scandinavian Political Studies
37(2): 196–214.

Hefetz, Amir, and M ildred Warner. 2004. Privatization and Its Reverse: Explaining
the Dynamics of the Government Contracting Process. Journal of Public
Administration Research and Th eory 14(2): 171–90.

———. 2012. Contract ing or Public Delivery? Th e Importance of Service, Market,
and Management Characteristics. Journal of Public Administration Research and
Th eory 22(2): 289–317.

Jackson, Susan E., Aparna Joshi, and Niclas L. Erhardt. 2003. Recent Research on
Team and Organizational Diversity: SWOT Analysis and Implications. Journal of
Management 29(6): 801–30.

Jacobson, Willow S. , Christine Kelleher Palus, and Cynthia J. Bowling. 2010.
A Woman’s Touch? Gendered Management and Performance in State
Administration. Journal of Public Administration Research and Th eory 20(2):
477–504.

Johansen, Morgen S. 2007. Th e Eff ect of Female Strategic Managers on
Organizational Performance. Public Organization Review 7(3): 269–79.

Juenke, Eric G. 200 5. Management Tenure and Network Time: How Experience
Aff ects Bureaucratic Dynamics. Journal of Public Administration Research and
Th eory 15(1): 113 –31.

Keiser, Lael R., Vi cky M. Wilkins, Kenneth J. Meier, and Catherine A.
Holland. 2002. Lipstick and Logarithms: Gender, Institutional Context,
and Representative Bureaucracy. American Political Science Review 96(3):
553–64.

Kent, Russell L., a nd Sherry E. Moss. 1994. Eff ects of Sex and Gender Role on
Leader Emergence. Academy of Management Journal 37(5): 1335–46.

Ki ngsley, John Dona ld. 1944. Representative Bureaucracy: An Interpretation of the
British Civil Service. Yellow Springs, OH: Antioch Press.

Knight, Don, Craig L. Pearce, Ken G. Smith, Judy D. Olian, Henry P. Sims,
Ken A. Smith, and Patrick Flood. 1999. Top Management Team Diversity,
Group Process, and Strategic Consensus. Strategic Management Journal 20(5):
445–65.

Kochan, Th omas, Kat erina Bezrukova, Robin Ely, Susan Jackson, Aparna Joshi,
Karen Jehn, Jonathan Leonard, David Levine, and David Th omas. 2003. Th e
Eff ects of Diversity on Business Performance: Report of the Diversity Research
Network. Human Resource Management 42(1): 3–21.

Krishnan, Hema A., and Daewoo Park. 2005. A Few Good Women—On Top
Management Teams. Journal of Business Research 58(12): 1712–20.

Lubatkin, Michael H ., Zeki Simsek, Yan Ling, and John F Veiga. 2006.
Ambidexterity and Performance in Small- to Medium-Sized Firms: Th e Pivotal
Role of Top Management Team Behavioral Integration. Journal of Management
32(5): 646–72.

Meier, Kenneth J., and John Bohte. 2001. Structure and Discretion: Missing Links in
Representative Bureaucracy. Journal of Public Administration Research and Th eory
11(4): 455–70.

Meier, Kenneth J., and Laurence J. O’Toole, Jr. 2001. Managerial Strategies and
Behavior in Networks: A Model with Evidence from U.S. Public Education.
Journal of Public Administration Research and Th eory 11(3): 271–94.

———. 2003. Public M anagement and Educational Performance: Th e Impact of
Managerial Networking. Public Administration Review 63(6): 689–99.

Meier, Kenneth J., Laurence J. O’Toole, Jr., and Holly T. Goerdel. 2006.
Management Activity and Program Performance: Gender as Management
Capital. Public Administration Review 66(1): 24–36.

Miller, Toyah, and María del Carmen Triana. 2009. Demographic Diversity in the
Boardroom: Mediators of the Board Diversity–Firm Performance Relationship.
Journal of Management Studies 46(5): 755–86.

Mintzberg, Henry. 19 79. Th e Structuring of Organizations: A Synthesis of the Research.
Englewood Cliff s, NJ: Prentice Hall.