The effect of Enron Andersen and Sarbanes Oxley on the US market for audit services

Description
The purpose of this paper is to examine the effect of the Enron scandal, Arthur
Andersen’s demise and the Sarbanes-Oxley Act on audit fees.

Accounting Research Journal
The effect of Enron, Andersen, and Sarbanes-Oxley on the US market for audit services
Sharad Asthana Steven Balsam Sungsoo Kim
Article information:
To cite this document:
Sharad Asthana Steven Balsam Sungsoo Kim, (2009),"The effect of Enron, Andersen, and Sarbanes-Oxley
on the US market for audit services", Accounting Research J ournal, Vol. 22 Iss 1 pp. 4 - 26
Permanent link to this document:
http://dx.doi.org/10.1108/10309610910975306
Downloaded on: 24 January 2016, At: 21:07 (PT)
References: this document contains references to 49 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 3035 times since 2009*
Users who downloaded this article also downloaded:
J oseph A. Petrick, Robert F. Scherer, (2003),"The Enron Scandal and the Neglect of
Management Integrity Capacity", American J ournal of Business, Vol. 18 Iss 1 pp. 37-50 http://
dx.doi.org/10.1108/19355181200300003
Hsi-An Shih, Yun-Hwa Chiang, Chu-Chun Hsu, (2006),"Can high performance work systems really
lead to better performance?", International J ournal of Manpower, Vol. 27 Iss 8 pp. 741-763 http://
dx.doi.org/10.1108/01437720610713530
J oseph Falzon, David Lanzon, (2013),"Comparing alternative house price indices: evidence from asking
prices in Malta", International J ournal of Housing Markets and Analysis, Vol. 6 Iss 1 pp. 98-135 http://
dx.doi.org/10.1108/17538271311306048
Access to this document was granted through an Emerald subscription provided by emerald-srm:115632 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.
*Related content and download information correct at time of download.
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
0
7

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
The effect of Enron, Andersen,
and Sarbanes-Oxley on the US
market for audit services
Sharad Asthana
Department of Accounting-COB, University of Texas at San Antonio,
San Antonio, Texas, USA
Steven Balsam
Department of Accounting, Fox School of Business and Management,
Temple University, Philadelphia, Pennsylvania, USA, and
Sungsoo Kim
Department of Accounting, School of Business, Rutgers University,
Camden, New Jersey, USA
Abstract
Purpose – The purpose of this paper is to examine the effect of the Enron scandal, Arthur
Andersen’s demise and the Sarbanes-Oxley Act on audit fees.
Design/methodology/approach – The paper uses empirical methodology (univariate and
multivariate).
Findings – Audit fees and the Big-4 premium increased in 2002. Increase was larger for bigger and
riskier clients. Evidence is also consistent with a competitive market for former Andersen clients.
Research limitations/implications – Data requirements might bias the sample towards larger
sized ?rms. Data availability limits the number of observations.
Practical implications – The research ?ndings on audit fees in post-Enron and Arthur Andersen
period reported in this paper are important for policy makers.
Originality/value – It is found that the premium charged by Big 4 over non-Big 4 has increased in
2002, and that the ability of an auditor to charge a premiumis adversely affected when its reputation is
tarnished. It is also reported that the frequency of voluntary switches within the Big 4 is lowest in
19 years. The audit fee model was also re?ned by adding two ownership variables to control for
agency aspect of client ?rms; inside and institutional ownership.
Keywords United States of America, Auditor’s fees, Legislation, Auditing standards
Paper type Research paper
1. Introduction
Since 2001, the market for audit services has changed dramatically, both in the USA
and abroad. Scandals involving large public companies such as Enron and WorldCom
highlighted the need for more intensive audits. Choices by the judicial and
legislative/executive branches of the US Government reduced the number of suppliers
in the market for audit services. In particular, the decision by US Department of Justice
to prosecute Andersen as a ?rm reduced the already small number of audit ?rms
capable of auditing large multinational corporations from ?ve to four. Reducing the
number of audit ?rms limits choice and increases the power of the remaining audit
?rms vis-a` -vis their clients. At about the same time, the additional work required by the
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1030-9616.htm
ARJ
22,1
4
Accounting Research Journal
Vol. 22 No. 1, 2009
pp. 4-26
qEmerald Group Publishing Limited
1030-9616
DOI 10.1108/10309610910975306
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
0
7

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
Sarbanes-Oxley Act also increased the costs to accounting ?rms of auditing public
corporations, for example, by requiring they register with and meeting the
requirements of the Public Company Accounting Oversight Board (PCAOB), which
likely caused a number of small ?rms to exit the market, decreasing competition for
small audits (General Accounting Of?ce, 2003)[1]. The effect of all these changes is to
increase the amount of work required of auditors and to reduce competition in the audit
market. Both have the potential to increase audit fees.
In this paper, we examine the effect of the con?uence of events beginning with the
disclosures relating to the audit of Enron, continuing with the trial and conviction of
Andersen, and ending with the passage of the Sarbanes-Oxley Act on the market for
audit services[2]. In particular, we examine the effect of these events on the fees paid by
clients for their audits.
As a percentage of total assets, the average audit fee increased from 0.070 percent in
2000 to 0.079 percent in 2001 to 0.107 percent in 2002. We attribute the 2001 increase to
the more intensive audits brought on by the scandals highlighted above. It may have
also been driven by the disarray in the audit market, that is, at the time most 2001
audits were being completed, switching to Andersen was no longer an option and most
Andersen clients were in the process of selecting new auditors. Consequently, the
increase may also have been driven by a lack of bargaining power on the part of clients
vis-a` -vis audit ?rms. The 2002 increase can be attributed to the requirements of
Sarbanes-Oxley, as well as the effect of Andersen’s demise on the supply of audit
services.
We begin by cross-sectionally examining audit fees for the years 2000-2002. We
document that fees were higher in 2002 vis-a` -vis 2000 and 2001 after controlling for the
previously found determinants of audit fees. We also con?rm the existence of a Big 4/5
premiumand ?nd that it increased in 2002. While this premiumdoes not differ between
Andersen and the Big 4 ?rms in 2000, in 2001 the premium charged by Andersen was
signi?cantly lower than that charged by the Big 4, which is consistent with Andersen
charging less because of its, by then lessened reputation, and/or weakened bargaining
power. Interestingly, former Andersen clients who switched to a Big 4 auditor in 2002
paid lower audit fees than pre-existing clients of the Big 4, consistent with DeAngelo’s
(1981) low-balling scenario.
We then move to a change model to examine whether the increases observed over
time differentially affected any one group of clients. First, we document that the
increase in audit fees charged by the Big 4 was greater than the increase by non-Big 4
auditors in 2002. Then, we ?nd that bigger and riskier clients had larger increases. We
believe that even after controlling for auditor identity, larger clients had a bigger
increase because they had fewer alternatives. That is, while a smaller public company
could potentially switch to a non-Big 4 auditor, a bigger company would not have that
option. We believe that riskier clients may have been subject to larger increases for a
number of reasons. For example, audit ?rms may have increased the amount of time
spent on these audits to reduce their risk. Another potential explanation is that audit
?rms increased the risk premium charged these clients because of changes in their
perceived risks.
This paper continues with theory and hypothesis development in Section 2. Section
3 discusses our data. Section 4 presents our model and Section 5 the empirical results.
The paper concludes in Section 6 with a discussion of our results.
Enron,
Andersen, and
Sarbanes-Oxley
5
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
0
7

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
2. Theory hypothesis development
2.1 Existing research
A string of papers beginning with Chaney and Philipich (2002) have documented the
consequences of the Enron debacle on share prices (i.e. declines) of Andersen’s
clients (Callen and Morel, 2002) and share prices of clients of other Big 4 auditors
(Asthana et al., 2004; Dooger et al., 2003). Asthana et al. (2009) follow up on
this literature by showing that share prices recovered, that is, there was a positive
reaction, when Andersen clients announced a switch to a new audit ?rm. Several
working papers address the effect of these events on earnings management (Balsam,
2003; Cohen et al., 2003; Lai, 2003), and audit opinions (Lai, 2003) ?nding some evidence
that earnings quality improved and auditors were more likely to give modi?ed
opinions afterwards. In another vein, Eisenberg and Macey (2003) examine earnings
restatements and do not ?nd evidence that Andersen audits were of inferior quality.
More recent research further examines the outcome of the Andersen/Enron incident
on the market for audit services. Hamilton et al. (2008) report evidence that the markets
are still competitive and Big N continue to charge a premium for their services in the
post-Enron era. Krishnan and Visvanathan (2008) conclude that Andersen, particularly
the Houston of?ce, was more tolerant of earnings management relative to other Big 5
auditors. As a result, the new auditors of former Andersen clients were more
conservative towards them (Krishnan and Chaney, 2007) and large former Andersen
clients were more likely to get a going-concern opinion (Krishnan et al., 2007) and had
lower levels of and larger decreases in abnormal accruals (Cahan and Zhang, 2006).
These ?ndings suggest that successor auditors viewed former Andersen clients as a
source of litigation risk. Additionally, Kealey et al. (2007) show that the fee charged by
the successor auditor varied positively with the clients tenure with Andersen. Blouin
et al. (2007) examine the factors involved in the selection of new auditors by former
Andersen clients. They show that ?rms with greater agency concerns were more likely
to sever ties with their former auditor, whereas those with greater switching costs were
more likely to follow their former audit team. Chen and Zhou (2007) conclude that ?rms
with more independent audit committees, audit committees with greater ?nancial
expertise, and larger and more independent boards dismissed Andersen earlier. They
also ?nd that ?rms with larger and more active audit committees as well as more
independent boards were more likely to choose a Big 4 successor auditor.
On a similar line of audit fee research Hay et al. (2006) conduct cross-country survey
of determinants of audit fees, ?nding many of the variables to be consistent across the
country, with some variables including ?rm losses, governance, audit opinion are
signi?cant in more recent years. Raghunandan and Rama (1999) study the differences
between auditor dismissal and resignation and who succeeds the resigned auditors.
They report Big N auditors are less likely to succeed the resigned auditors due to
potential accounting problems. In Sarbanes-Oxley related research, Markelevich et al.
(2005) report a strong positive correlation between discretionary accruals and
non-audit fees before the Sarbanes-Oxley, but not after due to its implementation.
Hogan and Wilkins (2008) examine the relations between internal control and ?nancial
reporting and its impact on audit fees.
In Table I, we summarize the more recent research on audit fee premiums and low
balling and compare their results to our research. In addition to con?rming prior
empirical ?ndings of low balling for former AA clients (Kohlbeck et al., 2008;
ARJ
22,1
6
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
0
7

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
P
a
p
e
r
D
a
t
a
a
n
d
m
e
t
h
o
d
o
l
o
g
y
B
i
g
4
p
r
e
m
i
u
m
f
o
r
f
o
r
m
e
r
A
A
c
l
i
e
n
t
s
N
o
n
-
B
i
g
4
p
r
e
m
i
u
m
f
o
r
f
o
r
m
e
r
A
A
c
l
i
e
n
t
s
L
o
w
b
a
l
l
i
n
g
f
o
r
A
A
c
l
i
e
n
t
s
R
e
m
a
r
k
s
O
u
r
p
a
p
e
r
S
&
P
A
u
d
i
t
F
e
e
D
a
t
a
2
0
0
0
,
2
0
0
1
,
2
0
0
2
P
r
e
m
i
u
m
i
n
2
0
0
1
a
n
d
2
0
0
2
m
o
r
e
t
h
a
n
n
o
n
-
B
i
g
4
P
r
e
m
i
u
m
b
u
t
l
e
s
s
t
h
a
n
B
i
g
4
L
o
w
b
a
l
l
i
n
g
V
o
l
u
n
t
a
r
y
s
w
i
t
c
h
w
i
t
h
i
n
B
i
g
4
l
o
w
e
s
t
i
n
1
9
y
e
a
r
s
M
o
r
e
p
r
e
m
i
u
m
f
o
r
l
a
r
g
e
r
a
n
d
r
i
s
k
i
e
r
?
r
m
s
A
b
i
d
i
n
e
t
a
l
.
(
2
0
0
8
)
U
K
D
a
t
a
(
1
9
9
8
-
2
0
0
3
)
;
d
e
s
c
r
i
p
t
i
v
e
s
t
a
t
s
N
o
p
r
e
m
i
u
m
,
o
n
l
y
i
n
?
a
t
i
o
n
a
d
j
u
s
t
m
e
n
t
N
o
p
r
e
m
i
u
m
,
o
n
l
y
i
n
?
a
t
i
o
n
a
d
j
u
s
t
m
e
n
t
N
o
e
v
i
d
e
n
c
e
o
f
l
o
w
b
a
l
l
i
n
g
K
o
h
l
b
e
c
k
e
t
a
l
.
(
2
0
0
8
)
2
0
0
2
a
u
d
i
t
a
n
a
l
y
t
i
c
a
n
d
E
d
g
a
r
8
-
K
F
o
c
u
s
e
d
o
n
f
o
r
m
e
r
A
A
c
l
i
e
n
t
s
O
L
S
,
b
u
t
m
i
s
s
i
n
g
i
n
s
i
d
e
o
w
n
e
r
s
h
i
p
v
a
r
i
a
b
l
e
N
o
t
e
x
a
m
i
n
e
d
N
o
t
e
x
a
m
i
n
e
d
F
o
r
m
e
r
A
A
c
l
i
e
n
t
s
w
i
t
c
h
e
r
s
:
l
o
w
b
a
l
l
i
n
g
T
h
o
s
e
w
h
o
m
o
v
e
d
a
l
o
n
g
w
i
t
h
t
h
e
i
r
f
o
r
m
e
r
A
A
t
e
a
m
t
o
a
n
e
w
a
u
d
i
t
o
r
:
n
o
l
o
w
b
a
l
l
i
n
g
F
o
c
u
s
e
d
o
n
a
u
d
i
t
f
e
e
s
f
o
r
f
o
r
m
e
r
A
A
c
l
i
e
n
t
s
b
y
o
t
h
e
r
B
i
g
4
S
a
n
k
a
r
a
g
u
r
u
s
w
a
m
y
a
n
d
W
h
i
s
e
n
a
n
t
(
2
0
0
5
)
S
&
P
:
a
u
d
i
t
f
e
e
d
a
t
a
T
e
s
t
D
y
e
(
1
9
9
1
)
m
o
d
e
l
,
n
o
t
r
e
l
a
t
e
d
t
o
A
n
d
e
r
s
e
n
N
o
t
e
x
a
m
i
n
e
d
N
o
t
e
x
a
m
i
n
e
d
F
e
e
d
i
s
c
o
u
n
t
i
n
g
e
v
e
n
w
i
t
h
p
u
b
l
i
c
d
i
s
c
l
o
s
u
r
e
s
(
t
e
s
t
e
d
D
y
e
’
s
(
1
9
9
1
)
t
h
e
o
r
y
)
C
o
n
s
i
s
t
e
n
t
w
i
t
h
D
e
A
n
g
e
l
o
(
1
9
8
1
)
N
o
t
r
e
l
a
t
e
d
t
o
A
n
d
e
r
s
e
n
o
r
S
a
r
b
a
n
e
s
-
O
x
l
e
y
A
c
t
(
2
0
0
2
)
C
h
i
(
2
0
0
4
)
S
&
P
a
u
d
i
t
f
e
e
d
a
t
a
;
O
L
S
,
b
u
t
m
i
s
s
i
n
g
i
n
s
i
d
e
o
w
n
e
r
s
h
i
p
v
a
r
i
a
b
l
e
Y
e
s
,
g
r
e
a
t
e
r
t
h
a
n
n
o
n
-
B
i
g
4
Y
e
s
,
b
u
t
s
m
a
l
l
e
r
(
s
i
m
i
l
a
r
?
n
d
i
n
g
)
T
h
e
r
e
i
s
l
o
w
b
a
l
l
i
n
g
H
o
a
n
d
W
a
n
g
(
2
0
0
7
)
S
&
P
A
u
d
i
t
F
e
e
d
a
t
a
O
L
S
,
b
u
t
n
o
i
n
s
i
d
e
o
w
n
e
r
s
h
i
p
v
a
r
i
a
b
l
e
N
o
B
i
g
4
p
r
e
m
i
u
m
i
n
2
0
0
1
2
0
0
2
l
o
w
b
a
l
l
i
n
g
N
o
f
e
e
d
i
f
f
e
r
e
n
c
e
b
e
t
w
e
e
n
B
i
g
4
a
n
d
s
e
c
o
n
d
t
i
e
r
,
b
u
t
f
e
e
d
i
f
f
e
r
e
n
c
e
b
e
t
w
e
e
n
s
e
c
o
n
d
a
n
d
t
h
i
r
d
t
i
e
r
s
G
h
o
s
h
a
n
d
L
u
s
t
g
a
r
t
e
n
(
2
0
0
6
)
S
&
P
A
u
d
i
t
F
e
e
d
a
t
a
O
L
S
(
l
e
v
e
l
a
n
d
c
h
a
n
g
e
)
,
n
o
i
n
s
i
d
e
o
w
n
e
r
s
h
i
p
v
a
r
i
a
b
l
e
N
o
t
e
x
a
m
i
n
e
d
N
o
t
e
x
a
m
i
n
e
d
M
o
r
e
d
i
s
c
o
u
n
t
f
r
o
m
s
m
a
l
l
e
r
a
u
d
i
t
o
r
s
(
1
5
p
e
r
c
e
n
t
)
t
h
a
n
l
a
r
g
e
a
u
d
i
t
o
r
s
(
1
1
p
e
r
c
e
n
t
)
C
o
n
s
i
s
t
e
n
t
w
i
t
h
D
e
A
n
g
e
l
o
(
1
9
8
1
)
a
n
d
C
h
a
n
(
1
9
9
9
)
Table I.
Summary of current
research on
premium/lowballing
Enron,
Andersen, and
Sarbanes-Oxley
7
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
0
7

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
Sankaraguruswamy and Whisenant, 2005; Chi, 2004; Ho and Wang, 2007; Ghosh and
Lustgarten, 2006), and Big 4 and non-Big 4 premium (Chi, 2004) we found the premium
charged by Big 4 auditors to have increased in 2002. Moreover, we show that the
ability of an auditor to charge a premium is adversely effected when its reputation is
tarnished. We also report that the frequency of voluntary switches within the Big 4 is
the lowest in 19 years, a ?nding we attribute to increased sensitivity to risk on the part
of auditors.
In a re?nement of the audit fee model used in prior research, we add two ownership
variables to control for the agency aspect of clients ?rms; inside and institutional
ownership. Given the structural changes, the auditing industry has gone through
recently, we believe these two variables should be necessary determinants in future
research on audit fees.
2.2 Hypotheses development
As is evident from the research discussed above, the market for audit services of public
corporations changed dramatically from late 2001 to mid-2002. On October 16, 2001 the
Enron Corporation disclosed a major asset write-down. While the price of Enron’s
shares plummeted immediately in response, there was no broad effect on the market
until the disclosures related to Andersen’s shredding of documents on January 10, 2002
(Chaney and Philipich, 2002). Shortly thereafter, the exodus of clients from Andersen
began (Asthana et al., 2009), an exodus that was accelerated by Andersen’s conviction
on June 15, 2002, which ended any hope of Andersen surviving as an auditor of public
corporations. Then on July 30, 2002 President Bush signed into law the Sarbanes-Oxley
Act. The Sarbanes-Oxley Act both prohibited independent auditors from performing
certain services for their clients as well as required them to perform additional audit
and audit-related services.
The disclosure of accounting irregularities at Enron, which followed earlier scandals
at other large public corporations including Waste Management and Sunbeam, led
many inside and outside of the accounting profession to conclude that audits needed to
be more exhaustive, with less reliance on management and more testing. For example,
McNamee and Weber (2002) note that “Enron’s failure has inspired directors and
shareholders to demand more from their auditors.” Consequently, audits for the 2001
?scal year may have cost more because of increased hours put in by audit personnel.
Further to the extent that audit fees are negotiated by the auditor and client after the
audit is complete, the disarray in the audit profession and the exodus of clients from
Andersen in early 2002 may have given the auditor the upper hand in the negotiations.
Similarly, the demise of Andersen removed one competitor from a market that, at least
for large public corporations, previously consisted of at most ?ve competitors[3].
Section 201 of Sarbanes-Oxley sharply limited the services accounting ?rms could
offer their clients, that is, consulting. If auditing was used as a “loss leader,” limiting
consulting will in equilibriumcause accounting ?rms to increase the price of their audit
services, that is, they no longer can offset their audit losses against the pro?ts from
their consulting (Coleman and Bryan-Low, 2002). In addition, Section 404 of
Sarbanes-Oxley requires management to report on internal controls and their auditor
to attest to management’s report, increasing the number of hours the audit would take.
Assuming that audit ?rms pass these costs on to their clients, these additional
requirements would also have the effect of increasing audit fees[4].
ARJ
22,1
8
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
0
7

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
Individually and jointly, the events described above should increase the fees
charged for auditing services:
H1. Audit fees increased post-Enron, Andersen, and Sarbanes-Oxley.
While we expect audit fees to increase across the board, the possibility exists that these
increases will differ across auditors and clients. Prior research has shown that the large
multinational audit ?rms that is, Big 5/6/8, charge higher fees, presumably because of
their greater reputations (Francis, 1984; Palmrose, 1986; Simunic, 1980). To the extent
that Andersen’s reputation was damaged due to the disclosures related to its audit of
Enron, its ability to charge a premiumwas diminished subsequent to those disclosures:
H2. The premium charged by Andersen decreased after the disclosures pertaining
to its audit of Enron.
Under normal circumstances auditors compete for new clients, potentially going so far
as to offer lower fees to attract business (DeAngelo, 1981; Schatzberg, 1990). They do
this despite the asymmetric information that exists between potential client and
auditor, and the potential for lemons analogous to the scenario described by Akerlof
(1970). In the new post-Enron, Andersen, and Sarbanes-Oxley environment, we expect
that auditors may be more circumspect in accepting/competing for new clients, and
potentially because of the asymmetric information, charge them to compensate for the
additional risk. Compounding this is the vast number of Andersen clients seeking new
audit ?rms, straining the capacity of the remaining Big 4 and making those ?rms
extremely selective in taking on new clients[5].
Clients that switch auditors in 2002 fall into two categories, Andersen clients forced
to switch auditors, and non-Andersen clients voluntarily switching auditors[6]. The
remaining audit ?rms may be more willing to take on Andersen clients as the true
reason for switching is known, that is, there is less asymmetric information.
Consequently, they may charge a lower fee to former Andersen clients. However, while
former Andersen clients have no choice but to ?nd a new auditor, clients of other
auditors do not have to switch[7]. Consequently, any increase in audit fee may dissuade
them from switching. So while we would predict, because of asymmetric information,
clients of non-Andersen auditors would be less desirable as new clients than Andersen
clients, the switches we do observe among the non-Andersen clients will be a truncated
distribution, that is, those that thought about changing and found the terms of the
switch acceptable. Consequently, we cannot say how audit fee increases will differ
across the two groups of switchers, nor can we say whether audit fees will differ
between switchers and non-switchers. Hence, we word hypothesis three in the null form:
H3. Audit fees will not differ for clients with new auditors.
The exit of Andersen from the market for audit services reduced the number of
auditors that could audit large public corporations. Consequently, the market power of
the remaining Big 4 auditors will increase, allowing them to charge higher fees:
H4. The premium charged by the Big 4 increased post Andersen.
While the exit of Andersen from the market for audit services reduced the number of
auditors that could audit large public corporations, the increase in costs associated
with Sarbanes-Oxley may have reduced the number of ?rms competing for the audit of
Enron,
Andersen, and
Sarbanes-Oxley
9
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
0
7

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
small corporations. However, there is still more competition for these audits, as while
only four audit ?rms can audit large multinational corporations, almost 800 audit ?rms
had registered with the PCAOB as of March 2004. Consequently, the effect of
Andersen’s exit and Sarbanes-Oxley may not be as pronounced on the market for
audits services for smaller corporations:
H5. Audit fee increases are greater for large clients.
The increase in audit fees may also depend on the risk of the client as the amount of
additional work the auditor has to do may depend on the risk of the client.
Additionally, in the new environment, the risk premium charged by the auditor may
change. Coleman and Bryan-Low (2002) quote John O’Connor, the Vice Chairman of
PricewaterhouseCoopers as saying, “Those companies that don’t have strong
management or represent high risk will see higher rates of increase”:
H6. Audit fee increases are greater for riskier clients.
3. Data
This paper utilizes audit fees disclosed in proxy statements ?led by public corporations
as compiled by Standard &Poors. The database began in 2000 following Securities and
Exchange Commission Release No. 33-7919 which mandated audit fee disclosure, and
had data through 2003 audits at the time we conducted our analysis. Unfortunately,
Securities and Exchange Commission Release No. 33-8183, which changed the reporting
format for audit fees, as well as what was to be included in the audit fee classi?cation
went into affect during this time period complicating our analysis. Consequently, to
ensure that our analysis is conducted using a constant de?nition of audit fees we exclude
all observations from 2003 (all companies had to adopt the expanded disclosure
provisions for ?scal years ending on or after May 6, 2003), as well as audit fees for
companies that adopted the new format in 2002 (early adopters). Since our audit fee
database reports the restated ?gures (SEC Release 33-8183 required ?rms on adoption of
the newdisclosure requirements to restate prior year ?gures), we manually collect all the
proxy statements for 2002 and overwrite the 2002 restated ?gures in the database with
the original disclosures. This insures that the data for the period 2000-2002 is consistent
and based on the pre-SEC Release 33-8183 disclosures.
We then merge the fee data with data extracted from Standard & Poors Compustat,
CRSP, and Compact Disclosure. After adding the requirement that ?rms have data for
the years 2000-2002, we are left with a sample of 771 ?rms and 2,313 ?rm-year
observations[8]. Panel A of Table II contains the sample selection procedure. Panel B of
Table II contains the industry distribution. The largest group of companies, about
39 percent of the sample, is in the manufacturing classi?cation, with the next largest
group of companies, about 25 percent of the sample, being in ?nancial services. As
shown in Panel B, in general the percentages in the sample mirror that of the Compustat
population. Panel C shows the sample distribution across auditors and years.
Table III provides descriptive statistics on audit fees and changes in audit fees over
the 2000-2002 period. While the overall in?ation rate was rather tame over the period
under examination, about 2 percent per year, the percentage change in audit fees was
much higher overall, 6.47 percent in 2001, and 21.04 percent in 2002 (see Panel B)[9]. In
dollar terms (Panel A), the average audit fee increased from $704,460 in 2000 to
$1,095,520 in 2002, whereas as a percentage of total assets it increased from
ARJ
22,1
10
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
0
7

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
Panel A: sample selection
Number of ?rm-years
Procedure Lost Left
Data available for 2000-2002 on 2003 audit
fee database 15,479
Firm is audited (7) 15,472
Auditor’s name available (21) 15,451
Data available on COMPUSTAT, CRSP,
and compact-disclosure databases (5,136) 10,315
Data available for all years during sample
period (6,817) 3,498
Proxy Statements available on
EDGAR.GOV (171) 3,327
Less early adopters of SEC Regulation
No. 33-8183 disclosure requirements (1,014) 2,313
a
Panel B: sample distribution across industries
Industry
Number
of ?rms Percentage
COMPUSTAT
percentage
1. Agriculture, forestry, and ?shing 1 0.13 0.36
2. Mining 30 3.89 4.12
3. Construction 7 0.91 0.98
4. Manufacturing 302 39.17 36.31
5. Transportation and utilities 85 11.02 10.02
6. Wholesale 20 2.59 3.11
7. Retail 31 4.02 5.26
8. Financial services 189 24.51 18.26
9. Services 103 13.36 20.50
10. Other 3 0.39 1.07
Total 771 100.00 100.00
The industry classi?cation is based on Dopuch et al. (1987), and includes the following SIC codes:
Agriculture, forestry, and ?shing 100-999
Mining 1,000-1,499
Construction 1,500-1,999
Manufacturing 2000-3,999
Transportation and utilities 4,000-4,999
Wholesale 5,000-5,199
Retail 5,200-5,999
Financial services 6,000-6,999
Services 7,000-8,999
Others ,100 and .8,999
Panel C: sample distribution across auditors and years
Year Andersen
Other
Big 4
Non-
Big 5 Total
2000 170 541 60 771
2001 175 532 64 771
2002 175
b
532 64 771
Total 520 1,605 188 2,313
Notes:
a
For 771 ?rms;
b
in 2002, these clients had switched to other auditors
Table II.
Sample selection
Enron,
Andersen, and
Sarbanes-Oxley
11
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
0
7

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
0.070 percent in 2000 to 0.107 percent in 2002. As shown in Panels A through C these
increases differed by auditor and by year[10].
While our focus was audit fees, we did examine the comparable increases in total
fees. In Panel D, we show that total fees as a percent of assets increased from
0.163 percent in 2000 to 0.174 in 2002. This indicates that while total fees are increasing
the amount spent on non-audit fees are decreasing, that is, from 0.093 percent of
assets in 2000 to 0.067 percent in 2002. As a percent of total fees (Panel E),
non-audit fees decrease from 52 percent in 2000 to 33 percent in 2002. The decrease
in non-audit fees in 2002 is consistent with the Sarbanes-Oxley prohibition on many
non-audit services[11].
4. Model
The descriptive statistics in Table III, Panel B show that increases in audit fees far
exceeded the rate of in?ation over the period under examination. In this section, we
examine the determinants of the level and changes in audit fees. Our goal is to examine
to what extent audit fees and changes in audit fees are attributable to changes in the
audit market. Consequently, our model needs to control for determinants of audit fees
found in the literature. These determinants include proxies for the size of the client, as
well as the clients’ complexity and riskiness to the auditor.
We begin with a pooled analysis of audit fees for the years 2000-2002 to examine the
effect of the events on the level of audit fees (H1), Andersen’s ability to charge a
premium (H2), whether Andersen clients pay a premium when they are forced to
Sub-samples by auditor
Full sample Andersen (AA) Other Big 4 (OB4) Non-Big 5 (NB5)
Year Obs. Mean Obs. Mean Obs. Mean Obs. Mean
Panel A: mean audit Fee ( in 000 ) ( N ¼ 2,313)
2000 771 704.46 170 1,071.32 541 652.86 60 130.32
2001 771 707.02 175 864.03 532 723.30 64 142.63
2002 771 1,095.52 175
a
1,296.65 532 1,134.80 64 196.81
Panel B: mean percentage increase in audit fee ( N ¼ 1,542 )
2001 771 6.47% 175 4.39% 532 7.09% 64 6.92%
2002 771 21.04% 175
a
15.43% 532 23.28% 64 17.98%
Panel C: mean audit fee per $1,000 of total assets ( N ¼ 2,313)
2000 771 0.6981 170 0.6763 541 0.6961 60 0.7772
2001 771 0.7857 175 0.7097 532 0.7909 64 0.9510
2002 771 1.0655 175
a
0.9592 532 1.0852 64 1.2001
Panel D: mean total fee per $1,000 of total assets ( N ¼ 2,313)
2000 771 1.6272 170 1.7311 541 1.6299 60 1.3090
2001 771 1.6358 175 1.5819 532 1.6764 64 1.4461
2002 771 1.7426 175
a
1.3611 532 1.8625 64 1.8112
Panel E: ratio of non-audit fee to total fee ( N ¼ 2,313)
2000 771 0.5223 170 0.5382 541 0.5279 60 0.4269
2001 771 0.5003 175 0.5092 532 0.5082 64 0.4101
2002 771 0.3275 175
a
0.2170 532 0.3595 64 0.3707
Notes:
a
In 2002, these Andersen clients had switched to other auditors; audit fee includes audit-related
fee; in Panel B, data for year 2000 is lost in differencing; all means are signi?cantly greater than zero at
5 percent level or better
Table III.
Distribution of audit fee
by auditor and year
ARJ
22,1
12
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
0
7

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
switch to new auditors in 2002 (H3), and whether the Big 4 premium increased
(H4)[12],[13],[14].We use the following model, which includes indicator variables for
Andersen (AA), the remaining Big 4 auditors (OB4), and year (D2000, D2001, and
D2002), as well as controls for the size, complexity and risk of clients, and other factors
which may in?uence audit fees:
LFEE
it
¼a
0
þ a
1
D2001 þ a
2
D2002 þ a
3
AA
*
D2000 þ a
4
OB4
*
D2000
þ a
5
AA
*
D2001 þ a
6
OB4
*
D2001 þ a
7
AA
*
D2002 þ a
8
OB4
*
D2002
þ a
9
LSIZE
it
þ a
10
RISK
it
þ a
11
COMPLEX
it
þ a
12
MSHARE
it
þ a
13
DOMINATE
it
þ a
14
INSTOWN
it
þ a
15
INSOWN
it
þ1
ð1Þ
where: i, client ?rm; t, year; LFEE, natural logarithm of audit plus audit-related
fees[15]; D2000/1/2, indicator variable taking the value of 1 for year 2000-2002,
respectively; 0 otherwise; AA, an indicator variable taking the value of 1 for clients of
Andersen in 2000 and 2001, or former Andersen clients that have switched to the Big 4
in 2002; and zero otherwise; OB4, indicator variable taking the value of 1 for clients of
one of the Big 4 (excluding ?rst time former Andersen clients in 2002) and zero
otherwise; LSIZE, natural logarithm of total assets; RISK, risk variable obtained froma
principal component analysis of the following nine variables:
.
LOSS. An indicator variable taking the value of 1 if net income is negative and
zero otherwise (Fields et al., 2004).
.
QOP. An indicator variable taking the value of 1 if audit opinion is a departure
from the standard unquali?ed report and zero otherwise (Abbott et al., 2003).
.
CA. Current assets de?ated by total assets (Geiger and Rama, 2003).
.
AR. Accounts receivable de?ated by total assets (Carcello et al., 2002; Willikens
and Achmadi, 2003).
.
INV. Total inventory de?ated by total assets (Carcello et al., 2002; Willikens and
Achmadi, 2003).
.
DAR. Total liabilities de?ated by total assets (Mayhewand Wilkins, 2003; Geiger
and Rama, 2003).
.
ROE. Net income de?ated by book value of equity (Geiger and Rama, 2003).
.
STDDEV. Standard deviation of daily stock returns for year t (Mayhew and
Wilkins, 2003; Fields et al., 2004).
.
IAR. Industry adjusted stock returns for year t (Geiger and Rama, 2003).
.
COMPLEX. Complexity variable obtained from a factor analysis of the following
four variables:
.
SEG. Number of industry segments (Mayhew and Wilkins, 2003).
.
SUB. Number of subsidiaries (Francis and Simon, 1987; Geiger and Rama, 2003).
.
FOR. Indicator variable taking the value of 1 if the client had foreign operations
and zero otherwise (Francis and Simon, 1987; Mayhew and Wilkins, 2003; Geiger
and Rama, 2003).
.
MERGER. Indicator variable taking the value of 1 if the client engaged in merger
activity during year t and zero otherwise (Geiger and Rama, 2003).
Enron,
Andersen, and
Sarbanes-Oxley
13
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
0
7

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
.
MSHARE. Ratio of sum of square root of assets of auditor’s clients to sum of
square root of assets of all ?rms in two-digit SIC code (Mayhew and Wilkins,
2003).
.
DOMINATE. Indicator variable taking the value of 1 if the client’s auditor is
dominant in the client’s industry, where dominance is de?ned as having a market
share at least ten percentage points greater than that of its next highest
competitor (Mayhew and Wilkins, 2003).
.
INSTOWN. Percentage of shares owned by institutions.
.
INSOWN. Percentage of shares owned by insiders.
If the events under examination had the effect of increasing audit fees, our expectation
(H1) is that the coef?cients on D2001 and/or D2002 will exceed zero (in our formulation
the coef?cient on 2000 is incorporated in the intercept). Our expectation is that in 2000
there will not be any difference between AA and OB4, that is, a
3
¼ a
4
, but in 2001 we
expect (H2) the coef?cient on AA will be less than that on OB4, that is, a
5
, a
6
. In all
three years we expect to ?nd the coef?cient on OB4 to be greater than 0, that is, a
4
, a
6
,
a
8
. 0, as we expect there will be premium paid to Big 4 auditors. If this premium
increases because of increasing pricing power by the Big 4, that is, because of
Andersen’s troubles and demise, we will see a
8
. a
6
. a
4
(H4). In 2002, there will be no
Andersen clients, therefore we use the interaction between AA and D2002 to examine if
former Andersen clients were treated differentially by their new auditors[16]. If
Andersen clients were at a disadvantage in the negotiation process because of their need
to ?nd a newauditor in the newenvironment we would expect to a
7
. a
8
. Alternatively,
if the remaining Big 4 auditors competed for the Andersen clients, we would see a
7
, a
8
,
a scenario consistent with the DeAngelo (1981) low-balling scenario.
In addition to the test variables, we include variables to control for other factors
found by prior literature to in?uence audit fees. We include log of assets (LSIZE), as the
audit fee will increase with the size of the ?rm (Mayhewand Wilkins, 2003; Fields et al.,
2004; Geiger and Rama, 2003). We include COMPLEX (obtained from a principal
component analysis of SEG, SUB, FOR, and MERGER) and RISK (obtained from a
principal component analysis of LOSS, QOP, CA, AR, INV, DAR, ROE, STDDEV, and
IAR) which we expect will be positively associated with the audit fee. We include the
industry market share of the auditor, MSHARE and auditor industry dominance,
DOMINATE (Mayhew and Wilkins, 2003) as specialization may affect the audit fee.
Finally, we include the percentage of shares owned by institutions, INSTOWN, and the
percentage of shares owned by insiders, INSOWN (Geiger and Rama, 2003) to control
for other alternative sources of monitoring of the ?rm, which may affect audit fees. Our
variables are further described in Table IV.
We then move on to a changes analysis, to better examine the effect of the events
under examination on changes in audit fees and to test the effects of client size (H5) and
risk (H6) on that change. In this analysis, we de?ne the dependent variable as the
change in audit fee, or to be more precise, the change in the log of audit fee. To
correspond to our dependent variable, we include the control variables from above
(i.e. size, complexity, risk, etc.) in the change form. To examine the effect of changes in
the audit market, we include variables to identify continuing Andersen clients in 2001
(AA
*
D2001), continuing clients of the remaining Big 4 auditors (OB4
*
D2001), and an
indicator variables for 2002 (D2002). As above, former Andersen clients switching to
ARJ
22,1
14
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
0
7

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
OB4 are still coded as AA in 2002 so the coef?cient on the interaction between AA and
D2002 represents the change in audit fee experienced by Andersen clients who retain a
Big 4 auditor in 2002[17]. Implicitly, the intercept controls for the average increase in
2001 for clients of non-Big 4 auditors as well as other omitted variables.
To test H5 and H6, we include the levels of size and risk, as well as their interactions
with D2002. We interact them with D2002 to allow differential effects of size and risk
on the change in audit fees in 2001 and 2002. That is, the coef?cients on LSIZE and
RISK measure the effects of those variables on the change in audit fees in 2001, while
the interactions measure the incremental effect of those variables on the change in
audit fees in 2002 above that in 2001.
Variable De?nition
AA Indicator variable taking the value of 1 for Andersen clients in 2000 and 2001 or
former Andersen clients that have switched to other Big 4 in 2002; 0 otherwise
AR Accounts receivable divided by total assets at end of ?scal year
CA Current assets divided by total assets at end of ?scal year
COMPLEX Measure of client complexity obtained from component analysis of FOR, MERGER,
SEG, and SUB
D2000/1/2 Indicator variables taking the value of 1 for year 2000-2002 observations,
respectively; 0 otherwise
DAR Debt to asset ratio at the end of ?scal year
DOMINATE Indicator variable taking the value of 1 if the client’s auditor is dominant in the
client’s industry (audit ?rm market share at least 10 percent higher than next market
share); 0 otherwise
FOR Indicator variable taking the value of 1 if the client has foreign subsidiaries; 0
otherwise
IAR Industry adjusted stock return during ?scal year
INSOWN Percentage of shares held by insiders at close of ?scal year
INSTOWN Percentage of shares held by institutions at close of ?scal year
INV Total inventory divided by total assets at end of ?scal year
LAFEE Natural logarithm of audit and audit-related fee (in thousands)
LOSS Indicator variable taking the value of 1 if net income ,0; 0 otherwise
LSIZE Natural logarithm of size (total assets in millions) at the end of ?scal year
MERGER Indicator variable taking the value of 1 if the client engaged in a mergers/acquisitions
in the current year; 0 otherwise
MSHARE Audit ?rm market share (based on square root of clients assets in two-digit SIC code)
NAF Non-audit fee divided by total fee at the end of previous ?scal year
OB4 Indicator variable taking the value of 1 if the auditor is one of the Big 4 and the ?rm
is not a former Andersen client; 0 otherwise
QOP Indicator variable taking the value of 1 if the audit opinion is a departure from the
standard unquali?ed report; 0 otherwise
RISK Measure of client risk obtained from a component analysis of AR, CA, DAR, IAR,
INV, LOSS, QOP, ROE, and STDDEV
ROE Return on equity during ?scal year
SEG Number of industry segments
SUB Number of subsidiaries
STDDEV Standard deviation of daily stock returns during ?scal year
X When suf?xed to a variable, implies value for current year minus value for previous
year
Note: In alphabetical order
Table IV.
Variable de?nitions
Enron,
Andersen, and
Sarbanes-Oxley
15
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
0
7

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
Our formal model is:
XLFEE
it
¼ a
0
þ a
1
D2002 þ a
2
AA
*
D2001 þ a
3
OB4
*
D2001 þ a
4
AA
*
D2002
þa
5
OB4
*
D2002 þ a
6
XLSIZE
it
þ a
7
XRISK
it
þ a
8
XCOMPLEX
it
þ a
9
XMSHARE
it
þ a
10
XDOMINATE
it
þ a
11
XINSTOWN
it
þ a
12
XINSOWN
it
þ a
13
LSIZE
it
þ a
14
LSIZE
it
*
D2002 þ a
15
RISK
it
þ a
16
RISK
it
*
D2002 þ a
17
NAF
it21
þ 1
ð2Þ
where NAF is the lagged value of the ratio of non-audit to total fees and is included to
control for the incentive[11], to shift non-audit to audit fees[18]. All other variables are
as de?ned above and X ¼ indicates the change version of the variable.
A positive coef?cient on OB4
*
D2001 and OB4
*
D2002 is consistent with audit fees
increasing faster for clients of the Big 4 audit ?rms than the non-Big 4 ?rms (H4), while
a positive coef?cient on D2002 would be consistent with the increase in audit fees being
greater in 2002 than in 2001. If former Andersen clients experienced a greater increase
in 2002, we will observe the coef?cient on AA
*
D2002 to be greater than that on
OB4
*
D2002, that is, a
4
. a
5
(H3). Positive coef?cients on LSIZE and LSIZE
*
D2002
(H5), are consistent with larger ?rms experiencing greater increases in audit fees in
2001 and 2002, respectively. Similarly, positive coef?cients on RISK and RISK
*
D2002
(H6), are consistent with more risky ?rms experiencing greater increases in audit fees.
5. Results
Panel A of Table V provides descriptive statistics for the variables used in our model.
We see that client ?rms are fairly large, with mean (median) assets of $5.7 (0.7) billion.
This is comparable to the mean of all ?rms on Compustat which is $5.6 billion.
RISK and COMPLEX, which are the factors associated with the risk and complexity of
the audit, by construction have a mean of zero and median of 1. MSHARE has a mean
of almost 36 percent, indicating the average auditor has a high degree of specialization
in his industry and DOMINATE has a mean of almost 33 percent, indicating that there
is a dominant auditor in one of three industries. We also observe that our sample ?rms
have higher institutional (INSTOWN) and lower insider (INSOWN) ownership than the
average ?rm on Compact Disclosure (which is consistent with our sample ?rms being
larger than the average ?rm on Compact Disclosure).
Panel B looks at the distribution of these variables by auditor type. As expected,
total assets for clients of Andersen and the Big 4 ?rms are much larger than those of
non-Big 4. Mean LAFEE, DLAFEE, RISK, MSHARE and DOMINATE for Andersen
are less than those for other Big 4. On the other hand, mean COMPLEX and INSTOWN
for Andersen are larger than those for other Big 4, providing some descriptive evidence
that there was a difference between clients of Andersen and those of the other Big 4.
INSOWN does not appear to differ signi?cantly by auditor type.
We present correlation coef?cients between the independent variables in equations
(1) and (2) in Table VI. Panel A contains the levels variables (equation (1)) and Panel B
contains the changes variables (equation (2)). None of the correlations are high (the
ARJ
22,1
16
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
0
7

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
magnitude of all the coef?cients are under 0.25). Thus, multicollinearity is not a
concern. However, as an additional precaution, we report the variance in?ation factors
(VIF) in Tables VII and VIII. The maximum VIF in either table is less than 10,
con?rming that multicollinearity is not a problem.
Table VII contains the results from estimating equation (1). The model is signi?cant
with an adjusted R
2
of over 83 percent. Looking at the coef?cients of interest, we see
that the coef?cient on D2002 is positive and signi?cantly different from zero, and
signi?cantly different from the coef?cient on D2001 (which is not different from zero),
providing evidence that audit fees increased in 2002 which is consistent with our
H1[19]. Also supporting this hypothesis, we ?nd the coef?cients on AA
*
D2002 and
OB4
*
D2002 to be statistically greater than their 2000 and 2001 counterparts[20].
Consistent with prior research, we also ?nd evidence of a Big 4 premium as the
coef?cients on OB4
*
D2000, OB4
*
D2001, and OB4
*
D2002, are all positive and
signi?cant. The coef?cients on AA
*
D2000 and AA
*
D2001 are also positive and
signi?cant indicating that Andersen was also able to charge audit fee premium.
In 2000, there is no difference between the Andersen and Big 4 premium. However,
consistent with our H2, we ?nd Andersen’s premium statistically lower than the
premium charged by the Big 4 in 2001.
The evidence on our H3 is somewhat mixed. The coef?cient on AA
*
D2002 is
statistically greater than the coef?cient on AA
*
D2001 indicating that when former
Andersen clients switched to a Big 4 auditor their fees went up by a statistically
Panel A Panel B
Variable Mean Median SD Andersen Other Big 4 Non-Big 5
Total Assets (in billion dollars) 5.7002 0.7145 20.9690 6.5748 6.5163 0.8254
LAFEE 5.8708 5.6937 1.1863 5.9107 5.9845 4.7145
XLAFEE
a
0.2068 0.1469 0.3670 0.0819 0.2286 0.0166
RISK 0 0.0525 1 20.0909 0.0159 0.0161
AR 0.2413 0.1552 0.2256 0.1961 0.2267 0.4630
CA 0.4506 0.4399 0.2367 0.4066 0.4583 0.4904
DAR 0.5948 0.6115 0.2769 0.5819 0.5878 0.6853
IAR 20.0447 20.1085 0.5933 20.0081 20.0513 20.0488
INV 0.0846 0.0321 0.1207 0.0855 0.0877 0.0527
LOSS 0.2356 0 0.4245 0.2348 0.2471 0.1283
QOP 0.2841 0 0.4511 0.2087 0.3200 0.0802
ROE 20.0141 0.1024 3.7188 0.0122 20.0251 0.0422
STDDEV 0.0357 0.0299 0.0185 0.0367 0.0360 0.0312
COMPLEX 0 20.0013 1 0.2111 0.0240 20.6152
FOR 0.2918 0 0.4547 0.3217 0.3088 0.0749
MERGER 0.1972 0 0.3979 0.2928 0.1915 0.0749
SEG 1.6308 1 1.5416 1.6435 1.6693 1.2406
SUB 32.8176 10 85.8484 49.0290 32.3549 7.3155
MSHARE (%) 35.66 26.33 29.21 30.88 37.95 5.41
DOMINATE (%) 32.82 0 46.97 28.70 35.09 0
INSTOWN (%) 43.14 45.57 25.55 51.16 49.97 21.90
INSOWN (%) 10.85 4.32 15.23 9.66 9.78 10.29
Notes: N ¼ 2,313; panel A, distribution of variables; panel B, distribution of variables across auditors;
see Table IV for variable de?nitions;
a
N ¼ 1,542, since 2000 data are lost in differencing
Table V.
Descriptive statistics
Enron,
Andersen, and
Sarbanes-Oxley
17
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
0
7

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
P
a
n
e
l
A
:
f
o
r
l
e
v
e
l
s
V
a
r
i
a
b
l
e
s
R
I
S
K
C
O
M
P
L
E
X
M
S
H
A
R
E
D
O
M
I
N
A
T
E
I
N
S
T
O
W
N
I
N
S
O
W
N
L
S
I
Z
E
0
.
1
4
1
6
(
,
0
.
0
0
0
1
)
0
.
2
3
7
9
(
,
0
.
0
0
0
1
)
0
.
2
4
6
6
(
,
0
.
0
0
0
1
)
0
.
1
7
7
3
(
,
0
.
0
0
0
1
)
0
.
2
0
3
2
(
,
0
.
0
0
0
1
)
2
0
.
2
1
5
7
(
,
0
.
0
0
0
1
)
R
I
S
K
0
.
1
0
5
8
(
,
0
.
0
0
0
1
)
0
.
0
2
6
4
(
,
0
.
0
0
0
1
)
0
.
0
2
2
7
(
,
0
.
0
0
0
1
)
0
.
1
3
6
9
(
,
0
.
0
0
0
1
)
2
0
.
0
5
8
5
(
,
0
.
0
0
0
1
)
C
O
M
P
L
E
X
2
0
.
1
1
2
9
(
,
0
.
0
0
0
1
)
2
0
.
0
4
7
0
(
,
0
.
0
0
0
1
)
0
.
2
4
9
8
(
,
0
.
0
0
0
1
)
2
0
.
1
5
0
1
(
,
0
.
0
0
0
1
)
M
S
H
A
R
E
0
.
1
7
9
2
(
,
0
.
0
0
0
1
)
2
0
.
0
8
3
2
(
,
0
.
0
0
0
1
)
2
0
.
0
8
8
5
(
,
0
.
0
0
0
1
)
D
O
M
I
N
A
T
E
2
0
.
0
3
9
5
(
,
0
.
0
0
0
1
)
2
0
.
0
6
4
1
(
,
0
.
0
0
0
1
)
I
N
S
T
O
W
N
2
0
.
1
6
5
2
(
,
0
.
0
0
0
1
)
P
a
n
e
l
B
:
f
o
r
c
h
a
n
g
e
s
V
a
r
i
a
b
l
e
s
X
R
I
S
K
X
C
O
M
P
L
E
X
X
M
S
H
A
R
E
X
D
O
M
I
N
A
T
E
X
I
N
S
T
O
W
N
X
I
N
S
O
W
N
X
L
S
I
Z
E
0
.
1
5
3
8
(
,
0
.
0
0
0
1
)
0
.
2
2
0
0
(
,
0
.
0
0
0
1
)
0
.
2
0
1
9
(
,
0
.
0
0
0
1
)
0
.
1
9
6
7
(
,
0
.
0
0
0
1
)
0
.
1
9
0
6
(
,
0
.
0
0
0
1
)
2
0
.
2
2
6
2
(
,
0
.
0
0
0
1
)
X
R
I
S
K
0
.
0
8
1
8
(
,
0
.
0
0
0
1
)
0
.
0
1
7
9
(
2
0
.
0
1
3
5
)
0
.
0
0
9
9
(
2
0
.
1
7
2
8
)
0
.
1
2
6
9
(
,
0
.
0
0
0
1
)
2
0
.
0
5
1
4
(
,
0
.
0
0
0
1
)
X
C
O
M
P
L
E
X
2
0
.
0
5
4
5
(
,
0
.
0
0
0
1
)
2
0
.
0
1
2
8
(
2
0
.
0
7
8
1
)
0
.
1
7
9
7
(
,
0
.
0
0
0
1
)
2
0
.
1
0
4
0
(
,
0
.
0
0
0
1
)
X
M
S
H
A
R
E
0
.
1
7
1
6
(
,
0
.
0
0
0
1
)
2
0
.
0
3
8
9
(
,
0
.
0
0
0
1
)
2
0
.
0
7
8
9
(
,
0
.
0
0
0
1
)
X
D
O
M
I
N
A
T
E
2
0
.
0
1
0
2
(
2
0
.
1
5
7
6
)
2
0
.
0
3
8
5
(
,
0
.
0
0
0
1
)
X
I
N
S
T
O
W
N
2
0
.
2
2
1
5
(
,
0
.
0
0
0
1
)
N
o
t
e
s
:
S
e
e
T
a
b
l
e
I
V
f
o
r
v
a
r
i
a
b
l
e
d
e
?
n
i
t
i
o
n
s
;
p
-
v
a
l
u
e
s
a
r
e
i
n
p
a
r
e
n
t
h
e
s
e
s
Table VI.
Correlation matrix among
independent variables
ARJ
22,1
18
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
0
7

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
signi?cant amount. However, the coef?cient on AA
*
D2002 is statistically lower than
that on OB4
*
D2002, which is consistent with low balling.
Providing support for our H4, we ?nd the coef?cient on OB4 to be signi?cantly
greater in 2002 than it was in either 2000 or 2001. Similarly the coef?cient on AA was
higher in 2002 than it was in 2001, but the difference between 2002 and 2000 is not
signi?cantly different.
Brie?y looking at the control variables, we ?nd, as expected, audit fees increase
with the size and risk of the auditee, as well as the complexity of the audit, that is,
LSIZE, RISK, and COMPLEX are all positive and signi?cant. The coef?cient on auditor
market share (MSHARE) is negative and signi?cant, while that on auditor dominance
(DOMINATE) is signi?cantly greater than zero. The negative coef?cient on MSHARE
is consistent with the evidence in Mayhew and Wilkins (2003) that auditors pass along
ef?ciencies in the form of lower audit fees. In contrast, the positive coef?cient on
DOMINATE indicates that once an auditor becomes dominant in an industry, he uses
that power to raise fees, that is, extract rents from clients. Finally, turning to the
ownership variables, we ?nd that the coef?cient on institutional ownership
(INSTOWN) is positive and signi?cant, while that on insider ownership (INSOWN)
is insigni?cantly different from zero. The positive coef?cient on institutional
Dependent variable LAFEE
Independent variables Estimate t-statistics
Intercept 2.5658
* *
35.11
D2001 0.0417 0.49
D2002 0.2038
*
2.38
AA
*
D2000 0.2611
* *
3.62
OB4
*
D2000 0.2464
* *
3.76
AA
*
D2001 0.1507
*
2.13
OB4
*
D2001 0.2664
* *
4.16
AA
*
D2002 0.3250
* *
4.49
OB4
*
D2002 0.4523
* *
6.95
LSIZE 0.4602
* *
66.55
RISK 0.0588
* *
4.44
COMPLEX 0.2866
* *
26.01
MSHARE 21.2723
* *
221.31
DOMINATE 0.1847
* *
5.45
INSTOWN 0.0034
* *
6.66
INSOWN 0.0013 1.54
Observations 2,313
Adjusted R
2
(%) 83.44
F-value 737.85
Prob. . F 0.0001
White’s x
2
176.59
Prob. . x
2
0.0001
Highest VIF 7.9035
In?uential outliers 49
Notes: Signi?cant at
*
5 and
* *
1 percent levels, respectively; see Table IV for variable de?nitions;
N ¼ 2,313
Table VII.
Regression of audit fee
level on ?rm and auditor
characteristics
Enron,
Andersen, and
Sarbanes-Oxley
19
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
0
7

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
ownership is consistent with institutional shareholders demanding a higher level of
monitoring via the audit.
Table VIII contains the results fromestimating equation (2). The model is signi?cant;
the adjusted R
2
is about 32 percent. Looking at the coef?cients of interest, we see that the
coef?cient on D2002 is negative and signi?cantly different fromzero, consistent with the
rate of increase in 2001 (incorporated in the intercept) exceeding that of 2002 for clients of
non-Big 4 auditors. While we ?nd that although the incremental increase in audit fee for
both Andersen clients and clients of the Big 4 in 2001 are insigni?cantly different from
zero, we ?nd the incremental increase for Andersen clients is signi?cantly lower than the
increase for clients of the Big 4 audit ?rms, that is, a
2
, a
4
, which provides support for
H2. In 2002, former Andersen clients that remained with a Big 4 auditor did see an
increase, that is, the coef?cient on the interaction between AAand D2002 is positive and
signi?cant, however, that increase that was less than the increase experienced by
existing clients of the Big 4, that is, a
3
, a
5
. This latter ?nding is consistent with
competition among the Big 4 for former Andersen clients, that is, low balling (H3). The
positive and signi?cant values for AA
*
D2002 and OB4
*
D2002 are consistent with the
Big 4 premium increasing in 2002 (H4).
Dependent variable XLAFEE
Independent variables Estimate t-statistics
Intercept 0.0828 0.62
D2002 23.4375
* *
217.85
AA
*
D2001 20.0387 20.34
OB4
*
D2001 20.0015 20.01
AA
*
D2002 0.8210
* *
6.79
OB4
*
D2002 0.9314
* *
8.46
XLSIZE 0.0390 0.41
XRISK 20.0439 21.14
XCOMPLEX 0.0068 0.22
XMSHARE 0.1262 0.44
XDOMINATE 0.0185 0.25
XINSTOWN 0.0011 0.53
XINSOWN 20.0006 20.23
LSIZE 20.0161 21.03
LSIZE
*
D2002 0.3745
* *
16.99
RISK 20.0025 20.07
RISK
*
D2002 0.1122
*
2.09
NAF 0.2398
*
2.37
Observations 1,542
a
Adjusted R
2
(%) 31.96
F-value 40.57
Prob. . F 0.0001
White’s x
2
426.27
Prob. . x
2
0.0001
Highest VIF 6.3199
In?uential outliers 37
Notes: Signi?cant at
*
5 and
* *
1 percent levels, respectively; see Table IV for variable de?nitions;
a
year 2000 data is lost in differencing
Table VIII.
Regression of audit fee
change on ?rm and
auditor characteristics
ARJ
22,1
20
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
0
7

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
Turning now to the coef?cients on XLSIZE and XRISK, we see that the coef?cients
on both are insigni?cantly different from zero. However, in 2002, the interactions of
LSIZE and RISK with D2002 are both positive and signi?cant. Consequently,
consistent with our H5, we ?nd that in 2002 even after controlling for other
determinants of the change in audit fees, including the change in ?rm size, larger ?rms
experienced greater increases in audit fees, perhaps because they have fewer
alternatives in an increasingly concentrated market. Consistent with our H6, we ?nd
that, in this new environment, riskier ?rms are experienced larger increases in audit
fees in 2002. Finally, we ?nd that the coef?cient on NAF is positive and signi?cant,
consistent with clients and their audit ?rms shifting non-audit fees to audit fees to
minimize their political impact.
We have not yet shown whether the increase in level of fees is greater than the audit
fee growth that would have occurred regardless of these events. To do this, we ?rst
estimate equation (2) for 2001. Then we use these estimates to predict the change in
audit fee for 2002. This would be the increase in audit fee if the “trend” continued
unaltered. Then we estimate equation (2) for 2002 and compare the predicted change in
audit fee from this regression with the “trend” value estimated before. The mean
difference of 0.1537 is signi?cant at 1 percent level (t-Statistic ¼ 5.0302,
p-value ¼ ,0.0001). Thus, the special circumstances discussed in the paper resulted
in an increase of over 15 percent in audit fee over and above the normal trend.
5.1 Regression diagnostics
The Belsley et al. (1980) test for multicollinearity is conducted on the regressions
reported in Tables VII and VIII. The highest VIF is below the critical level of 10. Hence,
multicollinearity does not appear to be driving our results. White’s (1980) test for
heteroskedasticity is conducted on the regressions in Tables VII and VIII. The null of
homoskedastic errors is rejected for both regressions. Heteroskedasticity corrected
t-statistics (not reported) do not alter any of the conclusions. Diagnostic tests are also
conducted for all regressions with in?uential outliers, which are 2.1 (2.4) percent of
observations in Table VIII, being removed based on the procedure outlined in
Belsley et al. (1980).
6. Conclusion
The above results, which show audit fee increases post-Enron, Andersen and
Sarbanes-Oxley are consistent with rent extraction on the part of the Big 4 audit ?rms.
As discussed above, with the demise of Andersen, large multinational corporations
saw the number of potential audit providers’ drop from ?ve to four. Additionally those
four were inundated with former Andersen clients beginning in the spring of 2002. The
lack of choice for existing clients means that auditors were able to push through price
increases as audit fees were negotiated for 2001 and 2002 audits.
Non-Big 4 ?rms had less ability to extract rents than the Big 4 auditors, as small
?rms had many more choices of audit ?rm. Thus, while the fees they charged increased
in 2002, the premium charged by the Big 4 over the non-Big 4 ?rms also increased.
Consequently, it appears that the additional concentration in the audit market for large
?rms brought on by the demise of Andersen had adverse consequences on the ?rms
purchasing audit services in that segment of the market.
Enron,
Andersen, and
Sarbanes-Oxley
21
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
0
7

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
Consistent with that theory, we observe that larger ?rms, which are more likely to
require the services of a Big 4 auditor, experienced a greater increase in audit fees. Also
consistent with our theory, we ?nd evidence suggesting that audit ?rms pay more
attention to risk post-Enron, Andersen, and Sarbanes-Oxley. That is, audit fees paid by
riskier clients increased at a greater rate.
Somewhat surprisingly, we ?nd evidence of a competitive market for former
Andersen clients. That is, we ?nd some evidence consistent with low balling on the
part of the new auditor and no evidence that would indicate audit ?rms were able to
take advantage of former Andersen clients.
Notes
1. For example, Section 104 of Sarbanes-Oxley requires annual quality reviews (inspections) of
?rms that audit more than 100 issuers, with other ?rms being reviewed every three years.
2. The timing of these events, in terms of their closeness in proximity, limits our ability to test
and make any inferences on each individually.
3. To the extent that not all of the Big 5 had of?ces in every city, the number of competitors
could have been less than ?ve.
4. Section 404 went into affect after our sample period (?scal years ending on or after
September 15, 2003), but may have had an effect on audit fees if clients and/or their auditors
started preparing for it earlier.
5. On average, over the period from 1991 to 2000, 368 clients switched from a Big 6/5 auditor
(this number was extracted from Compustat) in a given year. In contrast in 2002, 937 large
public Andersen clients had to ?nd new auditors (Asthana et al., 2009).
6. Some of the non-Andersen switches may be involuntary, as the previous auditor may have
resigned. Empirically resignations are not very common (Asthana et al., 2009).
7. Assuming their former auditor has not resigned.
8. While this requirement will lead to survivorship bias, we do not believe it will be severe as
our sample period is relatively short.
9. Both increases are signi?cantly different from zero.
10. We also examined the rate of increase for the individual Big 4 audit ?rms but do not report
them separately, as they are not signi?cantly different from each other in either 2001 or 2002.
11. In an environment where there were high political costs for reporting non-audit fees, clients
and their auditors had the incentive to shift fees from the non-audit to audit category.
This incentive was greatest for clients with the highest non-audit fees, who were primarily
clients of the Big 4. To control for this effect on our results, in our change model, we include
the lag ratio of non audit to total fees as an additional variable. The coef?cient on this
variable is positive and signi?cant as would be expected.
12. Given the disclosures relating to Enron, the demise of Andersen, and much of the discussion
leading up to the enactment of Sarbanes-Oxley occur near the end of 2001 and early in 2002,
we conduct sensitivity analyses excluding 2001 ?scal year ends. The results are identical to
those reported below except that we are unable to test the effect of the disclosures related to
Enron on Andersen’s ability to charge a premium (H2) in 2001. Separately, because a client
with for example, a June ?scal year end may not be affected in 2001 because their ?nancial
statements were issued before the events described above, we conduct an additional
sensitivity analysis using only December 31 ?scal year ends. The results reported below are
unaffected.
ARJ
22,1
22
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
0
7

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
13. We actually use log of audit fee to be consistent with the previous literature, for example,
Fields et al. (2004), Abbott et al. (2003) and Mayhew and Wilkins (2003).
14. To the extent that the disclosures pertaining to Enron affected the perceived quality of the
Big 4 auditors it too may have affected their ability to charge a premium.
15. Our conclusions are identical if we use the audit fee alone.
16. We do sensitivity analyses including additional variables to examine the effect of
non-Andersen changes within the Big 4 on audit fees. Since they do not affect the results
reported below (this may be because there were a relatively small number of changes,
17 changes within the Big 4 in 2001 and 11 in 2002) for brevity, we elect not to include those
variables in our model.
17. While a number of Andersen clients did switch to non-Big 4 auditors (Asthana et al., 2009) all
of the Andersen clients in our sample retained Big 4 auditors in 2002.
18. In an alternative formulation, we include both NAF and NAF
*
D2002 to allow the effect to
vary between the two years. Including the additional variable does not alter our conclusions.
19. x
2
-tests are conducted to compare coef?cients but are not reported for the sake of brevity.
20. The latter ?nding is not only consistent with audit fees increasing, but with the Big 4
premium also increasing over the sample period.
References
Abbott, L.J., Parker, S., Peters, G.F. and Raghunandan, K. (2003), “The association between audit
committee characteristics and audit fees”, Auditing: AJournal of Practice &Theory, Vol. 22
No. 2, pp. 17-32.
Abidin, S., Beattie, V. and Goodacre, A. (2008), “Audit market structure, fees and choice following
the Andersen break-up: evidence from the UK”, working paper, University of Glasgow,
Glasgow.
Akerlof, G.A. (1970), “The market for lemons”, Quarterly Journal of Economics, Vol. 84 No. 3,
pp. 488-500.
Asthana, S., Balsam, S. and Krishnan, J. (2004), “Audit ?rm reputation and client stock price
reactions: evidence from the Enron experience”, working paper, Temple University,
Philadelphia, PA.
Asthana, S., Balsam, S. and Krishnan, J. (2009), “Audit ?rm reputation, auditor switches, and
client stock price reactions: the Andersen experience”, working paper, Temple University,
Philadelphia, PA.
Balsam, S. (2003), “The effect of Andersen’s demise on earnings management”, working paper,
Fox School of Business and Management, Temple University, Philadelphia, PA.
Belsley, D.A., Kuh, E. and Welsch, R.E. (1980), Regression Diagnostics: Identifying In?uential
Data and Sources of Collinearity, Wiley, New York, NY.
Blouin, J., Grein, B.M. and Rountree, B.R. (2007), “An analysis of forced auditor change: the case
of former Arthur Andersen clients”, The Accounting Review, Vol. 82 No. 3, pp. 621-50.
Cahan, S.F. and Zhang, W. (2006), “After Enron: auditor conservatism and ex-Andersen clients”,
The Accounting Review, Vol. 81 No. 1, pp. 49-82.
Callen, J.L. and Morel, M. (2002), “The Enron-Andersen debacle: do equity markets react to
auditor reputation?”, working paper, University of Toronto, Toronto.
Carcello, J., Hermanson, D., Neal, T. and Riley, R. (2002), “Board characteristics and audit fees”,
Contemporary Accounting Research, Vol. 19 No. 3, pp. 365-84.
Enron,
Andersen, and
Sarbanes-Oxley
23
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
0
7

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
Chan, D. (1999), “Low-balling and ef?ciency in a two-period specialization model for auditing
competition”, Contemporary Accounting Research, Vol. 16 No. 4, pp. 609-42.
Chaney, P.K. and Philipich, K.L. (2002), “Shredded reputation: the cost of audit failure”, Journal of
Accounting Research, Vol. 40, pp. 1221-45.
Chen, K.Y. and Zhou, J. (2007), “Audit committee, board characteristics, and auditor switch
decisions by Andersen’s clients”, Contemporary Accounting Research, Vol. 24 No. 4,
pp. 1085-117.
Chi, W. (2004), “The effect of Enron-Andersen affairs on audit pricing”, working paper, National
Chengchi University, Taipei.
Cohen, D.A., Day, A. and Lys, T. (2003), “The effect of the Sarbanes-Oxley Act on earnings
management: what has changed?”, working paper, Kellogg School of Management,
Northwestern University, Evanston, IL.
Coleman, C. and Bryan-Low, C. (2002), “Audit fees rise, and investors may pay price”, The Wall
Street Journal, August 12, p. C1.
DeAngelo, L. (1981), “Auditor independence, ‘low balling,’ and disclosure regulation”, Journal of
Accounting and Economics, August, pp. 113-27.
Dooger, R., Sougiannis, T. and Xie, H. (2003), “The value of auditor credibility: evidence from the
Andersen debacle”, working paper, University of Illinois at Urbana-Champaign, Urbana, IL.
Dopuch, N., Holthausen, R.W. and Leftwich, R.W. (1987), “Predicting audit quali?cations with
?nancial and market variables”, The Accounting Review, Vol. 62, pp. 431-54.
Dye, R. (1991), “Informationally motivated auditor replacement”, Journal of Accounting and
Economics, Vol. 14 No. 4, pp. 347-74.
Eisenberg, T. and Macey, J.R. (2003), “Was Arthur Andersen different? An empirical
examination of major accounting ?rms’ audits of large clients”, Cornell Legal Studies
Research Paper No. 04-03.
Fields, L.P., Fraser, D.R. and Wilkins, M.S. (2004), “An investigation of the pricing of audit
services for ?nancial institutions”, Journal of Accounting &Public Policy, Vol. 23, pp. 53-77.
Francis, J.R. (1984), “The effect of audit ?rm size on audit prices: a study of the Australian
market”, Journal of Accounting and Economics, Vol. 6, pp. 133-51.
Francis, J.R. and Simon, D.T. (1987), “A test of audit pricing in the small-client segment of the US
audit market”, The Accounting Review, Vol. 62 No. 1, pp. 145-57.
Geiger, M.A. and Rama, D.V. (2003), “Audit fees, nonaudit fees, and auditor reporting on stressed
companies”, Auditing: A Journal of Practice & Theory, Vol. 22 No. 2, pp. 53-69.
General Accounting Of?ce (2003), “Public accounting ?rms: mandated study on consolidation
and competition”, Doc. No. GAO-03-864, July 30.
Ghosh, A. and Lustgarten, S.I. (2006), “Pricing of initial audit engagements by large and small
audit ?rms”, Contemporary Accounting Research, Vol. 23 No. 2, pp. 333-68.
Hamilton, J., Li, Y. and Stokes, D. (2008), “Is the audit services market competitive following
Arthur Andersen’s collapse?”, Accounting and Finance, Vol. 48 No. 2, p. 233.
Hay, D.C., Knechel, W.R. and Wong, N. (2006), “Audit fees: a meta-analysis of the effect of supply
and demand attributes”, Contemporary Accounting Research, Vol. 23 No. 1, pp. 141-91.
Ho, J.L.Y. and Wang, J. (2007), “Examinations of audit fee premiums and auditors switching pre
and post the demise of Arthur Andersen and the enactment of Sarbanes-Oxley Act”,
working paper, University of California, Irvine, CA.
ARJ
22,1
24
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
0
7

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
Hogan, C.E. and Wilkins, M.S. (2008), “Evidence on the audit risk model: do auditors increases
audit fees in the presence of internal control de?ciencies?”, Contemporary Accounting
Research, Vol. 25 No. 1, pp. 219-42.
Kealey, B.T., Lee, H.Y. and Stein, M.T. (2007), “The association between audit-?rm tenure and
audit fees paid to successor auditors: evidence from Arthur Andersen”, Auditing: A
Journal of Practice and Theory, Vol. 26 No. 2, pp. 95-116.
Kohlbeck, M., Mayhew, B., Murphy, P. and Wilkins, M. (2008), “Competition for Andersen’s
clients”, Contemporary Accounting Research, Vol. 25 No. 4, pp. 1137-46.
Krishnan, G.V. and Chaney, P.K. (2007), “Did earnings conservatism increase for former
Andersen clients?”, Journal of Accounting, Auditing & Finance, Vol. 22 No. 2, pp. 141-74.
Krishnan, G.V. and Visvanathan, G. (2008), “Was Arthur Andersen different? Further evidence
on earnings management by clients of Arthur Andersen”, International Journal of
Disclosure and Governance, Vol. 5 No. 1, pp. 36-47.
Krishnan, J., Raghunandan, K. and Yang, J.S. (2007), “Were former Andersen clients treated more
leniently than other clients? Evidence from going-concern modi?ed audit opinions”,
Accounting Horizons, Vol. 21 No. 4, pp. 423-35.
Lai, K. (2003), “The Sarbanes-Oxley Act and auditor independence: preliminary evidence from
audit opinion and discretionary accruals”, working paper, City University of Hong Kong,
Kowloon.
McNamee, M. and Weber, J. (2002), “Bloodied but rich with Andersen gone, the ?nal four will do
just ?ne”, Business Week, June 3, pp. 74-5.
Markelevich, A., Hoitash, R. and Barragato, C.A. (2005), “Auditor fees, abnormal fees and audit
quality before and after the Sarbanes-Oxley Act”, working paper, Long Island University,
Long Island, NY.
Mayhew, B.W. and Wilkins, M.S. (2003), “Audit ?rm industry specialization as a differentiation
strategy: evidence from fees charged to ?rms going public”, Auditing: A Journal of
Practice & Theory, Vol. 22 No. 2, pp. 33-52.
Palmrose, Z.-V. (1986), “Audit fees and auditor size”, Journal of Accounting Research, Vol. 24,
pp. 97-110.
Raghunandan, K. and Rama, D.V. (1999), “Auditor resignations and the market for audit
services”, Auditing: A Journal of Practice & Theory, Vol. 18 No. 1, pp. 124-34.
Sankaraguruswamy, S. and Whisenant, S. (2005), “Pricing initial audit engagement: empirical
evidence following public disclosure of audit fees”, working paper, University of Houston,
Houston, TX.
Schatzberg, J.W. (1990), “A laboratory market investigation low-balling in audit pricing”,
The Accounting Review, Vol. 65 No. 2, pp. 337-62.
Simunic, D.A. (1980), “The pricing of audit services: theory and evidence”, Journal of Accounting
Research, Vol. 18, pp. 161-90.
White, H. (1980), “A heteroskedasticity-consistent covariance matrix estimator and a direct test
for heteroskedasticity”, Econometrica, May, pp. 817-88.
Willikens, M. and Achmadi, C. (2003), “Pricing and supplier concentration in the private client
segment of the audit market: market power or competition?”, The International Journal of
Accounting, Vol. 38, pp. 431-55.
Enron,
Andersen, and
Sarbanes-Oxley
25
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
0
7

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
Further reading
Krishnamurthy, S., Zhou, J. and Zhou, N. (2006), “Auditor reputation, auditor independence, and
the stock-market impact of Andersen’s indictment on its client ?rms”, Contemporary
Accounting Research, Vol. 23 No. 2, p. 465.
Corresponding author
Sharad Asthana can be contacted at: [email protected]
ARJ
22,1
26
To purchase reprints of this article please e-mail: [email protected]
Or visit our web site for further details: www.emeraldinsight.com/reprints
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
0
7

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)
This article has been cited by:
1. Marion Brivot, Charles H. Cho, John R. Kuhn. 2015. Marketing or parrhesia: A longitudinal study of
AICPA's shifting languages in times of turbulence. Critical Perspectives on Accounting . [CrossRef]
2. John C. Coates, Suraj Srinivasan. 2014. SOX after Ten Years: A Multidisciplinary Review. Accounting
Horizons 28, 627-671. [CrossRef]
3. Arezoo Aghaei Chadegani, Zakiah Muhammaddun Mohamed. 2014. Reporting errors and misstatements:
a measurement for the quality of auditors' work. Asian Journal of Business Ethics 3, 83-96. [CrossRef]
4. Miles B. Gietzmann, Angela K. Pettinicchio. 2014. External Auditor Reassessment of Client Business
Risk Following the Issuance of a Comment Letter by the SEC. European Accounting Review 23, 57-85.
[CrossRef]
5. Joanna L. Ho, Fei Kang. 2013. Auditor Choice and Audit Fees in Family Firms: Evidence from the S&P
1500. AUDITING: A Journal of Practice & Theory 32, 71-93. [CrossRef]
6. Yang Xu, Elizabeth Carson, Neil Fargher, Liwei Jiang. 2013. Responses by Australian auditors to the
global financial crisis. Accounting & Finance 53, 301-338. [CrossRef]
7. Jeffrey R. Cohen, Colleen Hayes, Ganesh Krishnamoorthy, Gary S. Monroe, Arnold M. Wright. 2013.
The Effectiveness of SOX Regulation: An Interview Study of Corporate Directors. Behavioral Research
in Accounting 25, 61-87. [CrossRef]
8. Bin Srinidhi, Mahmud Hossain, Chee Yeow Lim. 2012. The Effect of Arthur Andersen's Demise on
Clients' Audit Fees and Auditor Conservatism: International Evidence. Journal of International Financial
Management & Accounting 23:10.1111/jifm.2012.23.issue-3, 208-244. [CrossRef]
9. Steven E. Kaplan, David D. Williams. 2012. The changing relationship between audit firm size and going
concern reporting. Accounting, Organizations and Society 37, 322-341. [CrossRef]
10. Timothy G Coville, Gary Kleinman. 2012. ‘Post-Sarbanes–Oxley changes in the composition of boards:
Have they impacted spending for audit services?’. International Journal of Disclosure and Governance 9,
36-51. [CrossRef]
11. Wieteke Numan, Marleen Willekens. 2012. An empirical test of spatial competition in the audit market.
Journal of Accounting and Economics 53, 450-465. [CrossRef]
12. Ahsan Habib. 2011. Audit firm industry specialization and audit outcomes: Insights from academic
literature. Research in Accounting Regulation 23, 114-129. [CrossRef]
13. Shamharir Abidin, Vivien Beattie, Alan Goodacre. 2010. Audit market structure, fees and choice in a
period of structural change: Evidence from the UK – 1998–2003. The British Accounting Review 42,
187-206. [CrossRef]
14. Anwer S. Ahmed, Mary Lea McAnally, Stephanie Rasmussen, Connie D. Weaver. 2010. How costly is
the Sarbanes Oxley Act? Evidence on the effects of the Act on corporate profitability. Journal of Corporate
Finance 16, 352-369. [CrossRef]
D
o
w
n
l
o
a
d
e
d

b
y

P
O
N
D
I
C
H
E
R
R
Y

U
N
I
V
E
R
S
I
T
Y

A
t

2
1
:
0
7

2
4

J
a
n
u
a
r
y

2
0
1
6

(
P
T
)

doc_180801039.pdf