Managerial Auditing Journal: Article Information

Managerial Auditing Journal
Auditor fees and audit quality

Rani Hoitash, Ariel Markelevich, Charles A. Barragato,
Article information:
To cite this document:
Rani Hoitash, Ariel Markelevich, Charles A. Barragato, (2007) "Auditor fees and audit quality", Managerial
Auditing Journal, Vol. 22 Issue: 8, pp.761-786, https://doi.org/10.1108/02686900710819634
Permanent link to this document:
https://doi.org/10.1108/02686900710819634
Downloaded by University of Sri Jayewardenepura At 03:54 12 January 2019 (PT)
Downloaded on: 12 January 2019, At: 03:54 (PT)

References: this document contains references to 38 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 18135 times since 2007*
Users who downloaded this article also downloaded:
(2005),"Determinants of audit time as a proxy of audit quality", Managerial Auditing Journal,
Vol. 20 Iss 5 pp. 460-478 <a href="https://doi.org/10.1108/02686900510598821">https://
doi.org/10.1108/02686900510598821</a>
(2010),"Auditor conservatism following audit failures", Managerial Auditing Journal, Vol.
25 Iss 7 pp. 639-658 <a href="https://doi.org/10.1108/02686901011061333">https://
doi.org/10.1108/02686901011061333</a>
Access to this document was granted through an Emerald subscription provided by emerald-srm:216788 []
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.

The current issue and full text archive of this journal is available at
www.emeraldinsight.com/0268-6902.htm
Auditor fees and

Auditor fees and audit quality audit quality
Rani Hoitash
Department of Accountancy, Bentley College,
Waltham, Massachusetts, USA, and
Ariel Markelevich and Charles A. Barragato 761
Long Island University, Brookville, New York, USA
Abstract
Purpose – The paper aims to examine the relation between fees paid to auditors and audit quality
during the period of 2000-2003.

Design/methodology/approach – The paper constructs a measure of auditor profitability that is
used as a proxy for auditor independence. The methodology is grounded in the notion that auditor
independence is influenced by effort and risk-adjusted fees, rather than the level of fees received from
clients. Since, risk and effort are unobservable, the paper uses proxies based on client size, complexity
and risk to estimate abnormal fees. Abnormal fees are derived using a fee estimation model drawn
from prior literature. The paper employs two metrics to assess audit quality – the standard deviation
of residuals from regressions relating current accruals to cash flows and the absolute value of
performance-adjusted discretionary accruals.
Findings – The paper documents a statistically significant negative association between total fees
and both audit quality proxies over all years. These findings are robust to a variety of additional tests
and several alternative design specifications. The results (pre- and post-SOX) are consistent with
economic bonding being a determinant of auditor behavior rather than auditor reputational concerns.
Research limitations/implications – The possibility that the empirical tests do not completely
capture the impact of unobserved risk cannot be ruled out, though the paper attempts to do so by
employing alternative specifications and sensitivity tests.
Practical implications – Policy makers should note that current restrictions on the provision of
non-audit services may not sufficiently resolve the issue of economic bonding and its impact on
auditor independence.
Originality/value – In contrast to previous studies whose results are ambiguous, the paper finds a
statistically significant positive association between several measures of total fees (it uses
size-adjusted and abnormal fees) and two metrics of accruals quality in all years (2000-2003),
consistent with economic bonding being a determinant of auditor behavior rather then auditor
reputation concerns.
Keywords Auditor’s fees, Auditors, Quality
Paper type Research paper
Introduction
The relation between auditor independence and an auditor’s ability to conduct
high-quality audits has been widely debated by regulators, legislators, financial
statement users and researchers. Much of this discussion has been fueled by dramatic
changes in the market for accounting services during the 1990s, as well as concerns
Managerial Auditing Journal
The authors thank Jean Bedard, Douglas Carmichael, Masako Darrough, Ross Fuerman, Vol. 22 No. 8, 2007
Rebecca Rosner and Joseph Weintrop for their comments. This paper also benefited from the pp. 761-786
q Emerald Group Publishing Limited
helpful comments of Debra Jeter and Mike Stein, who served as their discussants at the 2005 0268-6902
American Accounting Association Annual Conference. DOI 10.1108/02686900710819634
MAJ that auditors are less likely to enforce GAAP to the extent that they receive large fees
22,8 from their audit clients.
Fees paid to auditors can affect audit quality in two ways: large fees paid to auditors
may increase the effort exerted by auditors, hence, increasing audit quality.
Alternatively, large fees paid to auditors, particularly those that are related to
non-audit services, make auditors more economically dependent on their clients. Such
762 financial reliance may induce a relationship whereby the auditor becomes reluctant to
make appropriate inquiries during the audit for fear of losing highly profitable fees.
Conversely, the potential for audit failure imposes significant economic costs on the
auditor (DeAngelo, 1981; Simunic, 1984). Though a number of recent studies have
examined the relationship between audit and non-audit fees and independence, they
are ambiguous as to the relationship between audit fees and auditor behavior (Larcker
and Richardson, 2004). They also differ on how fee composition and client importance
affect auditor independence.
We believe that examining fees paid by firms in the context of auditor profitability
better captures the relation between audit quality and auditor independence. In this
regard and consistent with the discussion in Kinney and Libby (2002), we develop a
methodology that is grounded in the notion that auditor independence is influenced by
the amount of fees relative to their expected amounts; e.g. adjusted for auditor effort
and risk, rather than the level of fees received from clients. We therefore examine effort
and risk-adjusted fees rather than raw fees. Since, these attributes are unobservable,
we develop two proxies, one based on client size and the other based on estimates of
expected or normal fees paid to auditors. The former proxy assumes that a larger
company will, on average, require the auditor to exert more effort and creates more
reputation risk for the audit firm in the event of an audit failure. The latter proxy
further refines the financial linkage between the auditor and the client by estimating
expected or normal fees to be charged by the auditor predicated on client type. We
derive abnormal fees using a fee estimation model drawn from prior literature which
takes into account not only the company’s size, but also its complexity, risk, and other
factors that may affect the fees charged by the auditor. Thus, our principal objective is
to ascertain whether larger size-adjusted or abnormal fees result in a higher or lower
quality audits.
In this study, we examine fees paid to auditors during the period 2000-2003 and find
a significant positive relation between size-adjusted and abnormal total fees paid to the
auditor and two metrics used to assess audit quality – an accruals quality measure
developed by Dechow and Dichev (2002), as modified by McNichols (2002) and Francis
et al. (2005)[1] and the absolute value of performance-adjusted discretionary accruals.
We focus our analysis on the total fees paid to the auditor for two reasons. First, the
argument that non-audit fees may impair audit quality can, by extension, be applied to
all fees received by the auditor (Raghunandan et al., 2003). Second, since our sample
period spans a timeframe during which the SEC changed its classificatory scheme for
disclosing the components of fees paid to auditors, using total fees mitigates any
potential confounding fee disclosure effects. We also conduct similar tests by
separating total fees into audit and non-audit fees categories and find a positive and
significant association with both of our accruals quality metrics[2]. Our results (both
pre- and post-SOX) are consistent with economic bonding being a determinant of
auditor behavior, rather than auditor reputational concerns; however, we cannot rule
out the possibility that our empirical tests do not sufficiently capture the impact of Auditor fees and
unobserved risk, despite our attempts to do so utilizing a variety of alternative audit quality
specifications and sensitivity tests.
Our study complements and extends existing literature on several dimensions.
First, we examine fees paid to auditors and financial reporting quality over a period of
time consumed by sweeping changes in the business, regulatory and professional
environment faced by auditors (culminating in the passage of the SOX). Second, in 763
contrast to previous studies whose results are:
.
ambiguous as to the relation between auditor independence and audit quality;
and
.
in many instances fail to reject the null hypothesis of no association (indicative of
econometric tests that lack power), we find a statistically significant positive

association between total fees and two metrics of accruals quality in all years
(2000-2003), consistent with economic bonding being the primary determinant of
auditor behavior rather then auditor reputation concerns.
Third, as suggested by Kinney and Libby (2002), we introduce an approach to address

the impact of client importance and fee composition on auditor independence. This
methodology incorporates estimates of abnormal fees that are generated using a fee
prediction model rooted in existing audit fee research. Fourth, our paper extends the
audit quality literature by employing an accrual estimation error measure developed
by Dechow and Dichev (2002), as modified by McNichols (2002) and Francis et al.
(2005). We are unaware of any research that uses this measure to assess the association
between audit fees and audit quality.
The remainder of the paper is organized as follows: the next section provides
background on the regulatory landscape, describes prior literature, and develops our
empirical predictions. We then present our research design and information about our
sample. We next describe the estimation of abnormal fees, followed by a report of our
findings. We then describe our additional tests, and then conclude.
Regulatory landscape, prior research, and empirical predictions

Regulatory landscape
Prior to Sarbanes Oxley. Issues surrounding auditor independence and investor
confidence in the financial statements of public companies have been widely debated.
Much of the discussion has been fueled by sweeping changes that have taken place in
the accounting profession during the 1990s. More specifically, many accounting firms
(including some of the largest accounting firms in the world) merged and transformed
themselves into multi-specialty organizations that offered a variety of services to their
audit clients.
Contemporaneous with the accounting profession’s metamorphosis, public
companies were under increased pressure to meet or beat earnings expectations.
The confluence of these two events prompted the SEC to propose and then adopt rules
that:
.
better described and in some cases narrowed the scope of services that auditors
could provide; and
MAJ .
required public companies to disclose information about auditor fees in proxy
22,8 statements filed after February 5, 2001 (SEC Final Rule S7-13-00, November,
2000).
With respect to the scope of services provision, the Final Rule states that “we have
determined not to adopt a total ban on non-audit-services . . . we recognize that not all
764 non-audit services pose the same risk to independence” (SEC, 2000, Section I). The
scope of services section of the Final Rule was viewed by many as a relatively benign
change as it merely clarified and conformed restrictions that already existed in the
professional literature.
After Sarbanes-Oxley. In the wake of Enron’s failure and amidst the collapse of
Arthur Andersen, Congress stepped in and passed the Sarbanes-Oxley Act of 2002
(Sarbanes-Oxley Act, 2002). A major portion of this extensive legislation focused on
curtailing public accounting firms from rendering non-audit-related services to their
public clients. These banned non-audit services include:
.
financial information system implementation and design;
.
internal audit functions; and
.
“other services”[3].
SOX also requires public companies to make additional disclosures regarding the
services provided by, and fees paid to, their independent auditors. The new disclosure
rules, which are described in Release No. 34-47365 (SEC, 2003), expanded the existing
requirements outlined in the section above. Companies are now required to disclose
fees for the two most recent fiscal years (effective for annual filings for the first fiscal
year ending after December 15, 2003). In addition to the expanded disclosure
requirements, the new rules require audit committees to pre-approve all audit and
permissible non-audit services.
The foregoing suggests that the rendering of non-audit services impairs
independence, leads to lower quality audits and increases the likelihood of financial
reporting that violates generally accepted accounting principles (Kinney et al., 2004).
Prior research
Prior research has investigated several aspects surrounding the association between
auditor fees and audit quality. This research stream is based on the thesis that auditor
independence is vital to the production of high-quality audits. In recent years,
regulators have become increasingly focused on the reliance of audit firms on
non-audit fees (US House of Representatives, 2002; US Senate, 2002). The crux of the
issue is grounded in the notion that non-audit services provided by incumbent auditors
can negatively influence auditor judgment, which in turn impairs the auditors’ ability
to enforce GAAP. As previously described, in 2000 the SEC attempted to curtail the
types of non-audit services auditors could provide to their public clients. More recently,
Congress enacted legislation that specifically prohibits the rendering of specific
non-audit services.
Though unparalleled attention to this independence issue has persisted in the
post-Enron era, the empirical evidence surrounding auditor fees and audit quality
remains mixed. Frankel et al. (2002) find a positive association between non-audit fees
and the likelihood of reporting a small earnings surprise, the magnitude of Auditor fees and
discretionary accruals and the magnitudes of income-increasing and audit quality
income-decreasing accruals. They contend that their results provide some evidence
that firms procuring non-audit services manage earnings to a greater extent than other
firms. Conversely, Ashbaugh et al. (2003) find no association between firms’ total fees
and discretionary current accruals, nor any association between income
increasing-accruals and client fees. Similarly, Chung and Kallapur (2003) find no 765
association between several audit-fee metrics and their estimate of discretionary
accruals.
A number of papers incorporate the use of abnormal fees in their audit quality
studies. Frankel et al. (2002) estimated unexpected fee ratios but did not link
unexpected fees to earnings quality. DeFond et al. (2002) predict abnormal fees and the
ratio of non-audit to total fees, and observe R 2s between 32 and 58 percent for their
prediction models. Their sample is comprised of financially distressed firms and 86
firms that received going concern opinions from their auditors. Their results are
consistent with higher total fees being associated with better audit quality as measured
by the type of audit opinion issued (unqualified v. going concern). DeFond et al. (2002)
also examine a sample of financially distressed firms which by construction are more
likely to be the subject of future litigation (suggesting that auditors will generally be
more cautious when auditing such companies). Larcker and Richardson (2004) also
estimate expected total and non-audit fees. They use latent class mixture analysis and
find no indication of economic bonding. However, they use no controls in their accrual
regressions and their fees are deflated by the total revenue of the audit firm. Whisenant
et al. (2003) also used abnormal fee ratio and abnormal total fees and compared them
between 110 restating firms and the Compustat universe of non-restating firms. They
found no significant differences in abnormal fees between the two groups. Reynolds
et al. (2004) examined the finding obtained by Frankel et al. (2002) and observed that
these findings were driven by small- to medium-size high-growth firms. After
controlling for a number of factors related to assets growth and IPOs, they observed no
significant association between the fee ratio and the log of total fees to the absolute
value of discretionary accruals.
As more fully described in our research design section, our study is different in that
we examine the association between abnormal total fees and audit quality (measured
using two proxies of accruals quality) for a large number of companies over an
extended period of time.
Empirical predictions
Any relation between the fees paid to auditors and audit quality is an important input
to the ongoing debate on how the accounting profession should be organized and
monitored. Large fees paid to auditors may increase the effort exerted by auditors,
hence, increasing the audit quality. Critics contend that large fees paid to auditors,
particularly those that are related to non-audit services, make auditors more
economically dependent on their clients (Becker et al., 1998; Magee and Tseng, 1990).
Such financial reliance may induce a relationship whereby the auditor becomes
reluctant to make appropriate inquiries during the audit for fear of losing highly
profitable fees. Conversely, others suggest that the potential for audit failure imposes
significant economic costs on the auditor (DeAngelo, 1981; Simunic, 1984). Thus, the
MAJ relationship between audit fees and auditor behavior is ambiguous (Larcker and
22,8 Richardson, 2004).
The aforementioned studies are based on data and a regulatory environment that
pre-dates the unraveling of Enron and the ultimate passage of SOX. Our study
complements and extends existing literature by examining total fees, audit and
non-audit fees and audit quality over a period of time that spans an evolving business
766 and regulatory environment.
To summarize, though the full regulatory impact of SOX is yet to be determined, to
the extent that legislators and regulators are correct in their presumption that the
rendering of non-audit services:
.
impairs independence;
.
leads to lower quality audits; and
.
increases the likelihood of GAAP violations, then we suspect that the expanded
fee disclosures and restrictions on consulting services mandated by SOX should
be effective in curtailing auditor independence violations.
However, if auditor independence violations stem more from auditors’ dependency on

the fees received from audit clients, then we expect to see little change in auditor
independence breaches that result from the abusive application of GAAP.
Research design and sample description

Our approach
Our study complements previous research in a number of ways. First, we provide an
overview of audit and non-audit services by examining fees paid to auditors during the
period 2000-2003. This timeframe is of particular interest because fees paid to auditors
during this period occurred amidst sweeping changes in the business, regulatory and
professional environment faced by auditors; whereas prior studies have focused on
audit fee data collected in a pre-SOX environment.
Second, we assess audit quality utilizing both:
(1) an accruals quality measure developed by Dechow and Dichev (2002), as
modified by McNichols (2002) and Francis et al. (2005); and
(2) the magnitude of absolute discretionary current accruals (consistent with
Ashbaugh et al., 2003, among others).
Third, we incorporate these two proxies for audit quality into an empirical analysis
that focuses on abnormal fees. Kinney and Libby (2002) suggest that “better conceptual
definitions can improve measurement of the concepts in all empirical work.” They go
on to argue that the concept of economic bonding between clients and their auditors
can be further refined through building better models that distinguish between
unexpected non-audit and audit fees. Such a distinction may flush out the more
ominous impact on economic bonding that emanates from unexpected fees and may
better capture the effects of auditor profitability on independence. Since, it is likely that
auditor independence is influenced by the amount of the total fees relative to their
expected amounts, rather than the level of fees received from clients; i.e. clients with
unusually high or low fees may influence incumbent auditor judgment (DeFond et al.,
2002), incorporating these additional measures is economically intuitive and addresses
a number of concerns surrounding the use of the level of fees without controlling for Auditor fees and
their source (Larcker and Richardson, 2004). audit quality
Fourth, we use a fee prediction model drawn from Simunic (1984), Craswell and
Francis (1999) and Hay et al. (2006). We use this fee prediction model to estimate the
unexpected portion of total fees (which we deem to represent auditor profitability),
which is the primary focus of our analysis, and conduct ancillary tests on audit and
non-audit fees. Since, audit engagement profitability is not directly observable, we use 767
the residual term generated from estimating our fee prediction model as our proxy.
Though our estimation model for abnormal fees is not a perfect substitute for
unobservable audit engagement profits (losses), after adjusting for size, complexity,
risk, and imposing tests for multicollinearity and heteroscedasticity, it is well-specified
(e.g. explanatory power ranging from 73 to 78 percent).
Thus, with respect to the effects of client importance and auditor effort on audit
quality, we expect that including abnormal fees in our modeling will enhance our
controls beyond those obtained using only scaled fees or the ratio of non-audit fees to
total fees[4].
Sample description and variable definitions

Our initial sample consists of 21,522 observations for firms that reported audit and
non-audit fee data for fiscal years 2000-2003. These data were obtained from Standard
& Poor’s Audit Fee Database. We delete 4,305 observations of financial firms and 3,357
observations that lacked the necessary data for estimating our models, leaving us with
13,860 observations. Our final data set consists of 13,860 observations (2,476; 4,016;
3,949; 3,419 observations in 2000-2003, respectively).
We define total fees as the sum of all fees paid to the auditor. These fees are later
divided into two categories: audit fees and non-audit fees.
Measures of audit quality

We employ two measures of audit quality – the absolute value of
performance-adjusted discretionary accruals, which has been used in a variety of
studies to proxy for audit quality, and an accruals quality measure developed by
Dechow and Dichev (2002), as modified by McNichols (2002) and Francis et al. (2005),
hereinafter referred to as “FLOSAQ”. We choose these two metrics as our proxies for
audit quality because of their complementary characteristics, as more fully delineated
in the following sections.
Discretionary accruals. We use the magnitude of discretionary accruals as one of our
proxies for biasness. Discretionary accruals provide a metric for assessing the degree
of biasness infused into the financial statements by management and tolerated by the
auditor. Discretionary accruals have been used in a number of studies to proxy for
audit quality and for independence breaches (Ashbaugh et al., 2003; Chung and
Kallapur, 2003; Reynolds et al., 2004; Frankel et al., 2002, among others). Absent a
specific theory that predicts the direction of accruals (either income increasing or
income decreasing), we employ tests using the absolute value of abnormal accruals.
Using the unsigned value of abnormal accruals more completely identifies the
discretion afforded managers by their auditors and in this context does not require
assumptions about auditor bias with regard to the directional effect of an accounting
choice (Menon and Williams, 2004)[5].
MAJ Despite its popularity in a number of published studies, abnormal accruals
22,8 computed under the modified Jones model has been the subject of criticism. This
criticism stems from the fact that the Jones methodology has been found to estimate
discretionary accruals with considerable imprecision. This finding suggests that
caution must be exercised when interpreting results about management’s use of
accruals motivated by opportunism and/or a performance measure enhancement (see
768 Guay et al., 1996; Bernard and Skinner, 1996 for a complete review). More recently,
Francis et al. (2005) present a critical view of the modified Jones model because it
essentially relies on only two key financial statement attributes (property, plant and
equipment and changes in revenues) to identify accruals that are abnormal.
Based on the foregoing and since prior research documents a correlation between
discretionary accrual estimates and firm performance (Dechow et al., 1995; Kasnik,
1999; Kothari et al., 2005), we control for firm performance by including lagged ROA
and by constructing an accrual measure that is similar to the one in Ashbaugh et al.
(2003):
1
CA ¼ a þ bðDREVÞ þ gðROAt21 Þ þ 1t ð1Þ
TAt21
^ 1 ^ ^
ECA ¼ a þ bðDREV 2 DARÞ þ gðROAt21 Þ ð2Þ
TAt21
where current accruals (CA) is net income before extraordinary items (Compustat data
item 123) plus depreciation and amortization (Compustat data item 125) minus
operating cash flows (Compustat data item 308), scaled by beginning of year total
assets. TAt2 1 is total assets at the beginning of the fiscal year and DREV is equal to
net sales (Compustat data item 12) in year t less net sales in year t 2 1, scaled by
beginning of year total assets, and DAR is equal to accounts receivable (Compustat
data item 2) in year t less accounts receivable in year t 2 1, scaled by beginning of year
total assets. We first estimate equation (1) separately for each two-digit SIC code and
then use the parameters from equation (1) to estimate the expected current accruals
(ECA) using equation (2). REDCA is the difference between CA and ECA and
represents our measure for discretionary current accruals. We winsorize the 1 percent
extreme values from each tail and use the absolute value of REDCA, denoted
ABSREDCA, as our measure for the combined effect of income-increasing and
income-decreasing evidence of earnings management.
Accrual estimation error metric (FLOSAQ). Dechow and Dichev (2002) argue that
uncertainty in accruals can best be measured by the extent to which working capital
accruals map into cash flow realizations. The key insight to their model is that accruals
quality is affected by measurement error in accruals, irrespective of management’s
purpose (e.g. imposing intentional or unintentional errors in the estimation of accruals).
According to Francis et al. (2005) and from a practical perspective, the Dechow and
Dichev (D&D) model is limited to current accruals (because of the long lead/lags
between noncurrent accruals and ultimate cash flow realizations). Thus, consistent
with McNichols (2002), they augment the D&D model by incorporating the
fundamental variables included in the modified Jones model (e.g. property, plant and
equipment and changes in revenues). This augmented model produces a better
specified expectations model and, therefore, a better set of residuals. Following Francis Auditor fees and
et al. (2005), we estimate a cross-sectional accrual model as depicted in the following audit quality
equation:
TCAi;t ¼ at þ b1;i CFOi;t21 þ b1;i CFOi;t þ b1;i CFOi;tþ1 þ b1;i DARi;t

ð3Þ
þ b1;i PPEi;t þ 1i;t 769
Where total current accruals are calculated as:
TCAi;t ¼ ðDCAi;t 2 DCLi;t 2 DCASHi;t þ DCDEBTi;t Þ
and total accruals are calculated as:
TACCi;t ¼ ðDCAi;t 2 DCLi;t 2 DCASHi;t þ DCDEBTi;t 2 DDEPÞi;t

CFO is the cash flow from operations calculated consistent with Francis et al. (2005)
using the indirect method for each firm i year t (i.e. NIBEi,t – TACCi,t). Where NIBE i,t is
firm i’s income before extraordinary items (Compustat data item 18) and TACCi,t is
total current accruals, DCAi,t is change in current assets (Compustat data item 4), DCLi,t
is change in current liabilities (Compustat data item 5), DCASHi,t is change in cash
(Compustat data item 1), DCDEBTi,t is change in current debt (Compustat data item
34), and DDEPi,t is the change in depreciation and amortization (Compustat data item
14)[6]. All variables are scaled by the average total assets and all changes are
calculated between year t and year t 2 1.
Consistent with Francis et al. (2005) we estimate equation (3) separately for each
Fama and French (1997) industry group (a total of 48 industries) with at least 20 firms
in year t. We winsorize the 1 percent extreme values from each tail and use the residual
from equation (3) as our estimate for abnormal accruals. Subsequently, we use the
estimated firm year residual during the five-year period between t 2 4 and period t and
calculate the standard deviation of these residual. In turn, we term the calculated
standard deviation FLOSAQi,t and use it as our proxy for accruals quality. A large
standard deviation implies that discretionary accruals vary over time yielding
low-accrual quality. In contrast to REDCA, FLOSAQ is already stated in absolute
terms and therefore no further transformation is needed.
Descriptive statistics. Descriptive statistics for our sample are presented in Table I.
Panel A reflects descriptive statistics for non-deflated fees paid to audit firms. The
number of observations increases considerably from 2000 to 2003. Average total fees
declined substantially over our sample period (2 29 percent), while average audit fees
(non-audit fees) increased (decreased) by roughly 49 percent and (59 percent),
respectively. Overall, to the extent that raw total fees serve to proxy for economic
bonding between auditors and clients, the decline in total fees suggests that auditors
appear to be less dependent on their clients[7].
Panel B of Table I reflects descriptive statistics for the deflated variables used in our
analyses. We deflate fees by the square root of total assets following Kinney et al.
(2004) and Simunic (1980) to mitigate spurious correlations due to size and to reduce
heteroscedasticity. Our results in Panel B are consistent with those of Panel A. We
observe a moderate decline in average deflated total fees (2 8 percent), a sharper
increase in average deflated audit fees (69 percent) and slightly smaller decline in
22,8
770
MAJ
Table I.
Descriptive statistics
Variable Year N Mean Median SD
a
Panel A: non-deflated variables (in $000’s)
Total fees 2000 2,476 2,058.480 564.000 5,898.258
2001 4,016 1,454.994 362.760 4,666.295
2002 3,949 1,379.138 387.000 4,009.384
2003 3,419 1,469.476 465.000 3,906.040
Audit fees 2000 2,476 582.651 231.000 1,513.101
2001 4,016 494.585 188.173 1,187.006
2002 3,949 694.321 240.000 1,675.866
2003 3,419 870.702 305.570 2,347.812
Non-audit 2000 2,476 1,475.829 304.325 4,853.174
2001 4,016 960.409 148.000 3,745.256
2002 3,949 684.817 113.000 2,735.396
2003 3,419 598.773 123.894 2,029.480
Total assets 2000 2,476 2,391.321 308.278 9,296.745
2001 4,016 1,768.819 157.953 7,705.136
2002 3,949 1,862.065 162.306 7,636.657
2003 3,419 2,221.477 217.327 8,699.858
Panel B: deflated variables
Total fees 2000 2,476 47.775 33.335 55.458
2001 4,016 43.512 32.662 40.722
2002 3,949 44.918 33.906 42.713
2003 3,419 44.141 34.968 42.076
Audit fees 2000 2,476 17.044 13.801 14.244
2001 4,016 20.809 16.888 17.208
2002 3,949 27.254 21.198 28.295
2003 3,419 28.795 23.173 24.935
Non-audit 2000 2,476 30.732 17.169 48.667
2001 4,016 22.719 12.735 32.500
2002 3,949 17.691 9.696 26.806
2003 3,419 15.365 9.135 29.397
ABSREDCA 2000 2,476 0.502 0.099 1.120
2001 4,016 0.275 0.100 0.604
2002 3,949 0.278 0.076 0.655
2003 3,419 0.201 0.074 0.495
(continued)
FLOSAQ 2000 1,529 0.061 0.045 0.057

2001 2,865 0.076 0.054 0.078
2002 2,980 0.107 0.062 0.146
2003 2,607 0.101 0.058 0.143
Panel C: correlations between variables for the entire sample across years
Total fees Audit fees Non-audit fees ABSREDCA FLOSAQ
Total fees 1.0000 0.6545 * 0.8637 * 0.0961 * 0.0702 *
Audit fees 0.7537 * 1.0000 0.1843 * 0.1096 * 0.1688 *
Non-audit fees 0.7885 * 0.2773 * 1.0000 0.0525 * 20.0273 *
ABSREDCA 0.1288 * 0.1391 * 0.0388 * 1.0000 0.1999 *
FLOSAQ 0.1543 * 0.2688 * 20.0123b 0.3337 * 1.0000
Panel D: descriptive statistics of other variables used in our study
Variable Mean Median SD
Total assets 2,018.26 199.33 8,244.64
LNTA 5.336 5.295 2.167
ROA 20.214 0.014 3.516
LOSS 0.452 0 0.498
BIG5 0.85 1 0.357
MERGER 0.032 0 0.175
INSTOWN 43.586 42.62 43.432
SEG 1.374 1 0.485
TENURE 6.109 6 3.52
SUBS 3.392 2.449 3.783
FOREIGN 0.255 0 0.436
FINANCE 0.339 0 0.473
INVREC 0.26 0.225 0.195
LITIGATION 0.29 0 0.454
LIQ 3.339 2.015 8.932
L1Accrual 20.19 20.013 10.475
LNMVE 5.227 5.256 2.286
LEVERAGE 0.266 0.171 2.151
MB 3.214 1.769 92.075
CFO 20.014 0.067 1.439
(continued)
audit quality
Auditor fees and
771
Table I.
22,8
772
MAJ
Table I.
Panel E: variable definitions

Variable Definition
Audit fees Total audit fees paid to the auditor
Non-audit fees All other services fees paid to the auditor
Total fees Sum of audit and non-audit fees
Total assets Total assets
REDCA Performance adjusted abnormal discretionary current
accruals using Ashbaugh et al. (2003) model
ABSREDCA Absolute value of REDCA
FLOSAQ Accrual estimation error following Francis et al. (2005)
Total assets The firm’s total assets (data 6)
LNTA The natural log of the firm’s total assets
ROA Return on assets, defined as net income divided by total
assets
LOSS An indicator variable equal to one if the audit client
reported negative net income in either of the two
previous fiscal years (zero otherwise)
BIG5 An indicator variable equal to one when the auditor is a
member of the BIG5 (zero otherwise)
MERGER An indicator variable taking the value of 1 if the client
engaged in merger activity during year t (zero
otherwise)
INSTOWN The percentage of shares owned by institutions
(obtained from Research Insight)
SEG The square root of the number of business segments
TENURE The number of engagement years, measured as the
number of years under engagement over the last ten years
SUBS The square root of the number of subsidiaries
FOREIGN An indicator variable equal to 1 if the company has
foreign operations (zero otherwise)
FINANCE An indicator variable equal to one if the audit client
issues long-term equity (Data item 108 . 5 percent of
beginning total assets) in the current fiscal year (zero
otherwise)
(continued)
INVREC Calculated as inventory plus accounts receivable (Data

items 3 þ 2) divided by total assets
LITIGATION An indicator variable set at 1 if the firm operates in a
high-litigation industry, and 0 otherwise (high-litigation
industries are industries with SIC codes of 2833-2836,
3570-3577, 3600-3674, 5200-5961, and 7370-7370)
LIQ The ratio of current assets (Data 4) divided by current
liabilities (Data 5)
L1Accrual Last year’s total current accruals
LNMVE The natural log of the firm’s price per share at fiscal year
end (Compustat data item 199) multiplied by the number
of shares outstanding (Compustat data item 25)
LEVERAGE The ratio of total debt to total assets
MB The market-to-book ratio, which is defined as market
value of equity divided by book value
CFO Represents cash flow from operations (Compustat data
item 308), scaled by beginning of year total assets
Notes: See Panel E for variable definitions; aexcept for total assets, which are expressed in millions of dollars; all fees are deflated by the square root of
total assets. The Panel C presents Pearson (Spearman) correlation above (below) the diagonal. The correlation is calculated for the entire sample across all
years, N ¼ 13,860 and 9,981 for the FLOSAQ. *significant at 1 percent level; bnot statistically significant. Panel D presents the descriptive statistics for
the variables used in our fee prediction model and in our audit quality analysis. The descriptive statistics are calculated for the entire sample across all
years, N ¼ 13,860
audit quality
Auditor fees and
773
Table I.
MAJ average non-audit fees (2 50 percent). Overall, it appears that mean audit fees increase
22,8 monotonically over the sample period, while average non-audit fees decrease
monotonically over the same time frame. One potential reason for the observed
increase in audit fees and the observed decline in non-audit fees is the reclassification
of certain services rendered by the auditor and not the actual services (non-audit fees to
audit fees – see additional tests section).
774 Table I – Panel C presents correlations among variables used in our analyses. Total
fees are positively correlated with audit and non-audit fees. Audit fees and non-audit
fees are positively correlated, suggesting that a company with more audit services has
more non-audit services. Contrary to most previous studies, we find a significant
association between the level of deflated fees (total, audit, and non-audit) and the level
of absolute discretionary accruals. Similarly, we observe a positive and significant
correlation between total fees and audit fees and our measure of accruals quality
(FLOSAQ). The correlation between FLOSAQ and non-audit fees is not significant.
These results suggest that higher fees are associated with lower audit quality – which
is consistent with the presence of an auditor dependency problem.
Abnormal total fees

As described earlier, we utilize a fee prediction model drawn from prior literature to
estimate the unexpected portion of total fees, which we deem to represent either a
premium received or a discount granted by the auditor.
Methodology
Previous studies have modeled fees as a function of size, risk, complexity, auditor type
and tenure, monitoring, and additional work. Our fee prediction model draws on
Simunic (1984), Craswell and Francis (1999), Larcker and Richardson (2004) and Hay
et al. (2006). These prior studies have found audit and total fee prediction models to
have good explanatory power (adj. R 2 of 0.70 or better) and are robust to different
samples, time periods and sensitivity analyses for model misspecification (Chan et al.,
1993).
We expect that large, more complex clients will have to pay more to their auditors,
therefore we control for client size by including the natural log of total assets. We
capture audit complexity using the square root of the number of subsidiaries, the
square root of the number of business segments, and identify whether the company has
foreign operations. It is also expected that riskier clients will have to pay more to
auditors. We proxy for risk using liquidity and return on assets ratios, and identify
companies with net losses. Auditors often have to perform additional work related to
receivable confirmations, inventory-related activities, mergers and acquisitions and
equity issues. We capture them through mergers and acquisitions activity, the relative
size of accounts receivable and inventory, and new equity issues. We identify an
alternative source for firm monitoring by looking at the percentage of shares owned by
institutions and expect it to be positively associated with fees. Additionally, we proxy
for auditor characteristics by identifying BIG5 auditors and engagement tenure.
We model expected or normal fees using the following equation and consider the
residual term as our estimate of the abnormal portion of total fees:
LTFEE ¼ a þ b1 LNTA þ b2 SEG þ b3 SUBS þ b4 FOREIGN þ b5 LOSS Auditor fees and
audit quality
þ b6 LIQ þ b7 ROA þ b8 MERGER þ b9 INVREC þ b10 FINANCE
þ b11 INSTITOWN þ b12 BIG5 þ b13 TENURE þ 1 ð4Þ
775
LTFEE is the natural logarithm of total fees. LNTA is the natural logarithm of total
assets (Compustat data item 6), SEG is the square root of the number of segments,
where the number of segments is obtained from Research Insight, SUBS represents the
square root of the number of subsidiaries (retrieved from Compact Disclosure),
FOREIGN is an indicator variable equal to 1 if the audit client has foreign operations
(zero otherwise), LOSS is an indicator variable equal to one if the audit client reported
negative net income in either of the two previous fiscal years (zero otherwise), LIQ is
the ratio of current assets (Compustat data item 4) divided by current liabilities
(Compustat data item 5), ROA equals return on assets, defined as net income divided
by total assets. MERGER is an indicator variable taking the value of 1 if the client
engaged in merger activity during year t (zero otherwise), INVREC is calculated as
inventory plus accounts receivable (Compustat data items 2 þ 3) divided by total
assets, FINANCE is an indicator variable equal to one if the audit client issues
long-term equity (Compustat data item 108 . 5 percent of beginning total assets) in
the current fiscal year (zero otherwise), INSTITOWN is the percentage of shares owned
by institutions (obtained from Research Insight), BIG5 is an indicator variable equal to
one when the auditor is a member of the BIG5[8] (zero otherwise), and TENURE is the
number of engagement years, measured as the number of years under engagement
over the last ten years.
Estimation results
The results of our total fee (hereinafter referred to in this section as “fees”) estimation
model are presented in Table II. To capture changes in expected or normal fees over
time and to ensure that reported t-statistics are not unduly overstated due to
time-series correlation, we estimates our regressions separately each year[9]. Overall,
our model appears to be well specified, producing adjusted R 2 s ranging from 0.728 to
0.793[10]. We observe a positive association between fees and client size and audit
complexity. Our model documents a negative association between good performers
(ROA, LIQ) and fees and a positive association between fees and more risky firms
(LOSS). Our results suggest a premium to BIG5 auditors but are not consistent with
auditors having longer tenure[11]. Similarly, our model also supports the notion that
additional work is needed for firms with large receivables, inventories merger activity
and activities associated with new financing. We use the residuals generated from
estimating equation (4) as our measure of abnormal fees, which in turn is used to assess
the degree of bonding between the client and the auditor in subsequent analyses.
The association between total fees and audit quality

In this section, we examine the relation between total fees (hereinafter referred to as
“fees”) and audit quality. As described in previous sections, we employ two audit
quality measures. The first is the absolute value of performance adjusted current
22,8
776
MAJ
Table II.
predictions
Abnormal total fee
(lTOTAL) Dependent variable
Independent variables 2000 2001 2002 2003
INTERCEPT 2.1829, 23.94 * * * 2.6001, 48.43 * * * 2.9506, 58.17 * * * 2.8291, 55.98 * * *

LNTA 0.5852, 54.37 * * * 0.5023, 67.45 * * * 0.4303, 55.73 * * * 0.4393, 54.67 * * *
ROA 20.3579, 2 8.16 * * * 20.0841, 2 10.27 * * * 20.0102, 2 6.00 * * * 20.0550, 2 6.04 * * *
LOSS 0.1739, 4.84 * * * 0.2565, 11.22 * * * 0.2482, 10.95 * * * 0.2065, 8.66 * * *
BIG5 0.2855, 4.46 * * * 0.2708, 7.75 * * * 0.2436, 7.30 * * * 0.2647, 7.70 * * *
MERGER 0.1622, 2.54 * * 0.2842, 4.80 * * * 0.2409, 3.36 * * * 0.1480, 2.10 * *
INSTITOWN 20.0008, 2 1.88 * 20.0003, 2 1.48 0.0007, 2.04 * * 0.0006, 1.64
SEG 0.1925, 5.99 * * * 0.1762, 7.21 * * * 0.1476, 6.00 * * * 0.2039, 8.54 * * *
TENURE 20.0238, 2 4.67 * * * 20.0033, 2 0.92 0.0191, 6.13 * * * 0.0080, 2.44 * *
SUBS 0.0357, 8.32 * * * 0.0417, 11.36 * * * 0.0492, 13.51 * * * 0.0495, 14.09 * * *
FOREIGN 0.3355, 10.21 * * * 0.2896, 11.70 * * * 0.2639, 10.69 * * * 0.2852, 11.94 * * *
FINANCE 0.0651, 2.15 * * 0.0528, 2.37 * * 0.0537, 2.28 * * 0.0947, 4.07 * * *
INVREC 0.6679, 8.18 * * * 0.4749, 8.42 * * * 0.3298, 5.82 * * * 0.4173, 7.26 * * *
LIQ 20.0009, 2 0.98 20.0186, 2 7.85 * * * 20.0221, 2 9.24 * * * 20.0040, 2 3.73 * * *
N 2,475 4,015 3,948 3,418
Adj. R 2 0.7276 0.7772 0.7710 0.7934
Notes: We regress the natural log of total fees paid to the auditor (LTFEE) on various fees determinants to determine the positive and negative abnormal
or unexpected total fees: LTFEE ¼ a þ b1 LNTA þ b2 ROA þ b3 LOSS þ b4 BIG5 þ b5 MERGER þ b6 INSTITOWN þ b7 SEG þ b8 TENURE þ b9
SUBS þ b10 FOREIGN þ b11 FINANCE þ b12 INVREC þ b13 LIQ þ 1; *significant at 10 percent level; * *significant at 5 percent level; * * *significant at
1 percent level; see Table I Panel E for variable definitions
discretionary accruals (ABSREDCA). Our second measure is accruals quality Auditor fees and
(FLOSAQ).
audit quality
Regression analysis
To investigate the association between audit quality and client fees, we start by
examining the relation between deflated fees and our measures of audit quality, as
estimated using the following models: 777
FLOSAQ ¼ a þ b1 Total fees þ b2 LNTA þ b3 Cycle þ b4 PropLoss
þ b5 StdSales þ b6 StdCFO þ 1 ð5Þ

ABSREDCA ¼ a þ b1 Total fees þ b2 BIG5 þ b3 L1Accrual þ b4 LNMVE
þ b5 MERGER þ b6 FINANCING þ b7 LEVERAGE

ð6Þ
þ b8 MB þ b9 LITIGATION þ b10 INSTOWN þ b11 LOSS
þ b12 CFO þ 1
Additionally, we estimate similar models to equations (5) and (6) using previously
estimated abnormal fees instead of the deflated fees:
FLOSAQ ¼ a þ b1 Abnormal total fees þ b2 LNTA þ b3 Cycle þ b4 PropLoss
ð7Þ
þ b5 StdSales þ b6 StdCFO þ 1
ABSREDCA ¼a þ b1 Abnormal total fees þ b2 BIG5 þ b3 L1Accrual þ b4 LNMVE
þ b5 MERGER þ b6 FINANCING þ b7 LEVERAGE þ b8 MB ð8Þ
þ b9 LITIGATION þ b10 INSTOWN þ b11 LOSS þ b12 CFO þ 1

We posit that a positive association between our fee measures and both proxies for
audit quality is suggestive of increased economic bonding between a client and the
auditor (e.g. lower audit quality). We estimate all models separately for each year,
where total fees are total fees deflated by total assets (Compustat data item 6) and
abnormal total fees are the residual from equation (4).
Except for the fee variables, equations (5) and (7) are based on the model used in
Doyle et al. (2007). Following Dechow and Dichev (2002) and Doyle et al. (2007), we use
innate firm characteristics to control for the inherent difficulty that firms face in their
accrual estimation process. Similar to Dechow and Dichev (2002), we use firm size and
the length of the operating cycle as control variables that can impact the complexity of
accrual estimation. Additionally, we follow Doyle et al. (2007) and include control
variables for cash flow volatility, sales volatility, and proportion of loss. Where firm
size is LNTA and is as previously defined, operating cycle is defined as:
360/(sales/average AR) þ 360/(cost of goods sold)/(average inventory). Where sales
MAJ is Compustat data item 12, cost of goods sold is Compustat data item 41, AR is
22,8 Compustat data item 2 and inventory is Compustat data item 3. StdCFO is the standard
deviation of cash from operations (Compustat data item 308) scaled by average assets,
calculated over the previous ten years. StdSales is the standard deviation of the change
in sales (Compustat data item 2), scaled by average assets, calculated over the previous
ten years, and PropLoss is the ratio of the number of years of losses (Compustat data
778 item 123) calculated over the previous ten years.
Equations (6) and (8) are similar to the model used in Ashbaugh et al. (2003), with
the exception of the fee variables. L1Accrual is calculated as net income before
extraordinary items (Compustat data item 123) plus depreciation and amortization
(Compustat data item 125) minus operating cash flows (Compustat data item 308),
scaled by beginning of year total assets. LNMVE is the natural log of the firm’s price
per share at fiscal year end (Compustat data item 199) multiplied by the number of
shares outstanding (Compustat data item 25), LEVERAGE is defined as the firm’s total
assets less its book value (Compustat data item 60), divided by total assets. MB is the
market-to-book ratio, which is defined as market value of equity divided by book value.
LITIGATION is captured by an indicator variable set at 1 if the firm operates in a
high-litigation industry, and 0 otherwise (high-litigation industries are industries with
SIC codes of 2833-2836, 3570-3577, 3600-3674, 5200-5961, and 7370). LOSS is also an
indictor set at 1 if the firm reports a net loss in the current fiscal year, and 0 otherwise.
CFO represents cash flow from operations (Compustat data item 308), scaled by
beginning of year total assets. All other variables are as previously defined.
The results of estimating equations (5) and (6) are displayed in Table III (Panels A
and B). We observe that total fees are significant and positively related to both
FLOSAQ and ABSREDCA; i.e. the higher the deflated total fees the lower the quality.
This association is significant and consistent with the univariate results presented in
the previous section and remains stable throughout our sample period. The coefficients
on the control variables are, by and large, also consistent over time[12].
Since, the results in Table III may suffer from a scaling limitation (e.g. both large
and small clients could potentially have similar scaled fees), we expand our analysis by
substituting deflated total fees with abnormal total fees (equations (7) and (8)). As
previously discussed, since we consider the residual term as our estimate for the
abnormal portion of total fees (equation (4)), our measure represents a log dollar
amount and does not suffer from the limitation above. Nevertheless, we observe similar
results to those reported in Table III over the entire sample period. Table IV reports
that abnormal total fees are consistently positively and significantly associated with
both FLOSAQ and absolute abnormal accruals, consistent with audit quality being
compromised for clients who pay total fees beyond their expected values[13]. These
results provide evidence in support of the view delineated above and those suggested
in Kinney and Libby (2002); i.e. that economic bonding between clients and incumbent
auditors can be empirically discerned by segregating the unexpected portion of the
total fees. Our results are also consistent with higher fee premiums (abnormal fees)
being associated with lower audit quality and suggest that more profitable clients on
average have more influence over their auditors, who in turn may be more likely to
succumb to client pressure to misapply GAAP.
Auditor fees and
Independent
variables 2000 2001 2002 2003 audit quality
Panel A: association between total fees and accrual quality a
Intercept 0.0066, 0.79 0.0154, 2.02 * * 0.0291, 2.74 * * * 0.0192, 1.73 *
Total fees 0.00007, 4.35 * * * 0.00007, 3.55 * * * 0.0002, 6.67 * * * 0.0002, 5.37 * * *
LNTA 20.0046, 28.59 * * * 20.0046, 29.85 * * * 20.0060, 29.46 * * * 20.0056, 28.72 * * *
Cycle 0.0107, 7.64 * * * 0.0092, 6.97 * * * 0.0074, 3.98 * * * 0.0094, 4.86 * * * 779
PropLoss 0.0086, 2.50 * * 0.0162, 5.14 * * * 0.0254, 5.77 * * * 0.0189, 4.08 * * *
StdSales 0.0535, 12.96 * * * 0.0570, 14.31 * * * 0.0833, 15.91 * * * 0.0872, 14.89 * * *
StdOCF 0.0986, 8.69 * * * 0.0809, 9.98 * * * 0.0748, 7.72 * * * 0.0734, 6.36 * * *
N 1,320 2,414 2,470 2,179
Adj. R 2 0.4268 0.3596 0.3308 0.3005
Panel B: association between total fees and Absolute discretionary accruals b
Intercept 0.4459, 8.67 * * * 0.2253, 16.56 * * * 0.1717, 12.76 * * * 0.1872, 16.53 * * *

Total fees 0.0007, 3.94 * * * 0.0007, 6.95 * * * 0.0006, 6.31 * * * 0.0003, 3.64 * * *
BIG5 0.0186, 0.43 20.0186, 21.51 20.0286, 22.30 * * 20.0455, 24.66 * * *
L1Accrual 20.0069, 22.40 * * 0.0050, 1.98 * * 20.1802, 210.03 * * * 20.0012, 26.78 * * *
LNMVE 20.0167, 22.67 * * * 20.0108, 24.67 * * * 20.0153, 26.03 * * * 20.0061, 22.92 * * *
MERGER 0.1346, 3.11 * * * 0.0220, 1.00 20.0316, 21.15 0.0296, 1.41
FINANCING 0.1259, 6.08 * * * 0.0345, 4.08 * * * 0.0396, 4.34 * * * 0.0341, 4.89 * * *
LEVERAGE 20.5246, 212.22 * * * 20.0042, 20.50 20.0020, 21.75 * 0.0336, 5.05 * * *
MB 20.0003, 20.77 20.0002, 22.12 * * 20.0003, 20.94 0.0002, 1.02
LITIGATION 20.0858, 23.95 * * * 20.0049, 20.56 0.1562, 17.01 * * * 0.0012, 0.19
INSTOWN 20.0009, 23.09 * * * 20.0003, 22.04 * * 20.0002, 21.55 20.0005, 24.34 * * *
LOSS 0.0977, 4.20 * * * 0.0297, 3.29 * * * 0.0724, 8.10 * * * 0.0450, 6.20 * * *
CFO 20.4026, 215.97 * * * 20.1368, 28.25 * * * 20.1036, 28.96 * * * 20.0302, 24.46 * * *
N 2,377 3,909 3,812 3,332
Adj. R 2 0.2612 0.0822 0.3088 0.1279
Notes: aWe regress total fees on a measure of accrual quality using the following model: FLOSAQ ¼
a þ b1 Total fees þ b2 LNTA þ b3 Cycle þ b4 Pr opLoss þ b5 StdSales þ b6 StdCFO þ 1; bWe
regress total fees on the absolute value of REDCA using the following model (similar to Ashbaugh
et al., 2003): ABSREDCA ¼ a þ b1 Total fees þ b2 BIG5 þ b3 L1Accrualþ b4 LNMVE þ b5
MERGER þ b6 FINANCING þ b7 LEVERAGE þ b8 MBþ b9 LITIGATION þ b10 INSTOWN þ b11
LOSS þ b12 CFO þ 1; *significant at 10 percent level; * *significant at 5 percent level; * * *significant Table III.
at 1 percent level. Source: We control for outliers using the procedure suggested by Belsley et al. Association between total
(1980) fees and accruals
Additional tests
Audit and non-audit fees
The principal focus of our paper examines the association between total fees and audit
quality. By using total fees, we assume the impact of audit and non-audit fees on audit
quality is similar. To further enhance our understanding of the effects of client
importance and auditor effort on audit quality, we relax this assumption and extend
our analyses by:
. separating total fees into audit and non-audit fees categories; and
.
incorporating estimates of abnormal audit and non-audit fees, similar to our
analysis when using total fees[14].
We then modify equations (7) and (8) by separating total fees into its audit and
non-audit fee categories.
MAJ
Independent
22,8 variables 2000 2001 2002 2003
Panel A: association between abnormal total fees and accrual quality a

Intercept 0.0054, 0.65 0.0168, 2.19 * * 0.0361, 3.36 * * * 0.0247, 2.20 * *
Abnormal
total fees 0.0061, 5.24 * * * 0.0048, 3.91 * * * 0.0125, 7.24 * * * 0.0109, 5.77 * * *
780 LNTA 20.0040, 28.01 * * * 20.0043, 29.68 * * * 20.0056, 29.09 * * * 20.0054, 28.53 * *
Cycle 0.0108, 7.79 * * * 0.0092, 6.97 * * * 0.0072, 3.92 * * * 0.0095, 4.98 * * *
PropLoss 0.0093, 2.71 * * * 0.0168, 5.35 * * * 0.0267, 6.09 * * * 0.0191, 4.16 * * *
StdSales 0.0536, 13.03 * * * 0.0568, 14.27 * * * 0.0825, 15.77 * * * 0.0850, 14.55 * * *
StdCFO 0.0992, 8.77 * * * 0.0809, 10.00 * * * 0.0741, 7.66 * * * 0.0725, 6.31 * * *
N 1,310 2,414 2,470 2,178
Adj. R 2
0.4314 0.3603 0.3330 0.3036

Panel B: association between abnormal total fees and absolute discretionary accruals b
Intercept 0.4442, 8.66 * * * 0.2375, 17.45 * * * 0.1894, 14.43 * * * 0.1941, 17.34 * * *
Abnormal
total fees 0.0912, 6.55 * * * 0.0516, 8.54 * * * 0.0526, 8.58 * * * 0.0334, 6.52 * * *
BIG5 0.0264, 0.61 20.0187, 21.52 20.0264, 22.15 * * 20.0400, 24.12 * * *
L1Accrual 20.0067, 22.32 * * 0.0035, 1.37 20.0003, 20.65 20.0012, 26.82 * * *
LNMVE 20.0136, 22.26 * * 20.0088, 23.86 * * * 20.0141, 25.65 * * * 20.0062, 22.99 * * *
MERGER 0.1446, 3.35 * * * 0.0383, 1.73 * 20.0192, 20.71 0.0319, 1.52
FINANCING 0.1250, 6.06 * * * 0.0351, 4.14 * * * 0.0429, 4.78 * * * 0.0353, 5.09 * * *
LEVERAGE 20.4876, 211.36 * * * 0.0097, 1.15 20.0001, 20.14 0.0368, 5.57 * * *
MB 20.0003, 20.75 20.0002, 22.18 * * 20.0004, 21.04 0.00002, 1.05
LITIGATION 20.0905, 24.18 * * * 20.0089, 21.03 0.1522, 16.79 * * * 20.0020, 20.29
INSTOWN 20.0009, 23.07 * * * 20.0003, 22.05 * * 20.0002, 21.54 20.0005, 24.35 * * *
LOSS 0.1084, 4.68 * * * 0.0413, 4.59 * * * 0.0768, 8.79 * * * 0.0452, 6.30 * * *
CFO 20.3972, 215.78 * * * 20.1183, 27.11 * * * 20.1034, 229.28 * * * 20.0304, 24.52 * * *
N 2,378 3,911 3,809 3,332
Adj. R 2 0.2665 0.0853 0.3158 0. 1341
Notes: aWe regress total fees on a measure of accrual quality using the following model: FLOSAQ ¼
a þ b1 Abnormal total fees þ b2 LNTA þ b3 Cycle þ b4 Pr opLoss þ b5 StdSales þ b6 StdCFO þ 1;
b
We regress total fees on the absolute value of REDCA using the following model (similar to
Ashbaugh et al., 2003): ABSREDCA ¼ a þ b1 Abnormal total fees þ b2 BIG5 þ b3 L1Accrual þ b4
LNMVE þ b5 MERGER þ b6 FINANCING þ b7 LEVERAGEþ b8 MB þ b9 LITIGATION þ b10
Table IV. INSTOWN þ b11 LOSS þ b12 CFO þ 1; *significant at 10 percent level; * *significant at 5 percent
Association between level; * * *significant at 1 percent level
abnormal total fees and Source: We control for outliers using the procedure suggested by Belsley et al. (1980); we control for
accruals outliers using the procedure suggested by Belsley et al. (1980)
The results (untabulated) show that the coefficient on audit fees is positive and
significant in all years in both models (except for the year 2000 when regressing the
fees on ABSREDCA). The slope coefficient on non-audit fees is positive and significant
in 2002 and 2003 (when regressing the fees on FLOSAQ) and in 2000 and 2001 (when
regressing the fees on ABSREDCA), but not significant in all other years. Though
these results are not as robust as those presented in Table III (the total fee analysis),
they remain consistent with the economic bonding hypothesis. The consistent
significance of the coefficient on audit fees may suggest that economic bonding arises
primarily from audit services and not non-audit services (contrary to the assertions of
regulators and the subsequent passage of SOX). The signs and significance of the
control variable coefficients also remain similar to their counterparts in Table III.
We also jointly examine the association between abnormal audit and abnormal Auditor fees and
non-audit fees, and audit quality and estimate the following models: audit quality
FLOSAQ ¼ a þ b1 Abnormal Audit FEE þ b2 Abnormal Non – Audit FEE
ð9Þ
þ b3 LNTA þ b4 Cycle þ b5 PropLoss þ b6 StdSales þ b7 StdCFO þ 1
781
ABSREDCA ¼a þ b1 Abnormal Audit FEE þ b2 Abnormal Non 2 Audit FEE
þ b3 BIG5 þ b4 L1Accrual þ b5 LNMVE þ b6 MERGER

ð10Þ
þ b7 FINANCING þ b8 LEVERAGE þ b9 MB þ b10 LITIGATION
þ b11 INSTOWN þ b12 LOSS þ b13 CFO þ 1

In untabulated results we find that abnormal audit fees are significantly and positively
associated with both audit quality measures in all years. The coefficient on abnormal
non-audit fees is positive and significant in all years in both models (except for 2001 in
equation (9) and 2002 in equation (10), where it is not significant). These findings are
consistent with our primary results regarding total fees.
We do not observe any noteworthy changes in the signs and significance levels of
all coefficients in the pre- and post-SOX periods. Additionally, the adjusted R 2s are
comparable to those presented in Table IV. These results support the proposition that
economic bonding between auditors and their clients appears to emanate from both
audit and non-audit fees. Given the consistent significance of the coefficient on audit
services, the economic bonding may primarily result from the audit services.
An alternative interpretation of our results is that our results are driven by a
misspecification on the model on non-audit fees, or that they are driven from problem
in the reclassification of non-audit fees. Although we attempt to control for both
potential problems, we chose to present our main results using total fees and not audit
and non-audit fees separately.
The effect of fee reclassification

In November 2000, the SEC issued a directive requiring public companies to disclose
audit and audit-related fees paid to their outside auditors. These fee disclosure rules
became effective for proxy statements filed after February 5, 2001 (SEC Final Rule
S7-13-00, November, 2000). Following SOX, the SEC expanded (and in some instances
re-defined) these disclosure requirements and now mandates that fee paid to auditors
be broken down into the following four categories:
(1) audit fees;
(2) audit-related fees;
(3) tax fees; and
(4) all other fees.
One of the more significant definitional changes under the expanded guidelines relates to
audit fees. The initial rule adopted by the SEC (for proxies filed in 2000) required that
companies disclose fees paid for audits and quarterly reviews in the “audit fees” category.
MAJ The expanded rule requires companies to include any fees for services performed to fulfill
22,8 the accountant’s responsibility under GAAS[15]. Additionally, audit firms are now
prohibited from rendering certain “forbidden services,” such as:
.
financial information system implementation and design;
.
internal audit functions; and
782 .
“other services.”
When comparing audit and non-audit fees before and after the SOX, some concerns
may be raised regarding the comparability of fees given the reclassification and the
changes between the SEC rules in 2000 and 2003. To address the potential effects of
reclassification, we conduct additional analyses.
Fees paid to auditors in 2002 were reported under the old rules in 2002; however, in
2003, the SEC required companies to present fees paid to auditors in 2002 under a new
regulation (to enhance comparability with 2003). Since, fees paid by companies in 2002
were reported under both the old and new rules, we are presented with a unique
opportunity to study the effects of this reclassification. We contacted Standard & Poors
and requested the data as was reported under the old rules for 2002. We then conduct
our primary tests using the original data for 2002 (untabulated) and obtain exactly the
same results as those based on the new 2002 data. Both measures of the audit and
non-audit fees (both deflated and abnormal) are positively and significantly associated
with FLOSAQ in 2002. Additionally, consistent with the results discussed in the
previous section, audit (non-audit) fees are significantly (not significantly) associated
with absolute REDCA.
Additional tests for model specification

We perform several additional tests to ensure the robustness of our results (all
untabulated). First, to allay concerns that our results are driven by our accrual
specification, we re-estimate all of our regressions using accruals generated by the
“modified” Jones model as used in Frankel et al. (2002) and obtain similar results to
those reported in Tables III and IV. Specifically, using the same control variables as in
Tables III and IV, both abnormal total fees and deflated total fees are significant and
positive in all years at the 1 percent level.
Unobserved risk, size effects and validity of statistical assumptions

It is plausible that the reason we find a significant association between abnormal total
audit fees and our two audit quality metrics is that our empirical tests do not
adequately capture the effects of unobserved risk. To address this concern (and
incremental to the control variables already included in our model), we ranked our
sample based on proxies for financial risk, financial health, complexity, earnings
quality, and litigation risk and then re-ran our regressions separately by quintile. Our
results remain qualitatively the same across the entire range of distributions.
Though we have included size as a control variable in all of our analyses, such a
control will seldom fully capture the effect of size. Thus, we perform a sensitivity analysis
whereby we rerun our Table IV regressions (both panels) by size quartile. We partition
our sample into four groups (in each year) based on a firm’s total assets. We observe that
the abnormal total fee variable is positive and significant in all our regressions expect for
the 2003 regression using ABSREDCA for the quartile of large firms.
Replication of results – prior studies Auditor fees and
To ensure that our findings are not sample specific and to reconcile our results with audit quality
prior studies, we replicate the tests documented in Ashbaugh et al. (2003). Since,
Ashbaugh et al. (2003) limit their sample periods to the year 2000, we conduct
replications solely on our year 2000 data. Additionally, we use the FEERATIO
measure to replicate Frankel et al. (2002). Overall, despite minor differences in sample
sizes for the year 2000, our replication results remain consistent with those found in 783
prior literature. Hence, the results presented in the main body of our paper originate
from differences in fee measures and not differences in samples.
Conclusion
In this study, we provide an overview of fees paid to auditors during the period
2000-2003 and their association with audit quality. This timeframe is of particular
interest because fees paid to auditors during this period occurred amidst sweeping
changes in the business, regulatory and professional environment faced by auditors.
We study fees paid to auditors within the context that auditor profitability better
captures the relation between audit quality and auditor independence. Using client size
and estimates of expected fees paid to auditors to proxy for unobservable auditor risk
and effort, we find a statistically significant positive association between total fees and
two measures of audit quality – the standard deviation of residuals from regressions
relating current accruals to cash flows (FLOSAQ) and the absolute value of
performance-adjusted discretionary accruals (ABSREDCA) across all years
(2000-2003). We also conduct similar tests on the separate categories of audit and
non-audit fees and find a positive, but inconsistently significant association with our
two audit quality metrics. Overall, our results are consistent with economic bonding
rather then auditor reputation concerns.
Our study is subject to a number of limitations. First, we utilize accruals to
construct our measure of earnings quality. Using accruals might be a noisy proxy for
management’s discretion over earnings. Though we have attempted to control for the
effects of performance differences on accruals, any association we find between auditor
fees and abnormal accruals could be the result of measurement error rather than a
reflection of management behavior. Second, our sample spans a timeframe consumed
by sweeping business and regulatory changes and that requires public companies to
self-report information about auditor fees (whose classification involves subjective
judgments). Third, we include abnormal fees (total, audit and non-audit) in our
empirical analysis to address concerns relative to client importance and fee
composition. Though we compute abnormal fees using fee prediction models that
appear to be well-specified, we cannot rule out the possibility of an unknown degree of
model misstatement, and omitted variables, on our results. Lastly, it is conceivable that
our results are driven by the inability of our empirical analyses to adequately capture
the impact of unobservable risk. Though we attempt to explore this possibility
employing a variety of alternatives, our results remain qualitatively unchanged. Thus,
we defer further investigation of this limitation to future research.
Notes
1. This measure attempts to assess how well earnings map into cash flows. It utilizes the
standard deviation of residuals from regressions that relate current accruals to cash flows.
For purposes of this study, we refer to this measure as FLOSAQ.
MAJ 2. We also include sensitivity checks to control for the potential impact of the SEC’s change in
mandated fee component disclosures relative to fee paid to auditors by their clients.
22,8
3. “Other services” include bookkeeping; appraisal, valuation services and fairness opinions;
actuarial services; management functions and human resources; broker-dealer, investment
adviser, and investment banking; legal services unrelated to the audit; and any other
services that the PCAOB determines should be restricted (Sarbanes-Oxley Act, 2002).
784 4. Larcker and Richardson (2004) argue that “although RATIO has some intuitive appeal for
measuring the financial linkage between an auditor and client, the size of the fee payments to
the audit firm (both audit and non-audit) is not captured by this measure. That is, a client
with $1 of audit and non-audit payments produces the same score as a client with $10 million
of audit and non-audit payments”.
5. As in Menon and Williams (2004) among others, we also separate accruals into their income
increasing and income decreasing components in the additional tests section.
6. We re-calculate total accruals using cash flows extracted directly from the statement of cash
flows (as suggested by Hribar and Collins, 2002) and obtain qualitatively similar results.
7. The decline in mean total fees between years 2000 and 2001 could potentially be attributed to
the fact that our sample firms are different between these years, e.g. many smaller firms
reported audit and non-audit fees for the first time in 2001.
8. Even though Arthur Andersen did not exist throughout our sample period we use BIG5 to
capture auditing performed by any of the BIG5 auditing firms.
9. Since, our sample spans a turbulent period and is exposed to significant changes in
legislative and reporting requirements, it is likely that expected effort and risk, and hence,
expected fees, will change between the years examined.
10. The adjusted R 2s for our total fees prediction model are consistent with those found in
previous studies.
11. It is generally expected that an auditor with longer tenure will earn a premium. However, our
sample covers a sensitive time period wherein many of Arthur Andersen’s clients were
forced to shop for an auditor (and possibly pay a premium to new auditors with no tenure).
12. We assess the potential presence of multicollinearity by examining the significance of our
independent variables and calculating the Variance Inflation Factors. Our results suggest
that multicollinearity should not impact our results. We also use White’s (1980) consistent
standard errors to control for heteroscedasticity in the error terms for our main results. Our
results remain similar to those reported.
13. Consistent with Hribar and Nichols (2006), we investigate whether there is a difference in the
relationship between our fee variables and income increasing or income decreasing
discretionary accruals, we partition the sample into two groups based on the sign of a firm’s
discretionary accruals (REDCA). We replicate the results presented in Table IV for each of
the two groups and observe that abnormal total fees is positively related to positive
discretionary accruals and negatively related to negative discretionary accruals. Our
significant results indicate that economic bonding, as measured by abnormal total fees, is
associated with both income increasing and income decreasing earning management.
14. We estimate expected or normal audit fees by employing the same methodology used for
total fees. Our results (untabulated) are similar to the results of estimating total fees
presented in Table II.
15. Prior to our 2002 data, many such services were included in the NAS category.
References Auditor fees and
Ashbaugh, H., LaFond, R. and Mayhew, B. (2003), “Do non-audit services compromise auditor audit quality
independence? Further evidence”, The Accounting Review, Vol. 78 No. 3, pp. 611-39.
Becker, C., DeFond, M., Jiambalvo, J. and Subramanyam, K. (1998), “The effect of audit quality on
earnings management”, Contemporary Accounting Research, Vol. 15 No. 1, pp. 1-24.
Belsley, D.A., Kuh, E. and Welsh, R.E. (1980), Regression Diagnostics – Identifying Influential
Data and Sources of Collinearity, Wiley, New York, NY. 785
Bernard, V. and Skinner, D. (1996), “What motivates managers’ choice of discretionary
accruals?”, Journal of Accounting and Economics, Vol. 22 Nos 1/3, pp. 313-25.
Chan, P., Ezzamel, M. and Gwilliam, D. (1993), “Determinants of audit fees for quoted UK
companies”, Journal of Business, Finance and Accounting, Vol. 20 No. 6, pp. 765-86.
Chung, H. and Kallapur, S. (2003), “Client importance, non-audit fees, and abnormal accruals”,
working paper, Purdue University, West Lafayatte, IN.
Craswell, A.T. and Francis, R.J. (1999), “Pricing initial audit engagements: a test of competing
theories”, The Accounting Review, Vol. 74 No. 2, pp. 201-17.
DeAngelo, L.E. (1981), “Auditor size and audit quality”, Journal of Accounting and Economics,
Vol. 3 No. 3, pp. 183-99.
Dechow, P. and Dichev, I. (2002), “The quality of accruals and earnings”, The Accounting Review,
Vol. 77, Supplement, pp. 35-59.
Dechow, P., Sloan, R. and Sweeney, A. (1995), “Detecting earnings management”, The
Accounting Review, Vol. 70 No. 2, pp. 193-225.
DeFond, M.L., Raghunandan, K. and Subramayam, K.R. (2002), “Do non-audit service fees impair
auditor independence? Evidence from going concern opinions”, Journal of Accounting
Research, Vol. 40 No. 4, pp. 1247-74.
Doyle, J., Ge, W. and McVay, S. (2007), “Accrual quality and internal control over financial
reporting”, The Accounting Review, October, Forthcoming.
Fama, E. and French, K. (1997), “Industry costs of equity”, Journal of Financial Economics, Vol. 43
No. 2, pp. 153-93.
Francis, J., LaFond, R., Olsson, P. and Schipper, K. (2005), “The market pricing of accruals
quality”, Journal of Accounting and Economics, Vol. 39 No. 1, pp. 295-327.
Frankel, R., Johnson, M. and Nelson, K. (2002), “The relation between auditors’ fees for non-audit
services and earnings management”, The Accounting Review, Vol. 77, Supplement,
pp. 71-105.
Guay, W., Kothari, S.P. and Watts, R. (1996), “A market-based evaluation of discretionary accrual
models”, Journal of Accounting Research, Vol. 34, pp. 83-105, Supplement.
Hay, D., Knechel, R.W. 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.
Hribar, P. and Collins, D. (2002), “Errors in estimating accruals: implications for empirical
research”, Journal of Accounting Research, Vol. 40 No. 1, pp. 105-34.
Hribar, P. and Nichols, D.C. (2006), “The use of unsigned earnings quality measures in tests of
earnings management”, working paper, Cornell University, Ithaca, NY.
Kasnik, R. (1999), “On the association between voluntary disclosure and earnings management”,
Journal of Accounting Research, Vol. 37 No. 1, pp. 57-81.
Kinney, W. and Libby, R. (2002), “Discussion of the relation between auditors’ fees for nonaudit
services and earnings management”, The Accounting Review, Vol. 77, Supplement,
pp. 107-14.
MAJ Kinney, W., Palmrose, Z. and Scholz, S. (2004), “Auditor independence, non-audit services, and
restatements: was the US Government right?”, Journal of Accounting Research, Vol. 42
22,8 No. 3, pp. 561-88.
Kothari, S., Leone, A. and Wasley, C. (2005), “Performance matched discretionary accrual
measures”, Journal of Accounting and Economics, Vol. 39 No. 1, pp. 163-97.
Larcker, D.F. and Richardson, S.A. (2004), “Fees paid to audit firms, accrual choices, and
786 corporate governance”, Journal of Accounting Research, Vol. 42 No. 3, pp. 625-58.
McNichols, M. (2002), “Discussion of the quality of accruals and earnings: the role of accrual
estimation errors”, The Accounting Review, Vol. 77, pp. 61-9, Supplement.
Magee, R.P. and Tseng, M.C. (1990), “Audit pricing and independence”, The Accounting Review,
Vol. 65 No. 2, pp. 315-36.
Menon, K. and Williams, D. (2004), “Former audit partners and abnormal accruals”, The
Accounting Review, Vol. 79 No. 4, pp. 1095-118.
Raghunandan, K., Read, W.J. and Whisenant, J.S. (2003), “Initial evidence on the association
between nonaudit fees and restated financial statements”, Accounting Horizons, Vol. 17
No. 3, pp. 223-34.
Reynolds, J.K., Deis, D. and Francis, J. (2004), “Professional service fees and auditor objectivity”,
Auditing: A Journal of Practice & Theory, Vol. 21 No. 1, pp. 29-52.
Sarbanes-Oxley Act (2002) P.L. No. 207-204, Government Printing Office, Washington, DC.
SEC (2000), Final Rule: Revision of the Commission’s Auditor Independence Requirements,
Government Printing Office, Securities and Exchange Commission, Washington, DC.
SEC (2003), Final Rule: Strengthening the Commission’s Requirements Regarding Auditor
Independence, Government Printing Office, Securities and Exchange Commission,
Washington, DC.
Simunic, D.A. (1980), “The pricing of audit services: theory and evidence”, Journal of Accounting
Simunic, D.A. (1984), “Auditing, consulting, and auditor independence”, Journal of Accounting
US House of Representatives (2002), Hearings before the Committee on Financial Services, US
House of Representatives, Washington, DC, March 13.
US Senate (2002), Oversight Hearing on Accounting and Investor Protection Issues Raised by
Enron and other Companies, US Senate Committee on Banking and Urban Affairs
(February 12, 26 and 27; March 5, 6, 14), US Senate, Washington, DC.
Whisenant, S., Sankaraguruswamy, S. and Raghunandan, K. (2003), “Evidence on the joint
determination of audit and non-audit fees”, Journal of Accounting Research, Vol. 41 No. 4,
pp. 721-44.
White, H. (1980), “A heteroskedasticity-consistent covariance matrix estimator and a direct test
for heteroskedasticity”, Econometrica, Vol. 48 No. 4, pp. 817-38.
Corresponding author
Rani Hoitash can be contacted at: [email protected]
To purchase reprints of this article please e-mail: [email protected]

Or visit our web site for further details: www.emeraldinsight.com/reprints
This article has been cited by:
1. AlareeniBahaaeddin Ahmed, Bahaaeddin Ahmed Alareeni. The associations between audit firm attributes
and audit quality-specific indicators. Managerial Auditing Journal, ahead of print. [Abstract] [Full Text]
[PDF]
2. Sunhwa Choi, Youn-Sik Choi, Bum-Joon Kim. 2018. Auditors' Strategic Audit Pricing: Evidence from
the Pre- and Post-IFRS Periods. AUDITING: A Journal of Practice & Theory 37:4, 75-94. [Crossref]
3. Khurram Shahzad, Thierry Pouw, Ghulame Rubbaniy, Osama El-Temtamy. 2018. Audit quality during
the global financial crisis: The investors’ perspective. Research in International Business and Finance 45,
94-105. [Crossref]
4. Agyei-MensahBen Kwame, Ben Kwame Agyei-Mensah. The effect of audit committee effectiveness and
audit quality on corporate voluntary disclosure quality. African Journal of Economic and Management
Studies, ahead of print. [Abstract] [Full Text] [PDF]

5. AlhadabMohammad, Mohammad Alhadab. 2018. Abnormal audit fees and accrual and real earnings
management: evidence from UK. Journal of Financial Reporting and Accounting 16:3, 395-416. [Abstract]
[Full Text] [PDF]
6. María Consuelo Pucheta-Martínez, Inmaculada Bel-Oms, Lúcia Lima Rodrigues. 2018. The engagement
of auditors in the reporting of corporate social responsibility information. Corporate Social Responsibility
and Environmental Management 17. . [Crossref]
7. SalehiMahdi, Mahdi Salehi, TarighiHossein, Hossein Tarighi, SafdariSamaneh, Samaneh Safdari. 2018.
The relation between corporate governance mechanisms, executive compensation and audit fees.
Management Research Review 41:8, 939-967. [Abstract] [Full Text] [PDF]
8. Bin Li, Chen Ma. 2018. Can audit fees inhibit accounting misstatements? Moderating effects of auditor
reputation from Chinese experience. Asia-Pacific Journal of Accounting & Economics 88, 1-20. [Crossref]
9. KimSanghun, Sanghun Kim, KimTaewoo, Taewoo Kim, PaeSujin, Sujin Pae, KimSangphill, Sangphill
Kim. 2018. Audit fees via an indirect payment channel and professional skepticism. Managerial Auditing
Journal 33:5, 517-534. [Abstract] [Full Text] [PDF]
10. Maria I. Kyriakou, Augustinos I. Dimitras. 2018. Impact of auditor tenure on audit quality: European
evidence. Investment Management and Financial Innovations 15:1, 374-386. [Crossref]
11. K. Booker. 2018. Can clients of economically dependent auditors benefit from voluntary audit firm
rotation? An experiment with lenders. Research in Accounting Regulation 30:1, 63-67. [Crossref]
12. Geert Braam, Roy Peeters. 2018. Corporate Sustainability Performance and Assurance on Sustainability
Reports: Diffusion of Accounting Practices in the Realm of Sustainable Development. Corporate Social
Responsibility and Environmental Management 25:2, 164-181. [Crossref]
13. Maurizio Rija. Auditing Quality: Some Empirical Studies 3-20. [Crossref]
14. Lars Moratis, Max van Egmond. 2018. Concealing social responsibility? Investigating the relationship
between CSR, earnings management and the effect of industry through quantitative analysis. International
Journal of Corporate Social Responsibility 3:1. . [Crossref]
15. Silvia Ferramosca, Giulio Greco, Marco Allegrini. 2017. External audit and goodwill write-off. Journal of
Management & Governance 21:4, 907-934. [Crossref]
16. Seonghee Lee. 2017. Auditors’ Perception on Audit Risks－Focusing on Audit Environments Related to
Auditor－. Korea International Accounting Review null:76, 51-81. [Crossref]
17. Michael Ettredge, Elizabeth E. Fuerherm, Feng Guo, Chan Li. 2017. Client pressure and auditor
independence: Evidence from the “Great Recession” of 2007–2009. Journal of Accounting and Public Policy
36:4, 262-283. [Crossref]
18. Murya Habbash, Salim Alghamdi. 2017. Audit quality and earnings management in less developed
economies: the case of Saudi Arabia. Journal of Management & Governance 21:2, 351-373. [Crossref]
19. Carlos Miguel Amaral dos Prazeres, Claudio Antonio Figueiredo Pais. The audit quality and the audit
firm' caracteristicts in Portugal 1-6. [Crossref]
20. Vincent Pendehama, Nirupa Padia, Chris Callaghan. 2017. Audit fee premium: The potential effect of
King III. South African Journal of Accounting Research 31:2, 83-101. [Crossref]
21. Giuseppe Iuliano, Gaetano Matonti. 2017. Do big 4 audit companies detect earnings management
and report it in the audit opinion? Empirical evidence from italian non-listed firms. ESPERIENZE
D'IMPRESA :2, 5-43. [Crossref]
22. SalehiMahdi, Mahdi Salehi, JafarzadehAzadeh, Azadeh Jafarzadeh, NourbakhshhosseinyZeinab, Zeinab

Nourbakhshhosseiny. 2017. The effect of audit fees pressure on audit quality during the sanctions in Iran.
International Journal of Law and Management 59:1, 66-81. [Abstract] [Full Text] [PDF]
23. MeckfesselMichele D., Michele D. Meckfessel, SellersDrew, Drew Sellers. 2017. The impact of Big 4
consulting on audit reporting lag and restatements. Managerial Auditing Journal 32:1, 19-49. [Abstract]
[Full Text] [PDF]
24. NeifarSouhir, Souhir Neifar, HaliouiKhamoussi, Khamoussi Halioui, Ben AbdelazizFouad, Fouad Ben
Abdelaziz. 2016. The motivations of earnings management and financial aggressiveness in American firms
listed on the NASDAQ 100. Journal of Applied Accounting Research 17:4, 397-420. [Abstract] [Full Text]
[PDF]
25. Sarowar Hossain, Gary S. Monroe, Mark Wilson, Christine Jubb. 2016. The Effect of Networked Clients'
Economic Importance on Audit Quality. AUDITING: A Journal of Practice & Theory 35:4, 79-103.
[Crossref]
26. NaserKamal, Kamal Naser, HassanYousef Mohammad, Yousef Mohammad Hassan. 2016. Factors
influencing external audit fees of companies listed on Dubai Financial Market. International Journal of
Islamic and Middle Eastern Finance and Management 9:3, 346-363. [Abstract] [Full Text] [PDF]
27. Twaha K. Kaawaase, Mussa Juma Assad, Ernest G Kitindi, Stephen Korutaro Nkundabanyanga. 2016.
Audit quality differences amongst audit firms in a developing economy. Journal of Accounting in Emerging
Economies 6:3, 269-290. [Abstract] [Full Text] [PDF]
28. JungSoo-Jung, Soo-Jung Jung, KimBum-Joon, Bum-Joon Kim, ChungJu-Ryum, Ju-Ryum Chung.
2016. The association between abnormal audit fees and audit quality after IFRS adoption. International
Journal of Accounting & Information Management 24:3, 252-271. [Abstract] [Full Text] [PDF]
29. Domenico Campa, Ray Donnelly. 2016. Non-audit services provided to audit clients, independence of
mind and independence in appearance: latest evidence from large UK listed companies. Accounting and
Business Research 46:4, 422-449. [Crossref]
30. Sherliza Puat Nelson, Nurul Farha Mohamed-Rusdi. 2015. Ownership structures influence on audit fee.
Journal of Accounting in Emerging Economies 5:4, 457-478. [Abstract] [Full Text] [PDF]
31. Vivek Mande, Myungsoo Son. 2015. How Do Auditor Fees Affect Accruals Quality? Additional Evidence.
International Journal of Auditing 19:3, 238-251. [Crossref]
32. Md Khokan Bepari, Abu Taher Mollik. 2015. Effect of audit quality and accounting and finance
backgrounds of audit committee members on firms’ compliance with IFRS for goodwill impairment
testing. Journal of Applied Accounting Research 16:2, 196-220. [Abstract] [Full Text] [PDF]
33. Marius Gros, Christoph Wallek. 2015. Are different stock market transparency requirements associated
with different accounting quality levels? An analysis of bonding effects on the German stock market.
Journal of Business Economics 85:6, 597-633. [Crossref]
34. Hyoung-Joo Lim. 2015. A study on non-audit Service and Audit Quality: focused on the Comparison
between Big4 and Non-Big4 Audit Firm. The Journal of the Korea Contents Association 15:7, 477-488.
[Crossref]
35. Tânia Menezes Montenegro, Filomena Antunes Bras. 2015. Audit quality: does gender composition
of audit firms matter?. Spanish Journal of Finance and Accounting / Revista Española de Financiación y
Contabilidad 44:3, 264-297. [Crossref]
36. Hans-Joachim Böcking, Marius Gros, Daniel Worret. 2015. Enforcement of accounting standards: how
effective is the German two-tier system in detecting earnings management?. Review of Managerial Science
9:3, 431-485. [Crossref]
37. Jung-Kyo Kim, －－－. 2015. The Effect of Characteristics of External Audit on Accounting Enforcement
Actions. Korea International Accounting Review null:60, 309-340. [Crossref]
38. Karen Benson, Peter M Clarkson, Tom Smith, Irene Tutticci. 2015. A review of accounting research in
the Asia Pacific region. Australian Journal of Management 40:1, 36-88. [Crossref]
39. Patrick Krauß, Paul Pronobis, Henning Zülch. 2015. Abnormal audit fees and audit quality: initial evidence
from the German audit market. Journal of Business Economics 85:1, 45-84. [Crossref]
40. Allen D. Blay, Matthew Notbohm, Caren Schelleman, Adrian Valencia. 2014. Audit Quality Effects of an
Individual Audit Engagement Partner Signature Mandate. International Journal of Auditing 18:3, 172-192.
[Crossref]
41. Mahbub Zaman, Jaravee Chayasombat. 2014. Audit pricing and product differentiation in small private
firms: evidence from Thailand. Journal of Accounting in Emerging Economies 4:2, 240-256. [Abstract] [Full
Text] [PDF]
42. Mara Cameran, Pietro Perotti. 2014. Audit Fees and IAS/IFRS Adoption: Evidence from the Banking
Industry. International Journal of Auditing 18:2, 155-169. [Crossref]
43. Arezoo Aghaei Chadegani, Zakiah Muhammaddun Mohamed. 2014. Reporting errors and misstatements:
a measurement for the quality of auditors' work. Asian Journal of Business Ethics 3:1, 83-96. [Crossref]
44. Patrick Krauß, Benedikt M. Quosigk, Henning Zülch. 2014. Effects of Initial Audit Fee Discounts on
Audit Quality: Evidence from Germany. International Journal of Auditing 18:1, 40-56. [Crossref]
45. John Daniel Eshleman, Peng Guo. 2014. Abnormal Audit Fees and Audit Quality: The Importance of
Considering Managerial Incentives in Tests of Earnings Management. AUDITING: A Journal of Practice
& Theory 33:1, 117-138. [Crossref]
46. －－－. 2013. The Effect of Majority Shareholder and Foreign Investor Share on Audit Quality. Korea
International Accounting Review null:52, 377-396. [Crossref]
47. Ariel Markelevich, Rebecca L. Rosner. 2013. Auditor Fees and Fraud Firms. Contemporary Accounting
Research 30:4, 1590-1625. [Crossref]
48. Sarowar Hossain. 2013. Effect of Regulatory Changes on Auditor Independence and Audit Quality.
International Journal of Auditing 17:3, 246-264. [Crossref]
49. Allison K. Beck, Robert M. Fuller, Leah Muriel, Colin D. Reid. 2013. Audit Fees and Investor Perceptions
of Audit Characteristics. Behavioral Research in Accounting 25:2, 71-95. [Crossref]
50. Reiner Quick, Matthias Sattler, Daniela Wiemann. 2013. Agency conflicts and the demand for non‐audit
services. Managerial Auditing Journal 28:4, 323-344. [Abstract] [Full Text] [PDF]
51. Judith van Ravenstein, Georgios Georgakopoulos, Petros Kalantonis, Panagiotis Kaldis. 2013. Does Audit
Quality Influence the Relation between Earnings Management and Internal Control Weakness in the Post
–SOX Period. International Journal of Sustainable Economies Management 2:2, 70-100. [Crossref]
52. Gopal V. Krishnan, Gnanakumar Visvanathan, Wei Yu. 2013. Do Auditor-Provided Tax Services Enhance
or Impair the Value Relevance of Earnings?. The Journal of the American Taxation Association 35:1, 1-19.
[Crossref]
53. A. Brown, J. Ronen. Justifications for Audits of Financial Statements 345-356. [Crossref]
54. Marilena Mironiuc, Ioan-Bogdan Robu. 2012. Empirical Study on the Analysis of the Influence of the
Audit Fees and Non Audit Fees Ratio to the Fraud Risk. Procedia - Social and Behavioral Sciences 62,
179-183. [Crossref]
55. Marsha B. Keune, Karla M. Johnstone. 2012. Materiality Judgments and the Resolution of Detected
Misstatements: The Role of Managers, Auditors, and Audit Committees. The Accounting Review 87:5,
1641-1677. [Crossref]
56. Darius J. Fatemi. 2012. An Experimental Investigation of the Influence of Audit Fee Structure and Auditor
Selection Rights on Auditor Independence and Client Investment Decisions. AUDITING: A Journal of
Practice & Theory 31:3, 75-94. [Crossref]
57. Sharad C. Asthana, Jeff P. Boone. 2012. Abnormal Audit Fee and Audit Quality. AUDITING: A Journal
of Practice & Theory 31:3, 1-22. [Crossref]
58. Reiner Quick. 2012. EC Green Paper Proposals and Audit Quality. Accounting in Europe 9:1, 17-38.
[Crossref]
59. －－－, －－－, JeongSeokWoo. 2011. The Effect of External Audit on the Cost of Debt of Listed and Private
Firms. Korea International Accounting Review null:38, 143-172. [Crossref]
60. .Reiner Quick, Matthias Sattler. 2011. Beeinträchtigen Beratungsleistungen die Urteilsfreiheit des
Abschlussprüfers? Zum Einfluss von Beratungshonoraren auf diskretionäre Periodenabgrenzungen.
Schmalenbachs Zeitschrift für betriebswirtschaftliche Forschung 63:4, 310-343. [Crossref]
61. Arnold Schneider. 2011. Is investment decision-making influenced by perceptions relating to auditors’
client dependence and amount of audit fees?. Advances in Accounting 27:1, 75-80. [Crossref]
62. Reiner Quick, Matthias Sattler. 2011. Das Erfordernis der Umsatzunabhängigkeit und die Konzentration
auf dem deutschen Markt für Abschlussprüferleistungen. Zeitschrift für Betriebswirtschaft 81:1, 61-98.
[Crossref]
63. Antony Young, Yi Wang. 2010. Multi‐risk level examination of going concern modifications. Managerial
Auditing Journal 25:8, 756-791. [Abstract] [Full Text] [PDF]
64. Jean C. Bedard, Karla M. Johnstone, Edward F. Smith. 2010. Audit Quality Indicators: A Status Update
on Possible Public Disclosures and Insights from Audit Practice. Current Issues in Auditing 4:1, C12-
C19. [Crossref]
65. Arnold Schneider. 2010. Do client dependence and amount of audit fees affect lending decisions?.
Managerial Auditing Journal 25:5, 444-457. [Abstract] [Full Text] [PDF]
66. Jerry W. Lin, Mark I. Hwang. 2010. Audit Quality, Corporate Governance, and Earnings Management:
A Meta-Analysis. International Journal of Auditing 14:1, 57-77. [Crossref]
67. Baek, Mi-Kyung, －－－. 2010. Do Foreign Investors Consider Audit Quality at Investment Decision in
Korea Stock Market?. Korea International Accounting Review null:29, 131-159. [Crossref]
68. Fazlina Mohd Salman, Elizabeth Carson. 2009. The Impact of the Sarbanes-Oxley Act on the Audit Fees
of Australian Listed Firms. International Journal of Auditing 13:2, 127-140. [Crossref]
69. Jean C. Bedard, Donald R. Deis, Mary B. Curtis, J. Gregory Jenkins. 2008. Risk Monitoring and Control
in Audit Firms: A Research Synthesis. AUDITING: A Journal of Practice & Theory 27:1, 187-218.
[Crossref]

Managerial Auditing Journal: Article Information

Uploaded by

Copyright:

Available Formats

Managerial Auditing Journal: Article Information

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Managerial Auditing Journal: Article Information

Uploaded by

Copyright:

Available Formats

Managerial Auditing Journal

Auditor fees and audit quality

Downloaded on: 12 January 2019, At: 03:54 (PT)

*Related content and download information correct at time of download.

Auditor fees and

during the period of 2000-2003.

econometric tests that lack power), we find a statistically significant positive

Third, as suggested by Kinney and Libby (2002), we introduce an approach to address

Regulatory landscape, prior research, and empirical predictions

However, if auditor independence violations stem more from auditors’ dependency on

Research design and sample description

Sample description and variable definitions

Measures of audit quality

TCAi;t ¼ at þ b1;i CFOi;t21 þ b1;i CFOi;t þ b1;i CFOi;tþ1 þ b1;i DARi;t

TACCi;t ¼ ðDCAi;t 2 DCLi;t 2 DCASHi;t þ DCDEBTi;t 2 DDEPÞi;t

Variable Year N Mean Median SD

FLOSAQ 2000 1,529 0.061 0.045 0.057

Panel E: variable definitions

Variable Year N Mean Median SD

INVREC Calculated as inventory plus accounts receivable (Data

Abnormal total fees

þ b11 INSTITOWN þ b12 BIG5 þ b13 TENURE þ 1 ð4Þ

The association between total fees and audit quality

INTERCEPT 2.1829, 23.94 * * * 2.6001, 48.43 * * * 2.9506, 58.17 * * * 2.8291, 55.98 * * *

þ b5 StdSales þ b6 StdCFO þ 1 ð5Þ

ABSREDCA ¼ a þ b1 Total fees þ b2 BIG5 þ b3 L1Accrual þ b4 LNMVE

þ b5 MERGER þ b6 FINANCING þ b7 LEVERAGE

ABSREDCA ¼a þ b1 Abnormal total fees þ b2 BIG5 þ b3 L1Accrual þ b4 LNMVE

þ b5 MERGER þ b6 FINANCING þ b7 LEVERAGE þ b8 MB ð8Þ

þ b9 LITIGATION þ b10 INSTOWN þ b11 LOSS þ b12 CFO þ 1

Intercept 0.4459, 8.67 * * * 0.2253, 16.56 * * * 0.1717, 12.76 * * * 0.1872, 16.53 * * *

Panel A: association between abnormal total fees and accrual quality a

0.4314 0.3603 0.3330 0.3036

þ b3 BIG5 þ b4 L1Accrual þ b5 LNMVE þ b6 MERGER

þ b11 INSTOWN þ b12 LOSS þ b13 CFO þ 1

The effect of fee reclassification

Additional tests for model specification

Unobserved risk, size effects and validity of statistical assumptions

To purchase reprints of this article please e-mail: [email protected]

Studies, ahead of print. [Abstract] [Full Text] [PDF]

22. SalehiMahdi, Mahdi Salehi, JafarzadehAzadeh, Azadeh Jafarzadeh, NourbakhshhosseinyZeinab, Zeinab

You might also like