Description
We investigate whether women audit partners earn lower audit fees than their men colleagues. By
examining 2002e2011 fee data for audit engagements in Taiwan, where the names of signing audit
partners are disclosed, we find that audit engagements with women audit partners are related to
significantly lower audit fees than those with men counterparts. Furthermore, we document that such
fee difference is aggravated in the industries with fewer women audit partners, cannot be explained by
the differences in audit quality and audit reporting lags, and is robust to controlling for firm fixed effects.
Our finding provides evidence to support the existence of fee discrimination against women audit
partners in Taiwan's auditing industry. Our results should be of interests to audit firms with regard to
human resource decisions.
Lower audit fees for women audit partners in Taiwan and why
Ting-Chiao Huang
a, *
, Jeng-Ren Chiou
b, 1
, Hua-Wei Huang
b, 1
, Jeng-Fang Chen
b, 1
a
Department of Accounting, Monash University, Australia
b
Department of Accountancy, National Cheng Kung University, Taiwan, ROC
a r t i c l e i n f o
Article history:
Received 27 November 2014
Accepted 11 February 2015
Available online 19 June 2015
Keywords:
Sex discrimination
Auditing industry
Audit fees
Taiwan
a b s t r a c t
We investigate whether women audit partners earn lower audit fees than their men colleagues. By
examining 2002e2011 fee data for audit engagements in Taiwan, where the names of signing audit
partners are disclosed, we ?nd that audit engagements with women audit partners are related to
signi?cantly lower audit fees than those with men counterparts. Furthermore, we document that such
fee difference is aggravated in the industries with fewer women audit partners, cannot be explained by
the differences in audit quality and audit reporting lags, and is robust to controlling for ?rm ?xed effects.
Our ?nding provides evidence to support the existence of fee discrimination against women audit
partners in Taiwan's auditing industry. Our results should be of interests to audit ?rms with regard to
human resource decisions.
© 2015 College of Management, National Cheng Kung University. Production and hosting by Elsevier
Taiwan LLC. All rights reserved.
1. Introduction
The auditing profession around the world came under intense
public scrutiny after the collapse of Enron and the subsequent
demise of its auditor, Arthur Andersen. In the United States, the
government and the regulator have taken unprecedented steps to
restore stability and investors' con?dence in the capital markets.
For example, the U.S. Congress enacted the Sarbanes-Oxley Act of
2002, a stringent rules-based system widely considered to be the
most comprehensive economic regulation since the New Deal.
Yet it would be wrong to think that the auditing industry has
had its day, or to underplay its importance to the capital markets.
The audit ?rms still need to provide independent and objective
tests of the accounting policies, procedures, and subjective judg-
ment used by management inpreparing the ?nancial reports and to
issue audit opinions for the companies. Without the opinions
provided by the audit ?rms, creditors, bankers, investors, and
others cannot use the ?nancial reports with suf?cient con?dence.
In addition to auditing services, the audit ?rms also provide a wide
range of tax, advisory, and other professional services. In 2012,
revenues for the four largest global audit ?rms rose to a record
historic high level of $110 billion, up 6% from 2011. By service line,
auditing services accounted for 45% of total revenues and grew by
2.9% between 2011 and 2012. Tax-related services represented 23%
of total revenues and also rose by 5.6% between 2011 and 2012.
Advisory services have been the fastest growing service line,
however, and grew strongly by 12.2% between 2011 and 2012.
Because of the great demand for auditing services, the profes-
sion continues to attract talent from around the world, and has the
potential to continuously play an important role in the capital
markets. However, if that potential is to be realized, reform is
crucial. Regulators are debating newmethods of oversight to stop a
second audit failure, and to allow the auditing industry to develop
and prosper sustainably.
However, some believe that the process of reform in the
auditing industry will be a wasted opportunity if it does not largely
address the persistent and marked discrimination against women
that seems to permeate this industry. According to the survey
conducted by Schaefer and Zimmer (1995), the average income of
men accountants and auditors exceeds that of their women coun-
terparts by approximately 49%. In addition, the American Institute
of Certi?ed Public Accountants (AICPA) reports that managing di-
rectors are also predominately men, despite the fact that women
have entered the auditing profession in record numbers in recent
decades. In 2009, women made up 55% of newly-hired graduates
and 61.8% of all accountants and auditors. Despite comprising half
the workforce at audit ?rms, women account for only 23% of all
* Corresponding author. Department of Accounting, Monash University, Clayton
Campus, Wellington Road, Clayton, Victoria 3800, Australia.
E-mail addresses: [email protected], [email protected]
(T.-C. Huang).
Peer review under responsibility of College of Management, National Cheng
Kung University.
1
No. 1, University Road, Tainan City 701, Taiwan, ROC.
HOSTED BY
Contents lists available at ScienceDirect
Asia Paci?c Management Review
j ournal homepage: www. el sevi er. com/ l ocat e/ apmrvhttp://dx.doi.org/10.1016/j.apmrv.2015.02.001
1029-3132/© 2015 College of Management, National Cheng Kung University. Production and hosting by Elsevier Taiwan LLC. All rights reserved.
Asia Paci?c Management Review 20 (2015) 219e233
audit partners industry-wide. The statistics suggest that, across the
auditing industry, women are still facing the trials of discrimination
that take the form of unequal pay and lack of advancement in the
job place.
Likewise, the presence of women audit partners in audit
engagement is low in Taiwan. Within our sample, only 30% of the
leading auditors and 30% of the accompanying auditors are
women. This lowpresence of women audit partners is more severe
in the early period. For example, only 14% of the leading auditors
are women in 2002, and the number increases to 36% in 2011.
Based on the statistics provided by the Taiwan Financial Supervi-
sory Commission, the average salaries of the top 30 audit ?rms in
Taiwan are $904,769 NTD for women signing auditors and
$1,156,658 NTD for men signing auditors in self-owned audit ?rms,
and are $510,353 NTD for employed women auditors and $550,680
for employed men auditors. Overall, these statistics suggest that
both the presence and the salaries of women are low in Taiwan's
audit market.
Relatedly, Yang, Chen, and Yang (2013) show that in Taiwan
man-owned audit ?rms outperform woman-owned audit ?rms in
?nancial performance, and even the formers without professional
trainings on auditors have higher ?nancial performance than the
latters with professional trainings. The authors interpret these
?ndings as suggesting “the Chinese cultural values in social roles
against women”. Their results further reinforce that Taiwan's audit
industry is masculine and there exists discrimination against
women.
In this study, we try to extend prior research by further inves-
tigating whether the audited client pays lower audit fees to its
women audit partners. This issue is important because if women
audit partners cannot charge more for the quality services they
provide to a client, it would be hard for a woman to become a
partner or to be promoted. Similarly, if women partners cannot
make more pro?ts for the audit ?rm, they will have less negotiation
?exibility or room to maneuver in regard to salaries. Therefore, the
discrimination against women in relation to audit fees may
partially explain why women face unequal pay or opportunities for
promotion in an audit ?rm. Moreover, if the audited client pays
lower audit fees to its women audit partners, women would have
fewer resources that can be devoted to audit procedures, which
may in turn in?uence audit quality.
To address this issue, we employ a sample of publicly-listed
?rms in Taiwan to examine our research questions. The audit
report in Taiwan contains the names of two signing audit partners
as well as the name of the audit ?rm, in contrast to the U.S., where
the audit report only contains the audit ?rm's name. This provides
an opportunity to investigate the difference in audit fees between
men and women partners. By using a sample for the 2002e2011
period in which audit fees and audit partners' names were
observable, we ?nd that audit engagements with women signing
audit partners are related to signi?cantly lower audit fees than
those with men counterparts, suggesting the existence of
discrimination against women on audit fees in Taiwan's auditing
industry. Furthermore, we document evidence that such discrimi-
nation is more severe in the industries where the presence of
women audit partners is low. Further analyses suggest that the
difference in audit fees cannot be explained by the superior audit
quality of men auditors or by fewer audit hours of women audit
partners. In contrast, we ?nd that women audit partners are
associated with better client earnings quality and longer audit
reporting lags. The relation between women audit partners and
audit fees continues to be signi?cantly negative when we control
for client earnings quality and audit reporting lags. Finally, the
documented lower audit fees for women in this paper are robust to
controlling for ?rm ?xed effects.
In 2011, the PCAOB proposed an auditing standard about the
disclosure of the audit engagement partner (PCAOB, 2005). The
underlying reason would be the belief in that this would enhance
auditor accountability and audit quality. Since auditor sex is a sig-
ni?cant auditor characteristic, this paper provides insightful infor-
mation about the association between auditor sex and audit fees
(audit quality) to global regulators. We also contribute to the
literature examining the sex effects such as CEO, CFO, director, and
auditor sex. Finally, the results in the current study should be of
interest to audit ?rms in human resource decisions.
2. Related research
Sex discrimination has caught the eye of researchers in recent
decades. Many studies have emphasized the impact of sex
discrimination on pay, which is traditionally a vital social welfare
and equality issue (Berik, Rodgers, & Zveglich, 2004; Blinder, 1973;
Corcoran & Duncan, 1979; Goldin, 1990; Jarrell & Stanley, 2004;
Stanley & Jarrell, 1998). To reduce sex discrimination, the Civil
Rights Act of 1964 was enacted by the U.S. Senate and House of
Representatives to eliminate employment discrimination,
including discrimination according to sex, color, and race. More-
over, the Equal Pay Act of 1963 was also enacted with a view to
prohibiting sex differences in wages. However, the 2004 salary
survey conducted by the Institute of Management Accountants
shows that women earn lower wages than men regardless of work
experience. Although Adams and Harte (2000) posited that the
?eld of accounting has the potential by which to discover sex
discrimination, many studies have argued that the accounting
profession itself also has the same concern (Anderson-Gough, Grey,
& Robson, 2005; Marlow & Carter, 2004; Tinker & Fearfull, 2007;
Whiting & Wright, 2001). White and White (2006) also suggested
that women auditors are more likely to be devalued by their clients.
Researchers have indicated that sex in?uences auditors' job
satisfaction and employment, and that the glass ceiling prevents
women from moving higher up the hierarchy in audit ?rms. For
instance, while around 35%e50% of entrants into the ?eld are
women (Lehman, 1992), only 5% of partners in Big 6 accounting
?rms are women (Telberg, 1993). Likewise, 96% of the partners and
management in Australian accounting ?rms are men (Perera,
Fatseas, & Luckett, 1997), and only 3% of partners in the largest
accounting ?rms were found to be women in the 1980s (Hooks &
Cheramy, 1989). Although women entrants exceed men ones
(AICPA, 2008), sex discrimination may still be present in accounting
?rms. Examining Swedish audit industry, Måsson, Elg, and
Jonnergård (2013) indicate that women auditors are less likely to
be promoted. Moreover, they ?nd that having children increases
the likelihood for men auditors to be promoted but decreases the
likelihood for women auditors. Similarly, Jonnergård, Stafsudd, and
Elg (2010) show that in the Swedish audit industry, 50% of the new
employees and 92% of the audit partners are men, and that women
auditors have greater intentions to leave the audit ?rm.
Outside the auditing profession, researchers have also showed
that women executives are likely under-paid. For example, Lam,
McGuinness, and Vieito (2013) suggest that women CEOs receive
less favorable compensation terms than their men colleagues, and
Mohan and Ruggiero (2007) conclude that women CEOs are under-
paid even after controlling for performance. Kulich, Trojanowski,
Ryan, Haslam, and Renneboog (2011) examine the sex effect on
the compensation portfolio, and ?nd that men managers receive
more bonus and performance-sensitive compensation than
women. They interpret these ?ndings as suggesting that compared
to women, men would be more risk-taking. In contrast, Bugeja,
Matolcsy, and Spiropoulos (2012) ?nd insigni?cant relation be-
tween CEO sex and compensation.
T.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 219e233 220
Recent studies suggest that the differences in career promotion
and salaries between men and women auditors cannot be
explained by the differences in performance or independence. For
example, Niskanen, Karjalainen, Niskanen, and Karjalainen (2011)
show that in Finland, clients of women auditors report higher
magnitude of income-decreasing discretionary accruals than that
of men, suggesting that women are more conservative than their
men colleagues.
Recent studies outside the auditing industry also provide no
evidence that men executives outperform women. For example,
Huang and Kisgen (2013) indicate that men CEOs are more over-
con?dent than women, and the announcement returns to the ac-
quisitions made by men CEOs are 2% lower than that by women.
Khan and Vieito (2013) also show that ?rms managed by women
CEOs are less risky than those by men, consistent with Faccio,
Marchica, and Mura (2012) that ?rms with women CEOs have
lower leverage and less volatile earnings, and are more likely to
survive than ?rms with men CEOs. Similarly, Rose (2007) fails to
?nd signi?cant difference in ?rm performance between ?rms with
and without the presence of women directors.
Related business studies indicate that women are likely more
conservative, less risk-taking, and more ethical than men. These
studies generally showthat ?rms are more conservative in ?nancial
reporting when they have women CEOs, CFOs, or directors. For
example, Gul, Hutchinson, and Lai (2013) ?nd a positive association
between the presence of women directors and the accuracy of
analyst earnings forecasts, while Abbott, Parker, and Presley (2012)
showa negative relation between the presence of women directors
and the likelihood of ?nancial restatement. Consistently, Srinidhi,
Gul, and Tsui (2011) ?nd that women directors improve a ?rm's
earnings quality, and Barua, Davidson, Rama, and Thiruvadi (2010)
document that a ?rm's accruals quality is better when the CFO is
woman. Francis, Hasan, Park, and Wu (2009) also showthat women
CFOs are associated with more conservative ?nancial reporting, and
Peni and V€ ah€amaa (2010) ?nd that ?rms with women CFOs exhibit
more income-decreasing discretionary accruals than ?rms with
men CFOs. Gul, Srinidhi, and Ng (2011) ?nd that a ?rm's stock price
re?ects more ?rm-speci?c information and a ?rm's earnings is
more informative under the monitoring of a sex-diverse board,
especially for ?rms with weak corporate governance. Finally, Sun,
Liu, and Lan (2011) fail to ?nd evidence that women directors on
audit committees constrain earnings management, and Ye, Zhang,
and Rezaee (2010) also fail to document signi?cant difference in
earnings quality between ?rms with and without women top
executives.
Overall, the prior studies generally suggest that women are
under-paid and there is no evidence that men outperform women.
We contribute to the literature by providing evidence from Tai-
wan's audit industry that women audit partners earn lower audit
fees than men. Further analyses suggest that such difference in
audit fees cannot be attributed to the differences in audit quality
and audit efforts. Combined together, these results would suggest
that discrimination against women exist in Taiwan's audit industry.
3. Description of data
3.1. Taiwan auditing market
In Taiwan, the ?nancial reports (including the audit opinions) of
public ?rms are currently required to be signed by two individual
audit partners. The Taiwanese Securities and Futures Bureau (TSFB,
similar to the U.S. Securities and Exchange Commission) amended
Article II of the Criteria Governing Approval for Auditing and Cer-
ti?cation of Financial Reports of Public Companies by Certi?ed
Public Accountants (CGAAC) in 1982, which took effect in 1983, and
required that the ?nancial reports of listed ?rms be audited and
signed by two practicing certi?ed public accountants as well as by
the audit ?rm (Chin & Chi, 2009). In addition, the Statement of
Auditing Standards No. 33, “Auditor Report on Financial State-
ments,” requires that audit reports be signed by two independent
auditors and also by the audit ?rm (Accounting and Research
Development Foundation, ARDF, 1999). In contrast to the U.S.,
where the audit reports of public ?rms only disclose the name of
the audit ?rm and its location, the data from Taiwan identify the
names of the two engaged audit partners and that of the audit ?rm.
We use data from all public ?rms with audit fee information in
Taiwan for the period from 2002 to 2011. Financial data, audit ?rm
data, and individual audit partner names are all obtained from the
Taiwan Economic Journal (TEJ) database. The initial sample with
available audit fee data consists of 6052 ?rm-year observations. We
exclude observations with missing auditor information (3), ?nan-
cial ?rms (448), and ?rms with missing ?nancial data (658). Finally,
we have 4943 ?rm-year observations and 1511 unique ?rms for our
regression analysis.
It should be noted that ?rms in Taiwan are not mandated to
disclose the audit fee information. Only when non-audit fees are
higher than 25% of audit fees (Reason 1), when auditors are changed
with a reduction in audit fees (Reason 2), and when audit fees are
decreased by more than 15% compared to the fee for the previous
year (Reason 3) are ?rms required to disclose audit fees. Firms can
also disclose audit fees voluntarily (Reason 4). In the robustness
check, we divide our sample into ?rms that disclose audit fees due
to requirements (Reasons 1e3) or voluntarily (Reason 4).
The sample distribution across years and industries is presented
in Table 1. There is an increase in sample representation during our
sample period (Panel A). For instance, the number of observations
increases from 236 in 2002 to 1015 in 2011. Liao, Wang, and Chi
(2012) indicate that because of the need for audit ?rms in the
adoption of IFRS and because of the encouragement of voluntarily
disclosing audit fee information, the number of ?rms with audit fee
data increases after 2009. In the robustness checks, we restrict our
sample period to 2002e2008 and ?nd consistent results. Liao et al.
(2012) suggest that there is no signi?cant structural difference
between the sample period 2002e2008 and 2009e2010, and that
the sample selection bias is less likely to be large. Our results also
remain unchanged when we restrict the sample period to
2002e2009. Panel B of Table 1 presents the sample distribution by
industry. About 14.48% of the observations (716) operate in the
electronics components industry (TSE Industry Code ¼ 28), and the
remainder are evenly distributed in other industries. In the
regression analyses, we include year and industry indicators to
control for year and industry effects.
3.2. Measure of audit partner sex
We classify the audit partners' sex based on their names, or
make direct contact with audit ?rms to accurately identify auditor
sex sampled in this study. The results are robust to excluding those
observations for which audit partners' sex is unclear. Four in-
dicators are used to proxy for the sex effects of the signing audit
partners: (1) WOMEN equals 1 if at least one of the engaged audit
partners is woman, and 0 otherwise; (2) WOMENNUM denotes the
number of women audit partners engaged with a client, and is
distributed between 0 and 2; (3) CPA1WOMEN equals 1 if the
leading audit partner is woman, and 0 otherwise; (4) CPA2WOMEN
equals 1 if the accompanying audit partner is woman, and
0 otherwise. In the sensitivity test, we employ three additional
measures to capture audit partners' sex after considering the dif-
ferential effects of the leading partners and accompanying partners
and obtain similar results.
T.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 219e233 221
3.3. Distribution of women audit partners in Taiwan auditing
market
Table 1 also presents the distribution of women audit partners
across years and industries. We ?nd that 52% of the observations
have at least one women auditors, the average number of women
auditors is 0.60, and 30% of the leading auditors and the accom-
panying auditors are women. There is an increase in the presence of
women audit partners across years. For example, the average
number of women auditors increases from 0.46 in 2002 to 0.66 in
2011, and the presence of women leading auditors increases from
14% in 2002 to 0.36 in 2011. We also ?nd that, in most industries,
the presence of women audit partners is relatively low and
dispersed. Exceptions are the glass and ceramics industry, the
rubber industry and the oil and gas industry, where more than 75%
of ?rms engage with at least one women audit partner.
In contrast, lower than 40% of ?rms in the electric and ma-
chinery industry, the electrical industry, the papers and pulps in-
dustry, and the miscellaneous electronic industry engage with at
least one women audit partners. Overall, Table 1 suggests that the
presence of women audit partners in Taiwan is low and disperse
across years and industries, and reinforces the importance of con-
trolling for year and industry effects.
3.4. Firm characteristics
Descriptive statistics are presented in Table 2. The mean and the
median values of audit fees (LNAF) are 14.63 and 14.64, respectively,
and that of total assets (LNTA) are 21.76 and 21.57, respectively. Our
sample includes both young and old companies, and the mean
value of age since a ?rm was established (AGE) is 22 years. On
average, the receivables and the inventories account for 34% of total
assets (RECINV), and total liabilities account for 44% of total assets
(LEV). The mean ratio of foreign sales (FOREIGN) is 25%, and on
average current assets are 2.44 times of current liabilities (CUR-
RENT). Among the sample ?rms, 6% are newly listed ?rms (IPO), 36%
are traded over-the-counter (OTC), 8% are emerging stocks (ROTC),
and the remainders are publicly listed ?rms. In our sample, 26% of
sample ?rms suffer from net losses (LOSS), 3% receive going-
concern opinions (GC), 63% receive unclean audit opinions
Table 1
Distribution of women audit partners.
Panel A: Distribution by year
Year N WOMEN WOMENNUM CPA1WOMEN CPA2WOMEN
2002 236 0.43 0.46 0.14 0.32
2003 240 0.45 0.49 0.20 0.29
2004 191 0.50 0.58 0.26 0.32
2005 171 0.53 0.63 0.34 0.29
2006 477 0.49 0.57 0.26 0.31
2007 520 0.51 0.61 0.28 0.33
2008 505 0.50 0.59 0.30 0.29
2009 707 0.51 0.60 0.32 0.28
2010 881 0.54 0.63 0.33 0.31
2011 1015 0.57 0.66 0.36 0.30
Total 4943 0.52 0.60 0.30 0.30
Panel B: Distribution by industry
TSE Industry N WOMEN WOMENNUM CPA1WOMEN CPA2WOMEN
01: Cement 22 0.68 0.77 0.45 0.32
02: Foods 73 0.53 0.55 0.26 0.29
03: Plastics 101 0.60 0.66 0.34 0.33
04: Textiles 151 0.49 0.55 0.26 0.28
05: Electric and Machinery 254 0.37 0.45 0.19 0.26
06: Electrical 47 0.36 0.36 0.23 0.13
08: Glass and Ceramics 19 0.79 0.89 0.58 0.32
09: Paper and Pulp 13 0.23 0.23 0.15 0.08
10: Steel and Iron 141 0.61 0.76 0.29 0.47
11: Rubber 43 0.81 0.95 0.44 0.51
12: Automobile 24 0.71 0.71 0.38 0.33
13: Miscellaneous Electronic 112 0.38 0.41 0.21 0.21
14: Constructions 229 0.62 0.72 0.32 0.40
15: Transportations 76 0.47 0.59 0.34 0.25
16: Tourism 41 0.51 0.54 0.34 0.20
18: Trading 69 0.58 0.70 0.41 0.29
20: Others 224 0.53 0.63 0.31 0.32
21: Chemicals 148 0.53 0.62 0.37 0.25
22: Biotechnology 228 0.50 0.66 0.36 0.31
23: Oil and Gas 32 0.78 0.91 0.53 0.38
24: Semiconductor 492 0.49 0.57 0.29 0.28
25: Computer 369 0.59 0.70 0.36 0.35
26: Photoelectric 488 0.47 0.53 0.27 0.26
27: Communication Network 245 0.60 0.69 0.40 0.30
28: Electronics Components 716 0.47 0.55 0.26 0.29
29: E-Channel Industry 150 0.61 0.69 0.37 0.32
30: Information Services 160 0.56 0.61 0.35 0.26
31: Other Electronics 250 0.50 0.58 0.23 0.35
80: Management of Stock 26 0.54 0.58 0.08 0.50
Total 4943 0.52 0.60 0.30 0.30
T.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 219e233 222
(UNCLEAN), and 4% restate ?nancial statements (RESTATE). Similar
to other developed countries, most (85%) sample ?rms in Taiwan
are audited by Big4 audit ?rms (BIG4). On average, leading and
accompanying audit partners have 11 years of audit experience
(CPA1EXP and CPA2EXP) and have continuously served the client for
3 years (CPA1TENURE and CPA2TENURE). Each client's total assets
are on average 2% of the sum of the total assets of the audit ?rm's
clients (CIFIRM) and 11% of the sumof the total assets of the leading
partner's and the accompanying audit partner's clients (CICPA1 and
CICPA2). There are 7% of sample ?rms changing their engaged audit
?rm in the current year (INITIAL), and the average number of new
audit partners is 0.51 (NEWCPA). The mean and the median values
of the logarithm of non-audit fees (LNNAF) are 11.35 and 13.30,
respectively. Among the sample ?rms, 31% are audited by industry
specialized audit ?rms (EXPERT_FIRM), and 2% are audited by in-
dustry specialized leading or accompanying audit partners
(EXPERT_CPA1 and EXPERT_CPA2). Regarding the reason for audit
fee disclosure, 51% of sample ?rms disclose audit fees because non-
audit fees are higher than 25% of audit fees (REASON1), 2% because
there is an audit ?rm change with a reduction in audit fees
(REASON2), 5% because the reduction in audit fees is more than 15%
(REASON3), and 45% disclose audit fees voluntarily (REASON4). Note
that Reason 1 to 3 are not mutually exclusive. All continuous vari-
ables are winsorized at 1% and 99% to alleviate the problem of
outliers.
We further compare the differences in ?rm characteristics be-
tween ?rms with and without women audit partners (untabu-
lated). It shows that the clients audited by women audit partners
(WOMEN ¼ 1) are older (AGE), are more complex in the number of
related parties (RELATE), consist of fewer IPO ?rms (IPO), are more
likely to be audited by the Big 4 audit ?rms (BIG4) and by less
experienced accompanying auditors (CPA2EXP), involve in fewer
initial audit engagements (INITIAL), have longer audit ?rm tenure
(AFTENURE), and are less likely to be audited by industry specialized
audit ?rms (EXPERT_FIRM) than those by men audit partners
(WOMEN ¼ 0). Coupled with the fact that we ?nd no signi?cant
differences in size (LNTA), auditor specialization (EXPERT_CPA1 and
EXPERT_CPA2), and the reasons for audit fee disclosure (REASON1 to
REASON4), we conclude that our results of the difference in audit
fees between women and men auditors are unlikely driven by the
size effects, the premiums for the Big 4 audit ?rms and industry
specialized auditors, or the underlying reasons for ?rms to disclose
audit fees. Nevertheless, we control these ?rm and audit charac-
teristics in our regression analyses to alleviate such concerns.
4. Women audit partners and audit fees
The main question addressed by this study is whether there is
any evidence that women audit partners are discriminated against
upon providing and charging for their professional services. This
question is important, because women audit partners would be at a
disadvantage in terms of receiving higher compensation and pro-
motion in audit ?rms if they contribute less to audit ?rm earnings.
We do not assume that higher audit fees equal higher pro?ts, since
pro?ts are determined by audit fees, audit efforts, auditor's
compensation for risk, and other factors. However, we have dis-
cussed with audit partners about this issue, and they suggest that to
some extent audit fees are associated with pro?ts. Therefore, we
study the association between women audit partner and audit fees,
after controlling for other determinants.
Table 2
Descriptive statistics (N ¼ 4943).
Variables Mean STD Q1 Median Q3
Logarithm of audit fees (LNAF) 14.63 0.61 14.25 14.64 15.01
Logarithm of total assets (LNTA) 21.76 1.56 20.68 21.57 22.63
Firm age since setup (AGE) 22.22 12.70 12.00 20.00 31.00
Percentage of receivables and inventories (RECINV) 0.34 0.19 0.20 0.33 0.46
Square root of related parties (RELATE) 3.14 1.47 2.00 2.83 3.87
Foreign sales (FOREIGN) 0.25 0.37 0.00 0.00 0.57
Loss (LOSS) 0.26 0.44 0.00 0.00 1.00
Current ratio (CURRENT) 2.44 2.35 1.26 1.74 2.68
Leverage (LEV) 0.44 0.21 0.29 0.44 0.57
Returns on assets (ROA) 0.29 1.51 0.00 0.02 0.14
Going concern opinion (GC) 0.03 0.17 0.00 0.00 0.00
Unclean audit opinion (UNCLEAN) 0.63 0.48 0.00 1.00 1.00
Restatement (RESTATE) 0.04 0.19 0.00 0.00 0.00
Initial public offerings (IPO) 0.06 0.24 0.00 0.00 0.00
Over the counter (OTC) 0.36 0.48 0.00 0.00 1.00
Emerging stock (ROTC) 0.08 0.27 0.00 0.00 0.00
Big 4 audit ?rms (BIG4) 0.85 0.36 1.00 1.00 1.00
CPA1 experience (CPA1EXP) 11.59 5.88 7.00 11.00 16.00
CPA2 experience (CPA2EXP) 11.13 6.25 6.00 11.00 16.00
Initial audit engagements (INITIAL) 0.07 0.26 0.00 0.00 0.00
Number of new auditors (NEWCPA) 0.51 0.67 0.00 0.00 1.00
Logarithm of non-audit fees (LNNAF) 11.35 4.82 11.70 13.30 14.00
Client importance for audit ?rm (CIFIRM) 0.02 0.09 0.00 0.00 0.00
Client importance for CPA1 (CICPA1) 0.11 0.20 0.01 0.03 0.10
Client importance for CPA2 (CICPA2) 0.11 0.21 0.01 0.02 0.09
Audit ?rm tenure (AFTENURE) 10.25 6.71 5.00 9.00 15.00
CPA1 tenure (CPA1TENURE) 3.07 1.80 2.00 3.00 4.00
CPA2 tenure (CPA2TENURE) 2.64 1.57 1.00 2.00 4.00
Audit ?rm industry specialization (EXPERT_FIRM) 0.31 0.46 0.00 0.00 1.00
CPA1 industry specialization (EXPERT_CPA1) 0.02 0.15 0.00 0.00 0.00
CPA2 industry specialization (EXPERT_CPA2) 0.02 0.14 0.00 0.00 0.00
Non-audit fees higher than 25% of audit fees (REASON1) 0.51 0.50 0.00 1.00 1.00
Auditors are changed with a reduction in audit fees (REASON2) 0.02 0.14 0.00 0.00 0.00
Reduction in audit fees more than 15% (REASON3) 0.05 0.22 0.00 0.00 0.00
Voluntary disclosure of audit fees (REASON4) 0.45 0.50 0.00 0.00 1.00
T.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 219e233 223
4.1. Women audit partner and audit fees
We ?rst investigate the relationship between women audit
partners and audit fees. Both uni-variate analysis and regression
analysis are used to examine this association. First, the full sample
observations are classi?ed into two groups: (1) ?rms with zero
women audit partners and (2) ?rms with at least one women audit
partners. We then examine the mean and median differences for
two of the three groups, respectively. Based on t test and Wilcoxon
rank sum test, the untabulated results show that ?rms with zero
women audit partners and ?rms with at least one women audit
partners are close in audit fees. The mean and the median values of
the logarithmof audit fees are $14.62 and $14.63 for ?rms with zero
women audit partners and are $14.65 and $14.65 for ?rms with at
least one women audit partners. The difference in mean is insig-
ni?cant under the t test, while the difference in median is signi?-
cant under the Wilcoxon rank sum test.
Second, we follow prior audit pricing studies (e.g., Huang,
Raghunandan, & Rama, 2009; Kim, Liu, & Zheng, 2012; Dao,
Raghunandan, & Rama, 2012; Fung, Gul, & Krishnan, 2012) to
construct the audit pricing regression model as follows:
The dependent variable is the logarithm of audit fees (LNAF).
GENDER is the indicators of women audit partner sex discussed
above (WOMEN, WOMENNUM, CPA1WOMEN, and CPA2WOMEN).
We predict that b
1
is negative under the sex discrimination
assumption. That is, if women audit partners suffer from inequity,
they are remunerated with lower audit fees.
The model also includes several ?rm-speci?c control variables,
which account for the effects of factors on the cross-sectional dif-
ferences in audit fees. Accordingly, we control for the client size
effect by including the logarithm of total assets (LNTA) and the
number of years since the ?rm was established (AGE) and control
for client complexity by including the percentage of receivables and
inventories over total assets (RECINV), the square root of the
number of related parties (RELATE), and the percentage of foreign
sales (FOREIGN) (Chi, Huang, Liao, & Xie, 2009; Francis, 1984; Fung
et al., 2012; Simunic, 1980). Following Dao et al. (2012) and Kim
et al. (2012), client-speci?c litigation risks and ?nancial condi-
tions are controlled by including an indicator of reporting net loss
(LOSS), the ratio of current assets to current liabilities (CURRENT),
the debt-to-asset ratio (LEV), the return on assets (ROA), an indi-
cator of receiving a going concern opinion (GC), and an indicator of
receiving unclean audit opinions (UNCLEAN), which include un-
quali?ed audit opinions with explanatory notes, and an indicator of
?nancial restatements in the current year (RESTATE). Similar to
Ashbaugh, LaFond, and Mayhew (2003) and Kim et al. (2012), we
include an indicator of initial public offerings (IPO), an indicator of
the over-the-counter market (OTC), and an indicator of the
emerging stock market (ROTC) to control for the needs of additional
audit and consulting services. An indicator of a Big 4 client (BIG4) is
included, since previous studies have documented a Big 4 audit fee
premium (Choi, Kim, Liu, & Simunic, 2008; DeFond, Francis, &
Wong, 2000). In addition to the Big 4 indicator, we also include
the years of the leading audit partner's and the accompanying audit
partner's work experience (CPA1EXP1 and CPA2EXP). The prior
literature suggests an audit fee discount for an initial audit
engagement (Huang et al., 2009; Simon & Francis, 1988;
Whisenant, Sankaraguruswamy, & Raghunandan, 2003), so we
include an indicator of initial audit engagement (INITIAL) and
control for the number of new audit partners engaged (NEWCPA).
Similar to Huang, Liu, Raghunandan, and Rama (2007, 2009) and
Chen, Sun, and Wu (2010), we include the logarithm of non-audit
fees (LNNAF), the ratio of the audited client's total assets over the
sum of the total assets of all the audit ?rm's clients (CIFIRM), the
ratio of the audited client's total assets over the sum of the total
assets of all the leading audit partner's clients (CICPA1) and the
accompanying audit partner's clients (CICPA2), the number of years
of audit ?rm tenure since 1983 (AFTENURE), and the number of
continuous years that the leading audit partner and the accompa-
nying audit partner have been engaged with the client (CPA1TE-
NURE and CPA2TENURE) to capture client importance and the
client-auditor relationship. We also control for audit ?rm's and
auditor's industry specialization (EXPERT_FIRM, EXPERT_CPA1 and
EXPERT_CPA2) given the audit fees premiums for industry special-
ized documented in the prior literature (Francis, Reichelt, & Wang,
2005; Zerni, 2012). Finally, we control for the potential effects of
mandatory audit fee disclosure on audit pricing by including three
indicators of mandatory audit fee disclosure. That is, a ?rm is
required to disclose audit fee information if non-audit fees exceed
25% of audit fees (REASON1), a ?rm changes audit partners leading
to a reduction in audit fees (REASON2), or audit fees are reduced by
more than 15% of the fee for the previous year (REASON3). We also
include year and industry dummies to control for the potential time
and industrial effects on audit fees.
Table 3 reports the main regression analyses. Our audit fee
regression models have suitable explanatory power, and the
adjusted R-squared is around 54%. We ?nd that women audit
partners charge signi?cantly lower audit fees. Speci?cally, audit
fees are lower if there is at least one women audit partner in the
engagement (WOMEN: À0.032, p < 0.01). Similarly, audit fees are
negatively related to the number of women audit partners
(WOMENNUM: À0.026, p < 0.01). Consistent with Chin and Chi
(2009) and Chi and Chin (2011) that leading auditors are likely
more important than accompanying auditors in audit decisions, we
LNAF ¼ b
0
þb
1
GENDER þb
2
LNTA þb
3
AGE þb
4
RECINV þb
5
RELATE
þb
6
FOREIGN þb
7
LOSS þb
8
CURRENT þb
9
LEV þb
10
ROA þb
11
GC
þb
12
UNCLEAN þb
13
RESTATE þb
14
IPO þb
15
OTC þb
16
ROTC þb
17
BIG4
þb
18
CPA1EXP þb
19
CPA2EXP þb
20
INITIAL þb
21
NEWCPA þb
22
LNNAF
þb
23
CIFIRMþb
24
CICPA1 þb
25
CICPA2 þb
26
AFTENURE
þb
27
CPA1TENURE þb
28
CPA2TENURE þb
29
EXPERT FIRM
þb
30
EXPERT CPA1 þb
31
EXPERT CPA2 þb
32
REASON1 þb
33
REASON2
þb
34
REASON3 þYEAR þINDUSTRY þ 3
(1)
T.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 219e233 224
?nd that when the leading audit partner is women, there is a sig-
ni?cant reduction in audit fees (CPA1WOMEN: À0.036, p < 0.01). In
turns of economic signi?cance, Table 3 suggests that the audit fee
discounts for women audit partners are 3.2% when there is at least
one women audit partner, 5.2% when there are two women audit
partners, and 3.6% when the leading audit partner is women.
Overall, the results provide consistent evidence to support the view
that women audit partners are signi?cantly associated with lower
audit fees as compared to men partners.
The results of our control variables are mostly consistent with
the ?ndings of prior studies. The signi?cantly positive coef?cients
of AGE, RELATE, and FOREIGN suggest that older and larger ?rms
with higher complexity pay higher audit fees (p < 0.01), consistent
with the prior studies (Chi et al., 2009; Francis, 1984; Fung et al.,
2012; Simunic, 1980). We also ?nd that ?rms receiving going-
concern opinions (GC) and unclean audit opinions (UNCLEAN) and
?rms restating ?nancial statements (RESTATE) pay higher audit fees
(p < 0.01), consistent with the positive relation between audit risk
and audit fees (Simunic, 1980). BIG4 is signi?cantly positive
(p < 0.01), which re?ects an existing audit fee premium of BIG4
auditors (e.g., Francis &Simon, 1987; Ferguson &Stokes, 2002), and
the signi?cantly positive coef?cient of EXPERT_CPA1 (p < 0.01) also
corresponds to the audit fee premiums for industry specialized
auditors (Francis et al., 2005; Zerni, 2012). DeAngelo (1981) sug-
gested that incumbent auditors are likely to low-ball initial-year
audit fees in order to earn future quasi-rents fromclients. Likewise,
this initial audit fee discount is also discovered in our sample ?rms,
which is supported by a negatively coef?cient of INITIAL (p < 0.15)
and a positive coef?cient of AFTENURE (p < 0.01). Auditor experi-
ence is so valuable that it increases the demand for higher quality
audits (Craswell, Francis, & Taylor, 1995; Ward, Elder, & Kattelus,
1994). As we expected, both CPA1EXP and CPA2EXP are signi?-
cantly positive (p < 0.15). As expected, when a ?rm discloses audit
fees because of audit ?rm change and reduction in audit fees
(REASON1), audit fees are lower (p < 0.01). We are surprised that
the coef?cient of LNTA is negative, and interpret this ?nding as
suggesting that larger ?rms are less risky and have higher bargai-
ning power, resulting in lower audit fees. However, it is hard to
explain why the coef?cient of LOSS is negative.
4.2. Audit fee range
In 2009, Taiwanese ?rms can decide to report audit fee range
instead of the exactly amount of audit fees. To examine whether our
Table 3
Regression results of audit fees and women auditors.
Model Model 1 Model 2 Model 3
Dependent variable LNAF LNAF LNAF
Variables Coef?cient p-value Coef?cient p-value Coef?cient p-value
INTERCEPT 13.620 0.00 13.620 0.00 13.618 0.00
WOMEN À0.032 0.01
WOMENNUM À0.026 0.01
CPA1WOMEN À0.036 0.01
CPA2WOMEN À0.016 0.23
LNTA À0.008 0.08 À0.008 0.07 À0.008 0.07
AGE 0.004 0.00 0.004 0.00 0.004 0.00
RECINV À0.073 0.03 À0.074 0.02 À0.075 0.02
RELATE 0.110 0.00 0.110 0.00 0.110 0.00
FOREIGN 0.092 0.00 0.091 0.00 0.090 0.00
LOSS À0.024 0.10 À0.025 0.10 À0.025 0.09
CURRENT À0.004 0.27 À0.004 0.29 À0.004 0.29
LEV 0.054 0.17 0.055 0.17 0.056 0.16
ROA 0.004 0.35 0.004 0.35 0.004 0.36
GC 0.205 0.00 0.206 0.00 0.207 0.00
UNCLEAN 0.042 0.00 0.042 0.00 0.042 0.00
RESTATE 0.110 0.00 0.109 0.00 0.110 0.00
IPO À0.357 0.00 À0.358 0.00 À0.358 0.00
OTC À0.107 0.00 À0.106 0.00 À0.106 0.00
ROTC À0.322 0.00 À0.321 0.00 À0.321 0.00
BIG4 0.474 0.00 0.475 0.00 0.476 0.00
CPA1EXP 0.004 0.00 0.004 0.00 0.004 0.00
CPA2EXP 0.002 0.11 0.002 0.12 0.002 0.08
INITIAL À0.051 0.14 À0.051 0.14 À0.051 0.14
NEWCPA 0.018 0.16 0.018 0.16 0.019 0.16
LNNAF 0.027 0.00 0.027 0.00 0.027 0.00
CIFIRM À0.774 0.00 À0.776 0.00 À0.772 0.00
CICPA1 0.566 0.00 0.567 0.00 0.568 0.00
CICPA2 0.360 0.00 0.359 0.00 0.357 0.00
AFTENURE 0.007 0.00 0.007 0.00 0.007 0.00
CPA1TENURE 0.000 0.93 0.000 0.95 0.000 0.96
CPA2TENURE 0.001 0.84 0.001 0.83 0.001 0.83
EXPERT_FIRM 0.003 0.82 0.003 0.82 0.003 0.85
EXPERT_CPA1 0.118 0.00 0.117 0.00 0.117 0.00
EXPERT_CPA2 0.062 0.11 0.063 0.10 0.063 0.10
REASON1 À0.150 0.00 À0.150 0.00 À0.150 0.00
REASON2 0.054 0.29 0.053 0.30 0.053 0.31
REASON3 À0.005 0.86 À0.006 0.86 À0.006 0.84
Year Effects Controlled Controlled Controlled
Industry Effects Controlled Controlled Controlled
Clustering Firm-Year Firm-Year Firm-Year
Adjusted R-Square 54.06% 54.07% 54.07%
N 4943 4943 4943
T.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 219e233 225
results are robust to reporting audit fee as ranges, we use ordered
logistic regression models to re-examine the association between
audit fees and women audit partners. Audit fee ranges are coded as
follows: 1 for audit fees lower than $2 million, 2 for audit fees
between $2 million and $4 million, 3 for audit fees between $4
million and $6 million, 4 for audit fees between $6 million and $8
million, 5 for audit fees between $8 million and $10 million, and 6
for audit fees higher than $10 million (NTD). We continue to ?nd
that women audit partners charge lower audit fees than men
counterparts. In particular, the coef?cients of WOMEN, WOMEN-
NUM, CPA1WOMEN are À0.147, À0.118, and À0.130, respectively,
signi?cant at the 0.10 level (untabulated). The untabulated results
are available from the authors.
4.3. Masculine industries
Because discrimination against women would vary with the
extent to which an organization or an industry is masculine or not
(Eagly & Carli, 2003; Eagly, Makhijani, & Klonsky, 1992), we also
study whether the discrimination against women audit partners is
more severe in traditionally masculine industries. We identify
masculine industries as those where lower than 40% of the ?rms in
that industry engage with at least one women audit partner,
including the electric and machinery industry, the electrical in-
dustry, the paper and pulp industry, and the miscellaneous elec-
tronic industry. Our results remain unchanged when we identify
masculine industries as those where lower than 50% of the ?rms in
that industry engage with at least one women audit partner. The
proxies for women auditors are interacted with the indicator of
masculine industries, MIND. MIND equals one if a ?rm operates in
one of the masculine industries discussed above. If discrimination
is more severe in masculine industries, the interaction term of the
proxies for women partners and masculine industry is expected to
be negative.
The results are presented inTable 4. We ?nd signi?cant evidence
that the negative relation between women audit partner and audit
fees is more severe in masculine industries. Speci?cally, the audit
fees are further lower in the masculine industries discussed above
whenthere is at least onewomenaudit partner, whenthe number of
women audit partner increases, and when the leading audit partner
is women. The coef?cient of WOMEN*MIND, WOMENNUM*MIND,
and CPA1WOMEN*MIND are À0.095, À0.075, and À0.169, respec-
tively (p < 0.05). To illustrate the economic signi?cance, Table 4
suggests that the audit fee discounts for women audit partners in-
crease around 9.5%e16.9% in masculine industries. In sum, the re-
sults indicate that women audit partners charge further lower audit
fees in masculine industries.
4.4. Women audit partner and audit quality
There are at least three reasons why women audit partners
charge lower audit fees than men audit partners, including (1)
discrimination against women auditors, (2) the poor audit quality
provided by women audit partners and (3) the ef?ciency of
women audit partners. To analyze the possibility that the latter
two reasons explain our ?ndings, we examine the association
between audit quality, measured as client earnings quality and
audit ef?ciency, and women audit partners. We follow prior
studies (e.g. Al-Ajmi, 2008; Bamber, Bamber, & Schoderbek, 1993;
Chen et al., 2010, 2011; Chi & Chin, 2011; Chin & Chi, 2009;
Henderson & Kaplan, 2000; Knechel & Payne, 2001; Lee,
Mande, & Son, 2008; Leventis & Weetman, 2004; Reichelt &
Wang, 2010) construct the audit quality regression models as
follows.
jDCACCj is the absolute value of discretionary current accruals
(DCACC). As in Chi and Chin (2011), we measure DCACC by esti-
mating the following model for each year-industry grouping.
CACC ¼ b
0
þb
1
INVERSEAT þb
2
DSALE þb
3
ROA þ 3 (4)
CACC is current accruals, de?ned as net income before extraor-
dinary items plus depreciation and amortization minus operating
cash ?ow, divided by lagged total assets. INVERSEAT is 1 divided by
lagged assets. DSALE is the change in sales divided by lagged assets.
ROA is net income divided by lagged assets. We do not deduct the
jDCACCj ¼ b
0
þb
1
GENDER þb
2
LNTA þb
3
LOSS þb
4
ROA þb
5
LEV þb
6
CURRENT
þb
7
AGE þb
8
ZMJSCORE þb
9
OCF þb
10
TACC þb
11
PE þb
12
MB þb
13
RAISE þb
14
IPO
þb
15
OTC þb
16
ROTC þb
17
INITIAL þb
18
NEWCPA þb
19
BIG4
þb
20
EXPERT FIRMþb
21
EXPERT CPA1 þb
22
EXPERT CPA2
þb
23
AFTENURE þb
24
CPA1TENURE þb
25
CPA2TENURE þb
26
CIFIRM
þb
27
CICPA1 þb
28
CICPA2 þb
29
CPA1EXP þb
30
CPA2EXP þb
31
RESTATE
þb
32
UNCLEAN þb
33
GC þYEAR þINDUSTRY þ 3
(2)
LNRLAG ¼ b
0
þb
1
GENDER þb
2
LNTA þb
3
LOSS þb
4
ROA þb
5
LEV þb
6
CURRENTþ
b
7
AGE þb
8
ZMJSCORE þb
9
OCF þb
10
TACC þb
11
PE þb
12
MB þb
13
RAISE þb
14
IPO
þb
15
OTC þb
16
ROTC þb
17
INITIAL þb
18
NEWCPA þb
19
BIG4
þb
20
EXPERT FIRMþb
21
EXPERT CPA1 þb
22
EXPERT CPA2
þb
23
AFTENURE þb
24
CPA1TENURE þb
25
CPA2TENURE þb
26
CIFIRM
þb
27
CICPA1 þb
28
CICPA2 þb
29
CPA1EXP þb
30
CPA2EXP þb
31
RESTATE
þb
32
UNCLEAN þb
33
GC þYEAR þINDUSTRY þ 3
(3)
T.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 219e233 226
change in receivables fromDSALE, and we control for current ROA as
Kothari, Leone, & Wasley (2005). Discretionary current accruals
(DCACC) are the residual values derived (4). LNRLAG is audit
reporting lags, de?ned as the logarithm of the number of days be-
tween the ?scal year end and the ending date of auditor ?eld work
(e.g. Al-Ajmi, 2008; Lee et al., 2008; Leventis & Weetman, 2004). If
the lower audit fees for womenaudit partners are resultedfromthat
women audit partners provide poor audit quality or that women
audit partners are more ef?cient, we predict b
1
to be positive in (2)
and negative in (3).
We control for many ?rm and audit characteristics in?uencing
client earnings quality and audit reporting lags. We control for ?rm
size effects by including the logarithmof total assets (LNTA) and the
number of years since the ?rm was established (AGE) as in Myers,
Myers, and Omer (2003) and Menon and Williams (2004). Firm
performance and the likelihood of ?nancial distress are controlled
by including an indicator of net loss (LOSS), returns on assets (ROA),
leverage (LEV), current ratio (CURRENT), and ?nancial risk scores
(ZMJSCORE) as in Chin and Chi (2009) and Chi and Chin (2011).
Following Carcello, Hermanson, and Huss (1995) and Liu and Wang
(2005), ZMJSCORE is computed as À4.803 À3.6 Â (net income/total
assets) þ 5.4 Â (total debt/total assets) À 0.1 Â (current assets/
current liabilities). Higher ZMJSCORE indicates higher likelihood
of bankruptcy. Lagged total accruals divided by total assets (TACC)
and operating cash ?ows divided by total assets (OCF) are
controlled because of their negative association with the current
accruals (Ashbaugh et al., 2003; Sloan, 1996). Firm growth, market
valuation, and ?nancing activities are controlled by including
price-earnings ratio (PE), market-to-book ratio (MB), and the
?nancing cash ?ows divided by total assets (RAISE) as in the
prior literature (e.g. Chin & Chi, 2009; Jagadison, Aier, Gunlock, &
Lee, 2005; Richardson, Tuna, & Wu, 2003). The superior audit
quality of the Big 4 audit ?rms (BIG4) and industry specialists
(EXPERT_FIRM, EXPERT_CPA1, and EXPERT_CPA2) are controlled
Table 4
Regression results of audit fees, women auditors, and industrial effects.
Model Model 1 Model 2 Model 3
Dependent variable LNAF LNAF LNAF
Variables Coef?cient p-value Coef?cient p-value Coef?cient p-value
INTERCEPT 13.616 0.00 13.614 0.00 13.611 0.00
WOMEN À0.031 0.02
WOMEN*MIND À0.095 0.03
WOMENNUM À0.026 0.01
WOMENNUM*MIND À0.075 0.03
CPA1WOMEN À0.031 0.03
CPA1WOMEN*MIND À0.169 0.00
CPA2WOMEN À0.021 0.14
CPA2WOMEN*MIND 0.009 0.85
MIND À0.029 0.28 À0.032 0.22 À0.032 0.22
LNTA À0.008 0.09 À0.008 0.08 À0.008 0.08
AGE 0.002 0.00 0.002 0.00 0.002 0.00
RECINV À0.088 0.01 À0.089 0.01 À0.089 0.01
RELATE 0.112 0.00 0.112 0.00 0.112 0.00
FOREIGN 0.197 0.00 0.196 0.00 0.195 0.00
LOSS À0.031 0.05 À0.031 0.04 À0.031 0.05
CURRENT À0.005 0.10 À0.005 0.11 À0.005 0.12
LEV 0.055 0.18 0.056 0.17 0.059 0.15
ROA 0.003 0.46 0.003 0.46 0.003 0.46
GC 0.113 0.01 0.115 0.01 0.115 0.01
UNCLEAN 0.048 0.00 0.048 0.00 0.048 0.00
RESTATE 0.076 0.04 0.076 0.04 0.076 0.04
IPO À0.377 0.00 À0.378 0.00 À0.377 0.00
OTC À0.143 0.00 À0.141 0.00 À0.141 0.00
ROTC À0.377 0.00 À0.375 0.00 À0.374 0.00
BIG4 0.511 0.00 0.513 0.00 0.515 0.00
CPA1EXP 0.003 0.01 0.003 0.01 0.003 0.01
CPA2EXP 0.002 0.09 0.002 0.10 0.002 0.07
INITIAL À0.066 0.06 À0.065 0.07 À0.065 0.07
NEWCPA 0.024 0.08 0.024 0.08 0.024 0.07
LNNAF 0.029 0.00 0.029 0.00 0.029 0.00
CIFIRM À0.788 0.00 À0.788 0.00 À0.787 0.00
CICPA1 0.563 0.00 0.564 0.00 0.568 0.00
CICPA2 0.379 0.00 0.378 0.00 0.374 0.00
AFTENURE 0.005 0.00 0.005 0.00 0.005 0.00
CPA1TENURE 0.000 0.93 0.000 0.96 0.000 1.00
CPA2TENURE 0.005 0.27 0.005 0.27 0.005 0.27
EXPERT_FIRM À0.003 0.84 À0.003 0.83 À0.003 0.83
EXPERT_CPA1 0.122 0.00 0.121 0.00 0.120 0.00
EXPERT_CPA2 0.071 0.08 0.071 0.08 0.071 0.08
REASON1 À0.144 0.00 À0.144 0.00 À0.145 0.00
REASON2 0.092 0.09 0.091 0.10 0.092 0.09
REASON3 À0.001 0.98 À0.001 0.97 À0.003 0.93
Year Effects Controlled Controlled Controlled
Industry Effects Controlled Controlled Controlled
Clustering Firm-Year Firm-Year Firm-Year
Adjusted R-Square 51.56% 51.56% 51.61%
N 4943 4943 4943
T.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 219e233 227
(e.g. Reichelt & Wang, 2010). Audit ?rm (AFTENURE) and
auditor tenure (CPA1TENURE and CPA2TENURE) as well as
initial audit engagement (INITIAL) and the number of new
audit partners (NEWCPA) are also controlled given the fact that
audit quality would be improved by longer client-auditor rela-
tionship (Carey & Simnett, 2006; Davis, Soo, & Trompeter, 2009).
We control for client importance (CIFIRM, CICPA1, and CICPA2)
since Chen et al. (2010) suggest that client importance may in?u-
ence auditor independence and thus audit quality. As in the audit
fee models, we control for auditor experience (CPA1EXP and
CPA2EXP), ?nancial restatement (RESTATE), unclean audit opinions
(UNCLEAN), going-concern opinions (GC), and a ?rm's listing status
(IPO, OTC, and ROTC). Year and industry effects are also controlled
for. Likewise, all the continuous variables are winsorized at 1%
and 99%.
In Table 5, we provide the descriptive statistics of the variables
used in (2) and (3). The sample size decreases to 3872 because of
missing necessary data. The mean and median values of discre-
tionary current accruals (DCACC) are 0.09 and 0.05, and that of the
audit reporting lags in logarithm (LNRLAG) are 4.19 and 4.30. On
average, ZMJSCORE is À2.73, operating cash ?ows (OCF) account for
7% of total assets, PE is 36.04, MB is 2.94, and the cash ?ows from
?nancing activities (RAISE) are 2% of total assets. Other variables are
similar as in Table 2.
We then classify our sample ?rms into (1) ?rms with no women
audit partners (N ¼ 1832) and (2) ?rms with at least one women
audit partners (N ¼ 2040), and compare their differences in ?rm
and audit characteristics (untabulated). We ?nd that the clients
with at least one women audit partner (WOMEN ¼ 1) have lower
absolute value of discretionary current accruals (jDCACCj) than
clients with no women audit partners (WOMEN ¼ 0), and that
women audit partners are associated with longer audit reporting
lags (LNRLAG). These ?ndings provide preliminary evidence that
our previous results of the negative association between audit fees
and women auditors are less likely driven by the poor audit quality
and shorter reporting lags by women audit partners. With respect
to control variables, we ?nd that clients with at least one women
audit partners are older (AGE), involve in fewer initial audit en-
gagements (INITIAL), have longer audit ?rmtenure (AFTENURE), and
are less likely to be audited by industry specialized accompanying
auditors (EXPERT_CPA2).
Table 6 reports the regression analyses of audit quality and
women audit partner. The dependent variable is the absolute value
of discretionary current accruals (jDCACCj) in Panel A and is the
logarithm of audit reporting lags (LNRLAG) in Panel B. The explan-
atory power of these regression models (26% in Panel A and 23% in
Panel B) is comparable to that in the prior studies (e.g. 16% in Chi &
Chin, 2011 and 27% in Krishnan & Yang, 2009). In Panel A, we ?nd
that clients audited by women audit partners report better earnings
quality. In particular, the absolute value of discretionary current
accruals (jDCACCj) is lower if there is at least one women audit
partner in the engagement (WOMEN: À0.007, p < 0.05). Likewise,
the number of women audit partners is associated with better
earnings quality (WOMENNUM: À0.005, p < 0.05). As the ?ndings in
Table 2, we ?nd that when the leading audit partner is women, the
client earnings quality is higher (CPA1WOMEN: À0.011, p < 0.01),
suggesting the importance of the leading auditors in audit en-
gagements. These results indicate that women auditors may pro-
vide better audit quality, in terms of higher client earnings quality,
than men auditors, and suggest that the negative relation between
audit fees and women auditors may not be explained by the su-
perior audit quality of men auditors.
Table 5
Descriptive statistics of audit quality model (N ¼ 3872).
Variables Mean STD Q1 Median Q3
Absolute value of discretionary current accruals (jDCACCj) 0.09 0.13 0.02 0.05 0.09
Logarithm of audit reporting lags (LNRLAG) 4.19 0.39 4.04 4.30 4.44
Logarithm of total assets (LNTA) 21.75 1.53 20.70 21.56 22.59
Loss (LOSS) 0.25 0.43 0.00 0.00 0.00
Returns on assets (ROA) 0.28 1.46 0.00 0.02 0.14
Leverage (LEV) 0.44 0.21 0.29 0.44 0.57
Current ratio (CURRENT) 2.46 2.33 1.27 1.76 2.71
Firm age since setup (AGE) 23.63 12.29 14.00 22.00 32.00
Financial distress scores (ZMJSCORE) À2.73 1.55 À3.73 À2.74 À1.89
Operating cash ?ows (OCF) 0.07 0.12 0.01 0.07 0.13
Lagged total accruals (TACC) 0.00 0.10 À0.04 0.01 0.05
Price-to-earnings ratio (PE) 36.04 149.78 0.13 6.67 23.54
Market-to-book ratio (MB) 2.94 3.05 1.13 2.02 3.55
Funds raised (RAISE) 0.02 0.16 À0.06 À0.01 0.07
Initial public offerings (IPO) 0.01 0.11 0.00 0.00 0.00
Over the counter (OTC) 0.33 0.47 0.00 0.00 1.00
Emerging stock (ROTC) 0.07 0.26 0.00 0.00 0.00
Initial audit engagements (INITIAL) 0.06 0.23 0.00 0.00 0.00
Number of new auditors (NEWCPA) 0.53 0.67 0.00 0.00 1.00
Big 4 audit ?rms (BIG4) 0.84 0.36 1.00 1.00 1.00
Audit ?rm industry specialization (EXPERT_FIRM) 0.30 0.46 0.00 0.00 1.00
CPA1 industry specialization (EXPERT_CPA1) 0.02 0.15 0.00 0.00 0.00
CPA2 industry specialization (EXPERT_CPA2) 0.02 0.14 0.00 0.00 0.00
Audit ?rm tenure (AFTENURE) 11.28 6.71 6.00 11.00 16.00
CPA1 tenure (CPA1TENURE) 3.02 1.77 2.00 3.00 4.00
CPA2 tenure (CPA2TENURE) 2.66 1.57 1.00 2.00 4.00
Client importance for audit ?rm (CIFIRM) 0.02 0.09 0.00 0.00 0.00
Client importance for CPA1 (CICPA1) 0.12 0.20 0.01 0.03 0.12
Client importance for CPA2 (CICPA2) 0.12 0.21 0.01 0.03 0.11
CPA1 experience (CPA1EXP) 11.85 5.96 7.00 12.00 16.00
CPA2 experience (CPA2EXP) 11.26 6.27 6.00 11.00 16.00
Restatement (RESTATE) 0.03 0.17 0.00 0.00 0.00
Unclean audit opinion (UNCLEAN) 0.66 0.47 0.00 1.00 1.00
Going concern opinion (GC) 0.03 0.16 0.00 0.00 0.00
T.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 219e233 228
Table 6
Regression results of audit quality and women auditors.
Panel A: Earnings quality and women auditors
Model Model 1 Model 2 Model 3
Dependent variable jDCACCj jDCACCj jDCACCj
Variables Coef?cient p-value Coef?cient p-value Coef?cient p-value
INTERCEPT 0.046 0.23 0.045 0.23 0.044 0.25
WOMEN À0.007 0.04
WOMENNUM À0.005 0.04
CPA1WOMEN À0.011 0.00
CPA2WOMEN 0.000 0.95
LNTA 0.003 0.05 0.003 0.04 0.003 0.04
LOSS 0.002 0.66 0.002 0.68 0.002 0.68
ROA À0.001 0.22 À0.001 0.21 À0.001 0.19
LEV À0.029 0.28 À0.029 0.28 À0.029 0.28
CURRENT À0.001 0.18 À0.001 0.19 À0.001 0.20
AGE 0.000 0.02 0.000 0.02 0.000 0.02
ZMJSCORE 0.001 0.76 0.001 0.75 0.001 0.75
OCF À0.045 0.18 À0.045 0.18 À0.045 0.17
TACC À0.012 0.69 À0.012 0.69 À0.012 0.68
PE 0.000 0.12 0.000 0.11 0.000 0.11
MB 0.005 0.00 0.005 0.00 0.005 0.00
RAISE 0.132 0.00 0.133 0.00 0.133 0.00
IPO 0.068 0.03 0.068 0.03 0.067 0.03
OTC 0.017 0.00 0.017 0.00 0.017 0.00
ROTC 0.002 0.78 0.002 0.79 0.002 0.79
INITIAL 0.024 0.07 0.024 0.07 0.024 0.07
NEWCPA À0.002 0.66 À0.002 0.66 À0.002 0.67
BIG4 À0.014 0.08 À0.014 0.08 À0.013 0.10
EXPERT_FIRM À0.011 0.01 À0.010 0.01 À0.011 0.01
EXPERT_CPA1 0.004 0.79 0.004 0.80 0.005 0.78
EXPERT_CPA2 À0.009 0.49 À0.009 0.50 À0.009 0.50
AFTENURE 0.000 0.50 0.000 0.48 0.000 0.48
CPA1TENURE À0.002 0.05 À0.002 0.06 À0.002 0.06
CPA2TENURE 0.000 0.76 0.000 0.75 0.000 0.74
CIFIRM 0.017 0.61 0.017 0.61 0.020 0.56
CICPA1 À0.032 0.01 À0.032 0.01 À0.032 0.01
CICPA2 À0.004 0.76 À0.004 0.75 À0.006 0.68
CPA1EXP 0.000 0.98 0.000 0.97 0.000 0.73
CPA2EXP 0.000 0.99 0.000 0.99 0.000 0.70
RESTATE À0.007 0.62 À0.007 0.61 À0.007 0.62
UNCLEAN À0.009 0.04 À0.009 0.04 À0.009 0.04
GC 0.028 0.20 0.028 0.20 0.028 0.20
Year Effects Controlled Controlled Controlled
Industry Effects Controlled Controlled Controlled
Clustering Firm-Year Firm-Year Firm-Year
Adjusted R-Square 26.20% 26.20% 26.24%
N 3872 3872 3872
Panel B: Audit reporting lags and women auditors
Model Model 1 Model 2 Model 3
Dependent variable LNRLAG LNRLAG LNRLAG
Variables Coef?cient p-value Coef?cient p-value Coef?cient p-value
INTERCEPT 4.202 0.00 4.211 0.00 4.214 0.00
WOMEN 0.046 0.00
WOMENNUM 0.027 0.00
CPA1WOMEN 0.042 0.00
CPA2WOMEN 0.012 0.33
LNTA 0.004 0.29 0.004 0.30 0.004 0.31
LOSS À0.013 0.46 À0.012 0.50 À0.012 0.50
ROA À0.005 0.23 À0.005 0.26 À0.004 0.27
LEV À0.098 0.25 À0.097 0.25 À0.097 0.26
CURRENT 0.001 0.71 0.001 0.79 0.001 0.80
AGE 0.005 0.00 0.005 0.00 0.005 0.00
ZMJSCORE 0.008 0.54 0.007 0.57 0.007 0.58
OCF À0.415 0.00 À0.417 0.00 À0.415 0.00
TACC 0.009 0.90 0.009 0.90 0.010 0.88
PE 0.000 0.04 0.000 0.04 0.000 0.04
MB À0.009 0.00 À0.009 0.00 À0.009 0.00
RAISE 0.024 0.58 0.022 0.62 0.021 0.63
IPO 0.039 0.50 0.039 0.50 0.040 0.49
OTC 0.026 0.05 0.026 0.05 0.027 0.05
ROTC 0.157 0.00 0.158 0.00 0.158 0.00
(continued on next page)
T.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 219e233 229
In Panel B, we ?nd that women audit partners are more con-
servative in that they take more time to collect audit evidence,
resulting in longer audit reporting lags (LNRLAG). Speci?cally, the
audit reporting lags are signi?cantly longer when there is at least
one women audit partners (WOMEN: 0.046, p < 0.01). Similarly, the
association between audit reporting lags and the number of
women audit partners is also positive and signi?cant (WOMEN-
NUM: 0.027, p < 0.01). We further ?nd that the leading auditors play
a more important role in determining audit reporting lags (CPA1-
WOMEN: 0.042, p < 0.05). These ?ndings show that women audi-
tors are related to longer audit reporting lags, and suggest that the
negative correlation between audit fees and women auditors is not
driven by shorter audit reporting lags required by women audit
partners.
With respect to control variables, we ?nd that ?rms audited by
the Big 4 audit ?rms (BIG4) and industry specialized audit ?rms
(EXPERT_FIRM), ?rms with longer audit ?rm tenure (AFTENURE),
and ?rms receiving non-going-concern opinions (GC) are associ-
ated with higher earnings quality (jDCACCj), consistent with Chi
and Chin (2011). We also ?nd that ?rms audited by the industry
specialized audit ?rms (EXPERT_FIRM) and leading auditors
(EXPERT_CPA1), and ?rms audited by more experienced leading
auditors (CPA1EXP) have shorting audit reporting lags, whereas
?rms restating ?nancial statements (RESTATE) and receiving un-
clean audit opinions and going-concern opinions (GC) have longer
audit reporting lags, consistent with the previous studies (e.g. Lee
et al., 2008; Krishnan & Yang, 2009; Habib & Bhuiyan, 2011).
Overall, the results in Table 6 suggest that it is unlikely that the
differences inaudit qualityandaudit reporting lags betweenwomen
andmenaudit partners drive our ?nding that womenaudit partners
charge lower audit fees than men colleagues. Instead, we ?nd that
women audit partners are associated with better client earnings
quality and longer audit reporting lags, which should increase the
audit fees chargedbywomenaudit partners. Therefore, weconclude
that the reasons why women audit partners charge lower audit fees
than men audit partners in Taiwan are likely due to the masculine
audit industry and the discriminationagainst women. Nevertheless,
we acknowledge that there would be other explanations for the
negative relation between women audit partners and audit fees,
which deserve further research in the future.
As an additional analysis, we augment our audit fee models by
controlling for client earnings quality and audit reporting lags. We
?nd that the negative relation between women audit partners and
audit fees remains signi?cant when we include jDCACCj and
LNRLAG in the regression models (untabulated). In particular, the
coef?cients of WOMEN, WOMENNUM, and CPA1WOMEN
are À0.031, À0.025, and À0.033, respectively (p < 0.05). Therefore,
it is less likely that the negative correlation between audit fees and
women auditors is driven by the differences in audit quality and
audit ef?ciency between women and men audit partners.
4.5. Differences between ?rms with and without audit fee data
Because audit fees are required under certain situations in
Taiwan, it is likely that there is sample bias in this paper, which may
have great in?uence on our results. However, we argue that as long
as there is no signi?cant difference between clients with and
without women audit partners, such bias should not signi?cantly
in?uence our ?ndings. Moreover, it is unclear ex ante whether such
sample bias will lead to a positive or a negative relation between
women audit partners and audit fees. As discussed above, there are
few differences in ?rm and audit characteristics between clients
with and without women audit partners. Rather, untabulated re-
sults suggests that clients with women audit partners are older,
more complex, more likely to be audited by the Big 4 audit ?rms,
involve in fewer initial audit engagements, and have longer audit
?rm tenure, which should lead to higher audit fees. Furthermore,
there are no signi?cant differences in ?rm size, auditor speciali-
zation, and the reasons for audit fee disclosure. Consistently, Liao
et al. (2012) indicate that there is an increase in the number of
Taiwanese ?rms disclosing audit fees after 2009 because of the
demand for the IFRS adoption service provided by audit ?rms and
because of the encouragement of voluntarily disclosing audit fees.
They compare the differences in several ?rm and audit character-
istics between ?rms disclosing audit fees under certain situations
(2002e2008) and ?rms disclosing audit fees voluntarily
Table 6 (continued)
Panel B: Audit reporting lags and women auditors
Model Model 1 Model 2 Model 3
Dependent variable LNRLAG LNRLAG LNRLAG
Variables Coef?cient p-value Coef?cient p-value Coef?cient p-value
INITIAL À0.022 0.51 À0.022 0.50 À0.022 0.50
NEWCPA 0.008 0.50 0.009 0.49 0.008 0.50
BIG4 0.123 0.00 0.123 0.00 0.121 0.00
EXPERT_FIRM À0.101 0.00 À0.102 0.00 À0.101 0.00
EXPERT_CPA1 À0.082 0.07 À0.082 0.08 À0.083 0.07
EXPERT_CPA2 À0.023 0.65 À0.023 0.64 À0.023 0.64
AFTENURE 0.000 0.72 0.000 0.72 0.000 0.72
CPA1TENURE 0.000 0.98 0.000 0.98 0.000 0.96
CPA2TENURE 0.000 0.99 0.000 0.97 0.000 0.97
CIFIRM 0.059 0.54 0.060 0.53 0.053 0.58
CICPA1 À0.063 0.13 À0.063 0.13 À0.065 0.12
CICPA2 0.025 0.49 0.026 0.47 0.030 0.41
CPA1EXP À0.002 0.04 À0.002 0.04 À0.002 0.08
CPA2EXP 0.000 0.71 0.000 0.70 À0.001 0.46
RESTATE 0.144 0.00 0.145 0.00 0.145 0.00
UNCLEAN 0.060 0.00 0.060 0.00 0.060 0.00
GC 0.261 0.00 0.261 0.00 0.261 0.00
Year Effects Controlled Controlled Controlled
Industry Effects Controlled Controlled Controlled
Clustering Firm-Year Firm-Year Firm-Year
Adjusted R-Square 22.94% 22.79% 22.83%
N 3872 3872 3872
T.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 219e233 230
(2009e2010), and conclude that there is no signi?cant structural
change in the determinants of audit fees and that the problem of
selection bias is less likely to be severe. Therefore, it is less likely
that our results are signi?cantly biased by the sample bias. Never-
theless, we construct a sample consisting of ?rms with (N ¼ 4943)
and without (N ¼ 7447) audit fee data, and compare their ?rm and
audit characteristics.
The untabulated results show that there are signi?cant differ-
ences in ?rm and audit characteristics between ?rms with and
without audit fee data. In particular, ?rms with audit fee data are
more likely to be audited by women auditors (WOMEN, WOMEN-
NUM, CPA1WOMEN, and CPA2WOMEN), older (AGE), more complex
(RELATE) and export-oriented (FOREIGN), stronger in ?nancial
conditions (CURRENT and LEV), less likely to restate ?nancial
statements (RESTATE), more likely to receive unclean audit opinions
(UNCLEAN), consist of more newly listed ?rms (IPO, OTC, and ROTC),
are more likely to be audited by the Big 4 audit ?rms (BIG4) and
more experienced auditors (CPA1EXP and CPA2EXP), consists of
fewer initial audit engagements (INITIAL) and new audit partners
(NEWCPA), have higher non-audit fees (LNNAF), have longer audit
?rm tenure (AFTENURE) but shorter leading auditor tenure (CPA1-
TENURE), and are more likely to be audited by industry specialized
audit ?rms (EXPERT_FIRM).
As a robustness check, we control for ?rm ?xed effects in our
audit fee models to alleviate the concern that our results are driven
by some omitted variables. The adjusted R-square increases sub-
stantially to 90% when we control for ?rm ?xed effects, and we
continue to ?nd that women audit partners are associated with
lower audit fees (untabulated). Speci?cally, audit fees are signi?-
cantly lower when there is at least one women auditor in the
engagement (WOMEN: À0.019, p < 0.10). Similarly, audit fees are
signi?cantly and negatively related to the number of women audit
partners (WOMENNUM: À0.016, p < 0.10). We also continue to ?nd
that the leading auditors are more important in audit engagements.
Audit fees are signi?cantly lower when the leading auditor is
women (CPA1WOMEN: À0.028, p < 0.05). Therefore, we conclude
that it is less likely that the sample bias and the omitted variables
result in our ?ndings that women audit partners charge lower audit
fees than men audit partners. Nevertheless, we cannot fully rule out
these possibilities, and acknowledge that our results should be
explained with cautions.
4.6. Sensitivity analysis
Our results are robust to several sensitivity analyses. First, in
order to totally exclude the in?uence of Big 4 audit fee premiums,
we exclude non-Big 4 clients in additional regression tests. All co-
ef?cients of our women audit partner proxies continue to be
negative, and mostly signi?cant. Second, as discussed earlier, ?rms
publicly disclose audit fees for different reasons. Firms that are
required to disclose audit fees (Reasons 1, 2, and 3) may have
distinct characteristics from ?rms that disclose audit fees volun-
tarily (Reason 4). To prevent our results from being driven by these
different characteristics, we conduct separate regression analyses
for the two types of ?rms, and obtain consistent results. Third, we
exclude observations from the electronics industry and obtain
robust evidence. Finally, we employ additional measures of women
audit partners. Speci?cally, these measures consider (1) the dif-
ferential effect between the leading and accompanying audit
partners, (2) the composition of women audit partners, and (3) the
differential effect between one and two women audit partners. Our
results are robust to these different measures. Overall, we ?nd
robust evidence that audit fees are signi?cantly lower for women
audit partners, and this discrimination against women audit part-
ners is more pronounced in masculine industries and cannot be
explained by the differences in audit quality and audit reporting
lags between women and men audit partners.
5. Conclusion
This paper examines whether women audit partners earn lower
audit fees by using a sample of public companies inTaiwan. We ?nd
a signi?cant association between women audit partner and audit
fees, after controlling for the client attributes. This suggests that
discrimination against women audit partners currently exists in
Taiwan. In addition, the discrimination against women is more
severe in masculine industries. Moreover, we provide evidence that
it is unlikely that our results are driven by the differences in audit
quality and audit reporting lags between women and men audit
partners and by the sample bias and omitted variables. The results
provide exploratory insights into the question of how the audit
partner's sex may affect audit pricing. Our results should be of in-
terest to audit ?rms in designing human resource programs and
compensation packages and to regulators in setting labor policies.
Two relevant studies investigating companies in Belgium and in
three Nordic countries found that women audit partners earn
higher audit fees, perhaps due to higher independence (Hardies,
Breesch, & Branson, 2010), better diligence, and less over-
con?dence (Ittonen and Peni, 2012). These ?ndings suggest that sex
stereotypes may vary across cultures and that discrimination
against women in the workplace is less severe in Northern Euro-
pean countries as compared to Taiwan, where Confucianism ad-
vocates the masculine social value that women should stay at home
and be responsible for the household, while men are the bread-
winners and leaders of the family (Chan et al., 2002; Yang et al.,
2013). For example, Penner and Paret (2008) suggested that Asian
kindergarten boys perform best on entering kindergartens, while
Latino kindergarten girls perform better than Latino boys. With
U.S., U.K. and German data, Schein and Mueller (1992) supported
the view that sex differences differ across cultures. They found that
German women and German men were considered to have almost
the same ability to become managers, while, compared to German
women, British women were found to be considered to be less
likely to serve as managers. Besides, this study indicated that U.S.
residents do not express such sex stereotypes in regard to mana-
gerial positions, and both women and men are viewed as equally
capable of becoming managers.
Research related to audit fee differentials in the public ac-
counting profession is likely to impact public policy in the years
ahead. One important question is whether women audit partners
earn less just because they are discriminated. Our results indicate
that, while a pay gap between women and men clearly exists, the
extent of discrimination is neither consistent from industry to in-
dustry, nor from audit partner to audit partner. Since the pattern of
audit fee differences appears to be suf?ciently complex, policy-
makers will ?nd it necessary to approach the problem more ho-
listically. Clearly, there is a need for additional research on this
issue.
Our study is subject to a number of limitations. First, caution is
needed when applying the ?ndings in this paper to other countries,
since the culture and the development of welfare will differ inter-
nationally. However, our results suggest that international regula-
tors should consider mandatory disclosure of audit partner
information including sex, which may effectively signal the severity
of discrimination against women. Second, the explanatory power of
audit fee models in Taiwan is usually lower (Chen & Wu, 2004)
when compared to the adjusted R-square of U.S. audit fee studies. It
cannot be ruled out that our regressions can be made signi?cantly
better by the inclusion of some other omitted variables. Finally,
although we have shown some evidence that the lower audit fees
T.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 219e233 231
for women auditors should not be driven by the differences in audit
quality and audit reporting lags, we cannot fully rule out the pos-
sibility that other factors explain our ?ndings.
References
Abbott, L. J., Parker, S., & Presley, T. J. (2012). Women board presence and the
likelihood of ?nancial restatement. Accounting Horizons, 26(4), 607e629.
Accounting Research Development Foundation (ARDF). (1999). Auditing standard
no.33: Auditor report on ?nancial statements. Taiwan: ARDF.
Adams, C., & Harte, G. (2000). Making discrimination visible: the potential for social
accounting. Accounting Forum, 24(1), 56e79.
AICPA. (2008). Trends in the supply of accounting graduates and the demand for public
accounting recruits. New York: American Institute of Certi?ed Public
Accountants.
Al-Ajmi, J. (2008). Audit and reporting delays: evidence from an emerging market.
Advances in Accounting, 24(2), 217e226.
Anderson-Gough, F., Grey, C., & Robson, K. (2005). “Helping them to forget..”: the
organizational embedding of gender relations in public audit ?rms. Accounting,
Organizations and Society, 30(5), 469e490.
Ashbaugh, H., LaFond, R., & Mayhew, B. W. (2003). Do nonaudit services compro-
mise auditor independence? Further evidence. The Accounting Review, 78(3),
611e639.
Bamber, E. M., Bamber, L. S., & Schoderbek, M. P. (1993). Audit structure and other
determinants of audit report lag: an empirical analysis. Auditing: A Journal of
Practice & Theory, 12(1), 1e23.
Barua, A., Davidson, L. F., Rama, D. V., & Thiruvadi, S. (2010). CFO gender and ac-
cruals quality. Accounting Horizons, 24(1), 25e39.
Berik, G., Rodgers, Y., & Zveglich, J. E. (2004). International trade and gender wage
discrimination: evidence from East Asia. Review of Development Economics, 8(2),
237e254.
Blinder, A. S. (1973). Wage discrimination: reduced form and structural estimates.
Journal of Human Resources, 8(4), 436e455.
Bugeja, M., Matolcsy, Z. P., & Spiropoulos, H. (2012). Is there a gender gap in CEO
compensation? Journal of Corporate Finance, 18(4), 849e859.
Carcello, J. V., Hermanson, D. R., & Huss, H. F. (1995). Temporal changes in
bankruptcy-related reporting. Auditing: A Journal of Practice and Theory, 14(2),
133e143.
Carey, P., & Simnett, R. (2006). Audit partner tenure and audit quality. The Ac-
counting Review, 81(3), 653e676.
Chan, C. L. W., Yip, P. S. F., Ng, E. H. Y., Ho, P. C., Chan, C. H. Y., & Au, J. K. S. (2002).
Gender selection in China: Its meanings and implications. Journal of Assisted
Reproduction and Genetics, 19(9), 426e430.
Chen, H., Chen, J. Z., Lobo, G. J., & Wang, Y. (2011). Effects of audit quality on earnings
management and cost of equity capital: evidence from China. Contemporary
Accounting Research, 28(3), 892e925.
Chen, S., Sun, S. Y. J., & Wu, D. (2010). Client importance, institutional improve-
ments, and audit quality in China: an of?ce and individual auditor level anal-
ysis. The Accounting Review, 85(1), 127e158.
Chen, K. Y., & Wu, S. (2004). Industry specialists, audit fees and auditor size: evi-
dence from Taiwan. Taiwan Accounting Review, 5(2), 41e69.
Chi, H. Y., & Chin, C. L. (2011). Firm versus partner measures of auditor industry
expertise and effects on auditor quality. Auditing: A Journal of Practice and
Theory, 30(2), 201e229.
Chi, W., Huang, H., Liao, Y., & Xie, H. (2009). Mandatory audit partner rotation, audit
quality, and market perception: evidence from Taiwan. Contemporary Ac-
counting Research, 26(2), 359e391.
Chin, C. L., & Chi, H. Y. (2009). Reducing restatements with increased industry
expertise. Contemporary Accounting Research, 26(3), 729e765.
Choi, J. H., Kim, J. B., Liu, X., & Simunic, D. A. (2008). Audit pricing, legal liability
regimes, and Big 4 premiums: theory and cross-country evidence. Contempo-
rary Accounting Research, 25(1), 55e99.
Corcoran, M., & Duncan, G. J. (1979). Work history, labor force attachment and
earnings differences between the races and sexes. Journal of Human Resources,
14(1), 3e20.
Craswell, A. T., Francis, J. R., & Taylor, S. L. (1995). Auditor brand name reputations
and industry specializations. Journal of Accounting and Economics, 20(3),
297e322.
Dao, M., Raghunandan, K., & Rama, D. V. (2012). Shareholder voting on auditor
selection, audit fees, and audit quality. The Accounting Review, 87(1), 149e171.
Davis, L. R., Soo, B. S., & Trompeter, G. M. (2009). Auditor tenure and the ability to
meet or beat earnings forecasts. Contemporary Accounting Research, 26(2),
517e548.
DeAngelo, L. E. (1981). Auditor size and audit quality. Journal of Accounting and
Economics, 3(3), 183e199.
DeFond, M. L., Francis, J. R., & Wong, T. J. (2000). Auditor industry specialization and
market segmentation: evidence from Hong Kong. Auditing: A Journal of Practice
and Theory, 19(1), 49e66.
Eagly, A. H., & Carli, L. L. (2003). The women leadership advantage: an evaluation of
the evidence. The Leadership Quarterly, 14(6), 807e834.
Eagly, A. H., Makhijani, M. G., & Klonsky, G. B. (1992). Gender and the evaluation of
leaders: a meta-analysis. Psychological Bulletin, 111(1), 3e22.
Faccio, M., Marchica, M. T., & Mura, R. (2012). CEO gender, corporate risk-taking, and
the ef?ciency of capital allocation. (Unpublished working paper). Available fromhttp://ssrn.com/abstract¼2021136.
Ferguson, A., & Stokes, D. (2002). Brand name audit pricing, industry specialization,
and leadership premiums post-Big 8 and Big 6 mergers. Contemporary Ac-
counting Research, 19(1), 77e110.
Francis, J. R. (1984). The effect of audit ?rm size on audit prices. Journal of Ac-
counting and Economics, 6(2), 133e151.
Francis, B. B., Hasan, I., Park, J. C., & Wu, Q. (2009). Gender differences in ?nancial
reporting decision making: Evidence from accounting conservatism. (Unpublished
working paper). Available fromhttp://ssrn.com/abstract¼1471059.
Francis, J. R., Reichelt, K., & Wang, D. (2005). The pricing of national and cit-
yespeci?c reputations for industry expertise in the U.S. audit market. The Ac-
counting Review, 80(1), 113e136.
Francis, J. R., & Simon, D. T. (1987). A test of audit ?rm pricing in the small client
segment of the U.S. audit market. The Accounting Review, 62(1), 145e167.
Fung, S. Y. K., Gul, F. A., & Krishnan, J. (2012). City-level auditor industry speciali-
zation, economies of scale, and audit pricing. The Accounting Review, 87(4),
1281e1307.
Goldin, C. (1990). Understanding the gender gap: An economic history of American
women. New York, NY: Oxford University Press.
Gul, F. A., Hutchinson, M., & Lai, K. M. Y. (2013). Gender-diverse boards and prop-
erties of analyst earnings forecasts. Accounting Horizons, 27(3), 511e538.
Gul, F. A., Srinidhi, B., & Ng, A. C. (2011). Does board gender diversity improve the
informativeness of stock prices? Journal of Accounting and Economics, 51(3),
314e338.
Habib, A., & Bhuiyan, Md. B. U. (2011). Audit ?rm industry specialization and the
audit report lag. Journal of International Accounting, Auditing and Taxation, 20(1),
32e44.
Hardies, K., Breesch, D., & Branson, J. (2010). Audit quality: Paying more for a woman?
(Unpublished working paper). Available fromhttp://papers.ssrn.com/sol3/
papers.cfm?abstract_id¼1619911.
Henderson, B. C., &Kaplan, S. E. (2000). Anexaminationof audit report lag for banks: a
panel data approach. Auditing: A Journal of Practice and Theory, 19(2), 159e174.
Hooks, K., & Cheramy, S. J. (1989). Coping with women's expanding role in public
accounting. Journal of Accountancy, 167(2), 66e70.
Huang, J., & Kisgen, D. J. (2013). Gender and corporate ?nance: are male executives
overcon?dent relative to women executives? Journal of Financial Economics,
108(3), 822e839.
Huang, H. W., Liu, L. L., Raghunandan, K., & Rama, D. (2007). Auditor industry
specialization, client bargaining power, and audit fees: further evidence.
Auditing: A Journal of Practice and Theory, 26(1), 147e158.
Huang, H. W., Raghunandan, K., & Rama, D. (2009). Audit fees for initial audit en-
gagements before and after SOX. Auditing: A Journal of Practice and Theory, 28(1),
171e190.
Ittonen, K., & Peni, E. (2012). Auditor's gender and audit fees. International Journal of
Auditing, 16(1), 1e18.
Jagadison, K., Aier, J. C., Gunlock, M. T., & Lee, D. (2005). The ?nancial expertise of
CFOs and accounting restatements. Accounting Horizons, 19(3), 123e135.
Jarrell, S. B., & Stanley, S. D. (2004). Declining bias and gender wage discrimination?
A meta-regression analysis. Journal of Human Resources, 39(3), 828e838.
Jonnergård, K., Stafsudd, A., & Elg, U. (2010). Performance evaluations as gender
barriers in professional organizations: a study of auditing ?rms. Gender, Work
and Organization, 17(6), 721e747.
Khan, W. A., & Vieito, J. P. (2013). CEO gender and ?rm performance. Journal of
Economics and Business, 67(MayeJune), 55e66.
Kim, J. B., Liu, X., & Zheng, L. (2012). The impact of mandatory IFRS adoption on
audit fees: theory and evidence. The Accounting Review, 87(6), 2061e2094.
Knechel, W. R., & Payne, J. L. (2001). Additional evidence on audit report lag.
Auditing: A Journal of Practice and Theory, 20(1), 137e146.
Kothari, S., Leone, A., & Wasley, C. (2005). Performance matched discretionary
accrual measures. Journal of Accounting and Economics, 39(1), 163e197.
Krishnan, J., & Yang, J. S. (2009). Recent trends in audit report and earnings
announcement lags. Accounting Horizons, 23(3), 265e288.
Kulich, C., Trojanowski, G., Ryan, M. K., Haslam, S. A., & Renneboog, L. D. R. (2011).
Who gets the carrot and who gets the stick? Evidence of gender disparities in
executive remuneration. Strategic Management Journal, 32(3), 301e321.
Lam, K. C. K., McGuinness, P. B., & Vieito, J. P. (2013). CEO gender, executive
compensation and ?rm performance in Chinese-listed enterprises. Paci?c-Basin
Finance Journal, 21(1), 1136e1159.
Lee, H. Y., Mande, V., & Son, M. (2008). A comparison of reporting lags of multi-
national and domestic ?rms. Journal of International Financial Management and
Accounting, 19(1), 28e56.
Lehman, C. R. (1992). Herstory in accounting: the ?rst eighty years. Accounting,
Organizations and Society, 17(3e4), 261e297.
Leventis, S., & Weetman, P. (2004). Timeliness of ?nancial reporting: applicability of
disclosure theories in an emerging capital market. Accounting and Business
Research, 34(1), 43e56.
Liao, H. M., Wang, C. C., & Chi, W. C. (2012). What does the complete disclosure of
audit fee information tell us in Taiwan? Taiwan Accounting Review, 8(1), 49e88
(in Chinese).
Liu, C., & Wang, T. (2005). Going-concern opinions: before and after SAS No. 33.
Journal of Management, 22(4), 525e548.
Marlow, S., & Carter, S. (2004). Accounting for change: professional status, gender
disadvantage and self-employment. Women in Management Review, 19(1), 5e17.
T.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 219e233 232
Måsson, J., Elg, U., & Jonnergård, K. (2013). Gender-based career differences among
young auditors in Sweden. International Journal of Manpower, 34(6), 572e583.
Menon, K., & Williams, D. D. (2004). Former audit partners and abnormal accruals.
The Accounting Review, 79(4), 1095e1118.
Mohan, N., & Ruggiero, J. (2007). In?uence of ?rm performance and gender on CEO
compensation. Applied Economics, 39(9), 1107e1113.
Myers, J. N., Myers, L. A., & Omer, T. C. (2003). Exploring the term of the auditor-
client relationship and the quality of earnings: a case for mandatory auditor
rotation. The Accounting Review, 78(3), 779e799.
Niskanen, J., Karjalainen, J., Niskanen, M., & Karjalainen, J. (2011). Auditor gender
and corporate earnings management behavior in private Finnish ?rms. Mana-
gerial Auditing Journal, 26(9), 778e793.
Peni, E., & V€ ah€ amaa, S. (2010). Women executives and earnings management.
Managerial Finance, 36(7), 629e645.
Penner, A. M., & Paret, M. (2008). Gender differences in mathematics achievement:
exploring the early grades and the extremes. Social Science Research, 37(1),
239e253.
Perera, S., Fatseas, V., & Luckett, P. (1997). Performance evaluation of audit assistants
in chartered accounting ?rms. Australian Accounting Review, 7(1), 62e69.
Public Company Accounting Oversight Board (PCAOB). (2005). Improving trans-
parency through disclosure of engagement partner and certain other participants
in audits. Washington, D.C.: PCAOB.
Reichelt, K. J., & Wang, D. (2010). National and of?ce-speci?c measures of auditor
industry expertise and effects on audit quality. Journal of Accounting Research,
48(3), 647e686.
Richardson, S., Tuna, Y., & Wu, M. (2003). Capital market pressures and earnings
management: The case of earnings restatements (Unpublished working paper).
Philadelphia: University of Pennsylvania.
Rose, C. (2007). Does women board representation in?uence ?rm performance?
The Danish evidence. Corporate Governance: An International Review, 15(2),
404e413.
Schaefer, J., & Zimmer, M. (1995). Occupational licensure in the accounting pro-
fession: effects of public regulation on accountants' earnings. Journal of Applied
Business Research, 11(2), 9e16.
Schein, V. E., & Mueller, R. (1992). Sex role stereotyping and requisite management
characteristics: a cross-cultural look. Journal of Organizational Behavior, 13(5),
439e447.
Simon, D. T., & Francis, J. R. (1988). The effects of auditor change on audit fees: tests
of price cutting and price recovery. The Accounting Review, 63(2), 255e269.
Simunic, D. A. (1980). The pricing of audit services: theory and evidence. Journal of
Accounting Research, 18(1), 161e190.
Sloan, R. (1996). Does stock fully re?ect information in accruals and cash ?ows
about future earnings? The Accounting Review, 71(3), 289e315.
Srinidhi, B., Gul, F. A., & Tsui, J. (2011). Women directors and earnings quality.
Contemporary Accounting Research, 28(5), 1610e1644.
Stanley, T. D., & Jarrell, J. B. (1998). Gender wage discrimination bias? A meta-
regression analysis. Journal of Human Resources, 33(4), 947e973.
Sun, J., Liu, G., & Lan, G. (2011). Does women directorship on independent audit
committees constrain earnings management? Journal of Business Ethics, 99(3),
369e382.
Telberg, R. (1993, May 27). Deloitte and touche raises stakes on women's issues.
Accounting Today, p. 3.
Tinker, T., & Fearfull, A. (2007). The workplace politics of U.S. accounting race, class
and gender discrimination at Baruch College. Critical Perspectives on Accounting,
18(1), 123e138.
Ward, D. D., Elder, R. J., & Kattelus, S. C. (1994). Further evidence on the de-
terminants of municipal audit fees. The Accounting Review, 69(2), 399e411.
Whisenant, S., Sankaraguruswamy, S., & Raghunandan, K. (2003). Evidence on the
joint determination of audit and non-audit fees. Journal of Accounting Research,
41(4), 721e744.
White, M. J., & White, G. B. (2006). Implicit and explicit occupational gender ste-
reotypes. Sex Roles, 55(3e4), 259e266.
Whiting, R., & Wright, C. (2001). Explaining gender inequity in the New Zealand
accounting profession. British Accounting Review, 33(2), 191e222.
Yang, Y. F., Chen, Y. S., & Yang, L. W. (2013). Gender gap, training and ?nancial
performance: evidence from public accounting industry. The International
Journal of Human Resource Management, 24(19), 3697e3718.
Ye, K., Zhang, R., & Rezaee, Z. (2010). Does top executive gender diversity affect
earnings quality? A large sample analysis of Chinese listed ?rms. Advances in
Accounting, 26(1), 47e57.
Zerni, M. (2012). Audit partner specialization and audit fees: some evidence from
Sweden. Contemporary Accounting Research, 29(1), 312e340.
T.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 219e233 233
doc_581403973.pdf
We investigate whether women audit partners earn lower audit fees than their men colleagues. By
examining 2002e2011 fee data for audit engagements in Taiwan, where the names of signing audit
partners are disclosed, we find that audit engagements with women audit partners are related to
significantly lower audit fees than those with men counterparts. Furthermore, we document that such
fee difference is aggravated in the industries with fewer women audit partners, cannot be explained by
the differences in audit quality and audit reporting lags, and is robust to controlling for firm fixed effects.
Our finding provides evidence to support the existence of fee discrimination against women audit
partners in Taiwan's auditing industry. Our results should be of interests to audit firms with regard to
human resource decisions.
Lower audit fees for women audit partners in Taiwan and why
Ting-Chiao Huang
a, *
, Jeng-Ren Chiou
b, 1
, Hua-Wei Huang
b, 1
, Jeng-Fang Chen
b, 1
a
Department of Accounting, Monash University, Australia
b
Department of Accountancy, National Cheng Kung University, Taiwan, ROC
a r t i c l e i n f o
Article history:
Received 27 November 2014
Accepted 11 February 2015
Available online 19 June 2015
Keywords:
Sex discrimination
Auditing industry
Audit fees
Taiwan
a b s t r a c t
We investigate whether women audit partners earn lower audit fees than their men colleagues. By
examining 2002e2011 fee data for audit engagements in Taiwan, where the names of signing audit
partners are disclosed, we ?nd that audit engagements with women audit partners are related to
signi?cantly lower audit fees than those with men counterparts. Furthermore, we document that such
fee difference is aggravated in the industries with fewer women audit partners, cannot be explained by
the differences in audit quality and audit reporting lags, and is robust to controlling for ?rm ?xed effects.
Our ?nding provides evidence to support the existence of fee discrimination against women audit
partners in Taiwan's auditing industry. Our results should be of interests to audit ?rms with regard to
human resource decisions.
© 2015 College of Management, National Cheng Kung University. Production and hosting by Elsevier
Taiwan LLC. All rights reserved.
1. Introduction
The auditing profession around the world came under intense
public scrutiny after the collapse of Enron and the subsequent
demise of its auditor, Arthur Andersen. In the United States, the
government and the regulator have taken unprecedented steps to
restore stability and investors' con?dence in the capital markets.
For example, the U.S. Congress enacted the Sarbanes-Oxley Act of
2002, a stringent rules-based system widely considered to be the
most comprehensive economic regulation since the New Deal.
Yet it would be wrong to think that the auditing industry has
had its day, or to underplay its importance to the capital markets.
The audit ?rms still need to provide independent and objective
tests of the accounting policies, procedures, and subjective judg-
ment used by management inpreparing the ?nancial reports and to
issue audit opinions for the companies. Without the opinions
provided by the audit ?rms, creditors, bankers, investors, and
others cannot use the ?nancial reports with suf?cient con?dence.
In addition to auditing services, the audit ?rms also provide a wide
range of tax, advisory, and other professional services. In 2012,
revenues for the four largest global audit ?rms rose to a record
historic high level of $110 billion, up 6% from 2011. By service line,
auditing services accounted for 45% of total revenues and grew by
2.9% between 2011 and 2012. Tax-related services represented 23%
of total revenues and also rose by 5.6% between 2011 and 2012.
Advisory services have been the fastest growing service line,
however, and grew strongly by 12.2% between 2011 and 2012.
Because of the great demand for auditing services, the profes-
sion continues to attract talent from around the world, and has the
potential to continuously play an important role in the capital
markets. However, if that potential is to be realized, reform is
crucial. Regulators are debating newmethods of oversight to stop a
second audit failure, and to allow the auditing industry to develop
and prosper sustainably.
However, some believe that the process of reform in the
auditing industry will be a wasted opportunity if it does not largely
address the persistent and marked discrimination against women
that seems to permeate this industry. According to the survey
conducted by Schaefer and Zimmer (1995), the average income of
men accountants and auditors exceeds that of their women coun-
terparts by approximately 49%. In addition, the American Institute
of Certi?ed Public Accountants (AICPA) reports that managing di-
rectors are also predominately men, despite the fact that women
have entered the auditing profession in record numbers in recent
decades. In 2009, women made up 55% of newly-hired graduates
and 61.8% of all accountants and auditors. Despite comprising half
the workforce at audit ?rms, women account for only 23% of all
* Corresponding author. Department of Accounting, Monash University, Clayton
Campus, Wellington Road, Clayton, Victoria 3800, Australia.
E-mail addresses: [email protected], [email protected]
(T.-C. Huang).
Peer review under responsibility of College of Management, National Cheng
Kung University.
1
No. 1, University Road, Tainan City 701, Taiwan, ROC.
HOSTED BY
Contents lists available at ScienceDirect
Asia Paci?c Management Review
j ournal homepage: www. el sevi er. com/ l ocat e/ apmrvhttp://dx.doi.org/10.1016/j.apmrv.2015.02.001
1029-3132/© 2015 College of Management, National Cheng Kung University. Production and hosting by Elsevier Taiwan LLC. All rights reserved.
Asia Paci?c Management Review 20 (2015) 219e233
audit partners industry-wide. The statistics suggest that, across the
auditing industry, women are still facing the trials of discrimination
that take the form of unequal pay and lack of advancement in the
job place.
Likewise, the presence of women audit partners in audit
engagement is low in Taiwan. Within our sample, only 30% of the
leading auditors and 30% of the accompanying auditors are
women. This lowpresence of women audit partners is more severe
in the early period. For example, only 14% of the leading auditors
are women in 2002, and the number increases to 36% in 2011.
Based on the statistics provided by the Taiwan Financial Supervi-
sory Commission, the average salaries of the top 30 audit ?rms in
Taiwan are $904,769 NTD for women signing auditors and
$1,156,658 NTD for men signing auditors in self-owned audit ?rms,
and are $510,353 NTD for employed women auditors and $550,680
for employed men auditors. Overall, these statistics suggest that
both the presence and the salaries of women are low in Taiwan's
audit market.
Relatedly, Yang, Chen, and Yang (2013) show that in Taiwan
man-owned audit ?rms outperform woman-owned audit ?rms in
?nancial performance, and even the formers without professional
trainings on auditors have higher ?nancial performance than the
latters with professional trainings. The authors interpret these
?ndings as suggesting “the Chinese cultural values in social roles
against women”. Their results further reinforce that Taiwan's audit
industry is masculine and there exists discrimination against
women.
In this study, we try to extend prior research by further inves-
tigating whether the audited client pays lower audit fees to its
women audit partners. This issue is important because if women
audit partners cannot charge more for the quality services they
provide to a client, it would be hard for a woman to become a
partner or to be promoted. Similarly, if women partners cannot
make more pro?ts for the audit ?rm, they will have less negotiation
?exibility or room to maneuver in regard to salaries. Therefore, the
discrimination against women in relation to audit fees may
partially explain why women face unequal pay or opportunities for
promotion in an audit ?rm. Moreover, if the audited client pays
lower audit fees to its women audit partners, women would have
fewer resources that can be devoted to audit procedures, which
may in turn in?uence audit quality.
To address this issue, we employ a sample of publicly-listed
?rms in Taiwan to examine our research questions. The audit
report in Taiwan contains the names of two signing audit partners
as well as the name of the audit ?rm, in contrast to the U.S., where
the audit report only contains the audit ?rm's name. This provides
an opportunity to investigate the difference in audit fees between
men and women partners. By using a sample for the 2002e2011
period in which audit fees and audit partners' names were
observable, we ?nd that audit engagements with women signing
audit partners are related to signi?cantly lower audit fees than
those with men counterparts, suggesting the existence of
discrimination against women on audit fees in Taiwan's auditing
industry. Furthermore, we document evidence that such discrimi-
nation is more severe in the industries where the presence of
women audit partners is low. Further analyses suggest that the
difference in audit fees cannot be explained by the superior audit
quality of men auditors or by fewer audit hours of women audit
partners. In contrast, we ?nd that women audit partners are
associated with better client earnings quality and longer audit
reporting lags. The relation between women audit partners and
audit fees continues to be signi?cantly negative when we control
for client earnings quality and audit reporting lags. Finally, the
documented lower audit fees for women in this paper are robust to
controlling for ?rm ?xed effects.
In 2011, the PCAOB proposed an auditing standard about the
disclosure of the audit engagement partner (PCAOB, 2005). The
underlying reason would be the belief in that this would enhance
auditor accountability and audit quality. Since auditor sex is a sig-
ni?cant auditor characteristic, this paper provides insightful infor-
mation about the association between auditor sex and audit fees
(audit quality) to global regulators. We also contribute to the
literature examining the sex effects such as CEO, CFO, director, and
auditor sex. Finally, the results in the current study should be of
interest to audit ?rms in human resource decisions.
2. Related research
Sex discrimination has caught the eye of researchers in recent
decades. Many studies have emphasized the impact of sex
discrimination on pay, which is traditionally a vital social welfare
and equality issue (Berik, Rodgers, & Zveglich, 2004; Blinder, 1973;
Corcoran & Duncan, 1979; Goldin, 1990; Jarrell & Stanley, 2004;
Stanley & Jarrell, 1998). To reduce sex discrimination, the Civil
Rights Act of 1964 was enacted by the U.S. Senate and House of
Representatives to eliminate employment discrimination,
including discrimination according to sex, color, and race. More-
over, the Equal Pay Act of 1963 was also enacted with a view to
prohibiting sex differences in wages. However, the 2004 salary
survey conducted by the Institute of Management Accountants
shows that women earn lower wages than men regardless of work
experience. Although Adams and Harte (2000) posited that the
?eld of accounting has the potential by which to discover sex
discrimination, many studies have argued that the accounting
profession itself also has the same concern (Anderson-Gough, Grey,
& Robson, 2005; Marlow & Carter, 2004; Tinker & Fearfull, 2007;
Whiting & Wright, 2001). White and White (2006) also suggested
that women auditors are more likely to be devalued by their clients.
Researchers have indicated that sex in?uences auditors' job
satisfaction and employment, and that the glass ceiling prevents
women from moving higher up the hierarchy in audit ?rms. For
instance, while around 35%e50% of entrants into the ?eld are
women (Lehman, 1992), only 5% of partners in Big 6 accounting
?rms are women (Telberg, 1993). Likewise, 96% of the partners and
management in Australian accounting ?rms are men (Perera,
Fatseas, & Luckett, 1997), and only 3% of partners in the largest
accounting ?rms were found to be women in the 1980s (Hooks &
Cheramy, 1989). Although women entrants exceed men ones
(AICPA, 2008), sex discrimination may still be present in accounting
?rms. Examining Swedish audit industry, Måsson, Elg, and
Jonnergård (2013) indicate that women auditors are less likely to
be promoted. Moreover, they ?nd that having children increases
the likelihood for men auditors to be promoted but decreases the
likelihood for women auditors. Similarly, Jonnergård, Stafsudd, and
Elg (2010) show that in the Swedish audit industry, 50% of the new
employees and 92% of the audit partners are men, and that women
auditors have greater intentions to leave the audit ?rm.
Outside the auditing profession, researchers have also showed
that women executives are likely under-paid. For example, Lam,
McGuinness, and Vieito (2013) suggest that women CEOs receive
less favorable compensation terms than their men colleagues, and
Mohan and Ruggiero (2007) conclude that women CEOs are under-
paid even after controlling for performance. Kulich, Trojanowski,
Ryan, Haslam, and Renneboog (2011) examine the sex effect on
the compensation portfolio, and ?nd that men managers receive
more bonus and performance-sensitive compensation than
women. They interpret these ?ndings as suggesting that compared
to women, men would be more risk-taking. In contrast, Bugeja,
Matolcsy, and Spiropoulos (2012) ?nd insigni?cant relation be-
tween CEO sex and compensation.
T.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 219e233 220
Recent studies suggest that the differences in career promotion
and salaries between men and women auditors cannot be
explained by the differences in performance or independence. For
example, Niskanen, Karjalainen, Niskanen, and Karjalainen (2011)
show that in Finland, clients of women auditors report higher
magnitude of income-decreasing discretionary accruals than that
of men, suggesting that women are more conservative than their
men colleagues.
Recent studies outside the auditing industry also provide no
evidence that men executives outperform women. For example,
Huang and Kisgen (2013) indicate that men CEOs are more over-
con?dent than women, and the announcement returns to the ac-
quisitions made by men CEOs are 2% lower than that by women.
Khan and Vieito (2013) also show that ?rms managed by women
CEOs are less risky than those by men, consistent with Faccio,
Marchica, and Mura (2012) that ?rms with women CEOs have
lower leverage and less volatile earnings, and are more likely to
survive than ?rms with men CEOs. Similarly, Rose (2007) fails to
?nd signi?cant difference in ?rm performance between ?rms with
and without the presence of women directors.
Related business studies indicate that women are likely more
conservative, less risk-taking, and more ethical than men. These
studies generally showthat ?rms are more conservative in ?nancial
reporting when they have women CEOs, CFOs, or directors. For
example, Gul, Hutchinson, and Lai (2013) ?nd a positive association
between the presence of women directors and the accuracy of
analyst earnings forecasts, while Abbott, Parker, and Presley (2012)
showa negative relation between the presence of women directors
and the likelihood of ?nancial restatement. Consistently, Srinidhi,
Gul, and Tsui (2011) ?nd that women directors improve a ?rm's
earnings quality, and Barua, Davidson, Rama, and Thiruvadi (2010)
document that a ?rm's accruals quality is better when the CFO is
woman. Francis, Hasan, Park, and Wu (2009) also showthat women
CFOs are associated with more conservative ?nancial reporting, and
Peni and V€ ah€amaa (2010) ?nd that ?rms with women CFOs exhibit
more income-decreasing discretionary accruals than ?rms with
men CFOs. Gul, Srinidhi, and Ng (2011) ?nd that a ?rm's stock price
re?ects more ?rm-speci?c information and a ?rm's earnings is
more informative under the monitoring of a sex-diverse board,
especially for ?rms with weak corporate governance. Finally, Sun,
Liu, and Lan (2011) fail to ?nd evidence that women directors on
audit committees constrain earnings management, and Ye, Zhang,
and Rezaee (2010) also fail to document signi?cant difference in
earnings quality between ?rms with and without women top
executives.
Overall, the prior studies generally suggest that women are
under-paid and there is no evidence that men outperform women.
We contribute to the literature by providing evidence from Tai-
wan's audit industry that women audit partners earn lower audit
fees than men. Further analyses suggest that such difference in
audit fees cannot be attributed to the differences in audit quality
and audit efforts. Combined together, these results would suggest
that discrimination against women exist in Taiwan's audit industry.
3. Description of data
3.1. Taiwan auditing market
In Taiwan, the ?nancial reports (including the audit opinions) of
public ?rms are currently required to be signed by two individual
audit partners. The Taiwanese Securities and Futures Bureau (TSFB,
similar to the U.S. Securities and Exchange Commission) amended
Article II of the Criteria Governing Approval for Auditing and Cer-
ti?cation of Financial Reports of Public Companies by Certi?ed
Public Accountants (CGAAC) in 1982, which took effect in 1983, and
required that the ?nancial reports of listed ?rms be audited and
signed by two practicing certi?ed public accountants as well as by
the audit ?rm (Chin & Chi, 2009). In addition, the Statement of
Auditing Standards No. 33, “Auditor Report on Financial State-
ments,” requires that audit reports be signed by two independent
auditors and also by the audit ?rm (Accounting and Research
Development Foundation, ARDF, 1999). In contrast to the U.S.,
where the audit reports of public ?rms only disclose the name of
the audit ?rm and its location, the data from Taiwan identify the
names of the two engaged audit partners and that of the audit ?rm.
We use data from all public ?rms with audit fee information in
Taiwan for the period from 2002 to 2011. Financial data, audit ?rm
data, and individual audit partner names are all obtained from the
Taiwan Economic Journal (TEJ) database. The initial sample with
available audit fee data consists of 6052 ?rm-year observations. We
exclude observations with missing auditor information (3), ?nan-
cial ?rms (448), and ?rms with missing ?nancial data (658). Finally,
we have 4943 ?rm-year observations and 1511 unique ?rms for our
regression analysis.
It should be noted that ?rms in Taiwan are not mandated to
disclose the audit fee information. Only when non-audit fees are
higher than 25% of audit fees (Reason 1), when auditors are changed
with a reduction in audit fees (Reason 2), and when audit fees are
decreased by more than 15% compared to the fee for the previous
year (Reason 3) are ?rms required to disclose audit fees. Firms can
also disclose audit fees voluntarily (Reason 4). In the robustness
check, we divide our sample into ?rms that disclose audit fees due
to requirements (Reasons 1e3) or voluntarily (Reason 4).
The sample distribution across years and industries is presented
in Table 1. There is an increase in sample representation during our
sample period (Panel A). For instance, the number of observations
increases from 236 in 2002 to 1015 in 2011. Liao, Wang, and Chi
(2012) indicate that because of the need for audit ?rms in the
adoption of IFRS and because of the encouragement of voluntarily
disclosing audit fee information, the number of ?rms with audit fee
data increases after 2009. In the robustness checks, we restrict our
sample period to 2002e2008 and ?nd consistent results. Liao et al.
(2012) suggest that there is no signi?cant structural difference
between the sample period 2002e2008 and 2009e2010, and that
the sample selection bias is less likely to be large. Our results also
remain unchanged when we restrict the sample period to
2002e2009. Panel B of Table 1 presents the sample distribution by
industry. About 14.48% of the observations (716) operate in the
electronics components industry (TSE Industry Code ¼ 28), and the
remainder are evenly distributed in other industries. In the
regression analyses, we include year and industry indicators to
control for year and industry effects.
3.2. Measure of audit partner sex
We classify the audit partners' sex based on their names, or
make direct contact with audit ?rms to accurately identify auditor
sex sampled in this study. The results are robust to excluding those
observations for which audit partners' sex is unclear. Four in-
dicators are used to proxy for the sex effects of the signing audit
partners: (1) WOMEN equals 1 if at least one of the engaged audit
partners is woman, and 0 otherwise; (2) WOMENNUM denotes the
number of women audit partners engaged with a client, and is
distributed between 0 and 2; (3) CPA1WOMEN equals 1 if the
leading audit partner is woman, and 0 otherwise; (4) CPA2WOMEN
equals 1 if the accompanying audit partner is woman, and
0 otherwise. In the sensitivity test, we employ three additional
measures to capture audit partners' sex after considering the dif-
ferential effects of the leading partners and accompanying partners
and obtain similar results.
T.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 219e233 221
3.3. Distribution of women audit partners in Taiwan auditing
market
Table 1 also presents the distribution of women audit partners
across years and industries. We ?nd that 52% of the observations
have at least one women auditors, the average number of women
auditors is 0.60, and 30% of the leading auditors and the accom-
panying auditors are women. There is an increase in the presence of
women audit partners across years. For example, the average
number of women auditors increases from 0.46 in 2002 to 0.66 in
2011, and the presence of women leading auditors increases from
14% in 2002 to 0.36 in 2011. We also ?nd that, in most industries,
the presence of women audit partners is relatively low and
dispersed. Exceptions are the glass and ceramics industry, the
rubber industry and the oil and gas industry, where more than 75%
of ?rms engage with at least one women audit partner.
In contrast, lower than 40% of ?rms in the electric and ma-
chinery industry, the electrical industry, the papers and pulps in-
dustry, and the miscellaneous electronic industry engage with at
least one women audit partners. Overall, Table 1 suggests that the
presence of women audit partners in Taiwan is low and disperse
across years and industries, and reinforces the importance of con-
trolling for year and industry effects.
3.4. Firm characteristics
Descriptive statistics are presented in Table 2. The mean and the
median values of audit fees (LNAF) are 14.63 and 14.64, respectively,
and that of total assets (LNTA) are 21.76 and 21.57, respectively. Our
sample includes both young and old companies, and the mean
value of age since a ?rm was established (AGE) is 22 years. On
average, the receivables and the inventories account for 34% of total
assets (RECINV), and total liabilities account for 44% of total assets
(LEV). The mean ratio of foreign sales (FOREIGN) is 25%, and on
average current assets are 2.44 times of current liabilities (CUR-
RENT). Among the sample ?rms, 6% are newly listed ?rms (IPO), 36%
are traded over-the-counter (OTC), 8% are emerging stocks (ROTC),
and the remainders are publicly listed ?rms. In our sample, 26% of
sample ?rms suffer from net losses (LOSS), 3% receive going-
concern opinions (GC), 63% receive unclean audit opinions
Table 1
Distribution of women audit partners.
Panel A: Distribution by year
Year N WOMEN WOMENNUM CPA1WOMEN CPA2WOMEN
2002 236 0.43 0.46 0.14 0.32
2003 240 0.45 0.49 0.20 0.29
2004 191 0.50 0.58 0.26 0.32
2005 171 0.53 0.63 0.34 0.29
2006 477 0.49 0.57 0.26 0.31
2007 520 0.51 0.61 0.28 0.33
2008 505 0.50 0.59 0.30 0.29
2009 707 0.51 0.60 0.32 0.28
2010 881 0.54 0.63 0.33 0.31
2011 1015 0.57 0.66 0.36 0.30
Total 4943 0.52 0.60 0.30 0.30
Panel B: Distribution by industry
TSE Industry N WOMEN WOMENNUM CPA1WOMEN CPA2WOMEN
01: Cement 22 0.68 0.77 0.45 0.32
02: Foods 73 0.53 0.55 0.26 0.29
03: Plastics 101 0.60 0.66 0.34 0.33
04: Textiles 151 0.49 0.55 0.26 0.28
05: Electric and Machinery 254 0.37 0.45 0.19 0.26
06: Electrical 47 0.36 0.36 0.23 0.13
08: Glass and Ceramics 19 0.79 0.89 0.58 0.32
09: Paper and Pulp 13 0.23 0.23 0.15 0.08
10: Steel and Iron 141 0.61 0.76 0.29 0.47
11: Rubber 43 0.81 0.95 0.44 0.51
12: Automobile 24 0.71 0.71 0.38 0.33
13: Miscellaneous Electronic 112 0.38 0.41 0.21 0.21
14: Constructions 229 0.62 0.72 0.32 0.40
15: Transportations 76 0.47 0.59 0.34 0.25
16: Tourism 41 0.51 0.54 0.34 0.20
18: Trading 69 0.58 0.70 0.41 0.29
20: Others 224 0.53 0.63 0.31 0.32
21: Chemicals 148 0.53 0.62 0.37 0.25
22: Biotechnology 228 0.50 0.66 0.36 0.31
23: Oil and Gas 32 0.78 0.91 0.53 0.38
24: Semiconductor 492 0.49 0.57 0.29 0.28
25: Computer 369 0.59 0.70 0.36 0.35
26: Photoelectric 488 0.47 0.53 0.27 0.26
27: Communication Network 245 0.60 0.69 0.40 0.30
28: Electronics Components 716 0.47 0.55 0.26 0.29
29: E-Channel Industry 150 0.61 0.69 0.37 0.32
30: Information Services 160 0.56 0.61 0.35 0.26
31: Other Electronics 250 0.50 0.58 0.23 0.35
80: Management of Stock 26 0.54 0.58 0.08 0.50
Total 4943 0.52 0.60 0.30 0.30
T.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 219e233 222
(UNCLEAN), and 4% restate ?nancial statements (RESTATE). Similar
to other developed countries, most (85%) sample ?rms in Taiwan
are audited by Big4 audit ?rms (BIG4). On average, leading and
accompanying audit partners have 11 years of audit experience
(CPA1EXP and CPA2EXP) and have continuously served the client for
3 years (CPA1TENURE and CPA2TENURE). Each client's total assets
are on average 2% of the sum of the total assets of the audit ?rm's
clients (CIFIRM) and 11% of the sumof the total assets of the leading
partner's and the accompanying audit partner's clients (CICPA1 and
CICPA2). There are 7% of sample ?rms changing their engaged audit
?rm in the current year (INITIAL), and the average number of new
audit partners is 0.51 (NEWCPA). The mean and the median values
of the logarithm of non-audit fees (LNNAF) are 11.35 and 13.30,
respectively. Among the sample ?rms, 31% are audited by industry
specialized audit ?rms (EXPERT_FIRM), and 2% are audited by in-
dustry specialized leading or accompanying audit partners
(EXPERT_CPA1 and EXPERT_CPA2). Regarding the reason for audit
fee disclosure, 51% of sample ?rms disclose audit fees because non-
audit fees are higher than 25% of audit fees (REASON1), 2% because
there is an audit ?rm change with a reduction in audit fees
(REASON2), 5% because the reduction in audit fees is more than 15%
(REASON3), and 45% disclose audit fees voluntarily (REASON4). Note
that Reason 1 to 3 are not mutually exclusive. All continuous vari-
ables are winsorized at 1% and 99% to alleviate the problem of
outliers.
We further compare the differences in ?rm characteristics be-
tween ?rms with and without women audit partners (untabu-
lated). It shows that the clients audited by women audit partners
(WOMEN ¼ 1) are older (AGE), are more complex in the number of
related parties (RELATE), consist of fewer IPO ?rms (IPO), are more
likely to be audited by the Big 4 audit ?rms (BIG4) and by less
experienced accompanying auditors (CPA2EXP), involve in fewer
initial audit engagements (INITIAL), have longer audit ?rm tenure
(AFTENURE), and are less likely to be audited by industry specialized
audit ?rms (EXPERT_FIRM) than those by men audit partners
(WOMEN ¼ 0). Coupled with the fact that we ?nd no signi?cant
differences in size (LNTA), auditor specialization (EXPERT_CPA1 and
EXPERT_CPA2), and the reasons for audit fee disclosure (REASON1 to
REASON4), we conclude that our results of the difference in audit
fees between women and men auditors are unlikely driven by the
size effects, the premiums for the Big 4 audit ?rms and industry
specialized auditors, or the underlying reasons for ?rms to disclose
audit fees. Nevertheless, we control these ?rm and audit charac-
teristics in our regression analyses to alleviate such concerns.
4. Women audit partners and audit fees
The main question addressed by this study is whether there is
any evidence that women audit partners are discriminated against
upon providing and charging for their professional services. This
question is important, because women audit partners would be at a
disadvantage in terms of receiving higher compensation and pro-
motion in audit ?rms if they contribute less to audit ?rm earnings.
We do not assume that higher audit fees equal higher pro?ts, since
pro?ts are determined by audit fees, audit efforts, auditor's
compensation for risk, and other factors. However, we have dis-
cussed with audit partners about this issue, and they suggest that to
some extent audit fees are associated with pro?ts. Therefore, we
study the association between women audit partner and audit fees,
after controlling for other determinants.
Table 2
Descriptive statistics (N ¼ 4943).
Variables Mean STD Q1 Median Q3
Logarithm of audit fees (LNAF) 14.63 0.61 14.25 14.64 15.01
Logarithm of total assets (LNTA) 21.76 1.56 20.68 21.57 22.63
Firm age since setup (AGE) 22.22 12.70 12.00 20.00 31.00
Percentage of receivables and inventories (RECINV) 0.34 0.19 0.20 0.33 0.46
Square root of related parties (RELATE) 3.14 1.47 2.00 2.83 3.87
Foreign sales (FOREIGN) 0.25 0.37 0.00 0.00 0.57
Loss (LOSS) 0.26 0.44 0.00 0.00 1.00
Current ratio (CURRENT) 2.44 2.35 1.26 1.74 2.68
Leverage (LEV) 0.44 0.21 0.29 0.44 0.57
Returns on assets (ROA) 0.29 1.51 0.00 0.02 0.14
Going concern opinion (GC) 0.03 0.17 0.00 0.00 0.00
Unclean audit opinion (UNCLEAN) 0.63 0.48 0.00 1.00 1.00
Restatement (RESTATE) 0.04 0.19 0.00 0.00 0.00
Initial public offerings (IPO) 0.06 0.24 0.00 0.00 0.00
Over the counter (OTC) 0.36 0.48 0.00 0.00 1.00
Emerging stock (ROTC) 0.08 0.27 0.00 0.00 0.00
Big 4 audit ?rms (BIG4) 0.85 0.36 1.00 1.00 1.00
CPA1 experience (CPA1EXP) 11.59 5.88 7.00 11.00 16.00
CPA2 experience (CPA2EXP) 11.13 6.25 6.00 11.00 16.00
Initial audit engagements (INITIAL) 0.07 0.26 0.00 0.00 0.00
Number of new auditors (NEWCPA) 0.51 0.67 0.00 0.00 1.00
Logarithm of non-audit fees (LNNAF) 11.35 4.82 11.70 13.30 14.00
Client importance for audit ?rm (CIFIRM) 0.02 0.09 0.00 0.00 0.00
Client importance for CPA1 (CICPA1) 0.11 0.20 0.01 0.03 0.10
Client importance for CPA2 (CICPA2) 0.11 0.21 0.01 0.02 0.09
Audit ?rm tenure (AFTENURE) 10.25 6.71 5.00 9.00 15.00
CPA1 tenure (CPA1TENURE) 3.07 1.80 2.00 3.00 4.00
CPA2 tenure (CPA2TENURE) 2.64 1.57 1.00 2.00 4.00
Audit ?rm industry specialization (EXPERT_FIRM) 0.31 0.46 0.00 0.00 1.00
CPA1 industry specialization (EXPERT_CPA1) 0.02 0.15 0.00 0.00 0.00
CPA2 industry specialization (EXPERT_CPA2) 0.02 0.14 0.00 0.00 0.00
Non-audit fees higher than 25% of audit fees (REASON1) 0.51 0.50 0.00 1.00 1.00
Auditors are changed with a reduction in audit fees (REASON2) 0.02 0.14 0.00 0.00 0.00
Reduction in audit fees more than 15% (REASON3) 0.05 0.22 0.00 0.00 0.00
Voluntary disclosure of audit fees (REASON4) 0.45 0.50 0.00 0.00 1.00
T.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 219e233 223
4.1. Women audit partner and audit fees
We ?rst investigate the relationship between women audit
partners and audit fees. Both uni-variate analysis and regression
analysis are used to examine this association. First, the full sample
observations are classi?ed into two groups: (1) ?rms with zero
women audit partners and (2) ?rms with at least one women audit
partners. We then examine the mean and median differences for
two of the three groups, respectively. Based on t test and Wilcoxon
rank sum test, the untabulated results show that ?rms with zero
women audit partners and ?rms with at least one women audit
partners are close in audit fees. The mean and the median values of
the logarithmof audit fees are $14.62 and $14.63 for ?rms with zero
women audit partners and are $14.65 and $14.65 for ?rms with at
least one women audit partners. The difference in mean is insig-
ni?cant under the t test, while the difference in median is signi?-
cant under the Wilcoxon rank sum test.
Second, we follow prior audit pricing studies (e.g., Huang,
Raghunandan, & Rama, 2009; Kim, Liu, & Zheng, 2012; Dao,
Raghunandan, & Rama, 2012; Fung, Gul, & Krishnan, 2012) to
construct the audit pricing regression model as follows:
The dependent variable is the logarithm of audit fees (LNAF).
GENDER is the indicators of women audit partner sex discussed
above (WOMEN, WOMENNUM, CPA1WOMEN, and CPA2WOMEN).
We predict that b
1
is negative under the sex discrimination
assumption. That is, if women audit partners suffer from inequity,
they are remunerated with lower audit fees.
The model also includes several ?rm-speci?c control variables,
which account for the effects of factors on the cross-sectional dif-
ferences in audit fees. Accordingly, we control for the client size
effect by including the logarithm of total assets (LNTA) and the
number of years since the ?rm was established (AGE) and control
for client complexity by including the percentage of receivables and
inventories over total assets (RECINV), the square root of the
number of related parties (RELATE), and the percentage of foreign
sales (FOREIGN) (Chi, Huang, Liao, & Xie, 2009; Francis, 1984; Fung
et al., 2012; Simunic, 1980). Following Dao et al. (2012) and Kim
et al. (2012), client-speci?c litigation risks and ?nancial condi-
tions are controlled by including an indicator of reporting net loss
(LOSS), the ratio of current assets to current liabilities (CURRENT),
the debt-to-asset ratio (LEV), the return on assets (ROA), an indi-
cator of receiving a going concern opinion (GC), and an indicator of
receiving unclean audit opinions (UNCLEAN), which include un-
quali?ed audit opinions with explanatory notes, and an indicator of
?nancial restatements in the current year (RESTATE). Similar to
Ashbaugh, LaFond, and Mayhew (2003) and Kim et al. (2012), we
include an indicator of initial public offerings (IPO), an indicator of
the over-the-counter market (OTC), and an indicator of the
emerging stock market (ROTC) to control for the needs of additional
audit and consulting services. An indicator of a Big 4 client (BIG4) is
included, since previous studies have documented a Big 4 audit fee
premium (Choi, Kim, Liu, & Simunic, 2008; DeFond, Francis, &
Wong, 2000). In addition to the Big 4 indicator, we also include
the years of the leading audit partner's and the accompanying audit
partner's work experience (CPA1EXP1 and CPA2EXP). The prior
literature suggests an audit fee discount for an initial audit
engagement (Huang et al., 2009; Simon & Francis, 1988;
Whisenant, Sankaraguruswamy, & Raghunandan, 2003), so we
include an indicator of initial audit engagement (INITIAL) and
control for the number of new audit partners engaged (NEWCPA).
Similar to Huang, Liu, Raghunandan, and Rama (2007, 2009) and
Chen, Sun, and Wu (2010), we include the logarithm of non-audit
fees (LNNAF), the ratio of the audited client's total assets over the
sum of the total assets of all the audit ?rm's clients (CIFIRM), the
ratio of the audited client's total assets over the sum of the total
assets of all the leading audit partner's clients (CICPA1) and the
accompanying audit partner's clients (CICPA2), the number of years
of audit ?rm tenure since 1983 (AFTENURE), and the number of
continuous years that the leading audit partner and the accompa-
nying audit partner have been engaged with the client (CPA1TE-
NURE and CPA2TENURE) to capture client importance and the
client-auditor relationship. We also control for audit ?rm's and
auditor's industry specialization (EXPERT_FIRM, EXPERT_CPA1 and
EXPERT_CPA2) given the audit fees premiums for industry special-
ized documented in the prior literature (Francis, Reichelt, & Wang,
2005; Zerni, 2012). Finally, we control for the potential effects of
mandatory audit fee disclosure on audit pricing by including three
indicators of mandatory audit fee disclosure. That is, a ?rm is
required to disclose audit fee information if non-audit fees exceed
25% of audit fees (REASON1), a ?rm changes audit partners leading
to a reduction in audit fees (REASON2), or audit fees are reduced by
more than 15% of the fee for the previous year (REASON3). We also
include year and industry dummies to control for the potential time
and industrial effects on audit fees.
Table 3 reports the main regression analyses. Our audit fee
regression models have suitable explanatory power, and the
adjusted R-squared is around 54%. We ?nd that women audit
partners charge signi?cantly lower audit fees. Speci?cally, audit
fees are lower if there is at least one women audit partner in the
engagement (WOMEN: À0.032, p < 0.01). Similarly, audit fees are
negatively related to the number of women audit partners
(WOMENNUM: À0.026, p < 0.01). Consistent with Chin and Chi
(2009) and Chi and Chin (2011) that leading auditors are likely
more important than accompanying auditors in audit decisions, we
LNAF ¼ b
0
þb
1
GENDER þb
2
LNTA þb
3
AGE þb
4
RECINV þb
5
RELATE
þb
6
FOREIGN þb
7
LOSS þb
8
CURRENT þb
9
LEV þb
10
ROA þb
11
GC
þb
12
UNCLEAN þb
13
RESTATE þb
14
IPO þb
15
OTC þb
16
ROTC þb
17
BIG4
þb
18
CPA1EXP þb
19
CPA2EXP þb
20
INITIAL þb
21
NEWCPA þb
22
LNNAF
þb
23
CIFIRMþb
24
CICPA1 þb
25
CICPA2 þb
26
AFTENURE
þb
27
CPA1TENURE þb
28
CPA2TENURE þb
29
EXPERT FIRM
þb
30
EXPERT CPA1 þb
31
EXPERT CPA2 þb
32
REASON1 þb
33
REASON2
þb
34
REASON3 þYEAR þINDUSTRY þ 3
(1)
T.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 219e233 224
?nd that when the leading audit partner is women, there is a sig-
ni?cant reduction in audit fees (CPA1WOMEN: À0.036, p < 0.01). In
turns of economic signi?cance, Table 3 suggests that the audit fee
discounts for women audit partners are 3.2% when there is at least
one women audit partner, 5.2% when there are two women audit
partners, and 3.6% when the leading audit partner is women.
Overall, the results provide consistent evidence to support the view
that women audit partners are signi?cantly associated with lower
audit fees as compared to men partners.
The results of our control variables are mostly consistent with
the ?ndings of prior studies. The signi?cantly positive coef?cients
of AGE, RELATE, and FOREIGN suggest that older and larger ?rms
with higher complexity pay higher audit fees (p < 0.01), consistent
with the prior studies (Chi et al., 2009; Francis, 1984; Fung et al.,
2012; Simunic, 1980). We also ?nd that ?rms receiving going-
concern opinions (GC) and unclean audit opinions (UNCLEAN) and
?rms restating ?nancial statements (RESTATE) pay higher audit fees
(p < 0.01), consistent with the positive relation between audit risk
and audit fees (Simunic, 1980). BIG4 is signi?cantly positive
(p < 0.01), which re?ects an existing audit fee premium of BIG4
auditors (e.g., Francis &Simon, 1987; Ferguson &Stokes, 2002), and
the signi?cantly positive coef?cient of EXPERT_CPA1 (p < 0.01) also
corresponds to the audit fee premiums for industry specialized
auditors (Francis et al., 2005; Zerni, 2012). DeAngelo (1981) sug-
gested that incumbent auditors are likely to low-ball initial-year
audit fees in order to earn future quasi-rents fromclients. Likewise,
this initial audit fee discount is also discovered in our sample ?rms,
which is supported by a negatively coef?cient of INITIAL (p < 0.15)
and a positive coef?cient of AFTENURE (p < 0.01). Auditor experi-
ence is so valuable that it increases the demand for higher quality
audits (Craswell, Francis, & Taylor, 1995; Ward, Elder, & Kattelus,
1994). As we expected, both CPA1EXP and CPA2EXP are signi?-
cantly positive (p < 0.15). As expected, when a ?rm discloses audit
fees because of audit ?rm change and reduction in audit fees
(REASON1), audit fees are lower (p < 0.01). We are surprised that
the coef?cient of LNTA is negative, and interpret this ?nding as
suggesting that larger ?rms are less risky and have higher bargai-
ning power, resulting in lower audit fees. However, it is hard to
explain why the coef?cient of LOSS is negative.
4.2. Audit fee range
In 2009, Taiwanese ?rms can decide to report audit fee range
instead of the exactly amount of audit fees. To examine whether our
Table 3
Regression results of audit fees and women auditors.
Model Model 1 Model 2 Model 3
Dependent variable LNAF LNAF LNAF
Variables Coef?cient p-value Coef?cient p-value Coef?cient p-value
INTERCEPT 13.620 0.00 13.620 0.00 13.618 0.00
WOMEN À0.032 0.01
WOMENNUM À0.026 0.01
CPA1WOMEN À0.036 0.01
CPA2WOMEN À0.016 0.23
LNTA À0.008 0.08 À0.008 0.07 À0.008 0.07
AGE 0.004 0.00 0.004 0.00 0.004 0.00
RECINV À0.073 0.03 À0.074 0.02 À0.075 0.02
RELATE 0.110 0.00 0.110 0.00 0.110 0.00
FOREIGN 0.092 0.00 0.091 0.00 0.090 0.00
LOSS À0.024 0.10 À0.025 0.10 À0.025 0.09
CURRENT À0.004 0.27 À0.004 0.29 À0.004 0.29
LEV 0.054 0.17 0.055 0.17 0.056 0.16
ROA 0.004 0.35 0.004 0.35 0.004 0.36
GC 0.205 0.00 0.206 0.00 0.207 0.00
UNCLEAN 0.042 0.00 0.042 0.00 0.042 0.00
RESTATE 0.110 0.00 0.109 0.00 0.110 0.00
IPO À0.357 0.00 À0.358 0.00 À0.358 0.00
OTC À0.107 0.00 À0.106 0.00 À0.106 0.00
ROTC À0.322 0.00 À0.321 0.00 À0.321 0.00
BIG4 0.474 0.00 0.475 0.00 0.476 0.00
CPA1EXP 0.004 0.00 0.004 0.00 0.004 0.00
CPA2EXP 0.002 0.11 0.002 0.12 0.002 0.08
INITIAL À0.051 0.14 À0.051 0.14 À0.051 0.14
NEWCPA 0.018 0.16 0.018 0.16 0.019 0.16
LNNAF 0.027 0.00 0.027 0.00 0.027 0.00
CIFIRM À0.774 0.00 À0.776 0.00 À0.772 0.00
CICPA1 0.566 0.00 0.567 0.00 0.568 0.00
CICPA2 0.360 0.00 0.359 0.00 0.357 0.00
AFTENURE 0.007 0.00 0.007 0.00 0.007 0.00
CPA1TENURE 0.000 0.93 0.000 0.95 0.000 0.96
CPA2TENURE 0.001 0.84 0.001 0.83 0.001 0.83
EXPERT_FIRM 0.003 0.82 0.003 0.82 0.003 0.85
EXPERT_CPA1 0.118 0.00 0.117 0.00 0.117 0.00
EXPERT_CPA2 0.062 0.11 0.063 0.10 0.063 0.10
REASON1 À0.150 0.00 À0.150 0.00 À0.150 0.00
REASON2 0.054 0.29 0.053 0.30 0.053 0.31
REASON3 À0.005 0.86 À0.006 0.86 À0.006 0.84
Year Effects Controlled Controlled Controlled
Industry Effects Controlled Controlled Controlled
Clustering Firm-Year Firm-Year Firm-Year
Adjusted R-Square 54.06% 54.07% 54.07%
N 4943 4943 4943
T.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 219e233 225
results are robust to reporting audit fee as ranges, we use ordered
logistic regression models to re-examine the association between
audit fees and women audit partners. Audit fee ranges are coded as
follows: 1 for audit fees lower than $2 million, 2 for audit fees
between $2 million and $4 million, 3 for audit fees between $4
million and $6 million, 4 for audit fees between $6 million and $8
million, 5 for audit fees between $8 million and $10 million, and 6
for audit fees higher than $10 million (NTD). We continue to ?nd
that women audit partners charge lower audit fees than men
counterparts. In particular, the coef?cients of WOMEN, WOMEN-
NUM, CPA1WOMEN are À0.147, À0.118, and À0.130, respectively,
signi?cant at the 0.10 level (untabulated). The untabulated results
are available from the authors.
4.3. Masculine industries
Because discrimination against women would vary with the
extent to which an organization or an industry is masculine or not
(Eagly & Carli, 2003; Eagly, Makhijani, & Klonsky, 1992), we also
study whether the discrimination against women audit partners is
more severe in traditionally masculine industries. We identify
masculine industries as those where lower than 40% of the ?rms in
that industry engage with at least one women audit partner,
including the electric and machinery industry, the electrical in-
dustry, the paper and pulp industry, and the miscellaneous elec-
tronic industry. Our results remain unchanged when we identify
masculine industries as those where lower than 50% of the ?rms in
that industry engage with at least one women audit partner. The
proxies for women auditors are interacted with the indicator of
masculine industries, MIND. MIND equals one if a ?rm operates in
one of the masculine industries discussed above. If discrimination
is more severe in masculine industries, the interaction term of the
proxies for women partners and masculine industry is expected to
be negative.
The results are presented inTable 4. We ?nd signi?cant evidence
that the negative relation between women audit partner and audit
fees is more severe in masculine industries. Speci?cally, the audit
fees are further lower in the masculine industries discussed above
whenthere is at least onewomenaudit partner, whenthe number of
women audit partner increases, and when the leading audit partner
is women. The coef?cient of WOMEN*MIND, WOMENNUM*MIND,
and CPA1WOMEN*MIND are À0.095, À0.075, and À0.169, respec-
tively (p < 0.05). To illustrate the economic signi?cance, Table 4
suggests that the audit fee discounts for women audit partners in-
crease around 9.5%e16.9% in masculine industries. In sum, the re-
sults indicate that women audit partners charge further lower audit
fees in masculine industries.
4.4. Women audit partner and audit quality
There are at least three reasons why women audit partners
charge lower audit fees than men audit partners, including (1)
discrimination against women auditors, (2) the poor audit quality
provided by women audit partners and (3) the ef?ciency of
women audit partners. To analyze the possibility that the latter
two reasons explain our ?ndings, we examine the association
between audit quality, measured as client earnings quality and
audit ef?ciency, and women audit partners. We follow prior
studies (e.g. Al-Ajmi, 2008; Bamber, Bamber, & Schoderbek, 1993;
Chen et al., 2010, 2011; Chi & Chin, 2011; Chin & Chi, 2009;
Henderson & Kaplan, 2000; Knechel & Payne, 2001; Lee,
Mande, & Son, 2008; Leventis & Weetman, 2004; Reichelt &
Wang, 2010) construct the audit quality regression models as
follows.
jDCACCj is the absolute value of discretionary current accruals
(DCACC). As in Chi and Chin (2011), we measure DCACC by esti-
mating the following model for each year-industry grouping.
CACC ¼ b
0
þb
1
INVERSEAT þb
2
DSALE þb
3
ROA þ 3 (4)
CACC is current accruals, de?ned as net income before extraor-
dinary items plus depreciation and amortization minus operating
cash ?ow, divided by lagged total assets. INVERSEAT is 1 divided by
lagged assets. DSALE is the change in sales divided by lagged assets.
ROA is net income divided by lagged assets. We do not deduct the
jDCACCj ¼ b
0
þb
1
GENDER þb
2
LNTA þb
3
LOSS þb
4
ROA þb
5
LEV þb
6
CURRENT
þb
7
AGE þb
8
ZMJSCORE þb
9
OCF þb
10
TACC þb
11
PE þb
12
MB þb
13
RAISE þb
14
IPO
þb
15
OTC þb
16
ROTC þb
17
INITIAL þb
18
NEWCPA þb
19
BIG4
þb
20
EXPERT FIRMþb
21
EXPERT CPA1 þb
22
EXPERT CPA2
þb
23
AFTENURE þb
24
CPA1TENURE þb
25
CPA2TENURE þb
26
CIFIRM
þb
27
CICPA1 þb
28
CICPA2 þb
29
CPA1EXP þb
30
CPA2EXP þb
31
RESTATE
þb
32
UNCLEAN þb
33
GC þYEAR þINDUSTRY þ 3
(2)
LNRLAG ¼ b
0
þb
1
GENDER þb
2
LNTA þb
3
LOSS þb
4
ROA þb
5
LEV þb
6
CURRENTþ
b
7
AGE þb
8
ZMJSCORE þb
9
OCF þb
10
TACC þb
11
PE þb
12
MB þb
13
RAISE þb
14
IPO
þb
15
OTC þb
16
ROTC þb
17
INITIAL þb
18
NEWCPA þb
19
BIG4
þb
20
EXPERT FIRMþb
21
EXPERT CPA1 þb
22
EXPERT CPA2
þb
23
AFTENURE þb
24
CPA1TENURE þb
25
CPA2TENURE þb
26
CIFIRM
þb
27
CICPA1 þb
28
CICPA2 þb
29
CPA1EXP þb
30
CPA2EXP þb
31
RESTATE
þb
32
UNCLEAN þb
33
GC þYEAR þINDUSTRY þ 3
(3)
T.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 219e233 226
change in receivables fromDSALE, and we control for current ROA as
Kothari, Leone, & Wasley (2005). Discretionary current accruals
(DCACC) are the residual values derived (4). LNRLAG is audit
reporting lags, de?ned as the logarithm of the number of days be-
tween the ?scal year end and the ending date of auditor ?eld work
(e.g. Al-Ajmi, 2008; Lee et al., 2008; Leventis & Weetman, 2004). If
the lower audit fees for womenaudit partners are resultedfromthat
women audit partners provide poor audit quality or that women
audit partners are more ef?cient, we predict b
1
to be positive in (2)
and negative in (3).
We control for many ?rm and audit characteristics in?uencing
client earnings quality and audit reporting lags. We control for ?rm
size effects by including the logarithmof total assets (LNTA) and the
number of years since the ?rm was established (AGE) as in Myers,
Myers, and Omer (2003) and Menon and Williams (2004). Firm
performance and the likelihood of ?nancial distress are controlled
by including an indicator of net loss (LOSS), returns on assets (ROA),
leverage (LEV), current ratio (CURRENT), and ?nancial risk scores
(ZMJSCORE) as in Chin and Chi (2009) and Chi and Chin (2011).
Following Carcello, Hermanson, and Huss (1995) and Liu and Wang
(2005), ZMJSCORE is computed as À4.803 À3.6 Â (net income/total
assets) þ 5.4 Â (total debt/total assets) À 0.1 Â (current assets/
current liabilities). Higher ZMJSCORE indicates higher likelihood
of bankruptcy. Lagged total accruals divided by total assets (TACC)
and operating cash ?ows divided by total assets (OCF) are
controlled because of their negative association with the current
accruals (Ashbaugh et al., 2003; Sloan, 1996). Firm growth, market
valuation, and ?nancing activities are controlled by including
price-earnings ratio (PE), market-to-book ratio (MB), and the
?nancing cash ?ows divided by total assets (RAISE) as in the
prior literature (e.g. Chin & Chi, 2009; Jagadison, Aier, Gunlock, &
Lee, 2005; Richardson, Tuna, & Wu, 2003). The superior audit
quality of the Big 4 audit ?rms (BIG4) and industry specialists
(EXPERT_FIRM, EXPERT_CPA1, and EXPERT_CPA2) are controlled
Table 4
Regression results of audit fees, women auditors, and industrial effects.
Model Model 1 Model 2 Model 3
Dependent variable LNAF LNAF LNAF
Variables Coef?cient p-value Coef?cient p-value Coef?cient p-value
INTERCEPT 13.616 0.00 13.614 0.00 13.611 0.00
WOMEN À0.031 0.02
WOMEN*MIND À0.095 0.03
WOMENNUM À0.026 0.01
WOMENNUM*MIND À0.075 0.03
CPA1WOMEN À0.031 0.03
CPA1WOMEN*MIND À0.169 0.00
CPA2WOMEN À0.021 0.14
CPA2WOMEN*MIND 0.009 0.85
MIND À0.029 0.28 À0.032 0.22 À0.032 0.22
LNTA À0.008 0.09 À0.008 0.08 À0.008 0.08
AGE 0.002 0.00 0.002 0.00 0.002 0.00
RECINV À0.088 0.01 À0.089 0.01 À0.089 0.01
RELATE 0.112 0.00 0.112 0.00 0.112 0.00
FOREIGN 0.197 0.00 0.196 0.00 0.195 0.00
LOSS À0.031 0.05 À0.031 0.04 À0.031 0.05
CURRENT À0.005 0.10 À0.005 0.11 À0.005 0.12
LEV 0.055 0.18 0.056 0.17 0.059 0.15
ROA 0.003 0.46 0.003 0.46 0.003 0.46
GC 0.113 0.01 0.115 0.01 0.115 0.01
UNCLEAN 0.048 0.00 0.048 0.00 0.048 0.00
RESTATE 0.076 0.04 0.076 0.04 0.076 0.04
IPO À0.377 0.00 À0.378 0.00 À0.377 0.00
OTC À0.143 0.00 À0.141 0.00 À0.141 0.00
ROTC À0.377 0.00 À0.375 0.00 À0.374 0.00
BIG4 0.511 0.00 0.513 0.00 0.515 0.00
CPA1EXP 0.003 0.01 0.003 0.01 0.003 0.01
CPA2EXP 0.002 0.09 0.002 0.10 0.002 0.07
INITIAL À0.066 0.06 À0.065 0.07 À0.065 0.07
NEWCPA 0.024 0.08 0.024 0.08 0.024 0.07
LNNAF 0.029 0.00 0.029 0.00 0.029 0.00
CIFIRM À0.788 0.00 À0.788 0.00 À0.787 0.00
CICPA1 0.563 0.00 0.564 0.00 0.568 0.00
CICPA2 0.379 0.00 0.378 0.00 0.374 0.00
AFTENURE 0.005 0.00 0.005 0.00 0.005 0.00
CPA1TENURE 0.000 0.93 0.000 0.96 0.000 1.00
CPA2TENURE 0.005 0.27 0.005 0.27 0.005 0.27
EXPERT_FIRM À0.003 0.84 À0.003 0.83 À0.003 0.83
EXPERT_CPA1 0.122 0.00 0.121 0.00 0.120 0.00
EXPERT_CPA2 0.071 0.08 0.071 0.08 0.071 0.08
REASON1 À0.144 0.00 À0.144 0.00 À0.145 0.00
REASON2 0.092 0.09 0.091 0.10 0.092 0.09
REASON3 À0.001 0.98 À0.001 0.97 À0.003 0.93
Year Effects Controlled Controlled Controlled
Industry Effects Controlled Controlled Controlled
Clustering Firm-Year Firm-Year Firm-Year
Adjusted R-Square 51.56% 51.56% 51.61%
N 4943 4943 4943
T.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 219e233 227
(e.g. Reichelt & Wang, 2010). Audit ?rm (AFTENURE) and
auditor tenure (CPA1TENURE and CPA2TENURE) as well as
initial audit engagement (INITIAL) and the number of new
audit partners (NEWCPA) are also controlled given the fact that
audit quality would be improved by longer client-auditor rela-
tionship (Carey & Simnett, 2006; Davis, Soo, & Trompeter, 2009).
We control for client importance (CIFIRM, CICPA1, and CICPA2)
since Chen et al. (2010) suggest that client importance may in?u-
ence auditor independence and thus audit quality. As in the audit
fee models, we control for auditor experience (CPA1EXP and
CPA2EXP), ?nancial restatement (RESTATE), unclean audit opinions
(UNCLEAN), going-concern opinions (GC), and a ?rm's listing status
(IPO, OTC, and ROTC). Year and industry effects are also controlled
for. Likewise, all the continuous variables are winsorized at 1%
and 99%.
In Table 5, we provide the descriptive statistics of the variables
used in (2) and (3). The sample size decreases to 3872 because of
missing necessary data. The mean and median values of discre-
tionary current accruals (DCACC) are 0.09 and 0.05, and that of the
audit reporting lags in logarithm (LNRLAG) are 4.19 and 4.30. On
average, ZMJSCORE is À2.73, operating cash ?ows (OCF) account for
7% of total assets, PE is 36.04, MB is 2.94, and the cash ?ows from
?nancing activities (RAISE) are 2% of total assets. Other variables are
similar as in Table 2.
We then classify our sample ?rms into (1) ?rms with no women
audit partners (N ¼ 1832) and (2) ?rms with at least one women
audit partners (N ¼ 2040), and compare their differences in ?rm
and audit characteristics (untabulated). We ?nd that the clients
with at least one women audit partner (WOMEN ¼ 1) have lower
absolute value of discretionary current accruals (jDCACCj) than
clients with no women audit partners (WOMEN ¼ 0), and that
women audit partners are associated with longer audit reporting
lags (LNRLAG). These ?ndings provide preliminary evidence that
our previous results of the negative association between audit fees
and women auditors are less likely driven by the poor audit quality
and shorter reporting lags by women audit partners. With respect
to control variables, we ?nd that clients with at least one women
audit partners are older (AGE), involve in fewer initial audit en-
gagements (INITIAL), have longer audit ?rmtenure (AFTENURE), and
are less likely to be audited by industry specialized accompanying
auditors (EXPERT_CPA2).
Table 6 reports the regression analyses of audit quality and
women audit partner. The dependent variable is the absolute value
of discretionary current accruals (jDCACCj) in Panel A and is the
logarithm of audit reporting lags (LNRLAG) in Panel B. The explan-
atory power of these regression models (26% in Panel A and 23% in
Panel B) is comparable to that in the prior studies (e.g. 16% in Chi &
Chin, 2011 and 27% in Krishnan & Yang, 2009). In Panel A, we ?nd
that clients audited by women audit partners report better earnings
quality. In particular, the absolute value of discretionary current
accruals (jDCACCj) is lower if there is at least one women audit
partner in the engagement (WOMEN: À0.007, p < 0.05). Likewise,
the number of women audit partners is associated with better
earnings quality (WOMENNUM: À0.005, p < 0.05). As the ?ndings in
Table 2, we ?nd that when the leading audit partner is women, the
client earnings quality is higher (CPA1WOMEN: À0.011, p < 0.01),
suggesting the importance of the leading auditors in audit en-
gagements. These results indicate that women auditors may pro-
vide better audit quality, in terms of higher client earnings quality,
than men auditors, and suggest that the negative relation between
audit fees and women auditors may not be explained by the su-
perior audit quality of men auditors.
Table 5
Descriptive statistics of audit quality model (N ¼ 3872).
Variables Mean STD Q1 Median Q3
Absolute value of discretionary current accruals (jDCACCj) 0.09 0.13 0.02 0.05 0.09
Logarithm of audit reporting lags (LNRLAG) 4.19 0.39 4.04 4.30 4.44
Logarithm of total assets (LNTA) 21.75 1.53 20.70 21.56 22.59
Loss (LOSS) 0.25 0.43 0.00 0.00 0.00
Returns on assets (ROA) 0.28 1.46 0.00 0.02 0.14
Leverage (LEV) 0.44 0.21 0.29 0.44 0.57
Current ratio (CURRENT) 2.46 2.33 1.27 1.76 2.71
Firm age since setup (AGE) 23.63 12.29 14.00 22.00 32.00
Financial distress scores (ZMJSCORE) À2.73 1.55 À3.73 À2.74 À1.89
Operating cash ?ows (OCF) 0.07 0.12 0.01 0.07 0.13
Lagged total accruals (TACC) 0.00 0.10 À0.04 0.01 0.05
Price-to-earnings ratio (PE) 36.04 149.78 0.13 6.67 23.54
Market-to-book ratio (MB) 2.94 3.05 1.13 2.02 3.55
Funds raised (RAISE) 0.02 0.16 À0.06 À0.01 0.07
Initial public offerings (IPO) 0.01 0.11 0.00 0.00 0.00
Over the counter (OTC) 0.33 0.47 0.00 0.00 1.00
Emerging stock (ROTC) 0.07 0.26 0.00 0.00 0.00
Initial audit engagements (INITIAL) 0.06 0.23 0.00 0.00 0.00
Number of new auditors (NEWCPA) 0.53 0.67 0.00 0.00 1.00
Big 4 audit ?rms (BIG4) 0.84 0.36 1.00 1.00 1.00
Audit ?rm industry specialization (EXPERT_FIRM) 0.30 0.46 0.00 0.00 1.00
CPA1 industry specialization (EXPERT_CPA1) 0.02 0.15 0.00 0.00 0.00
CPA2 industry specialization (EXPERT_CPA2) 0.02 0.14 0.00 0.00 0.00
Audit ?rm tenure (AFTENURE) 11.28 6.71 6.00 11.00 16.00
CPA1 tenure (CPA1TENURE) 3.02 1.77 2.00 3.00 4.00
CPA2 tenure (CPA2TENURE) 2.66 1.57 1.00 2.00 4.00
Client importance for audit ?rm (CIFIRM) 0.02 0.09 0.00 0.00 0.00
Client importance for CPA1 (CICPA1) 0.12 0.20 0.01 0.03 0.12
Client importance for CPA2 (CICPA2) 0.12 0.21 0.01 0.03 0.11
CPA1 experience (CPA1EXP) 11.85 5.96 7.00 12.00 16.00
CPA2 experience (CPA2EXP) 11.26 6.27 6.00 11.00 16.00
Restatement (RESTATE) 0.03 0.17 0.00 0.00 0.00
Unclean audit opinion (UNCLEAN) 0.66 0.47 0.00 1.00 1.00
Going concern opinion (GC) 0.03 0.16 0.00 0.00 0.00
T.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 219e233 228
Table 6
Regression results of audit quality and women auditors.
Panel A: Earnings quality and women auditors
Model Model 1 Model 2 Model 3
Dependent variable jDCACCj jDCACCj jDCACCj
Variables Coef?cient p-value Coef?cient p-value Coef?cient p-value
INTERCEPT 0.046 0.23 0.045 0.23 0.044 0.25
WOMEN À0.007 0.04
WOMENNUM À0.005 0.04
CPA1WOMEN À0.011 0.00
CPA2WOMEN 0.000 0.95
LNTA 0.003 0.05 0.003 0.04 0.003 0.04
LOSS 0.002 0.66 0.002 0.68 0.002 0.68
ROA À0.001 0.22 À0.001 0.21 À0.001 0.19
LEV À0.029 0.28 À0.029 0.28 À0.029 0.28
CURRENT À0.001 0.18 À0.001 0.19 À0.001 0.20
AGE 0.000 0.02 0.000 0.02 0.000 0.02
ZMJSCORE 0.001 0.76 0.001 0.75 0.001 0.75
OCF À0.045 0.18 À0.045 0.18 À0.045 0.17
TACC À0.012 0.69 À0.012 0.69 À0.012 0.68
PE 0.000 0.12 0.000 0.11 0.000 0.11
MB 0.005 0.00 0.005 0.00 0.005 0.00
RAISE 0.132 0.00 0.133 0.00 0.133 0.00
IPO 0.068 0.03 0.068 0.03 0.067 0.03
OTC 0.017 0.00 0.017 0.00 0.017 0.00
ROTC 0.002 0.78 0.002 0.79 0.002 0.79
INITIAL 0.024 0.07 0.024 0.07 0.024 0.07
NEWCPA À0.002 0.66 À0.002 0.66 À0.002 0.67
BIG4 À0.014 0.08 À0.014 0.08 À0.013 0.10
EXPERT_FIRM À0.011 0.01 À0.010 0.01 À0.011 0.01
EXPERT_CPA1 0.004 0.79 0.004 0.80 0.005 0.78
EXPERT_CPA2 À0.009 0.49 À0.009 0.50 À0.009 0.50
AFTENURE 0.000 0.50 0.000 0.48 0.000 0.48
CPA1TENURE À0.002 0.05 À0.002 0.06 À0.002 0.06
CPA2TENURE 0.000 0.76 0.000 0.75 0.000 0.74
CIFIRM 0.017 0.61 0.017 0.61 0.020 0.56
CICPA1 À0.032 0.01 À0.032 0.01 À0.032 0.01
CICPA2 À0.004 0.76 À0.004 0.75 À0.006 0.68
CPA1EXP 0.000 0.98 0.000 0.97 0.000 0.73
CPA2EXP 0.000 0.99 0.000 0.99 0.000 0.70
RESTATE À0.007 0.62 À0.007 0.61 À0.007 0.62
UNCLEAN À0.009 0.04 À0.009 0.04 À0.009 0.04
GC 0.028 0.20 0.028 0.20 0.028 0.20
Year Effects Controlled Controlled Controlled
Industry Effects Controlled Controlled Controlled
Clustering Firm-Year Firm-Year Firm-Year
Adjusted R-Square 26.20% 26.20% 26.24%
N 3872 3872 3872
Panel B: Audit reporting lags and women auditors
Model Model 1 Model 2 Model 3
Dependent variable LNRLAG LNRLAG LNRLAG
Variables Coef?cient p-value Coef?cient p-value Coef?cient p-value
INTERCEPT 4.202 0.00 4.211 0.00 4.214 0.00
WOMEN 0.046 0.00
WOMENNUM 0.027 0.00
CPA1WOMEN 0.042 0.00
CPA2WOMEN 0.012 0.33
LNTA 0.004 0.29 0.004 0.30 0.004 0.31
LOSS À0.013 0.46 À0.012 0.50 À0.012 0.50
ROA À0.005 0.23 À0.005 0.26 À0.004 0.27
LEV À0.098 0.25 À0.097 0.25 À0.097 0.26
CURRENT 0.001 0.71 0.001 0.79 0.001 0.80
AGE 0.005 0.00 0.005 0.00 0.005 0.00
ZMJSCORE 0.008 0.54 0.007 0.57 0.007 0.58
OCF À0.415 0.00 À0.417 0.00 À0.415 0.00
TACC 0.009 0.90 0.009 0.90 0.010 0.88
PE 0.000 0.04 0.000 0.04 0.000 0.04
MB À0.009 0.00 À0.009 0.00 À0.009 0.00
RAISE 0.024 0.58 0.022 0.62 0.021 0.63
IPO 0.039 0.50 0.039 0.50 0.040 0.49
OTC 0.026 0.05 0.026 0.05 0.027 0.05
ROTC 0.157 0.00 0.158 0.00 0.158 0.00
(continued on next page)
T.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 219e233 229
In Panel B, we ?nd that women audit partners are more con-
servative in that they take more time to collect audit evidence,
resulting in longer audit reporting lags (LNRLAG). Speci?cally, the
audit reporting lags are signi?cantly longer when there is at least
one women audit partners (WOMEN: 0.046, p < 0.01). Similarly, the
association between audit reporting lags and the number of
women audit partners is also positive and signi?cant (WOMEN-
NUM: 0.027, p < 0.01). We further ?nd that the leading auditors play
a more important role in determining audit reporting lags (CPA1-
WOMEN: 0.042, p < 0.05). These ?ndings show that women audi-
tors are related to longer audit reporting lags, and suggest that the
negative correlation between audit fees and women auditors is not
driven by shorter audit reporting lags required by women audit
partners.
With respect to control variables, we ?nd that ?rms audited by
the Big 4 audit ?rms (BIG4) and industry specialized audit ?rms
(EXPERT_FIRM), ?rms with longer audit ?rm tenure (AFTENURE),
and ?rms receiving non-going-concern opinions (GC) are associ-
ated with higher earnings quality (jDCACCj), consistent with Chi
and Chin (2011). We also ?nd that ?rms audited by the industry
specialized audit ?rms (EXPERT_FIRM) and leading auditors
(EXPERT_CPA1), and ?rms audited by more experienced leading
auditors (CPA1EXP) have shorting audit reporting lags, whereas
?rms restating ?nancial statements (RESTATE) and receiving un-
clean audit opinions and going-concern opinions (GC) have longer
audit reporting lags, consistent with the previous studies (e.g. Lee
et al., 2008; Krishnan & Yang, 2009; Habib & Bhuiyan, 2011).
Overall, the results in Table 6 suggest that it is unlikely that the
differences inaudit qualityandaudit reporting lags betweenwomen
andmenaudit partners drive our ?nding that womenaudit partners
charge lower audit fees than men colleagues. Instead, we ?nd that
women audit partners are associated with better client earnings
quality and longer audit reporting lags, which should increase the
audit fees chargedbywomenaudit partners. Therefore, weconclude
that the reasons why women audit partners charge lower audit fees
than men audit partners in Taiwan are likely due to the masculine
audit industry and the discriminationagainst women. Nevertheless,
we acknowledge that there would be other explanations for the
negative relation between women audit partners and audit fees,
which deserve further research in the future.
As an additional analysis, we augment our audit fee models by
controlling for client earnings quality and audit reporting lags. We
?nd that the negative relation between women audit partners and
audit fees remains signi?cant when we include jDCACCj and
LNRLAG in the regression models (untabulated). In particular, the
coef?cients of WOMEN, WOMENNUM, and CPA1WOMEN
are À0.031, À0.025, and À0.033, respectively (p < 0.05). Therefore,
it is less likely that the negative correlation between audit fees and
women auditors is driven by the differences in audit quality and
audit ef?ciency between women and men audit partners.
4.5. Differences between ?rms with and without audit fee data
Because audit fees are required under certain situations in
Taiwan, it is likely that there is sample bias in this paper, which may
have great in?uence on our results. However, we argue that as long
as there is no signi?cant difference between clients with and
without women audit partners, such bias should not signi?cantly
in?uence our ?ndings. Moreover, it is unclear ex ante whether such
sample bias will lead to a positive or a negative relation between
women audit partners and audit fees. As discussed above, there are
few differences in ?rm and audit characteristics between clients
with and without women audit partners. Rather, untabulated re-
sults suggests that clients with women audit partners are older,
more complex, more likely to be audited by the Big 4 audit ?rms,
involve in fewer initial audit engagements, and have longer audit
?rm tenure, which should lead to higher audit fees. Furthermore,
there are no signi?cant differences in ?rm size, auditor speciali-
zation, and the reasons for audit fee disclosure. Consistently, Liao
et al. (2012) indicate that there is an increase in the number of
Taiwanese ?rms disclosing audit fees after 2009 because of the
demand for the IFRS adoption service provided by audit ?rms and
because of the encouragement of voluntarily disclosing audit fees.
They compare the differences in several ?rm and audit character-
istics between ?rms disclosing audit fees under certain situations
(2002e2008) and ?rms disclosing audit fees voluntarily
Table 6 (continued)
Panel B: Audit reporting lags and women auditors
Model Model 1 Model 2 Model 3
Dependent variable LNRLAG LNRLAG LNRLAG
Variables Coef?cient p-value Coef?cient p-value Coef?cient p-value
INITIAL À0.022 0.51 À0.022 0.50 À0.022 0.50
NEWCPA 0.008 0.50 0.009 0.49 0.008 0.50
BIG4 0.123 0.00 0.123 0.00 0.121 0.00
EXPERT_FIRM À0.101 0.00 À0.102 0.00 À0.101 0.00
EXPERT_CPA1 À0.082 0.07 À0.082 0.08 À0.083 0.07
EXPERT_CPA2 À0.023 0.65 À0.023 0.64 À0.023 0.64
AFTENURE 0.000 0.72 0.000 0.72 0.000 0.72
CPA1TENURE 0.000 0.98 0.000 0.98 0.000 0.96
CPA2TENURE 0.000 0.99 0.000 0.97 0.000 0.97
CIFIRM 0.059 0.54 0.060 0.53 0.053 0.58
CICPA1 À0.063 0.13 À0.063 0.13 À0.065 0.12
CICPA2 0.025 0.49 0.026 0.47 0.030 0.41
CPA1EXP À0.002 0.04 À0.002 0.04 À0.002 0.08
CPA2EXP 0.000 0.71 0.000 0.70 À0.001 0.46
RESTATE 0.144 0.00 0.145 0.00 0.145 0.00
UNCLEAN 0.060 0.00 0.060 0.00 0.060 0.00
GC 0.261 0.00 0.261 0.00 0.261 0.00
Year Effects Controlled Controlled Controlled
Industry Effects Controlled Controlled Controlled
Clustering Firm-Year Firm-Year Firm-Year
Adjusted R-Square 22.94% 22.79% 22.83%
N 3872 3872 3872
T.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 219e233 230
(2009e2010), and conclude that there is no signi?cant structural
change in the determinants of audit fees and that the problem of
selection bias is less likely to be severe. Therefore, it is less likely
that our results are signi?cantly biased by the sample bias. Never-
theless, we construct a sample consisting of ?rms with (N ¼ 4943)
and without (N ¼ 7447) audit fee data, and compare their ?rm and
audit characteristics.
The untabulated results show that there are signi?cant differ-
ences in ?rm and audit characteristics between ?rms with and
without audit fee data. In particular, ?rms with audit fee data are
more likely to be audited by women auditors (WOMEN, WOMEN-
NUM, CPA1WOMEN, and CPA2WOMEN), older (AGE), more complex
(RELATE) and export-oriented (FOREIGN), stronger in ?nancial
conditions (CURRENT and LEV), less likely to restate ?nancial
statements (RESTATE), more likely to receive unclean audit opinions
(UNCLEAN), consist of more newly listed ?rms (IPO, OTC, and ROTC),
are more likely to be audited by the Big 4 audit ?rms (BIG4) and
more experienced auditors (CPA1EXP and CPA2EXP), consists of
fewer initial audit engagements (INITIAL) and new audit partners
(NEWCPA), have higher non-audit fees (LNNAF), have longer audit
?rm tenure (AFTENURE) but shorter leading auditor tenure (CPA1-
TENURE), and are more likely to be audited by industry specialized
audit ?rms (EXPERT_FIRM).
As a robustness check, we control for ?rm ?xed effects in our
audit fee models to alleviate the concern that our results are driven
by some omitted variables. The adjusted R-square increases sub-
stantially to 90% when we control for ?rm ?xed effects, and we
continue to ?nd that women audit partners are associated with
lower audit fees (untabulated). Speci?cally, audit fees are signi?-
cantly lower when there is at least one women auditor in the
engagement (WOMEN: À0.019, p < 0.10). Similarly, audit fees are
signi?cantly and negatively related to the number of women audit
partners (WOMENNUM: À0.016, p < 0.10). We also continue to ?nd
that the leading auditors are more important in audit engagements.
Audit fees are signi?cantly lower when the leading auditor is
women (CPA1WOMEN: À0.028, p < 0.05). Therefore, we conclude
that it is less likely that the sample bias and the omitted variables
result in our ?ndings that women audit partners charge lower audit
fees than men audit partners. Nevertheless, we cannot fully rule out
these possibilities, and acknowledge that our results should be
explained with cautions.
4.6. Sensitivity analysis
Our results are robust to several sensitivity analyses. First, in
order to totally exclude the in?uence of Big 4 audit fee premiums,
we exclude non-Big 4 clients in additional regression tests. All co-
ef?cients of our women audit partner proxies continue to be
negative, and mostly signi?cant. Second, as discussed earlier, ?rms
publicly disclose audit fees for different reasons. Firms that are
required to disclose audit fees (Reasons 1, 2, and 3) may have
distinct characteristics from ?rms that disclose audit fees volun-
tarily (Reason 4). To prevent our results from being driven by these
different characteristics, we conduct separate regression analyses
for the two types of ?rms, and obtain consistent results. Third, we
exclude observations from the electronics industry and obtain
robust evidence. Finally, we employ additional measures of women
audit partners. Speci?cally, these measures consider (1) the dif-
ferential effect between the leading and accompanying audit
partners, (2) the composition of women audit partners, and (3) the
differential effect between one and two women audit partners. Our
results are robust to these different measures. Overall, we ?nd
robust evidence that audit fees are signi?cantly lower for women
audit partners, and this discrimination against women audit part-
ners is more pronounced in masculine industries and cannot be
explained by the differences in audit quality and audit reporting
lags between women and men audit partners.
5. Conclusion
This paper examines whether women audit partners earn lower
audit fees by using a sample of public companies inTaiwan. We ?nd
a signi?cant association between women audit partner and audit
fees, after controlling for the client attributes. This suggests that
discrimination against women audit partners currently exists in
Taiwan. In addition, the discrimination against women is more
severe in masculine industries. Moreover, we provide evidence that
it is unlikely that our results are driven by the differences in audit
quality and audit reporting lags between women and men audit
partners and by the sample bias and omitted variables. The results
provide exploratory insights into the question of how the audit
partner's sex may affect audit pricing. Our results should be of in-
terest to audit ?rms in designing human resource programs and
compensation packages and to regulators in setting labor policies.
Two relevant studies investigating companies in Belgium and in
three Nordic countries found that women audit partners earn
higher audit fees, perhaps due to higher independence (Hardies,
Breesch, & Branson, 2010), better diligence, and less over-
con?dence (Ittonen and Peni, 2012). These ?ndings suggest that sex
stereotypes may vary across cultures and that discrimination
against women in the workplace is less severe in Northern Euro-
pean countries as compared to Taiwan, where Confucianism ad-
vocates the masculine social value that women should stay at home
and be responsible for the household, while men are the bread-
winners and leaders of the family (Chan et al., 2002; Yang et al.,
2013). For example, Penner and Paret (2008) suggested that Asian
kindergarten boys perform best on entering kindergartens, while
Latino kindergarten girls perform better than Latino boys. With
U.S., U.K. and German data, Schein and Mueller (1992) supported
the view that sex differences differ across cultures. They found that
German women and German men were considered to have almost
the same ability to become managers, while, compared to German
women, British women were found to be considered to be less
likely to serve as managers. Besides, this study indicated that U.S.
residents do not express such sex stereotypes in regard to mana-
gerial positions, and both women and men are viewed as equally
capable of becoming managers.
Research related to audit fee differentials in the public ac-
counting profession is likely to impact public policy in the years
ahead. One important question is whether women audit partners
earn less just because they are discriminated. Our results indicate
that, while a pay gap between women and men clearly exists, the
extent of discrimination is neither consistent from industry to in-
dustry, nor from audit partner to audit partner. Since the pattern of
audit fee differences appears to be suf?ciently complex, policy-
makers will ?nd it necessary to approach the problem more ho-
listically. Clearly, there is a need for additional research on this
issue.
Our study is subject to a number of limitations. First, caution is
needed when applying the ?ndings in this paper to other countries,
since the culture and the development of welfare will differ inter-
nationally. However, our results suggest that international regula-
tors should consider mandatory disclosure of audit partner
information including sex, which may effectively signal the severity
of discrimination against women. Second, the explanatory power of
audit fee models in Taiwan is usually lower (Chen & Wu, 2004)
when compared to the adjusted R-square of U.S. audit fee studies. It
cannot be ruled out that our regressions can be made signi?cantly
better by the inclusion of some other omitted variables. Finally,
although we have shown some evidence that the lower audit fees
T.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 219e233 231
for women auditors should not be driven by the differences in audit
quality and audit reporting lags, we cannot fully rule out the pos-
sibility that other factors explain our ?ndings.
References
Abbott, L. J., Parker, S., & Presley, T. J. (2012). Women board presence and the
likelihood of ?nancial restatement. Accounting Horizons, 26(4), 607e629.
Accounting Research Development Foundation (ARDF). (1999). Auditing standard
no.33: Auditor report on ?nancial statements. Taiwan: ARDF.
Adams, C., & Harte, G. (2000). Making discrimination visible: the potential for social
accounting. Accounting Forum, 24(1), 56e79.
AICPA. (2008). Trends in the supply of accounting graduates and the demand for public
accounting recruits. New York: American Institute of Certi?ed Public
Accountants.
Al-Ajmi, J. (2008). Audit and reporting delays: evidence from an emerging market.
Advances in Accounting, 24(2), 217e226.
Anderson-Gough, F., Grey, C., & Robson, K. (2005). “Helping them to forget..”: the
organizational embedding of gender relations in public audit ?rms. Accounting,
Organizations and Society, 30(5), 469e490.
Ashbaugh, H., LaFond, R., & Mayhew, B. W. (2003). Do nonaudit services compro-
mise auditor independence? Further evidence. The Accounting Review, 78(3),
611e639.
Bamber, E. M., Bamber, L. S., & Schoderbek, M. P. (1993). Audit structure and other
determinants of audit report lag: an empirical analysis. Auditing: A Journal of
Practice & Theory, 12(1), 1e23.
Barua, A., Davidson, L. F., Rama, D. V., & Thiruvadi, S. (2010). CFO gender and ac-
cruals quality. Accounting Horizons, 24(1), 25e39.
Berik, G., Rodgers, Y., & Zveglich, J. E. (2004). International trade and gender wage
discrimination: evidence from East Asia. Review of Development Economics, 8(2),
237e254.
Blinder, A. S. (1973). Wage discrimination: reduced form and structural estimates.
Journal of Human Resources, 8(4), 436e455.
Bugeja, M., Matolcsy, Z. P., & Spiropoulos, H. (2012). Is there a gender gap in CEO
compensation? Journal of Corporate Finance, 18(4), 849e859.
Carcello, J. V., Hermanson, D. R., & Huss, H. F. (1995). Temporal changes in
bankruptcy-related reporting. Auditing: A Journal of Practice and Theory, 14(2),
133e143.
Carey, P., & Simnett, R. (2006). Audit partner tenure and audit quality. The Ac-
counting Review, 81(3), 653e676.
Chan, C. L. W., Yip, P. S. F., Ng, E. H. Y., Ho, P. C., Chan, C. H. Y., & Au, J. K. S. (2002).
Gender selection in China: Its meanings and implications. Journal of Assisted
Reproduction and Genetics, 19(9), 426e430.
Chen, H., Chen, J. Z., Lobo, G. J., & Wang, Y. (2011). Effects of audit quality on earnings
management and cost of equity capital: evidence from China. Contemporary
Accounting Research, 28(3), 892e925.
Chen, S., Sun, S. Y. J., & Wu, D. (2010). Client importance, institutional improve-
ments, and audit quality in China: an of?ce and individual auditor level anal-
ysis. The Accounting Review, 85(1), 127e158.
Chen, K. Y., & Wu, S. (2004). Industry specialists, audit fees and auditor size: evi-
dence from Taiwan. Taiwan Accounting Review, 5(2), 41e69.
Chi, H. Y., & Chin, C. L. (2011). Firm versus partner measures of auditor industry
expertise and effects on auditor quality. Auditing: A Journal of Practice and
Theory, 30(2), 201e229.
Chi, W., Huang, H., Liao, Y., & Xie, H. (2009). Mandatory audit partner rotation, audit
quality, and market perception: evidence from Taiwan. Contemporary Ac-
counting Research, 26(2), 359e391.
Chin, C. L., & Chi, H. Y. (2009). Reducing restatements with increased industry
expertise. Contemporary Accounting Research, 26(3), 729e765.
Choi, J. H., Kim, J. B., Liu, X., & Simunic, D. A. (2008). Audit pricing, legal liability
regimes, and Big 4 premiums: theory and cross-country evidence. Contempo-
rary Accounting Research, 25(1), 55e99.
Corcoran, M., & Duncan, G. J. (1979). Work history, labor force attachment and
earnings differences between the races and sexes. Journal of Human Resources,
14(1), 3e20.
Craswell, A. T., Francis, J. R., & Taylor, S. L. (1995). Auditor brand name reputations
and industry specializations. Journal of Accounting and Economics, 20(3),
297e322.
Dao, M., Raghunandan, K., & Rama, D. V. (2012). Shareholder voting on auditor
selection, audit fees, and audit quality. The Accounting Review, 87(1), 149e171.
Davis, L. R., Soo, B. S., & Trompeter, G. M. (2009). Auditor tenure and the ability to
meet or beat earnings forecasts. Contemporary Accounting Research, 26(2),
517e548.
DeAngelo, L. E. (1981). Auditor size and audit quality. Journal of Accounting and
Economics, 3(3), 183e199.
DeFond, M. L., Francis, J. R., & Wong, T. J. (2000). Auditor industry specialization and
market segmentation: evidence from Hong Kong. Auditing: A Journal of Practice
and Theory, 19(1), 49e66.
Eagly, A. H., & Carli, L. L. (2003). The women leadership advantage: an evaluation of
the evidence. The Leadership Quarterly, 14(6), 807e834.
Eagly, A. H., Makhijani, M. G., & Klonsky, G. B. (1992). Gender and the evaluation of
leaders: a meta-analysis. Psychological Bulletin, 111(1), 3e22.
Faccio, M., Marchica, M. T., & Mura, R. (2012). CEO gender, corporate risk-taking, and
the ef?ciency of capital allocation. (Unpublished working paper). Available fromhttp://ssrn.com/abstract¼2021136.
Ferguson, A., & Stokes, D. (2002). Brand name audit pricing, industry specialization,
and leadership premiums post-Big 8 and Big 6 mergers. Contemporary Ac-
counting Research, 19(1), 77e110.
Francis, J. R. (1984). The effect of audit ?rm size on audit prices. Journal of Ac-
counting and Economics, 6(2), 133e151.
Francis, B. B., Hasan, I., Park, J. C., & Wu, Q. (2009). Gender differences in ?nancial
reporting decision making: Evidence from accounting conservatism. (Unpublished
working paper). Available fromhttp://ssrn.com/abstract¼1471059.
Francis, J. R., Reichelt, K., & Wang, D. (2005). The pricing of national and cit-
yespeci?c reputations for industry expertise in the U.S. audit market. The Ac-
counting Review, 80(1), 113e136.
Francis, J. R., & Simon, D. T. (1987). A test of audit ?rm pricing in the small client
segment of the U.S. audit market. The Accounting Review, 62(1), 145e167.
Fung, S. Y. K., Gul, F. A., & Krishnan, J. (2012). City-level auditor industry speciali-
zation, economies of scale, and audit pricing. The Accounting Review, 87(4),
1281e1307.
Goldin, C. (1990). Understanding the gender gap: An economic history of American
women. New York, NY: Oxford University Press.
Gul, F. A., Hutchinson, M., & Lai, K. M. Y. (2013). Gender-diverse boards and prop-
erties of analyst earnings forecasts. Accounting Horizons, 27(3), 511e538.
Gul, F. A., Srinidhi, B., & Ng, A. C. (2011). Does board gender diversity improve the
informativeness of stock prices? Journal of Accounting and Economics, 51(3),
314e338.
Habib, A., & Bhuiyan, Md. B. U. (2011). Audit ?rm industry specialization and the
audit report lag. Journal of International Accounting, Auditing and Taxation, 20(1),
32e44.
Hardies, K., Breesch, D., & Branson, J. (2010). Audit quality: Paying more for a woman?
(Unpublished working paper). Available fromhttp://papers.ssrn.com/sol3/
papers.cfm?abstract_id¼1619911.
Henderson, B. C., &Kaplan, S. E. (2000). Anexaminationof audit report lag for banks: a
panel data approach. Auditing: A Journal of Practice and Theory, 19(2), 159e174.
Hooks, K., & Cheramy, S. J. (1989). Coping with women's expanding role in public
accounting. Journal of Accountancy, 167(2), 66e70.
Huang, J., & Kisgen, D. J. (2013). Gender and corporate ?nance: are male executives
overcon?dent relative to women executives? Journal of Financial Economics,
108(3), 822e839.
Huang, H. W., Liu, L. L., Raghunandan, K., & Rama, D. (2007). Auditor industry
specialization, client bargaining power, and audit fees: further evidence.
Auditing: A Journal of Practice and Theory, 26(1), 147e158.
Huang, H. W., Raghunandan, K., & Rama, D. (2009). Audit fees for initial audit en-
gagements before and after SOX. Auditing: A Journal of Practice and Theory, 28(1),
171e190.
Ittonen, K., & Peni, E. (2012). Auditor's gender and audit fees. International Journal of
Auditing, 16(1), 1e18.
Jagadison, K., Aier, J. C., Gunlock, M. T., & Lee, D. (2005). The ?nancial expertise of
CFOs and accounting restatements. Accounting Horizons, 19(3), 123e135.
Jarrell, S. B., & Stanley, S. D. (2004). Declining bias and gender wage discrimination?
A meta-regression analysis. Journal of Human Resources, 39(3), 828e838.
Jonnergård, K., Stafsudd, A., & Elg, U. (2010). Performance evaluations as gender
barriers in professional organizations: a study of auditing ?rms. Gender, Work
and Organization, 17(6), 721e747.
Khan, W. A., & Vieito, J. P. (2013). CEO gender and ?rm performance. Journal of
Economics and Business, 67(MayeJune), 55e66.
Kim, J. B., Liu, X., & Zheng, L. (2012). The impact of mandatory IFRS adoption on
audit fees: theory and evidence. The Accounting Review, 87(6), 2061e2094.
Knechel, W. R., & Payne, J. L. (2001). Additional evidence on audit report lag.
Auditing: A Journal of Practice and Theory, 20(1), 137e146.
Kothari, S., Leone, A., & Wasley, C. (2005). Performance matched discretionary
accrual measures. Journal of Accounting and Economics, 39(1), 163e197.
Krishnan, J., & Yang, J. S. (2009). Recent trends in audit report and earnings
announcement lags. Accounting Horizons, 23(3), 265e288.
Kulich, C., Trojanowski, G., Ryan, M. K., Haslam, S. A., & Renneboog, L. D. R. (2011).
Who gets the carrot and who gets the stick? Evidence of gender disparities in
executive remuneration. Strategic Management Journal, 32(3), 301e321.
Lam, K. C. K., McGuinness, P. B., & Vieito, J. P. (2013). CEO gender, executive
compensation and ?rm performance in Chinese-listed enterprises. Paci?c-Basin
Finance Journal, 21(1), 1136e1159.
Lee, H. Y., Mande, V., & Son, M. (2008). A comparison of reporting lags of multi-
national and domestic ?rms. Journal of International Financial Management and
Accounting, 19(1), 28e56.
Lehman, C. R. (1992). Herstory in accounting: the ?rst eighty years. Accounting,
Organizations and Society, 17(3e4), 261e297.
Leventis, S., & Weetman, P. (2004). Timeliness of ?nancial reporting: applicability of
disclosure theories in an emerging capital market. Accounting and Business
Research, 34(1), 43e56.
Liao, H. M., Wang, C. C., & Chi, W. C. (2012). What does the complete disclosure of
audit fee information tell us in Taiwan? Taiwan Accounting Review, 8(1), 49e88
(in Chinese).
Liu, C., & Wang, T. (2005). Going-concern opinions: before and after SAS No. 33.
Journal of Management, 22(4), 525e548.
Marlow, S., & Carter, S. (2004). Accounting for change: professional status, gender
disadvantage and self-employment. Women in Management Review, 19(1), 5e17.
T.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 219e233 232
Måsson, J., Elg, U., & Jonnergård, K. (2013). Gender-based career differences among
young auditors in Sweden. International Journal of Manpower, 34(6), 572e583.
Menon, K., & Williams, D. D. (2004). Former audit partners and abnormal accruals.
The Accounting Review, 79(4), 1095e1118.
Mohan, N., & Ruggiero, J. (2007). In?uence of ?rm performance and gender on CEO
compensation. Applied Economics, 39(9), 1107e1113.
Myers, J. N., Myers, L. A., & Omer, T. C. (2003). Exploring the term of the auditor-
client relationship and the quality of earnings: a case for mandatory auditor
rotation. The Accounting Review, 78(3), 779e799.
Niskanen, J., Karjalainen, J., Niskanen, M., & Karjalainen, J. (2011). Auditor gender
and corporate earnings management behavior in private Finnish ?rms. Mana-
gerial Auditing Journal, 26(9), 778e793.
Peni, E., & V€ ah€ amaa, S. (2010). Women executives and earnings management.
Managerial Finance, 36(7), 629e645.
Penner, A. M., & Paret, M. (2008). Gender differences in mathematics achievement:
exploring the early grades and the extremes. Social Science Research, 37(1),
239e253.
Perera, S., Fatseas, V., & Luckett, P. (1997). Performance evaluation of audit assistants
in chartered accounting ?rms. Australian Accounting Review, 7(1), 62e69.
Public Company Accounting Oversight Board (PCAOB). (2005). Improving trans-
parency through disclosure of engagement partner and certain other participants
in audits. Washington, D.C.: PCAOB.
Reichelt, K. J., & Wang, D. (2010). National and of?ce-speci?c measures of auditor
industry expertise and effects on audit quality. Journal of Accounting Research,
48(3), 647e686.
Richardson, S., Tuna, Y., & Wu, M. (2003). Capital market pressures and earnings
management: The case of earnings restatements (Unpublished working paper).
Philadelphia: University of Pennsylvania.
Rose, C. (2007). Does women board representation in?uence ?rm performance?
The Danish evidence. Corporate Governance: An International Review, 15(2),
404e413.
Schaefer, J., & Zimmer, M. (1995). Occupational licensure in the accounting pro-
fession: effects of public regulation on accountants' earnings. Journal of Applied
Business Research, 11(2), 9e16.
Schein, V. E., & Mueller, R. (1992). Sex role stereotyping and requisite management
characteristics: a cross-cultural look. Journal of Organizational Behavior, 13(5),
439e447.
Simon, D. T., & Francis, J. R. (1988). The effects of auditor change on audit fees: tests
of price cutting and price recovery. The Accounting Review, 63(2), 255e269.
Simunic, D. A. (1980). The pricing of audit services: theory and evidence. Journal of
Accounting Research, 18(1), 161e190.
Sloan, R. (1996). Does stock fully re?ect information in accruals and cash ?ows
about future earnings? The Accounting Review, 71(3), 289e315.
Srinidhi, B., Gul, F. A., & Tsui, J. (2011). Women directors and earnings quality.
Contemporary Accounting Research, 28(5), 1610e1644.
Stanley, T. D., & Jarrell, J. B. (1998). Gender wage discrimination bias? A meta-
regression analysis. Journal of Human Resources, 33(4), 947e973.
Sun, J., Liu, G., & Lan, G. (2011). Does women directorship on independent audit
committees constrain earnings management? Journal of Business Ethics, 99(3),
369e382.
Telberg, R. (1993, May 27). Deloitte and touche raises stakes on women's issues.
Accounting Today, p. 3.
Tinker, T., & Fearfull, A. (2007). The workplace politics of U.S. accounting race, class
and gender discrimination at Baruch College. Critical Perspectives on Accounting,
18(1), 123e138.
Ward, D. D., Elder, R. J., & Kattelus, S. C. (1994). Further evidence on the de-
terminants of municipal audit fees. The Accounting Review, 69(2), 399e411.
Whisenant, S., Sankaraguruswamy, S., & Raghunandan, K. (2003). Evidence on the
joint determination of audit and non-audit fees. Journal of Accounting Research,
41(4), 721e744.
White, M. J., & White, G. B. (2006). Implicit and explicit occupational gender ste-
reotypes. Sex Roles, 55(3e4), 259e266.
Whiting, R., & Wright, C. (2001). Explaining gender inequity in the New Zealand
accounting profession. British Accounting Review, 33(2), 191e222.
Yang, Y. F., Chen, Y. S., & Yang, L. W. (2013). Gender gap, training and ?nancial
performance: evidence from public accounting industry. The International
Journal of Human Resource Management, 24(19), 3697e3718.
Ye, K., Zhang, R., & Rezaee, Z. (2010). Does top executive gender diversity affect
earnings quality? A large sample analysis of Chinese listed ?rms. Advances in
Accounting, 26(1), 47e57.
Zerni, M. (2012). Audit partner specialization and audit fees: some evidence from
Sweden. Contemporary Accounting Research, 29(1), 312e340.
T.-C. Huang et al. / Asia Paci?c Management Review 20 (2015) 219e233 233
doc_581403973.pdf