Journal of Business Research: Josep Garcia-Blandon, Josep Maria Argilés-Bosch, Diego Ravenda T

Journal of Business Research 96 (2019) 238–249
Contents lists available at ScienceDirect
Journal of Business Research

journal homepage: www.elsevier.com/locate/jbusres
Is there a gender effect on the quality of audit services? T

a,⁎ b c
Josep Garcia-Blandon , Josep Maria Argilés-Bosch , Diego Ravenda
a
IQS School of Management, Universitat Ramon Llull, Spain
b
Department of Accounting, Universitat de Barcelona, Spain
c
Management Control, Accounting and Auditing, Toulouse Business School, Spain
ARTICLE INFO ABSTRACT
Keywords: Should we expect a gender effect on the quality of audit services? Both the behavioural economics literature and
Female auditor the glass ceiling for women in the audit profession suggest an affirmative answer to this question. We conduct
Gender differences the study with a sample of Spanish-listed companies for the period between 2008 and 2015 and use the quality of
Glass ceiling the client's financial statements as a proxy for the quality of audit services. The results provide sound and
Audit quality
consistent support for a positive female auditor effect on the quality of audit services. Although our figures
Gender specialization
suggest that men and women tend to audit different types of clients, the results are not driven by these dif-
ferences. Moreover, the gender effect occurs from the very first year of the appointment of a female partner to
replace a male auditor. This study may contribute to challenging stereotypes and may be informative for the
ongoing political debate regarding the under-representation of women in senior management levels.
1. Introduction Following the Enron scandal, regulators and policy makers became
particularly concerned about the quality of audit services (i.e., the
The term gender, in the psychology and sociology fields, refers to Sarbanes-Oxley Act and, more recently, the Directive 2014/56/EU on
the efforts of feminists to distinguish between the biological differences audit quality). We investigate whether there exists a gender effect on
between men and women and the differences that are determined by the quality of audit services. Such an effect could exist, as the psy-
social and cultural forces (Welsh, 1992). Hence, while sex differences chology and behavioural economics literature generally supports that
suggest the existence of stable intrinsic traits for men and women, the men and women tend to behave differently (e.g., Costa, Terracciano, &
concept of gender explains the effects of social norms on the differential McCrae, 2001; Croson & Gneezy, 2009; Feingold, 1994; Nettle, 2007;
perceptions between them (Unger, 1990; Welsh, 1992). In the guest Schmitt, Realo, Voracek, & Allik, 2008). The investigation of gender
editorial to an accounting and gender special issue, Broadbent and differences in risk aversion is particularly meaningful for this study. In a
Kirkham (2008, p. 465) make the following controversial statement: review article, Charness and Gneezy (2012) conclude that women are
“(….) though the World has, for some women, changed much over the more risk averse than men. This conclusion has potentially serious
last 30 years, it has, in some ways, stayed the same, and- for that reason implications for the job market, mainly in financial and accounting
the issue of gender remains just as relevant as it was then”. While it is fields, where risk aversion is a fundamental issue. According to Ittonen,
true that recent decades have observed the entry of women into the Vähämaa, and Vähämaa (2013), gender-based differences in cognitive
accounting profession in substantial numbers and that accounting firms information-processing, diligence, conservatism, overconfidence and
have implemented targeted policies towards supporting women, the top risk tolerance may impact auditor judgements, and consequently, the
ranks of the accounting profession remain resolutely gendered, showing final quality of audit services. Using a sample of Swedish and Finnish
a clear underrepresentation of women at senior levels (Broadbent & companies, the researchers investigate the association between the
Kirkham, 2008). This situation occurs not only in the accounting pro- gender of the auditor and the quality of the financial reports of clients
fession, as the ongoing political debate within the European Union and conclude that companies with female auditors tend to have higher
concerning the under-representation of women in firm management financial reporting quality.
positions demonstrates (Khlif & Achek, 2017). The research question of this study states: Do female auditors pro-
Auditors play a fundamental role in the classical corporate gov- vide higher-quality audit services than do male auditors? We conduct
ernance scheme in guaranteeing the quality of financial statements. the empirical analysis with a sample of Spanish listed companies for the
⁎
Corresponding author at: IQS School of Management, Universitat Ramon Llull, Via Augusta, 390, 08017 Barcelona, Spain.
E-mail addresses: josep.garcia@iqs.edu (J. Garcia-Blandon), josep.argiles@ub.edu (J.M. Argilés-Bosch).
https://doi.org/10.1016/j.jbusres.2018.11.024
Received 3 April 2018; Received in revised form 12 November 2018; Accepted 14 November 2018
0148-2963/ © 2018 Elsevier Inc. All rights reserved.
J. Garcia-Blandon et al. Journal of Business Research 96 (2019) 238–249
research period between 2008 and 2015. Similar to prior related studies these skills. Thereafter, we discuss how the existence of a glass ceiling
(Ittonen et al., 2013; Niskanen, Karjalainen, Niskanen, & Karjalainen, for women auditors may also provide sound arguments to support our
2011), we use the quality of the financial statements of the auditor's hypothesis. Finally, we discuss the evidence reported by related studies.
client (as measured by discretionary accruals), as a proxy for the quality
of audit services. According to the discussion in the former paragraph, 2.1. Gender differences in skills for the accounting profession
the audit sector provides an ideal setting in which to observe a gender
effect on the quality of the services provided. However, thus far, the Accounting conservatism, which is defined “as the accountant's
scarce available evidence is limited to the Scandinavian region and has tendency to require a higher degree of verification to recognize good
not provided sound and conclusive results. Therefore, we extend and news as gains than to recognize bad news as losses” (Basu, 1997, p. 7),
refine the work of Ittonen et al. (2013) for Scandinavia to the Spanish has significantly influenced the accounting profession for a while. In
audit market. Both the implications of the issue on various fields (i.e., accordance with Watts (2003), the available evidence supports that
gender studies, accounting and audit), and the limited available re- conservatism benefits financial reporting quality, at least from the
search make this study timely and potentially interesting. Additionally, perspective of the users of financial reports. Moreover, in countries such
it should be noted that, because of the importance of the institutional as Spain, which requires the signature of the auditor in the audit report,
context for the level of accountability of auditors (Bedard, 2012), the auditors are personally accountable for the audit report. Accordingly,
results of single country studies cannot be generalized. In our specific more conservative auditors are expected to be more committed to fi-
case, this assertion is stressed by the clear country (or region) compo- nancial reporting quality. The overconfidence of auditors may also af-
nent of gender issues. As an example, an important finding supporting fect audit quality (e.g., Messier, Owhoso, & Rakovski, 2008; Owhoso &
the idea that firms with female auditors would likely manipulate the Weickgenannt, 2009). According to Owhoso & Weickgenannt (2009, p.
financial statements less is that women appear to be more ethical than 17), “an overconfident auditor may fail to properly gauge the effec-
men (Ruegger & King, 1992). However, Chen, Velasquez Tuliao, Cullen, tiveness of a client's systems and procedures, believing that he or she
and Chang (2016) argue that gender differences in ethics are more has the capacity to discover whatever weaknesses the client may have.”
pronounced under the cultural dimension of gender egalitarianism. As information-processing is intrinsically related to the auditing task
Accordingly, as the Scandinavian region is regarded as more gender (e.g., Brown & Solomon, 1991; Maletta & Kida, 1993), auditors with
egalitarian than Spain, we may observe that companies with female stronger information-processing skills are likely to provide higher-
auditors manipulate financial statements less in Scandinavia, whereas quality audit services. Finally, it should be noted that the audit pro-
the situation is not necessarily the same in Spain. Therefore, we intend fession plays a social trust function by representing a mechanism for
to contribute to the literature by providing the first study on the re- holding managers accountable for their actions (Jensen, 2006).
lationship between auditor gender and financial reporting quality out- Therefore, auditors confront a serious conflict of interest as they owe a
side the Scandinavian region. We also strive to contribute to the on- professional duty to the company stockholders and to society in gen-
going political debate regarding the under-representation of women in eral, although their remuneration is determined by the managers of the
management positions in firms, providing updated evidence on the si- audited companies (Citron & Taffler, 2001). In sum, conservatism, in-
tuation in Spain and its evolution during the last decade, and more formation-processing skills and ethical behaviour should positively
importantly, on the implications of this situation for the quality of audit impact the quality of audit services, while overconfidence should have
services. Moreover, in contrast to prior related studies, we control for a negative impact.
the different types of firms that men and women generally audit, as well Behavioural research has generally shown gender differences in the
as analyse the impact of the substitution of a male auditor for a female skills previously discussed. For conservatism, Byrnes, Miller, and
auditor on financial reporting quality. Schafer's (1999) meta-analysis on 150 studies shows that women are, in
Our results provide sound and consistent support for a gender effect general, more conservative than men. In this regard, focusing specifi-
on the quality of audit services in Spain. Specifically, we find that fe- cally on financial risk, there appears to be a consensus regarding the
male auditors are positively and significantly associated with higher following: 1) women are more risk averse than men (e.g., Charness &
financial reporting quality. This result appears robust and consistent Gneezy, 2012; Jianakoplos & Bernasek, 1998; Sunden & Surette, 1998);
across a variety of additional analyses and checks and, in particular, for 2) men are more overconfident than women (e.g., Bengtsson, Persson, &
the observation that men and women generally audit different types of Willenhag, 2005; Levin, Snyder, & Chapman, 1988; Lundeberg, Fox, &
firms. We also observe that the positive gender effect on the quality of Puncochar, 1994); and 3) women possess stronger information-pro-
audit services occurs from the very first year of the appointment of a cessing skills than men (e.g., Chung & Monroe, 2001; Darley & Smith,
woman to replace a male auditor. According to Dambrin and Lambert 1995). Regarding this latter issue, O'Donnell and Johnson (2001) con-
(2012), the lack of women in the higher ranks of the accounting pro- clude that female auditors demonstrate significantly greater efficiency
fession suggests that, until stereotypes are challenged, gender in- than men in complex analytical procedures tasks. Finally, although
equality will remain. The researchers link the need for critical and re- there is no agreement on gender differences in ethical behaviour, cer-
flective research on gender with an ongoing struggle to improve the tain studies show that women tend to be more ethical than men in the
position of women's representation in the accounting profession perception of business ethical situations (Ruegger & King, 1992) and
(Haynes, 2017). The results reported here intend to contribute to this less willing to engage in unethical behaviour in the workplace to obtain
goal. financial rewards (Betz, O'Connell, & Shepard, 1989). Focusing speci-
The remainder of the paper is organized as follows. Section two fically on the audit profession, Bernardi and Arnold (1997) conclude
outlines a review of the literature and develops our hypothesis. In that female auditors are at a significantly higher average level of moral
section three, we present the design of the empirical research, followed development than are male auditors. However, evidence reported by
by the discussion of the results in section four. Finally, in the last sec- other studies is not as conclusive (Roxas & Stoneback, 2004).
tion, we derive the conclusions and limitations of our research, as well
as the implications of the findings. 2.2. Barriers to entry and glass ceiling
2. Background and hypothesis development Although women appear to possess strong skills for the audit pro-
fession, barriers to entry have led to the under-representation of women
To answer the research question of this study, first, we examine the at senior levels of audit firms. As proffered by Haynes (2017), the in-
personal characteristics and skills that are perceived as the most im- corporation of women into the accounting profession occurred over
portant in the audit profession and infer potential gender differences in many decades of struggle. Kirkham and Loft (1993) stress the gender
239
hierarchy of accountancy dominated by men when they claim that the modified opinion in the audit report. The authors interpret this result in
masculine qualities required contrasted markedly with the image of terms of higher audit quality associated with female auditors.2 Another
women as being weak, dependent and emotional. After acknowledging related issue is how the presence of women on audit committees im-
advancements in recent decades in accountancy and other service pacts financial reporting quality. The same reasons that support a
professions, Empson, Muzio, Broschak, and Hinings (2015) note that gender effect on financial reporting quality may also explain higher
barriers to real advancement remain important. The changing gender financial reporting quality associated with the presence of female di-
balance has not been straightforward, and quantitative changes in the rectors on the audit committee. Although the available evidence gen-
numbers of women have not been accompanied by gender transfor- erally supports this view (e.g., Pucheta-Martinez, Bel-Oms, & Olcina-
mation in the hierarchy of the profession (Ciancanelli, Gallhofer, Sempere, 2016; Thiruvadi & Huang, 2011), Sun, Liu, and Lan (2011)
Humphrey, & Kirkham, 1990; Haynes, 2017). According to Anderson- are unable to identify any association between the proportion of female
Gough, Grey, & Robson (2005), women confront stronger difficulties to directors on audit committees and financial reporting quality. Finally,
attain leading positions in the audit profession because the evaluation the available evidence generally supports that firms with female chief
of career promotion opportunities does not only involve technical and financial officers tend to show higher financial reporting quality (e.g.,
managerial skills but also the integration into informal and formal or- Barua, Davidson, Rama, & Thiruvadi, 2010; Francis, Hasan, Park, & Wu,
ganizational processes that systematically disadvantage female auditors 2015; Peni & Vähämaa, 2010).
with family commitments. Similarly, Guillaume and Pochic (2009)
argue that promotion criteria generally involve disadvantages for
women. Finally, Haynes (2017) summarizes the main areas in which 2.4. Hypothesis
important challenges remain: gendered career hierarchies (Broadbent &
Kirkham, 2008; Dambrin & Lambert, 2012), interaction with mother- As we previously discussed in this section, women would appear to
hood (Dambrin & Lambert, 2012; Haynes, 2008), work-life issues and outperform men in the skills that are considered the most important for
choices (Anderson-Gough et al., 2005; Ladva & Andrew, 2014), and the audit profession. Moreover, a female auditor gender effect on the
feminisation and segmentation in the profession (Almer, Lightbody, & quality of audit work may also be expected due to the important bar-
Single, 2012; Khalifa, 2013; Lupu, 2012). riers of entry in the audit profession, which means that female auditors
in partnership positions must demonstrate extra competence (Ittonen
et al., 2013). Accordingly, we should expect these female partners to
2.3. Results of related empirical studies
provide, on average, higher-quality audit services than male auditors.
Regarding the empirical evidence, prior related research provides
Likely because most countries do not require the signature of the
support for a positive relationship between having a female auditor and
auditor on the audit report, the available empirical evidence is very
financial reporting quality, and there is also evidence that both the
limited and focuses on the Scandinavian region. Using a sample of small
presence of women on audit committees and in chief financial officer
and medium-size private Finnish firms, Niskanen et al. (2011) study the
positions is associated with higher financial reporting quality. There-
impact of auditor gender on discretionary accruals, a usual proxy for
fore, we pose the hypothesis of this study as follows:
audit quality. As the authors acknowledge, their results were twofold.
On the one hand, the authors observe that companies with female au- Hypothesis. Female auditors are positively and significantly associated
ditors show higher discretionary accruals in absolute values, thus sug- with financial reporting quality.
gesting more accounting manipulation and lower financial reporting
quality. However, on the other hand, in the analysis conducted with
separate sub-samples of income-increasing (earnings overstatements)
3. Research design and sample selection
and income-decreasing (earnings understatements) accruals, female
auditors appear to be more conservative than men, which is an in-
3.1. Research design
dicator of higher financial reporting quality. Subsequently, Ittonen
et al. (2013) conduct a similar investigation on a sample of public
In accordance with prior related studies (Ittonen et al., 2013;
Swedish and Finnish firms and observe a positive and significant direct
Niskanen et al., 2011), we measure financial reporting quality with
relationship between having a female auditor and the quality of fi-
discretionary accruals, a usual indicator of accounting manipulation.
nancial reports as measured by the absolute value of discretionary ac-
Similar to these studies, we assume that high-quality auditors should
cruals. However, the evidence supporting this conclusion was not
lead to high financial reporting quality by reducing the client's use of
overly strong, as it was not consistent for both univariate and multi-
discretionary accruals for accounting manipulation purposes. To esti-
variate analyses or across different measures of discretionary accruals.
mate discretionary accruals, we utilize the approach proposed by
Other studies have addressed related issues that are also of interest
Francis and Wang (2008). Accordingly, we first calculate total accruals
for our research. Specifically, Menezes and Bras (2015) find evidence
as operating net income minus cash flows from operations scaled by
that a predominant presence of female certified public accountants in
lagged total assets. Then, we define discretionary accruals as total ac-
partnership positions in audit firms is positively associated with the
cruals in year t minus predicted accruals for year t as defined below.
quality of financial reports of the clients of the audit firm. Hardies,
Similar to Francis and Wang (2008), we prefer this measure of accruals
Breesch, and Branson (2015) observe that Belgian firms pay higher
over the Jones-type models (Dechow, Sloan, & Sweeney, 1995, Jones,
audit fees (by approximately 7%) to female auditors. This evidence may
1991, Kothari, Leone, & Wasley, 2005) due to the generally low number
be interpreted in terms of audit quality (higher audit fees may indicate
of observations per year, industry and country in our sample.3
higher-quality audit services). However, there were also alternative
explanations, as the hiring of a female auditor may be demand-driven
without necessarily involving higher audit quality.1 In addition for
Belgium, Hardies, Breesch, and Branson (2016) observe that firms with
2
female auditors have a higher likelihood of receiving a going-concern The higher propensity to issue going-concern opinions would indicate that
female auditors are more independent.
3
The standard approach in the literature is to conduct separate cross-section
1
For example, given that women are generally found to be more agreeable, estimations of Jones (1991) type models at the industry level. However, as
tenderminded, warm and open to feelings, female auditors may be associated Francis and Wang (2008) note, this approach may not be feasible in many
with higher client satisfaction (Hardies et al., 2015, p. 174). country-studies because of the low number of firms in certain industries.
240
Predicted accruals={[salest × (current accrualst 1/salest 1)] GROWTH (sales growth). Sales in year t minus sales in t − 1 and
[gross PPEt × (depreciationt 1/gross PPEt 1]} scaled by sales in year t − 1; and
ΔPPE (growth rate of gross property plant and equipment). Gross
/total assetst 1
PPE in year t minus gross PPE in t − 1 and scaled by gross PPE in
where t − 1.
Current accruals (change in non cash working capital)=

Year and industry dummy variables are also included as fixed ef-
[total current assets fects.
cash and short term investments According to the hypothesis of this study, we predict a negative and
treasury stock shown as current assets] significant coefficient for our variable of interest, FEMALE. This pre-
[total current liabilities diction would indicate that female auditors are associated with lower
total amount of debt in current liabilities levels of discretionary accruals and, thus, with higher financial re-
proposed dividends] porting quality.
gross PPE: grossproperty, plant and equipment. Next, we briefly discuss the control variables in our model in similar
terms as Francis and Wang (2008). We include BIG4 because prior
After computing discretionary accruals, the next step is to decide
studies generally show that large audit firms provide higher quality
which form of this variable (the absolute or the raw values) is the most
services (e.g., Becker, DeFond, Jiambalvo, & Subramanyam, 1998;
appropriate measure of financial reporting quality. This issue is not
Francis, Maydew, & Sparks, 1999). According to Francis and Wang
minor, as occasionally results are inconsistent across different defini-
(2008), large firms (LSALES) show lower levels of accruals than smaller
tions of accruals (e.g., Niskanen et al., 2011). To report robust results,
firms, including after scaling accruals by firm size. With CFO, we intend
we use both measures of discretionary accruals, as well as income-in-
to control for a well-documented inverse relationship between oper-
creasing and income-decreasing discretionary accruals. However, the
ating cash flows and accruals (Dechow, 1994). We include LEV, as
main analysis is conducted with the raw value of discretionary ac-
highly leveraged firms have stronger incentives to manipulate earnings
cruals.4
to avoid debt covenant violations (Becker et al., 1998). The reporting of
As a preliminary step, we conduct a univariate analysis of the mean
losses (LAG_LOSS) constitutes an indicator of poor financial health, and
and median differences of discretionary accruals by auditor gender.
according to DeAngelo, DeAngelo, and Skinner (1994), troubled com-
Subsequently, for the multivariate analysis, we estimate the model
panies confront strong incentives to manipulate earnings. Finally, as
given by Eq. (1) below, which includes discretionary accruals (ACCR-
Francis and Wang (2008), we include GROWTH and ΔPPE to control for
UALS) as the dependent variable, our variable of interest FEMALE in-
the growth and property and the plant and equipment of firms, which
dicating auditor gender, as well as the usual control variables (Francis &
are two recognized drivers of accruals.
Wang, 2008).
ACCRUALSi, t = 0 + 1 FEMALEi, t + 2 BIG 4i, t + 3 LSALESi, t + 4 CFOi,t
+ 5LEVi, t + 6 LAG _LOSSi, t + 7 GROWTHi, t 3.2. Sample selection
+ 8 PPEi, t + fixed effectsi, t + i, t (1)
We conduct the empirical analysis on a sample of non-financial
companies5 listed on the Spanish Stock Exchange (Sistema de Inter-
Dependent variable:
conexión Bursátil Español) for the period between 2008 and 2015. In-
ACCRUALS (discretionary accruals). Defined as the raw value of
formation regarding our variable of interest (FEMALE) is hand collected
discretionary accruals (DA) in the main model. The absolute value
from financial reports available at the National Securities Market
(ABSDA), as well as income-increasing (IIDA) and income-de-
Commission (Comisión Nacional del Mercado de Valores - CNMV) web-
creasing (IDDA) accruals are also used as dependent variables for
site. Data for control variables are obtained from Capital IQ. Our sample
additional analyses.
consists of 91 firms and, given the eight-year research period, 728 firm-
Variable of interest:
year observations. However, due to lack of data for seven firm-year
FEMALE (female auditor). A dummy variable that takes the value
observations, the final sample consists of 721 firm-year observations.
of 1 when the firm has a female auditor, and 0 otherwise.
Table 1 summarizes the presence of female audit partners in Spain
Controls:
during the research period. As it can be observed, firms with female
BIG4 (type of audit firm). A dummy variable that takes the value
auditors only represent 10% of the total sample. This percentage is
of 1 if the auditor belongs to a Big 4 audit firm, and 0 otherwise;
slightly lower than in Ittonen et al. (2013) (11%). However, it should be
LSALES (client's sales). The log of the client's sales in millions of €;
noted that the research period in Ittonen et al. (2013) was 2005–2007.
CFO (cash-flow from operations). Operating cash flows of the
For 2008, the first year of our research period, the presence of women
client scaled by lagged total assets;
in our sample is only 5%.6 Thus, a first conclusion is that the presence
LEV (leverage). The client's financial leverage measured as total
of female audit partners in Spain is lower (or at least was lower in the
liabilities divided by total assets;
past) than in the Scandinavian region. A second interesting figure is
LAG_LOSS (lagged losses). A dummy variable that takes the value
that the number of audit reports signed by women has tripled during
of 1 if the client reports negative income before extraordinary
the research period, meaning an average annual rate of growth of 17%.
items in year t − 1, and 0 otherwise;
Finally, the presence of women is not homogeneously distributed across
industries; for instance, “financial services and real estate”, “technology
4
Similar to Francis and Wang (2008), we base our decision on the fact that and telecommunications” and “consumer services” are the industries
the use of the absolute value of discretionary accruals does not allow one to with the highest percentage of female auditors, whereas “petrol and
control for the different implications of earnings overstatements and earnings power” and “basic materials, industry and construction” are the in-
understatements. The limitations of absolute discretionary accruals as a reliable dustries with the lowest percentage. These figures indicate that men
measure of financial reporting quality are clearly illustrated by the fact that,
although Niskanen et al. (2011) found a positive and significant relationship
5
between having a female auditor and absolute discretionary accruals and As is usual in the audit literature, we exclude financial companies because
Ittonen et al. (2013) reported a negative and significant relationship, both we use financial ratios as control variables in Eq. (1).
6
authors derived the same conclusion; firms with female auditors had higher We do not compare our figures with Niskanen et al. (2011) or Hardies et al.
financial reporting quality. (2016) because both papers use samples of private firms.
241
Table 1
Number of audit reports per year and industry. In parentheses, the number of audit reports signed by female auditors.
2008 2009 2010 2011 2012 2013 2014 2015 Total
Petrol and power 8 (0) 8 (0) 8 (0) 8 (0) 8 (0) 8 (1) 8 (2) 8 (1) 64 (4)
Basic materials, industry and construction 29 (1) 28 (0) 28 (0) 28 (2) 29 (3) 29 (3) 29 (3) 29 (3) 229 (15)
Consumer goods 23 (1) 23 (1) 23 (2) 23 (2) 23 (3) 23 (3) 22 (3) 22 (3) 182 (18)
Consumer services 14 (2) 14 (2) 14 (2) 14 (2) 14 (2) 14 (2) 13 (1) 13 (3) 110 (16)
Financial services and real estate 11 (1) 11 (2) 11 (1) 11 (1) 11 (1) 11 (1) 11 (3) 11 (3) 88 (13)
Technology and telecommunications 6 (0) 6 (0) 6 (0) 6 (0) 6 (1) 6 (2) 6 (2) 6 (2) 48 (7)
Total 91 (5) 90 (5) 90 (5) 90 (7) 91 (10) 91 (12) 89 (14) 89 (15) 721 (73)
Table 2 Table 4
Descriptive statistics for our sample. Univariate analysis. Mean and median values of discretionary accruals for the
subsamples of firms with male and with female auditors.
Variable Mean Median Standard dev. Maximum Minimum
Variable Mean p-Value Median p-Value
FEMALE 0.10 0.00 0.30 1.00 0.00
BIG4 0.94 1.00 0.23 1.00 0.00 Female Male Female Male
LSALES 2.77 2.80 0.88 4.92 0.18 auditor auditor auditor auditor
CFO 0.07 0.06 0.08 0.50 −0.34
LEV 0.74 0.67 1.03 21.17 0.10 DA −0.068 −0.037 0.027 −0.048 −0.040 0.088
LAG_LOSS 0.28 0.00 0.45 1.00 0.00 ABSDA 0.111 0.089 0.069 0.075 0.061 0.452
GROWTH 0.02 0.00 0.45 6.82 −0.94 IIDA 0.082 0.083 0.959 0.035 0.048 0.638
ΔPPE 0.04 −0.02 0.95 20.71 −2.34 IDDA −0.111 −0.086 0.025 −0.076 −0.066 0.288
Variables: FEMALE (female auditor); BIG4 (type of audit firm); LSALES (client's The t-test is used to assess statistical significance of mean accruals while the
sales); CFO (cash-flow from operations); LEV (leverage); LAG_LOSS (lagged Mann-Whitney test is used for median accruals.
losses); GROWTH (sales growth); and ΔPPE (growth rate of gross property plant Variables: DA (raw value of discretionary accruals); ABSDA (absolute value of
and equipment). discretionary accruals); IIDA (income-increasing discretionary accruals); and
IDDA (income-decreasing discretionary accruals).
and women tend to audit different types of firms and show industry
specialization of auditors by gender. 4. Results and discussion
Table 2 provides descriptive statistics for our sample. The most in-
teresting feature is the extreme concentration of the Spanish audit 4.1. Results of the univariate analysis
market by Big 4 audit firms, representing 94% of the market for listed
companies. This percentage is slightly above the 90% value in Ittonen We begin with a univariate analysis of the mean and median dif-
et al. (2013). ferences of discretionary accruals by auditor gender, the results dis-
Table 3 provides Pearson correlation coefficients with significance played in Table 4. Although the main analysis is conducted with raw
levels. We are particularly interested in the correlation patterns of our discretionary accruals (DA), we also provide information for the alter-
gender variable FEMALE with discretionary accruals. The most re- native definitions of accruals (ABSDA, IIDA and IDDA). We use a t-test
markable result is the negative and significant correlation between both to assess the statistical significance of the mean differences and the
variables, suggesting that firms with female auditors show lower dis- Mann-Whitney test for median differences. The figures indicate that
cretionary accruals and, thus, higher financial reporting quality. An- firms with female auditors show significantly lower discretionary ac-
other interesting feature is the negative and significant correlation be- cruals; thus, they present higher financial reporting quality. The same
tween having a female auditor and client's sales, indicating that women result is observed with both the t-test and the Mann-Whitney test; al-
tend to audit smaller firms. Regarding the dependent variable, Table 3 though in the latter case, significance is reported only at marginal levels
shows the expected correlation pattern of discretionary accruals with (p-value < 0.10). This result reinforces the negative correlation be-
cash flows, leverage and losses. However, for the type of audit firm and tween FEMALE and DA observed in Table 3. Regarding the alternative
GROWTH, the sign of the correlation is opposite to our expectations. definitions of accruals, significant differences in the means are also
Finally, given the relatively low correlation coefficients (maximum observed for income-decreasing discretionary accruals and for the ab-
value is 0.32 for the correlation between sales and audit firm type), we solute value of discretionary accruals. In sum, data displayed in Table 4
do not expect serious multicollinearity problems in the dataset. suggest that female auditors appear to show stronger accounting con-
servatism, as the firms they audit have lower income-decreasing
Table 3
Pearson correlations and levels of significance.
DA FEMALE BIG4 LSALES CFO LEV LAG_LOSS GROWTH
FEMALE −0.13***
BIG4 0.10*** −0.04
LSALES −0.04 −0.07* 0.32***
CFO −0.25*** −0.02 0.11*** 0.19***
LEV 0.13*** −0.01 −0.15*** −0.13*** −0.11***
LAG_LOSS 0.09** 0.09** −0.18*** −0.25*** −0.30*** 0.15***
GROWTH −0.31*** 0.04 0.00 0.07* 0.15*** −0.08** −0.05
ΔPPE −0.06 0.10*** −0.02 0.02 0.15*** −0.07* −0.05 0.07*
*, **, ***Significant at 10%, 5% and 1% respectively.

Variables: DA (raw value of discretionary accruals); FEMALE (female auditor); BIG4 (type of audit firm); LSALES (client's sales); CFO (cash-flow from operations); LEV
(leverage); LAG_LOSS (lagged losses); GROWTH (sales growth); and ΔPPE (growth rate of gross property plant and equipment).
242
discretionary accruals eventually leading to lower raw discretionary associated with female auditors. The results are clearer for IDDA, as
accruals and larger absolute discretionary accruals.7 Prior evidence FEMALE shows a negative and statistically significant coefficient (p-
reported by Ittonen et al. (2013) does not show significant differences value < 0.05). This finding indicates that female auditors tend to be
in discretionary accruals by auditor gender.8 more conservative11 than male auditors, and as accounting con-
servatism is positively regarded in the financial reporting literature
4.2. Results of the multivariate analysis (Watts, 2003), it suggests higher financial reporting quality. The results
displayed in Table 5 stress the importance of considering the widest
Table 5 shows the estimates of Eq. (1). The first column displays the possible definition of discretionary accruals when this variable is used
results of the main analysis using raw discretionary accruals as a de- as a proxy for financial reporting quality. In sum, the results of the
pendent variable, whereas the remaining columns show the results of multivariate analysis are consistent with the preliminary univariate
the estimations with the alternative definitions of accruals. In ac- analysis displayed in Table 4.
cordance with the panel structure of our dataset, we use panel data To end this section, we investigate whether the positive association
techniques in all the estimations. As we detected heteroscedasticity and between female auditors and financial reporting quality may be ex-
autocorrelation in the dataset, we apply the Prais-Wistein regression plained by differences in the audit effort between male and female
with heteroscedastic panels corrected errors for the estimation of the auditors. Hence, we collected the amount of audit fees for our sample of
models with DA and ABSDA as the dependent variables. Regarding the firms during the research period12 and subsequently conducted a t-test
models with IIDA and IDDA, the fact that both variables are censored at and the Mann-Whitney test to seek significant gender differences in
zero advocates the use of Tobit panel data regression (Greene, 2000). audit fees. The results of both tests, which are displayed in Table 6
All four estimations are statistically significant at the usual levels (p- (panel B), do not provide support for significant differences in audit fees
value < 0.01) with relatively high explanatory power.9 Although the due to the gender of the audit partner. Therefore, the relationship be-
Pearson correlation coefficients displayed in Table 3 do not suggest tween auditor gender and financial reporting quality does not appear to
multicollinearity problems, after the estimations, we calculate variance be driven by gender differences in audit effort.
inflation factors (untabulated) to further assess potential multi-
collinearity problems. The low values of these factors (average value of
4.3. Robustness checks
1.16 with a maximum of 1.27 for LAG_LOSS) support the view of no
serious multicollinearity in our dataset.
In this subsection, we discuss the results of various checks con-
The results of the main analysis (first column of Table 5) show a
ducted to assess the robustness of the main result of this study that
negative and significant coefficient for FEMALE. Accordingly, female
female auditors provide higher-quality audit services.
auditors are associated with lower discretionary accruals and thus with
First, we check for potential endogeneity problems in our estima-
higher financial reporting quality. This result supports our hypothesis of
tions. The results of the Durbin–Wu–Hausman test (augmented regres-
a positive female auditor effect on the quality of financial statements.
sion test) do not suggest endogeneity problems in the estimations of Eq.
The results for control variables generally meet expectations. Hence,
(1) (results untabulated). Specifically, when we re-estimate the equa-
with the only exception of the type of audit firm (marginally significant
tion after including the residuals of the potentially endogenous variable
with a positive sign), whenever a significant effect is observed, it has
(FEMALE) as a new independent variable in the model, the coefficient
the predicted sign. We report significant effects for cash flow and
of the new variable is statistically insignificant (p-value = 0.861), while
growth (p-value < 0.01) and marginally significant results for leverage
FEMALE remains significant with the same sign as in the original esti-
(p-value < 0.1). Conversely, we report statistically insignificant re-
mations.
sults, at conventional levels, for sales, lagged losses and increase in
Second, after the estimation of Eq. (1) with the definition of dis-
property, plant and equipment.
cretionary accruals provided by Francis and Wang (2008), we check the
Regarding the additional analyses, we observe positive and sig-
robustness of our results to other definitions of discretionary accruals.
nificant coefficient for FEMALE in the model with discretionary accruals
Hence, we re-estimate Eq. (1), first with the definition of discretionary
in absolute values. This observation indicates that, when we ignore the
accruals proposed by DeFond and Park (2001) and Carey and Simnett
sign of accruals, firms with female auditors show significantly higher
(2006), among others, which essentially captures abnormal working
levels of discretionary accruals. Prior research has provided conflicting
capital accruals13 and thereafter, with discretionary accruals defined as
results on the impact auditor gender on absolute discretionary accruals.
in Kothari et al. (2005),14 controlling for firm performance. Given the
In particular, while Niskanen et al. (2011) observe that female auditors
relatively low number of observations in certain industries, in the latter
are associated with higher accruals, Ittonen et al. (2013) conclude the
case, we conduct country-year estimations of the model. For the first
opposite.10 Therefore, although our study shows more similarities with
definition of discretionary accruals, the results (untabulated) are very
Ittonen et al. (2013), the results are more in accordance with Niskanen
similar to those reported in Table 5. Conversely, when we use the de-
et al. (2011). Our segmented analyses with income-increasing (IIDA)
finition of accruals used by Kothari et al. (2005), FEMALE shows
and income-decreasing (IDDA) accruals support that the higher levels of
absolute discretionary accruals for firms with female auditors suggest
that women are, in fact, more conservative than men. The coefficient of 11
As income-decreasing accruals (IDDA) have a negative sign, lower levels of
IIDA is negative and on the edge of being statistically significant at the accruals mean higher earnings understatement adjustments and, thus, stronger
10% level. This finding suggests lower earnings overstatements accounting conservatism.
12
This analysis was conducted with a subsample of 612 firm-year observa-
tions because, for 109 cases, we were not able to collect the amount of audit
7
Female auditors are associated with lower income-decreasing discretionary fees.
13
accruals. As these accruals have negative sign, this result means stronger ac- AWCA (abnormal working capital accruals) are defined as:
counting conservatism, and thus, higher financial reporting quality. AWCAt = WCt − [(WCt−1 / St−1) × St], where WC and S denote non-cash
8
Ittonen et al. (2013) conduct only a t-test of the differences of the means and working capital and sales, respectively.
14
reported non-significant differences. The model is similar to the Jones (1991) model; however, it also includes
9
32% R-squared. In Niskanen et al. (2011) and Ittonen et al. (2013), pseudo R- lagged return on assets. It is defined as: TAit = d0/
square is 6% and 20%, respectively. ASSETSit−1 + d1ΔSALESit + d2PPEit + d3ROAit−1 + uit, where TA, ASSETS,
10
It should be noted that Ittonen et al. (2013) reported significant results ΔSALES, PPE and ROA denote total accruals, total assets, increase in sales,
with one measure of discretionary accruals but insignificant results with the property plant and equipment and return on assets, respectively. Discretionary
other measure. accruals are captured by the error term of the model.
243
Table 5
Multivariate analysis. Results of the estimation of the model (z-values in parentheses). Main analysis with DA as the dependent variable. Additional analyses with
ABSDA, IIDA and IDDA as the dependent variables.
Variable Predicted sign DA ABSDA IIDA IDDA
FEMALE – −0.032 0.028 −0.042 −0.030

(−3.13)*** (2.25)** (−1.62) (−2.41)**
BIG4 − 0.032 −0.061 0.006 0.037
(1.82)* (−1.99)** (0.17) (1.98)**
LSALES − 0.005 −0.032 −0.027 0.015
(1.31) (−5.13)*** (−2.13)** (2.52)**
CFO − −0.651 0.243 −0.834 −0.605
(−9.98)*** (3.94)*** (−7.93)*** (−10.55)***
LEV + 0.029 0.064 0.078 −0.000
(1.80)* (2.90)*** (2.78)*** (−0.00)
LAG_LOSS + 0.001 −0.003 −0.008 −0.004
(0.84) (−0.41) (−0.51) (−0.46)
GROWTH + 0.057 0.079 0.128 0.015
(3.13)*** (4.36)*** (5.61)*** (1.07)
ΔPPE + −0.018 −0.008 −0.028 −0.012
(−1.37) (−0.61) (−1.31) (−1.09)
Industry effects YES YES YES YES
Year effects YES YES YES YES
Constant −0.102 0.198 −0.095 −0.124
(−4.29)*** (5.30)*** (−1.60) (−4.16)***
# of observ. 721 721 203 518
R-sq. 0.32 0.32
Wald-Chi sq. 223.46*** 125.53*** 122.96*** 184.01***

Variables: DA (raw value of discretionary accruals); ABSDA (absolute value of discretionary accruals); IIDA (income-increasing discretionary accruals); IDDA (in-
come-decreasing discretionary accruals); FEMALE (female auditor); BIG4 (type of audit firm); LSALES (client's sales); CFO (cash-flow from operations); LEV
Table 6
Clients characteristics (panels A and C) and audit effort (panel B) by auditor gender.
Panel A: The “big baths” explanation Female auditor Male auditor p-Value
% of clients with big losses 15 11 0.28

% of clients with big negative surprises 14 12 0.64
Mean p-Value Median p-Value
Female auditor Male auditor Female auditor Male auditor
Panel B: Differences in audit effort

Audit fees (in millions of €) 0.70 1.28 0.18 0.22 0.37 0.25
Panel C: Difference in client's characteristics

CFO 0.06 0.07 0.67 0.04 0.06 0.31
LEV 0.70 0.68 0.66 0.67 0.66 0.84
Size (sales in millions of €) 932.86 4584.08 0.01 597.96 643.60 0.14
# of business segments 3.81 4.66 0.05 4.00 4.00 0.11
p-Values correspond to Pearson Chi-squared test in panel A, and t-test (means) and Mann-Whitney test (medians) in panels B and C.
negative and significant coefficients in all the new estimations with the accruals used by Kothari et al. (2005) do not support that female au-
only exception being the model with income-decreasing accruals ditors allow more big baths than male auditors. Under the “big baths”
(β = −0.011 with p-value < 0.05 in the model with DA; β = −0.007 explanation, the interpretation that female auditors provide higher-
with p-value < 0.10 in the model with ABSDA; β = −0.019 with p- quality audit services could be put into question, as “big baths” indicate
value < 0.05 in the model with IIDA; and β = −0.003 with p- lower financial reporting quality. To conduct this analysis, we define
value = 0.77 in the model with IDDA). Therefore, the positive asso- two new variables: BIGLOSS (big losses): 1 if net income divided by
ciation between female auditors and financial reporting quality ob- lagged total assets is lower than −10%, and 0 otherwise; and BIGNE-
served in Table 5 is robust across alternative definitions of discretionary GSURP (big negative surprise): 1 if the reported net income was < 75%
accruals. the expected net income by the consensus of analysts, and 0 other-
In the third check, we examine whether the positive association wise.15 Thereafter, we apply the Pearson chi-squared test to assess
between female auditors and financial reporting quality may be driven whether gender differences in both variables are significant. The results
by the observation that women could allow more “big baths” (one-time of the test displayed in Table 6 (panel A) do not suggest that female
overstatement of expenses that would reduce future expenses) than auditors allow more “big baths” than male auditors, as we do not
male auditors. This allowance would be consistent with the results
displayed in Table 5 in the models with IIDA (insignificant results for
15
FEMALE) and IDDA (significant results with negative sign for FEMALE); We choose these cut-off points based on the distribution of both variables
however, the results obtained using the definition of discretionary in our sample. However, we checked for the −5% and 50% thresholds for
BIGLOSS and BIGNEGSURP, respectively, and results do not change.
244
observe significant differences in either BIGLOSS or BIGNEGSURP by Table 7

auditor gender. Therefore, it does not appear that our results supporting Results of the estimation of the logistic model with accounting restatements
female association with higher financial reporting quality would be (RESTATE) as the dependent variable (z-values in parentheses).
biased by female propensity for “big baths” practices. Variable Predicted sign RESTATE
The last check addresses whether the higher financial reporting
quality observed for firms with female auditors may be driven by the FEMALE – −0.869
(−2.20)**
observation that, as women appear to be more risk-adverse than men
BIG4 − −2.116
(e.g., Charness & Gneezy, 2012; Jianakoplos & Bernasek, 1998; Sunden (−5.18)***
& Surette, 1998), they may tend to audit those companies with lower LSALES +/− 0.394
levels of risk. Although Eq. (1) currently controls for client riskiness, (2.83)***
AGE − 0.024
and the results of the endogeneity test previously discussed do not
(1.47)
suggest this to be the case, we further check for this possibility. Ac- CFO − −2.429
cordingly, we choose cash-flows from operations (CFO) and financial (−1.73)*
leverage (LEV) as the proxies for client's riskiness; we then apply both a LEV + 0.078
t-test and the Mann-Whitney test for assessing differences in client (0.19)
LAG_LOSS + 0.136
riskiness by auditor gender. As shown by Table 6 (panel C), both tests
(0.56)
agree that there are no significant gender differences in either CFO or Industry effects YES
LEV. Therefore, we should discard the explanation that the association Year effects YES
between auditor gender and the quality of audit services is driven by Constant −0.128
(−0.18)
differences in client riskiness.
# of observ. 721
R-sq. 0.122
4.4. Additional analysis (I): financial reporting quality measured by Wald-Chi sq. 92.63***
accounting restatements
Variables: RESTATE (accounting restatements); FEMALE (female auditor); BIG4
This analysis addresses whether the auditor gender effect on fi- (type of audit firm); LSALES (client's sales); AGE (client's age); CFO (cash-flow
nancial reporting quality holds for a different proxy of financial re- from operations); LEV (leverage); and LAG_LOSS (lagged losses).
porting quality. This analysis is motivated by the observation that, as
DeFond and Zhang (2014) note, measuring audit quality is a challen- quality holds when financial reporting quality is proxied by accounting
ging task, and all the proposed proxies (including discretionary ac- restatements.
cruals) present one limitation or another. The issuance of modified
audit opinions to those clients who deserve it and accounting restate-
ments are other usual proxies in the accounting and auditing literature 4.5. Additional analysis (II): analysis with a matched sample
(DeFond & Zhang, 2014). In the latter case, the restatement of ac-
counting statements is interpreted as an indicator of lack of financial Results in Tables 3, 4, 5 and 7 indicate a positive association be-
reporting quality. Due to the very low number of modified opinions in tween female auditors and the quality of audit services. While the
our sample, which is insufficient for conducting a sound analysis based battery of robustness checks conducted thus far makes us conclude that
on this variable, we choose accounting restatements as the additional this result is sound, we want to explore more in depth whether this
proxy for financial reporting quality. Accordingly, we propose the result may be driven by differences in the types of clients generally
model given by Eq. (2) below: audited by men and women. The glass ceiling for women in the audit
profession provides justification for this analysis, as the stronger diffi-
RESTATEi, t = 0 + 1 FEMALEi, t + 2 BIG 4i, t + 3 LSALESi, t + 4 AGE + culties of female auditors to attain the top of the hierarchy in the audit
+ CFOi, t + LEVi, t + LAG _LOSSi, t + fixed effectsi, t firms (Anderson-Gough et al., 2005; Dambrin & Lambert, 2012) suggest
5 6 7
that they likely audit the smallest and less complex clients, usually
+ i, t involving lower audit difficulties. To control for this issue, we first need
(2) to assess whether male and female partners do in fact audit different
types of clients. Table 6 (panel C) displays the mean and median values
The dependent variable RESTATE is defined as 1 if the accounting
of sales and number of business segments (proxies for size and com-
statements of year t are later restated, and 0 otherwise.16 The frequency
plexity, respectively) for firms by auditor gender. The Table 6 also
of restatements in our sample (20%) is in accordance with prior studies
provides the results of both a t-test and the Mann-Whitney test to assess
for the US (e.g., 17.6% in Paterson & Valencia, 2011 and 16% in Blay &
the statistical significance of these differences. While the t-test results
Geiger, 2013). The variable of interest (FEMALE) is the same as in Eq.
support that women generally audit smaller and less complex firms, the
(1), and we include the usual control variables for restatements (e.g.,
Mann-Whitney test provides statistically insignificant results. More-
Kinney, Palmrose, & Scholz, 2004; Paterson & Valencia, 2011). Ac-
over, Table 1 shows that the presence of female auditors is not homo-
cording to the hypothesis of this study, we expect a positive relationship
geneously distributed across industries (i.e., women being openly un-
between having a female auditor and financial reporting quality and,
derrepresented in “petrol and power” and in “basic materials, industry
therefore, a negative relationship with the likelihood of restatements.
and construction”). Therefore, according to the figures in Tables 1 and
Table 7 displays the estimates of the logistic model represented by Eq.
6, it appears convenient to check whether the gender effect observed in
(2). The most interesting result in the Table is the negative and sig-
this study thus far is driven by the different types of firms that men and
nificant coefficient of FEMALE (p-value < 0.05), indicating that, as
women generally audit.
predicted, firms with female auditors show lower likelihood of ac-
Accordingly, we re-estimate the model with a matched sample to
counting restatements. Accordingly, the main conclusion from Table 5
control for the facts that: 1) women generally audit smaller and less
that female auditors are associated with higher financial reporting
complex firms, and 2) their clients are not homogeneously distributed
across industries. The standard technique for constructing a matched
16
Restatements (categorized as “RS” in the Capital IQ database) involve any sample with a panel dataset is to match each firm with a female auditor
situation in which net income or cash from operations are fundamentally dif- in year t with a similar firm (in terms of industry, size and number of
ferent from originally reported. business segments) with a male auditor the same year. However,
245
because of the low presence of female auditors in our sample, we utilize Table 9 become significant (p-value < 0.05) with the predicted nega-
an alternative approach that ignores the time dimension in the tive sign. This finding indicates that the substitution of a male auditor
matching procedure; thus, we match each firm with a female auditor for a female auditor is associated with less overstatement of earnings,
with a similar firm (same industry, similar size and same number of providing further support against the “big baths” explanation pre-
business segments) with a male auditor, resulting in a final sample of viously discussed.
349 firm-year observations.17 Once created, it is necessary to check Conversely, MALE-MALE presents statistically insignificant coeffi-
whether the new sample adequately controls for the different types of cients in all the estimations. Hence, in contrast to what occurs when a
clients of male and female auditors. Both the t-test and the Mann- woman substitutes for a man as the auditor of the firm, the replacement
Whitney test (untabulated) results show insignificant differences in of a male auditor for another male auditor involves no significant effect
client's size or complexity by auditor gender in the new sample. Simi- on discretionary accruals, and this holds independently of how discre-
larly, the Kruskal–Wallis test (untabulated) shows insignificant differ- tionary accruals are defined. Therefore, we should conclude that while
ences in the industry distribution of clients by auditor gender. There- the appointment of a female auditor to replace a male auditor appears
fore, we should conclude that the matched sample effectively controls to be associated with higher-quality audit services, the change of a male
for these issues. auditor for another male auditor does not. However, it should be noted
Table 8 shows the results of the new set of estimations of Eq. (1) that, because of the low number of male-female auditor changes in our
with the matched sample. All four estimations are statistically sig- sample (only 20), the results of these estimations need to be carefully
nificant at the usual levels (p-value < 0.01), with similar explanatory considered.
power as the estimations based on the whole sample. However, because
of the smaller size of the matched sample, the results in Table 8 are not 5. Conclusions, implications ad limitations
as reliable as those displayed in Table 5. Thus, the main result is the
negative and significant coefficient of FEMALE in the main model with Women are clearly underrepresented in the top ranks of audit firms.
raw discretionary accruals (DA) as the dependent variable. As in This situation, which occurs not only in the accounting profession but
Table 5, this result indicates that female auditors are associated with also in other areas of management, has motivated intense academic and
lower discretionary accruals, and thus with higher financial reporting political debates on its causes and, more importantly, on the possible
quality. This result reinforces our previous findings with the whole solutions. In the specific case of Spain, our figures show that although
sample and indicates that results in Table 5 were not driven by differ- the audit reports signed by women have tripled during our eight-year
ences in the types of firms than men and women generally audit. In research period, at the end of 2015, they continued to represent < 20%
contrast to Table 5, we do not observe significant results for FEMALE in of total reports. Such an unbalanced situation, although far from sur-
any of the additional estimations. Nevertheless, the sign of the coeffi- prising, appears particularly undesirable in the audit profession, as the
cients is always the same as in Table 5, and in the estimations with behavioural economics literature suggests that women possess strong
accruals in absolute values and, particularly with income-decreasing advantages in some of the skills that are considered the most important
accruals (the estimations which show significant results for FEMALE in for the audit profession. Our results indicate the existence of a gender
Table 5), the results are on the edge of significance at the 10% level. effect on the quality of audit services, as having a female auditor is
associated with significantly higher levels of financial reporting quality.
4.6. Additional analysis (III): the substitution of a male auditor for a female To minimize the risk of reporting purely spurious results, we have
auditor conducted a variety of robustness checks.
The evidence reported here extends and reinforces prior related
In the last analysis of this study, we investigate whether the ap- research for the Scandinavian region, where the presence of women in
pointment of a female auditor to replace a male auditor has any sig- the senior levels of audit firms began earlier than in Spain. Our figures
nificant impact on financial reporting quality. According to our hy- also indicate that the presence of female auditors is much higher in
pothesis and the results reported thus far, we should expect a positive certain industries (real estate, financial and consumer services) than in
impact. Hence, we re-estimate Eq. (1) with two new variables of in- others (petrol and power and basic materials, industry and construc-
terest: MALE-FEMALE (1 if a female auditor substitutes a male auditor, tion). We also observe that women tend to audit relatively smaller and
and 0 otherwise) and MALE-MALE (1 if a male auditor substitutes an- less complex firms. Therefore, we have controlled for potential en-
other male auditor, and 0 otherwise), instead of the original variable dogeneity problems, which may cause the positive female auditor effect
FEMALE.18 Table 9 presents the new set of estimates, which are strongly observed here to be driven by the different types of firms audited by
consistent with the figures displayed in Table 5. Hence, the new vari- men and women. Nevertheless, the results obtained with a matched
able MALE-FEMALE has a negative and significant coefficient in the sample, which intends to control for this issue, do not suggest this as-
main model, indicating that the substitution of a male auditor for a sertion to be the case. Furthermore, the results show that the positive
female auditor is associated with lower raw discretionary accruals female auditor impact on financial reporting quality occurs since the
(higher financial reporting quality). Moreover, as in Table 5, we report very first year of the appointment of a female partner to replace a male
a positive and statistically significant coefficient in the model with auditor.
accruals in absolute values and a negative and significant coefficient in These results openly support the claim of Jeacle (2011) that the
the model with income-decreasing accruals. Nevertheless, while in female gender, often viewed as a negative characteristic to achieving
Table 5, the impact of FEMALE on income-increasing accruals was on career success within the discriminatory environment of the accounting
the edge of marginal significance, the results for MALE-FEMALE in firm, may generally act as a positive attribute in business. Similarly,
Dambrin and Lambert (2012) argue the need to perform critical and
17
reflective research on gender to improve the position of women's re-
The standard matching procedure would have involved the loss of ap- presentation in the accounting profession since, until stereotypes are
proximately 80% of the dataset. The fact that our initial sample was not overly
challenged, gender inequality will remain. This study may contribute to
large precludes us from utilising this approach.
18 the challenging of these stereotypes and may be informative for the
To conduct these new estimations, we removed from the sample seven
observations corresponding to substitutions of female auditors for male auditors ongoing political debate regarding the under-representation of women
and one observation corresponding to a substitution of a female auditor for in the senior levels of the accounting and audit profession.
another female auditor. Because of the low number of cases, we have not in- Like prior related studies, the main limitation of this study is due to
cluded the corresponding FEMALE-FEMALE or FEMALE-MALE variables in the the low presence of female auditors in our sample, which may make our
new estimations. results sensitive to the particular characteristics of these relatively few
246
Table 8
Results of the estimations with a matched sample (z-values in parentheses). Main analysis with DA as the dependent variable. Additional analyses with ABSDA, IIDA
and IDDA as the dependent variables.
FEMALE – −0.034 0.018 −0.037 −0.022

(−2.37)** (1.33) (−0.99) (−1.53)
BIG4 − 0.061 −0.073 0.045 0.069
(1.45) (−1.95)* (0.72) (2.75)***
LSALES − 0.002 −0.051 −0.037 0.019
(0.24) (−4.77)*** (−1.32) (1.91)*
CFO − −0.772 0.173 −1.353 −0.690
(−7.29)*** (1.94)* (−7.27)*** (−7.67)***
LEV + 0.076 0.089 0.137 0.005
(2.11)** (3.13)*** (3.09)*** (0.23)
LAG_LOSS + 0.011 −0.012 −0.010 0.000
(0.63) (−0.85) (−0.31) (0.03)
GROWTH + 0.076 0.116 0.210 0.007
(2.42)** (3.93)*** (5.83)*** (0.40)
ΔPPE + −0.025 −0.000 −0.076 −0.018
(−1.83)* (−0.03) (−1.84)* (−1.92)
Constant −0.135 0.230 −0.129 −0.135
(−2.27)** (3.71)*** (−1.17) (−3.13)***
# of observ. 349 349 257 92
R-sq. 0.35 0.36
Wald-Chi sq. 132.97*** 94.02*** 100.31*** 120.84***

come-decreasing discretionary accruals); FEMALE (female auditor); BIG4 (type of audit firm); LSALES (client's sales); CFO (cash-flow from operations); LEV
Table 9
The substitution of audit partners (MALE-FEMALE and MALE-MALE). Main analysis with DA as the dependent variable (z-values in parentheses). Additional analyses
with ABSDA, IIDA and IDDA as the dependent variables.
MALE-FEMALE – −0.065 0.049 −0.106 −0.056

(−3.12)*** (2.37)** (−1.99)** (−2.91)***
MALE-MALE +/− −0.101 0.005 −0.20 −0.011
(−1.33) (0.71) (−1.25) (−1.28)
BIG4 − 0.023 −0.067 −0.000 0.033
(1.29) (−2.21)** (−0.00) (1.76)*
LSALES − 0.006 −0.029 −0.030 0.015
(1.34) (−4.76)*** (−2.35)** (2.50)**
CFO − −0.642 0.250 −0.849 −0.602
(−9.95)*** (4.05)*** (−7.83)*** (−10.44)***
LEV + 0.037 0.064 0.079 0.001
(2.23)** (3.00)*** (2.83)*** (0.06)
LAG_LOSS + 0.007 −0.004 −0.010 −0.004
(0.76) (−0.48) (−0.60) (−0.45)
GROWTH + 0.057 0.078 0.125 0.013
(3.04)*** (4.17)*** (5.40)*** (0.93)
ΔPPE + −0.016 −0.013 −0.023 −0.011
(−1.16) (−0.92) (−1.09) (−0.97)
Constant −0.101 0.193 −0.075 −0.119
(−4.28)*** (5.11)*** (−1.26) (−3.94)***
N 713 713 713 713
R-sq. 0.33 0.33
Wald-Chi sq. 221.45*** 134.62*** 125.11*** 187.88***

come-decreasing discretionary accruals); MALE-FEMALE (change from male to female auditor); MALE-MALE (change from male to male auditor); LSALES (client's
sales); CFO (cash-flow from operations); LEV (leverage); LAG_LOSS (lagged losses); GROWTH (sales growth); and ΔPPE (growth rate of gross property plant and
equipment).
female auditors, beyond the gender of the auditor. Additionally, the by a combination of both.
research design of this study does not allow us to address whether the This study could be extended by various means. First, it would be
gender effect reported here is explained by differences in skills between interesting to elaborate on the drivers of the relationship between au-
male and female auditors, by the glass ceiling in the audit profession or ditor gender and audit quality. While our results do not appear to be
247
explained by gender differences in audit effort, other issues, such as Psychology, 81, 322–331.
differences in accounting conservatism, could be further explored. Croson, R., & Gneezy, U. (2009). Gender differences in preferences. Journal of Economic
Literature, 47(2), 1–27. https://doi.org/10.1257/jel.47.2.448.
Similarly, additional research on the relative importance of explana- Dambrin, C., & Lambert, C. (2012). Who is she and who are we? A reflexive journey in
tions based on gender differences in skills versus those based on the research into the rarity of women in the highest ranks of accounting. Critical
glass ceiling would be needed. Finally, the interaction between female Perspectives on Accounting, 23(1), 1–16. https://doi.org/10.1016/j.cpa.2011.06.006.
Darley, W. K., & Smith, R. E. (1995). Gender differences in information processing stra-
auditors, female CEOs and CFOs, and the presence of women on the tegies: An empirical test of the selectivity model in advertising response. Journal of
audit committee and the financial reporting quality could also be an Advertising, 24(1), 41–56. https://doi.org/10.1080/00913367.1995.10673467.
interesting issue to address and investigate. DeAngelo, H., DeAngelo, L., & Skinner, D. J. (1994). Accounting choice in troubled
companies. Journal of Accounting and Economics, 17(1), 113–143. https://doi.org/10.
1016/0165-4101(94)90007-8.
Funding Dechow, P. M. (1994). Accounting earnings and cash flows as measures of firm perfor-
mance. Journal of Accounting and Economics, 18(1), 3–42. https://doi.org/10.1016/
0165-4101(94)90016-7.
This research received no specific grant from funding agencies in
Dechow, P. M., Sloan, R. G., & Sweeney, A. P. (1995). Detecting earnings management.
the public, commercial, or not-for-profit sectors. The Accounting Review, 70(2), 193–225. http://www.jstor.org/stable/248303.
DeFond, M., & Park, C. W. (2001). The reversal of abnormal accruals and the market
Acknowledgments valuation of earnings surprises. The Accounting Review, 76(3), 375–404. https://doi.
org/10.2308/accr.2001.76.3.375.
DeFond, M., & Zhang, J. (2014). A review of archival auditing research. Journal of
We thank Claire Crutchley Lending (associate editor) and two Accounting and Economics, 58(2–3), 275–326. https://doi.org/10.1016/j.jacceco.
anonymous referees for helpful comments. 2014.09.002.
Empson, L., Muzio, D., Broschak, J., & Hinings, B. (2015). Researching professional ser-
vice firms: An introduction and overview. The Oxford handbook of professional service
References firms (pp. 1–24). .
Feingold, A. (1994). Gender differences in personality: A meta-analysis. Psychological
Bulletin, 116, 429–456.
Almer, E. D., Lightbody, M. G., & Single, L. E. (2012). Successful promotion or segregation
Francis, B., Hasan, I., Park, J. C., & Wu, Q. (2015). Gender differences in financial re-
from partnership? An examination of the “Post-Senior Manager” position in public
porting decision making: Evidence from accounting conservatism. Contemporary
accounting and the implications for women's careers. Accounting Forum, 36(2),
Accounting Research, 32(3), 1285–1318. https://doi.org/10.1111/1911-3846.12098.
122–133. https://doi.org/10.1016/j.accfor.2011.06.002.
Francis, J. R., Maydew, E. L., & Sparks, H. C. (1999). The role of Big 6 auditors in the
Anderson-Gough, F., Grey, C., & Robson, K. (2005). ‘Helping them to forget’: The orga-
credible reporting of accruals. Auditing: A Journal of Practice & Theory, 18(2), 17–34.
nizational embedding of gender relations in public audit firms. Accounting,
https://doi.org/10.2308/aud.1999.18.2.17.
Organizations and Society, 30(5), 469–490. https://doi.org/10.1016/j.aos.2004.05.
Francis, J. R., & Wang, D. (2008). The joint effect of investor protection and big 4 audits
003.
on earnings quality around the world. Contemporary Accounting Research, 25(1),
Barua, A., Davidson, L., Rama, D., & Thiruvadi, S. (2010). CFO gender and accruals
157–191. https://doi.org/10.1506/car.25.1.6.
quality. Accounting Horizons, 24(1), 25–39. https://doi.org/10.2308/acch.2010.24.
Greene, W. H. (2000). Econometric analysis (4th ed.). New York: Prentice Hall.
1.25.
Guillaume, C., & Pochic, S. (2009). What would you sacrifice? Access to top management
Basu, S. (1997). The conservatism principle and the asymmetric timeliness of earnings.
and the work–life balance. Gender, Work and Organization, 16(1), 14–36. https://doi.
Journal of Accounting and Economics, 24(1), 3–37. https://doi.org/10.1016/S0165-
org/10.1111/j.1468-0432.2007.00354.x.
4101(97)00014-1.
Hardies, K., Breesch, D., & Branson, J. (2015). The female audit fee premium. Auditing: A
Becker, C. L., DeFond, M. L., Jiambalvo, J. J., & Subramanyam, K. R. (1998). The effect of
Journal of Practice & Theory, 34(4), 171–195. https://doi.org/10.2308/ajpt-51079.
audit quality on earnings management. Contemporary Accounting Research, 15(1),
Hardies, K., Breesch, D., & Branson, J. (2016). Do (fe) male auditors impair audit quality?
1–24. https://doi.org/10.1111/j.1911-3846.1998.tb00547.x.
Evidence from going-concern opinions. The European Accounting Review, 25(1), 7–34.
Bedard, J. (2012). Discussion of “Audit partner specialization and audit fees: Some evi-
https://doi.org/10.1080/09638180.2014.921445.
dence from Sweden”. Contemporary Accounting Research, 29(1), 341–348. https://doi.
Haynes, K. (2008). Transforming identities: Accounting professionals and the transition to
org/10.1111/j.1911-3846.2011.01153.x.
motherhood. Critical Perspectives on Accounting, 19(5), 620–642. https://doi.org/10.
Bengtsson, C., Persson, M., & Willenhag, P. (2005). Gender and overconfidence.
1016/j.cpa.2006.10.003.
Economics Letters, 86(2), 199–203. https://doi.org/10.1016/j.econlet.2004.07.012.
Haynes, K. (2017). Accounting as gendering and gendered: A review of 25 years of critical
Bernardi, R. A., & Arnold, D. F. (1997). An examination of moral development within
accounting research on gender. Critical Perspectives on Accounting, 43(2), 110–124.
public accounting by gender, staff level, and firm. Contemporary Accounting Research,
https://doi.org/10.1016/j.cpa.2016.06.004.
14(4), 653–668. https://doi.org/10.1111/j.1911-3846.1997.tb00545.x.
Ittonen, K., Vähämaa, E., & Vähämaa, S. (2013). Female auditors and accruals quality.
Betz, M., O'Connell, L., & Shepard, J. M. (1989). Gender differences in proclivity for
Accounting Horizons, 27(2), 205–228. https://doi.org/10.2308/acch-50400.
unethical behavior. Journal of Business Ethics, 8(5), 321–324. https://doi.org/10.
Jeacle, I. (2011). A practice of her own: Female career success beyond the accounting
1007/BF00381722.
firm. Critical Perspectives on Accounting, 22(3), https://doi.org/10.1016/j.cpa.2010.
Blay, A. D., & Geiger, M. A. (2013). Auditor fees and auditor independence: Evidence
12.002 (288-03).
from going concern reporting decisions. Contemporary Accounting Research, 30(2),
Jensen, M. (2006). Should we stay or should we go? Accountability, status anxiety, and
579–606. https://doi.org/10.1111/j.1911-3846.2012.01166.x.
client defections. Administrative Science Quarterly, 51(1), 97–128. https://doi.org/10.
Broadbent, J., & Kirkham, L. (2008). Glass ceilings, glass cliffs or new worlds? Accounting,
2189/asqu.51.1.97.
Auditing & Accountability Journal, 21(4), 465–473. https://doi.org/10.1108/
Jianakoplos, N. A., & Bernasek, A. (1998). Are women more risk averse? Economic Inquiry,
09513570810872888.
36(4), 620–630. https://doi.org/10.1111/j.1465-7295.1998.tb01740.x.
Brown, C. E., & Solomon, I. (1991). Configural data processing in auditing: The role of
Jones, J. J. (1991). Earnings management during import relief investigation. Journal of
domain specific knowledge. The Accounting Review, 66(1), 110–119. http://www.
Accounting Research, 29(2), 193–228. https://www.jstor.org/stable/2491047.
jstor.org/stable/247708.
Khalifa, R. (2013). Intra-professional hierarchies: The gendering of accounting special-
Byrnes, J. P., Miller, D. C., & Schafer, W. D. (1999). Gender differences in risk taking: A
isms in UK accountancy. Accounting, Auditing & Accountability Journal, 26(8),
meta-analysis. Psychological Bulletin, 125, 367–383.
1212–1245. https://doi.org/10.1108/AAAJ-05-2013-1358.
Carey, P. J., & Simnett, R. (2006). Audit partner tenure and audit quality. The Accounting
Khlif, H., & Achek, I. (2017). Gender in accounting research: A review. Managerial
Review, 81(3), 653–676. https://doi.org/10.2308/accr.2006.81.3.653.
Auditing Journal, 32(6), 627–655. https://doi.org/10.1108/MAJ-02-2016-1319.
Charness, G., & Gneezy, U. (2012). Strong evidence for gender differences in risk taking.
Kinney, W. R., Palmrose, Z. V., & Scholz, S. (2004). Auditor independence, non-audit
Journal of Economic Behavior & Organization, 83(1), 50–58. https://doi.org/10.1016/
services, and restatements: Was the U.S. Government right? Journal of Accounting
j.jebo.2011.06.007.
Research, 42(3), 561–588. https://doi.org/10.1111/j.1475-679X.2004.t01-1-
Chen, C. W., Velasquez Tuliao, K., Cullen, J. B., & Chang, Y. Y. (2016). Does gender
00141.x.
influence managers' ethics? A cross-cultural analysis. Business Ethics: A European
Kirkham, L., & Loft, A. (1993). Gender and the construction of the professional accoun-
Review, 25(4), 345–362. https://doi.org/10.1111/beer.12122.
tant. Accounting, Organizations and Society, 18(6), 507–558. https://doi.org/10.1016/
Chung, J., & Monroe, G. S. (2001). A research note on the effects of gender and task
0361-3682(93)90002-N.
complexity on an audit judgment. Behavioral Research in Accounting, 13(1), 111–125.
Kothari, S. P., Leone, A. J., & Wasley, C. E. (2005). Performance matched discretionary
https://doi.org/10.2308/bria.2001.13.1.111.
accrual measures. Journal of Accounting and Economics, 39(1), 163–197. https://doi.
Ciancanelli, P., Gallhofer, S., Humphrey, C., & Kirkham, L. (1990). Gender and ac-
org/10.1016/j.jacceco.2004.11.002.
countancy: Some evidence from the UK. Critical Perspectives on Accounting, 1(2),
Ladva, P., & Andrew, J. (2014). Weaving a web of control: “The promise of opportunity”
117–144. https://doi.org/10.1016/1045-2354(90)02011-7.
and work-life balance in multinational accounting firms. Accounting, Auditing &
Citron, D. B., & Taffler, R. J. (2001). Ethical behaviour in the UK audit profession: The
Accountability Journal, 27(4), 634–654. https://doi.org/10.1108/AAAJ-02-2012-
case of the self-fulfilling prophecy under going-concern uncertainties. Journal of
00955.
Business Ethics, 29(4), 353–363. https://doi.org/10.1023/A:101075220.
Levin, I., Snyder, M., & Chapman, D. (1988). The interaction of experimental and situa-
Costa, P., Terracciano, A., & McCrae, R. (2001). Gender differences in personality traits
tional factors and gender in a simulated risky decision-making task. Journal of
across cultures: Robust and surprising findings. Journal of Personality and Social
Psychology, 122, 173–181. https://doi.org/10.1080/00223980.1988.9712703.
248
Lundeberg, M. A., Fox, P. W., & Puncochar, J. (1994). Highly confident but wrong: Schmitt, D., Realo, A., Voracek, M., & Allik, J. (2008). Why can't a man be more like a
Gender differences and similarities in confidence judgments. Journal of Educational woman? Sex differences in big five personality traits across 55 cultures. Journal of
Psychology, 86(1), 114–121. https://doi.org/10.1037/0022-0663.86.1.114. Personality and Social Psychology, 94, 168–182.
Lupu, I. (2012). Approved routes and alternative paths: The construction of women's Sun, J., Liu, G., & Lan, G. (2011). Does female directorship on independent audit com-
careers in large accounting firms. Evidence from the French Big Four. Critical mittees constrain earnings management? Journal of Business Ethics, 99(3), 369–382.
Perspectives on Accounting, 23(4–5), 351–369. https://doi.org/10.1016/j.cpa.2012. https://doi.org/10.1007/s10551-010-0657-0.
01.003. Sunden, A. E., & Surette, B. J. (1998). Gender differences in the allocation of assets in
Maletta, M. J., & Kida, T. (1993). The effect of risk factors on auditors' configural in- retirement savings plans. American Economic Review, 88(2), 207–211. https://www.
formation processing. The Accounting Review, 68(3)https://www.jstor.org/stable/ jstor.org/stable/116920.
248209 (681-191). Thiruvadi, S., & Huang, H. W. (2011). Audit committee gender differences and earnings
Menezes, T., & Bras, F. A. (2015). Audit quality: Does gender composition of audit firms management. Gender in Management: An International Journal, 26(7), 483–498.
matter? Spanish Journal of Finance and Accounting, 44(3), 264–297. https://doi.org/ https://doi.org/10.1108/17542411111175469.
10.1080/02102412.2015.1035578. Unger, R. K. (1990). Imperfect reflections of reality: psychology constructs gender. In R.
Messier, W. F., Jr., Owhoso, V., & Rakovski, C. (2008). Can audit partners predict sub- T. HareMustin, & J. Maracek (Eds.). Making a difference: Psychology and the con-
ordinates' ability to detect errors? Journal of Accounting Research, 46(5), 1241–1264. struction of gender. New Haven, CO: Yale University Press.
https://doi.org/10.1111/j.1475-679X.2008.00307.x. Watts, R. L. (2003). Conservatism in accounting part I: Explanations and implications.
Nettle, D. (2007). Empathizing and systemizing: What are they, and what do they con- Accounting Horizons, 17(3), 207–221. https://doi.org/10.2308/acch.2003.17.3.207.
tribute to our understanding of psychological sex differences? British Journal of Welsh, M. J. (1992). The construction of gender: Some insights from feminist psychology.
Psychology, 98, 237–255. https://doi.org/10.1348/000712606X117612. Accounting, Auditing & Accountability Journal, 5(3), 120–132. https://doi.org/10.
Niskanen, J., Karjalainen, J., Niskanen, M., & Karjalainen, J. (2011). Auditor gender and 1108/09513579210017433.
corporate earnings management behavior in private Finnish firms. Managerial
Auditing Journal, 26, 778–793. https://doi.org/10.1108/02686901111171448. Josep Garcia-Blandon is currently Full Professor of Accounting and Finance at IQS
O'Donnell, E., & Johnson, E. (2001). The effects of auditor gender and task complexity on School of Management, Universitat Ramon Llull in Barcelona. He completed his PhD in
information processing efficiency. International Journal of Auditing, 5(2), 91–105. Economics and Business Administration at the Universitat de Barcelona in 1997. He leads
https://doi.org/10.1111/j.1099-1123.2001.00328.x. the Research Group on Corporate Governance at Universitat Ramon Llull. He has pub-
Owhoso, V., & Weickgenannt, A. (2009). Auditors' self-perceived abilities in conducting lished numerous articles and book chapters in the fields of corporate governance, ac-
domain audits. Critical Perspectives on Accounting, 20(1), 3–21. counting and audit.
Paterson, J. S., & Valencia, A. (2011). The effects of recurring and nonrecurring tax, audit-
related, and other nonaudit services on auditor independence. Contemporary
Accounting Research, 28(5), 1510–1536. https://doi.org/10.1016/j.cpa.2007.04.005. Josep Maria Argiles Bosch is Associate Professor of Accounting at the University of
Peni, E., & Vähämaa, S. (2010). Female executives and earnings management. Managerial Barcelona. He teaches in the Degree in Business Administration, the Master of Research in
Business and in Masters of practitioner profile. He completed his PhD in Economics and
Finance, 36(7), 629–645. https://doi.org/10.1108/03074351011050343.
Pucheta-Martinez, M. C., Bel-Oms, I., & Olcina-Sempere, G. (2016). Corporate govern- Business at the Universitat Autònoma de Barcelona and his MBA at ESADE (Barcelona).
ance, female directors and quality of financial information. Business Ethics: A He has published numerous research articles in academic JCR journals, books and book
European Review, 25(4), 363–385. https://doi.org/10.1111/beer.12123. chapters, and has participated in many congresses on various topics on accounting and
Roxas, M., & Stoneback, J. (2004). The importance of gender across cultures in ethical business.
decision making. Journal of Business Ethics, 50(2), 149–165. https://doi.org/10.1023/
B:BUSI.0000. Diego Ravenda is currently Professeur Resident at Toulouse Business School. He com-
Ruegger, D., & King, E. W. (1992). A study of the effect of age and gender upon student pleted his PhD in Business at the Universitat de Barcelona in 2014. He has published
business ethics. Journal of Business Ethics, 11(3), 179–186. https://doi.org/10.1007/ several articles in high impact journals such as Accounting, Auditing and Accountability
BF00871965. and the Journal of Business Ethics.
249

Journal of Business Research: Josep Garcia-Blandon, Josep Maria Argilés-Bosch, Diego Ravenda T

Uploaded by

Copyright:

Available Formats

Journal of Business Research: Josep Garcia-Blandon, Josep Maria Argilés-Bosch, Diego Ravenda T

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Journal of Business Research: Josep Garcia-Blandon, Josep Maria Argilés-Bosch, Diego Ravenda T

Uploaded by

Copyright:

Available Formats

Journal of Business Research 96 (2019) 238–249

Contents lists available at ScienceDirect

Journal of Business Research

Is there a gender effect on the quality of audit services? T

ARTICLE INFO ABSTRACT

Current accruals (change in non cash working capital)=

*, **, ***Significant at 10%, 5% and 1% respectively.

FEMALE – −0.032 0.028 −0.042 −0.030

*, **, ***Significant at 10%, 5% and 1% respectively.

% of clients with big losses 15 11 0.28

Mean p-Value Median p-Value

Female auditor Male auditor Female auditor Male auditor

Panel B: Differences in audit effort

Panel C: Difference in client's characteristics

observe significant differences in either BIGLOSS or BIGNEGSURP by Table 7

FEMALE – −0.034 0.018 −0.037 −0.022

*, **, ***Significant at 10%, 5% and 1% respectively.

MALE-FEMALE – −0.065 0.049 −0.106 −0.056

*, **, ***Significant at 10%, 5% and 1% respectively.

You might also like

, , Significant at 10%, 5% and 1% respectively.

, , Significant at 10%, 5% and 1% respectively.

, , Significant at 10%, 5% and 1% respectively.

, , Significant at 10%, 5% and 1% respectively.