Accounting critique of quantitative and qualitative research

Accounting critique of quantitative and qualitative research
The paper involves preparing a critique of a research paper after reading "Female Auditors and accruals quality" article.
After reading "Female Auditors and accruals quality" artticle, you need to analyze whether this article is well written or not according to 'critique of quantitative and qualitative research'.
Refer Exhibit 1,2,3 from 'critique of quantitative and qualitative research' hanout and I would like you to take an Exhibit 3 of "research critique framework" as your outline and include all of the criteria on it and reason out why you think so.
Female Auditors and Accruals Quality
Kim Ittonen, Emilia Va¨ha¨maa, and Sami Va¨ha¨maa
SYNOPSIS: This paper examines the association between accruals quality and the
gender of the firm’s audit engagement partner. In particular, given the documented
gender-based differences in diligence, conservatism, and risk tolerance, we postulate
that female auditors may improve accruals quality. Using a sample of Finnish and
Swedish NASDAQ OMX-listed firms, we run several alternative panel regressions of
abnormal accruals on female auditor variables and firm-specific controls. The results
suggest that firms with female audit engagement partners are associated with smaller
abnormal accruals, thereby implying that female auditors may have a constraining effect
on earnings management. In general, our findings indicate that the behavioral
differences between women and men may have important implications for the quality
of auditing and financial reporting.
Keywords: auditor gender; female auditors; gender differences; discretionary accruals;
abnormal accruals; accruals quality; earnings management.
Data Availability: The data used in this paper are derived from public sources.
INTRODUCTION
This paper examines the association between accruals quality and the gender of the firm’s
audit engagement partner. Specifically, we aim to assess whether and how female auditors
affect the quality of financial reporting. The motivation for this analysis is based upon the
behavioral differences between women and men that have been extensively documented in the
psychology and behavioral economics literature (see, e.g., Feingold 1994; Byrnes et al. 1999; Costa
et al. 2001; Dwyer et al. 2002; Eckel and Grossman 2002; Nettle 2007; Schmitt et al. 2008; Croson
Kim Ittonen and Emilia Va¨ha¨maa are Assistant Professors, and Sami Va¨ha¨maa is a Professor, all at the
University of Vaasa.
We thank two anonymous referees, Dana Hermanson (co-editor), Phyllis Keys, Emma-Riikka Myllyma¨ki, S. Gulfem
Bayram, and participants at the 2011 American Accounting Association Meeting, the 2011 Southern Finance Association
Meeting, and the 2010 Eastern Finance Association Meeting for helpful discussions and comments. We gratefully
acknowledge the support of the Academy of Finland (project no. 259676), Emil Aaltonen Foundation, the Finnish
Foundation for Advancement of Securities Markets, the Finnish Savings Banks Research Foundation, the Foundation for
Economic Education, the NASDAQ OMX Nordic Foundation, the Ostrobothnia Chamber of Commerce, and the Paulo
Foundation.
Part of this paper was written while E. and S. Va¨ha¨maa were visiting the University of Central Florida.
Submitted: October 2010
Accepted: July 2012
Published Online: January 2013
Corresponding author: Sami Va¨ha¨maa
Email: sami@uwasa.fi
205
and Gneezy 2009). In particular, we presume that gender-based differences in cognitive information
processing, diligence, conservatism, overconfidence, and risk tolerance may have important
implications for the audit process and auditor judgments and, thus, ultimately, for the quality of
audited financial information.
It is widely acknowledged that external auditors play a central role in ensuring the integrity of
the financial reporting process (see, e.g., Cohen et al. 2004; Watkins et al. 2004; Lin and Hwang
2010). Over the past few years, the relationship between auditors and the extent of earnings
management has been empirically examined in, for instance, Balsam et al. (2003), Kim et al.
(2003), Myers et al. (2003), Nagy (2005), Jenkins et al. (2006), Piot and Janin (2007), Srinidhi and
Gul (2007), Caramanis and Lennox (2008), Jenkins and Velury (2008), and Gul et al. (2009). These
studies provide considerable evidence to suggest that auditor attributes such as reputation,
independence, industry expertise, and tenure affect the quality of the client firm’s financial
reporting. In general, the prior literature shows that external auditors are effective monitors of
financial reporting quality.
Although the association between external auditors and earnings management has been
examined extensively in the literature, surprisingly little attention has been devoted to the potential
effects of auditor gender. Our idea to focus on auditor gender effects is not, however, completely
novel. Chung and Monroe (2001), O’Donnell and Johnson (2001), and Gold et al. (2009) conduct
experiments on the influence of auditor gender on audit judgments. These experimental studies
suggest that the gender of the auditor may affect the audit process. Chung and Monroe (2001)
document that female audit partners are more accurate and effective information processors in
complex audit tasks, while O’Donnell and Johnson (2001) show that female auditors may exhibit
greater efficiency in audit judgments. Gold et al. (2009) find that female auditors are less influenced
by unverified client explanations. However, in contrast to Chung and Monroe (2001), their
experiment also suggests that the audit adjustments proposed by female audit partners may be less
accurate. In this paper, we attempt to extend the above experimental literature by empirically
examining the relationship between female auditors and accruals quality.1
In brief, our empirical findings indicate that the gender of the audit engagement partner is
associated with the quality of financial reporting. Using a sample of Finnish and Swedish NASDAQ
OMX-listed firms, we document that the client firms of female audit engagement partners may have
higher accruals quality than firms audited by male audit partners. In contrast to the United States
and the United Kingdom, it is mandatory in Finland and Sweden to disclose the identity of the audit
partner(s) in the audit reports, and thereby the NASDAQ OMX firms provide an expedient setting
to empirically examine the effects of auditor gender on accruals quality. We use modified versions
of abnormal accruals measures originally developed by Jones (1991) and Dechow and Dichev
(2002) to estimate proxies for accruals quality, and run several alternative panel regressions of
abnormal accruals on female audit partner variables and firm-specific controls. These regressions
demonstrate that the client firms of female audit engagement partners are associated with
significantly smaller absolute abnormal accruals. Our results also indicate that female auditors may
constrain the use of both income-increasing and income-decreasing accruals. Thus, consistent with
the prior experimental studies, our empirical findings suggest that the inherent gender differences
1 While revising this paper, we became aware of a contemporaneous independent study by Niskanen et al. (2011)
that also examines the effects of auditor gender on earnings management. Niskanen et al. (2011) use data on
small- and medium-sized private Finnish firms, and document a positive association between female auditors and
earnings quality. In contrast to Niskanen et al. (2011), we use data on large, publicly listed Finnish and Swedish
firms, which are obliged to use professional, certified auditors, are required to follow the IFRS reporting
standards and, moreover, have to comply with a stricter legal environment, as well as with the listing
requirements of the NASDAQ OMX Nordic Exchange. Nevertheless, the results of Niskanen et al. (2011) can be
viewed as complementary to the empirical findings reported in this paper.
206 Ittonen, Va¨ha¨maa, and Va¨ha¨maa
Accounting Horizons
June 2013
may affect the audit process. Moreover, the results of this study are broadly consistent with the vast
body of literature on gender differences in diligence, conservatism, and risk tolerance. Overall, our
findings suggest that the behavioral differences between women and men may have important
implications for the quality of auditing and financial reporting.
The remainder of the paper proceeds as follows. The next section reviews the related literature
and presents our research hypothesis. The third section describes the data on the Finnish and
Swedish NASDAQ OMX firms, while the fourth section presents the methodology used in the
analysis. Our empirical findings on the relationship between accruals quality and female audit
partners are reported in the fifth section. Finally, the last section provides concluding remarks.
BACKGROUND AND HYPOTHESIS
External Auditors and Financial Reporting Quality
The quality of financial reporting can be compromised due to the management’s intentional
manipulation of accounting information by using the discretion provided in accounting principles
and financial reporting standards. The contracting-based theory of accounting posits that the
executives of the firm either efficiently report accurate financial information to stakeholders, or that
they opportunistically manipulate reported earnings for private gains or in order to mislead some
stakeholders about the firm’s financial performance (see, e.g., Holthausen 1990; Christie and
Zimmerman 1994). Although several measures of financial reporting quality have been proposed in
the literature, the most commonly used construct is a measure of abnormal accruals that intends to
assess discretionary components in reported earnings (see, e.g., Becker et al. 1998; Francis et al.
2005; Geiger and North 2006; Ali et al. 2007). These discretionary accruals are temporary
adjustments that resolve timing problems between accounting earnings and the underlying cash
flows at the cost of making assumptions and estimates.
Given that opportunistic, self-interested executives may have incentives to report overly
optimistic earnings, it is the role of the external auditors, board of directors, and other corporate
governance actors to attempt to constrain excessive earnings management (see, e.g., Christie and
Zimmerman 1994; Cohen et al. 2004; Lin and Hwang 2010). Recent legal reforms, such as the
Sarbanes-Oxley Act of 2002, have further emphasized the central role of the external auditors in
ensuring the integrity of the financial reporting process. It is, therefore, not surprising that a vast
body of empirical auditing literature has recently focused on the association between auditors and
earnings management (see, e.g., Balsam et al. 2003; Kim et al. 2003; Myers et al. 2003; Nagy 2005;
Jenkins et al. 2006; Piot and Janin 2007; Caramanis and Lennox 2008; Jenkins and Velury 2008;
Gul et al. 2009). In general, this literature indicates that the extent of earnings management is
influenced by auditor attributes such as reputation, independence, industry expertise, and tenure.
Becker et al. (1998) investigate the relationship between auditor quality and earnings
management, and report that the client firms of the Big 6 auditors have lower discretionary accruals
than the clients of other auditors. In a similar vein, Francis et al. (1999), Nelson et al. (2002), and
Kim et al. (2003) document that auditors with better reputations are more effective in curtailing
excessive earnings management behavior. Srinidhi and Gul (2007) focus on the effects of auditor
independence by examining the association between nonaudit fees and earnings management. Their
findings indicate that auditor independence improves the quality of reported earnings.
Johnson et al. (2002), Frankel et al. (2002), Myers et al. (2003), Ghosh and Moon (2005),
Nagy (2005), Jenkins and Velury (2008), and Gul et al. (2009) examine the relationship between
auditor tenure and earnings management, and document that auditor tenure is positively associated
with the quality of financial reporting. Balsam et al. (2003), Krishnan (2003), and Jenkins et al.
(2006) show that an auditor’s industry expertise affects discretionary accruals. In particular, Balsam
et al. (2003) find that the client firms of industry-specialist auditors have higher earnings quality
Female Auditors and Accruals Quality 207
Accounting Horizons
June 2013
than clients of non-specialists, while Jenkins et al. (2006) report that the decrease in earnings
quality during the late 1990s was significantly smaller for the clients of industry-specialist auditors.
Finally, Caramanis and Lennox (2008) take advantage of a unique Greek dataset on audit hours to
examine the effect of auditor effort on earnings quality. They find that low auditor effort increases
managerial opportunities for aggressive earnings management.
Gender Effects
It has long been acknowledged in the cognitive psychology and behavioral economics
literature that significant gender-based differences exist, e.g., in information processing, diligence,
conservatism, overconfidence, cautiousness, and risk tolerance (see, e.g., Levin et al. 1988;
Feingold 1994; Johnson and Powell 1994; Byrnes et al. 1999; Costa et al. 2001; Eckel and
Grossman 2002; Nettle 2007; Schmitt et al. 2008). Several studies have documented that women
are more conservative and risk-averse than men, and may exhibit less risky behavior in financial
decisions (see, e.g., Powell and Ansic 1997; Jianakoplos and Bernasek 1998; Barber and Odean
2001; Dwyer et al. 2002; Watson and McNaughton 2007). The prior literature also indicates that
women may have a higher propensity to comply with rules and regulations in taxation and financial
decision contexts (see Baldry 1987; Barnett et al. 1994; Bernardi and Arnold 1997; Fallan 1999;
Beu et al. 2003; Pierce and Sweeney 2010). It is conceivable that the gender traits related to
conservatism, risk tolerance, and compliance with rules may have implications for the integrity of
the financial reporting process.2
Furthermore, based on previous literature, it is plausible to expect that female audit
engagement partners are generally highly competent and hard-working. Given the well-known glass
ceiling phenomenon, women may have to demonstrate extra competence in order to reach the top
levels of the hierarchy, such as positions as partners in audit firms. Accordingly, recent empirical
studies by Green et al. (2009) and Kumar (2010) indicate that female financial analysts may need to
possess better-than-average skills in a financial expertise environment due to gender discrimination.
3 Moreover, Fondas and Sassalos (2000) argue that women tend to have higher expectations
regarding their responsibilities, which may induce them to expend more effort on their tasks. These
gender differences in skills and effort may be expected to affect the scope and performance of the
audit.
Although the documented behavioral differences between women and men may inherently
influence the audit process and auditor judgments, potential auditor gender effects have so far
received surprisingly little attention in the auditing literature. The few exceptions are the
experimental studies by Chung and Monroe (2001), O’Donnell and Johnson (2001), and Gold et al.
(2009), and a recent empirical study by Ittonen and Peni (2012). Chung and Monroe (2001)
examine the relationship between auditor gender on audit judgments, and document that female
audit partners are more accurate and effective information processors in complex audit tasks. In a
similar vein, O’Donnell and Johnson (2001) find that female auditors may exhibit greater efficiency
in their audit judgments. Gold et al. (2009) report that female auditors are less influenced by
unverified client explanations. However, in contrast to Chung and Monroe (2001), their experiment
also suggests that the audit adjustments proposed by female audit partners may be less accurate.
Finally, Ittonen and Peni (2012) examine the relationship between female auditors and audit fees in
an empirical setting using data from three Nordic countries. Their findings indicate that firms with
2 Anecdotal evidence from the recent high-profile accounting scandals indicates that females have often acted as
the whistleblowers (e.g., Sherron Watkins at Enron and Cynthia Cooper at WorldCom).
3 While Kumar (2010) reports that female analysts issue more accurate earnings forecasts, the findings of Green et
al. (2009) indicate that the forecasts of female analysts are less accurate. Nevertheless, Green et al. (2009) also
document that female analysts outperform male analysts in other aspects of job performance.
208 Ittonen, Va¨ha¨maa, and Va¨ha¨maa
Accounting Horizons
June 2013
female audit engagement partners have higher audit fees, suggesting that auditor gender may affect
audit effort. Overall, the previous studies suggest that the gender of the auditor may affect the audit
process. In this paper, we aim to contribute to the literature by empirically examining the
relationship between female auditors and accruals quality.
Closely related to the current study, Barua et al. (2010) and Peni and Va¨ha¨maa (2010) postulate
that gender-based behavioral differences may affect the quality of financial reporting. More
specifically, both Barua et al. (2010) and Peni and Va¨ha¨maa (2010) use data on large U.S. firms to
empirically examine the association between female executives and accruals quality, and document
that firms with female chief financial officers (CFOs) are associated with more conservative
financial reporting practices. Thus, these studies provide empirical evidence to suggest that female
executives may make more cautious and conservative decisions with respect to earnings
management. We attempt to extend Barua et al. (2010) and Peni and Va¨ha¨maa (2010) by
assessing the potential effects of gender differences on financial reporting quality in the context of
audit engagement partners.
Hypothesis
In this paper, we presume that the gender-based differences in cognitive information
processing, diligence, conservatism, and risk tolerance may have important implications for the
audit process and auditor judgments and, thus, ultimately, for the quality of financial reporting. The
documented gender differences in cognitive information processing and diligence may affect audit
effort and the intensity of monitoring. Therefore, it is plausible to expect that female audit
engagement partners may be more effective monitors of excessive earnings management behavior.
Furthermore, the gender-based differences in conservatism and risk tolerance imply that female
auditors may inherently promote more conservative financial reporting practices and make less
risky audit judgments. Thus, based on the behavioral differences between women and men, we
posit the following general research hypothesis:
H1: The client firms of female audit engagement partners have higher accruals quality than
firms audited by male audit partners.
We empirically assess the accruals quality by estimating the magnitude of abnormal accruals in
reported earnings.4 We focus on the association between female audit engagement partners and the
magnitude of abnormal accruals, and we also examine the effects of female auditors on incomeincreasing
and income-decreasing abnormal accruals. Given the gender-based differences in
information processing, diligence, conservatism, and risk tolerance, we posit that female auditors
constrain the degree of earnings management in general and, therefore, we expect that absolute
abnormal accruals are lower for the client firms of female auditors.
Previous studies indicate that auditors are more likely to disagree with their clients about
income-increasing than income-decreasing financial reporting practices (see, e.g., DeFond and
Jiambalvo 1993; Kinney and Martin 1994; Nelson et al. 2002). As discussed by Caramanis and
Lennox (2008), auditors face higher litigation and reputational risks stemming from overstated,
rather than understated, earnings. Hence, if the gender-based differences in conservatism and riskaversion
influence the audit process and the riskiness of audit judgments, we should observe the
firms with female audit engagement partners to be associated with lower income-increasing
abnormal accruals than the client firms of male auditors. Furthermore, the prior literature suggests
4 In our empirical analysis, we focus on abnormal accruals and ignore alternative measures of earnings quality,
such as earnings restatements, the timeliness of loss recognition, earnings response coefficients, and the
likelihood of meeting or beating analysts’ earnings forecasts.
Female Auditors and Accruals Quality 209
Accounting Horizons
June 2013
that increased audit effort may decrease income-increasing earnings management (Caramanis and
Lennox 2008). We posit that gender differences in information processing and diligence may affect
audit effort and, thus, we expect a negative relationship between female auditors and incomeincreasing
abnormal accruals.5
The association between female audit engagement partners and income-decreasing abnormal
accruals may be positive or negative. If female auditors have a lower tolerance toward earnings
management in general, we should observe that the magnitude of income-decreasing abnormal
accruals is lower for the client firms of female auditors. Moreover, female auditors may be more
prone to constrain income-decreasing financial reporting practices driven by tax reduction
incentives or minimization of litigation and regulatory costs. On the other hand, income-decreasing
abnormal accruals are often considered to reflect accounting conservatism, and may indicate
asymmetric treatment of unrealized gains and losses. If female audit engagement partners prefer
more conservative accounting choices, the client firms of female auditors may have larger incomedecreasing
abnormal accruals. Given the somewhat ambiguous theoretical relationship between
income-decreasing earnings management and female audit engagement partners, we make no
prediction about the effect of female auditors on income-decreasing abnormal accruals.
DATA ENVIRONMENT
The data used in the empirical analysis consist of the Finnish and Swedish firms listed in the
NASDAQ OMX Nordic exchange as of the end of 2007. The sample covers fiscal years
2005–2007. After excluding financial institutions (SIC codes 6000–6900) due to their unique
features, and after removing observations with insufficient financial data in Thomson Reuters
Worldscope, we obtain a sample consisting of 770 firm-year observations. For these 770 firm-year
observations, we manually collect the names of the audit engagement partners from the firms’ audit
reports. The firms listed in the NASDAQ OMX Nordic exchange are required to provide an audit
report signed by at least one audit engagement partner, even if an audit firm is appointed. Thus, by
reviewing the engagement auditors’ names from the audit report, the gender of the audit
engagement partner(s) can be determined.6 Table 1 reports the distribution of the data classified by
industry and auditor gender.
Since the audit engagement partner is identifiable from the audit reports, the institutional
setting in Finland and Sweden provides an expedient environment for studying whether and how
the gender of the audit engagement partner affects the quality of financial reporting. This is in
contrast with the United States and the United Kingdom, where only the name of the responsible
audit firm is public information, and the identity of the actual audit engagement partner is
unobtainable.7
The legal environments, as well as the listing requirements, concerning corporate governance,
financial reporting, and auditing are similar for Finnish and Swedish firms listed on the NASDAQ
OMX Nordic exchange and, thus, these firms can be analyzed as one group. Sinani et al. (2008)
find, however, some differences in the formal board and ownership structures between the Nordic
countries. The main difference in the ownership structure between Finland and Sweden is that the
Swedish firms have, on average, a higher proportion of family ownership, whereas state ownership
5 In a recent empirical study, Ittonen and Peni (2012) document that firms with female audit engagement partners
have higher audit fees. This finding indicates that the gender-based behavioral differences may affect audit effort.
6 Some clients have opted to engage two audit partners from the same firm. In these engagements, both auditors
have signed the audit report and we consider both auditors to be jointly responsible for the audit.
7 Interestingly, the issue of disclosing the actual audit engagement partner is on the agenda of the Public Company
Accounting Oversight Board (PCAOB 2009). Consequently, the audit engagement partner signatures may
become obligatory information also in the United States.
210 Ittonen, Va¨ha¨maa, and Va¨ha¨maa
Accounting Horizons
June 2013
is more common in Finland (La Porta et al. 1998). According to Sinani et al. (2008), corporate
governance practices bear strong similarities in Finland and Sweden. There are also similarities in
the key corporate governance characteristics, such as trust, quality of enforcement, absence of
corruption, quality of government, and freedom of speech, which are all results of the two
countries’ small size and ethnical homogeneity.
From the audit engagement partner perspective, it should be noted that the requirements for
becoming a certified auditor are very similar in the two countries, as they both comply with the
Eighth Directive of the European Union. In both countries, auditors are required to have theoretical
education, as well as professional experience, and they need to pass a practical examination before
being certified. Accordingly, the audit engagement partners in Finland and Sweden have very
similar educational backgrounds and professional experience.
METHODOLOGY
We examine the association between accruals quality and the gender of the firm’s audit
engagement partner with cross-sectional panel regressions. As the first step of the analysis, we
assess the extent of earnings management by estimating the abnormal accruals in reported earnings.
Since the estimation of abnormal accruals is model-dependent, we utilize two alternative models of
expected accruals to quantify the degree of earnings management. The first model used in this study
is the modified Jones (1991) model, and the second model is the modified version of the Dechow
and Dichev (2002) model.8 Both of these models have been extensively used in the prior literature
TABLE 1
Number of Female Auditors by Industries
SIC Code Industry Description
2005 2006 2007
# of
Firms
# of
Female
Auditors
# of
Firms
# of
Female
Auditors
# of
Firms
# of
Female
Auditors
0100–1400 Agriculture, Forestry, Fishing, and
Mining
4 0 7 0 5 0
1500–1900 Construction 0 0 6 0 1 0
2000–3900 Manufacturing 130 15 153 17 162 20
4000–4900 Transportation, Communications,
Electric, Gas, and Sanitary Services
17 3 20 3 23 4
5000–5400 Wholesale Trade 9 1 14 1 12 1
5500–5900 Retail Trade 5 0 6 0 7 0
7000–8900 Services 49 5 65 10 75 11
Total 214 24 271 31 285 36
The table presents the number of firms by standard industry classification (SIC) codes, and shows the industry-specific
segmentation of the female audit engagement partners. The sample consists of 770 firm-year observations of Finnish and
Swedish firms listed in the NASDAQ OMX Nordic exchange.
8 As an additional test, we have also followed the approach of Ashbaugh et al. (2003), and used the modified Jones
(1991) model to estimate abnormal current accruals. The results based on abnormal current accruals are
discussed later in the paper.
Female Auditors and Accruals Quality 211
Accounting Horizons
June 2013
(see, e.g., Dechow et al. 1995; McNichols 2002; Ashbaugh et al. 2003; Kothari et al. 2005; Jones et
al. 2008; Barua et al. 2010; Francis et al. 2012; Sharma et al. 2011).
We follow the approach of Francis et al. (2012), and use the following modified Jones (1991)
model to estimate the expected accruals for firm i in year t:
TAi;t ¼ k0 þ k11=Assetsi;t1 þ k2ðDREVi;t DARi;tÞ þ k3PPEi;t þ k4ROAi;t
þ Year Fixed Effects þ Industry Fixed Effects þ Country Fixed Effects þ ei;t; ð1Þ
where:
TAi,t ¼ total accruals for firm i at time t;
DREVi,t ¼ the change in revenues for firm i between year t1 and year t;
DARi,t ¼ the change in accounts receivable for firm i;
PPEi,t ¼ the gross property, plant, and equipment for firm i; and
ROAi,t ¼ the return on assets for firm i.
Following Kothari et al. (2005), we scale all variables by lagged total assets, Assetsi,t1. We include
year, industry, and country fixed effects in the accruals model in order to control for potential
heterogeneity in accruals quality across years, industries, and the two countries. The Jones (1991)
model abnormal accruals (AAJONi,j) for firm i in year t are defined as the residual term of Equation
(1), ei,t.
Our second measure of abnormal accruals is derived from the Dechow and Dichev (2002)
model modified by McNichols (2002), in which the expected accruals for firm i in year t are given
by:
TCAi;j ¼ u0 þ u1OCFi;t1 þ u2OCFi;t þ u3OCFi;tþ1 þ u4DREVi;t þ u5PPEi;t
þ Year Fixed Effects þ Industry Fixed Effects þ Country Fixed Effects þ ei;t; ð2Þ
where:
TCAi,t ¼ total current accruals for firm i at time t;
OCFi,t ¼ the operating cash flow of firm i;
DREVi,t ¼ the change in revenues for firm i between year t1 and year t; and
PPEi,j ¼ the gross property, plant, and equipment.
Following Dechow and Dichev (2002), all variables in Equation (2) are scaled by the three-year
average of total assets. Similar to the modified Jones (1991) model, we estimate the modified
Dechow and Dichev (2002) model with year, industry, and country fixed effects. The Dechow and
Dichev (2002) abnormal accruals (AADDi,j) for firm i in year t are defined as the residuals of
Equation (2).
After estimating the abnormal accruals, we examine the association between abnormal accruals
and the gender of the firm’s audit engagement partner with the following fixed-effects panel
regression:
AAi;t ¼ a0 þ b1FEMALEi;t þ b2SIZEi;t þ b3LEVi;t þ b4ROAi;t þ b5LOSSi;t þ b6OCFi;t
þ b7INVRECi;t þ b8FOROPRi;t þ b9SGRTHi;t þ b10MBi;t þ b11ACi;t þ b12BIG4i;t
þ b13AAGEi;t þ b14AINDEXPi;t þ b15STAOWNi;t þ b16FINi;t þX
n1
k¼1
akSICki
þ X
2007
y¼2006
xyYEARy
i þ ei;t;
ð3Þ
212 Ittonen, Va¨ha¨maa, and Va¨ha¨maa
Accounting Horizons
June 2013
where AAi,t denotes abnormal accruals for firm i in year t, based either on the Jones (1991)
(AAJONi,t) or the Dechow and Dichev (2002) (AADDi,t) model. In testing our research hypothesis,
we first use the absolute value of abnormal accruals (jAAJONi,tj or jAADDi,tj) to measure the extent
of earnings management, and in the subsequent analysis, we use positive and the absolute value of
negative abnormal accruals to assess the magnitude of income-increasing and income-decreasing
earnings management. The test variable in our regression specification is FEMALE, which is either
(1) FAUDITOR, a female auditor dummy equaling 1 if the firm has appointed a female auditor, or
(2) FRATIO, which is defined as the proportion of female auditors to all audit engagement partners
(designed to control for audits with two or more engagement partners). Thus, we estimate
alternative versions of Equation (3), where the included female variable is either FAUDITOR or
FRATIO.
Following prior literature, we include several firm-specific control variables in our model:
SIZE ¼ the natural logarithm of total assets;
LEV ¼ the financial leverage, measured as total liabilities divided by total assets;
ROA ¼ the return on assets;
LOSS ¼ a dummy variable that equals 1 for firms with negative net income;
OCF ¼ the cash flow from operations;
INVREC ¼ the total inventories plus total receivables divided by total assets;
FOROPR ¼ the proportion of foreign sales of total sales;
SGRTH ¼ sales growth between years t1 and t;
MB ¼ the ratio of market value of equity to book value;
AC ¼ a dummy for the existence of an audit committee;9
BIG4 ¼ a dummy for the Big 4 audit firms;
AAGE ¼ the log of the age of the audit engagement partner;
AINDEXP ¼ a dummy variable that equals 1 if the audit partner has over 10 percent market
share in a particular industry;
STAOWN ¼ a dummy variable equaling 1 if the Finnish or the Swedish government holds
shares of the firm; and
FIN ¼ a dummy variable that equals 1 for Finnish firms.
We winsorize the abnormal accruals and the firm-specific control variables at the 1 percent and 99
percent levels to moderate the effects of extreme observations and outliers.10 Finally, given that the
extent of earnings management may vary across industries and over time, we use dummy variables
to control for the potential industry and time effects. SIC in Equation (3) is a dummy variable
according to industry classification codes, and YEAR is a dummy for fiscal years.
Previous studies indicate that the above firm characteristics are useful predictors of earnings
management (see, e.g., Becker et al. 1998; Francis et al. 1999; Bartov et al. 2001; Frankel et al.
2002; Caramanis and Lennox 2008). LEV, ROA, OCF, and LOSS are proxies for the financial
condition of the firm. According to DeAngelo et al. (1994), troubled companies may have strong
incentives to use income-decreasing accruals. Moreover, accruals models may overestimate the
accruals for poorly performing firms (see, e.g., Dechow et al. 1995). Furthermore, we expect a
positive association between INVREC, FOROPR, SGRTH, MB, and abnormal accruals, because
firms with high proportions of inventory and receivables or foreign operations and high growth
9 According to the ‘‘comply or explain’’ rule of the NASDAQ OMX Nordic exchange, establishing an audit
committee is voluntary for the sample firms.
10 As a robustness check, we have also estimated Equation (3) without winsorizing and with different winsorizing
levels. The estimates of these additional regressions are consistent with the results reported in Tables 4 and 5,
and thereby suggest that our findings are not affected by a few extreme observations or outliers.
Female Auditors and Accruals Quality 213
Accounting Horizons
June 2013
rates may be less transparent, more difficult to audit, and may provide greater opportunities for
opportunistic earnings management. Previous studies indicate that SIZE, AC, BIG4, and STAOWN
are negatively associated with earnings management. Larger and state-owned firms may have
stronger governance structures, such as audit committees, and lower information asymmetries, and
are generally subject to greater monitoring by auditors and financial analysts (see, e.g., Meek et al.
2007). Moreover, Becker et al. (1998) argue that firm size may also be a surrogate for numerous
omitted variables. Finally, given that auditor attributes may affect the auditing process (see, e.g.,
Jenkins et al. 2006; Chi and Chin 2011) we control for audit partner experience and industry
expertise. We expect a negative association between AAGE, AINDEXP, and the level of abnormal
accruals.
EMPIRICAL FINDINGS
Descriptive Statistics and Univariate Tests
Table 2 reports the descriptive statistics for the different variables used in the analysis. Panel A
reports the statistics for the full sample of 770 firm-year observations. As can be noted from the
table, the sample firms are heterogeneous, for instance, in terms of size, financial performance, and
the extent of international operations. The log of total assets ranges from0.541 (€0.582 million) to
9.985 (€21.7 billion), with a mean of 4.792 (€120 million), while the proportion of debt financing
ranges from 0 percent to 69 percent. The proportion of foreign sales to total sales varies from 0 to
100 percent, with a mean of 33.5 percent. Furthermore, Panel A shows that a Big 4 audit firm is
engaged by over 90 percent of the sample firms, and about 40 percent of the firms have complied
with the NASDAQ OMX Nordic comply or explain rule regarding an audit committee
appointment. Finally, it can be noted that about 3.1 percent of the sample firms are state-owned,
and that Finnish firms comprise approximately 31 percent of the sample.
Panel B of Table 2 reports the descriptive statistics for the subsample of 91 firm-year
observations with female audit partners. Panel C of Table 2 presents the t-statistics for the null
hypothesis that there is no difference between firms with female and male auditors. Some
interesting features can be noted from these statistics. First, the mean values of absolute abnormal
accruals, based both on the Jones (1991) model and the Dechow and Dichev (2002) model, are
slightly lower in the subsample of firms with female auditors, suggesting that female auditors may
constrain earnings management. A simple t-test indicates, however, that the differences in the
abnormal accruals between female- and male-audited firms are statistically insignificant. Thus, in a
univariate setting, we are unable to find any association between earnings management and the
gender of the firm’s audit engagement partner. Panel B further shows that the client firms of female
auditors are relatively similar to other firms in terms of firm characteristics. Nevertheless, the t-tests
indicate that the client firms of female audit partners are significantly smaller and have lower
profitability and sales growth. Interestingly, Panel B also indicates that the propensity to appoint a
female auditor is significantly lower in state-owned firms.11
Pairwise correlation coefficients demonstrate that the absolute abnormal accruals based on the
Jones (1991) model and the Dechow and Dichev (2002) model are strongly positively correlated
with each other.12 Our experimental variables, FAUDITOR and FRATIO, appear to correlate
negatively with both measures of abnormal accruals, thereby indicating that the firms with female
auditors may be associated with less earnings management. These correlations, however, are
11 In fact, there are no female audit engagement partners in state-owned firms.
12 For brevity, we do not tabulate the pairwise correlation coefficients. The correlation matrix is available from the
authors upon request.
214 Ittonen, Va¨ha¨maa, and Va¨ha¨maa
Accounting Horizons
June 2013
TABLE 2
Descriptive Statistics
Panel A: Descriptive Statistics for All Firms (n ¼ 770)
Variable Mean Median Max Min Std. Dev.
AAJON 0.073 0.041 0.416 0.000 0.087
AADD 0.066 0.040 0.350 0.000 0.073
SIZE 4.792 4.530 9.985 0.541 2.127
LEV 0.194 0.171 0.690 0.000 0.164
ROA 0.037 0.077 0.351 1.002 0.178
LOSS 0.229 0.000 1.000 0.000 0.420
OCF 0.048 0.073 0.359 0.707 0.149
INVREC 0.357 0.358 0.792 0.015 0.188
FOROPR 0.335 0.296 1.000 0.000 0.340
SGRTH 0.231 0.119 2.823 0.584 0.475
MB 3.365 2.663 17.421 0.231 2.692
AC 0.401 0.000 1.000 0.000 0.490
BIG4 0.904 1.000 1.000 0.000 0.295
AAGE 3.927 3.932 4.143 3.584 0.124
AINDEXP 0.170 0.000 1.000 0.000 0.376
STAOWN 0.031 0.000 1.000 0.000 0.174
FIN 0.308 0.000 1.000 0.000 0.462
Panel B: Descriptive Statistics for Firms with Female Auditors (n ¼ 91)
Variable Mean Median Max Min Std. Dev.
AAJON 0.067 0.035 0.391 0.001 0.079
AADD 0.059 0.044 0.341 0.001 0.059
SIZE 4.496 4.177 9.628 0.541 2.371
LEV 0.194 0.157 0.616 0.000 0.164
ROA 0.027 0.077 0.351 0.525 0.177
LOSS 0.275 0.000 1.000 0.000 0.449
OCF 0.045 0.069 0.359 0.443 0.142
INVREC 0.310 0.325 0.650 0.015 0.162
FOROPR 0.335 0.277 1.000 0.000 0.348
SGRTH 0.171 0.090 2.823 0.584 0.456
MB 3.528 2.533 11.488 0.539 2.584
AC 0.385 0.000 1.000 0.000 0.489
BIG4 0.857 1.000 1.000 0.000 0.352
AAGE 3.899 3.932 4.143 3.584 0.133
AINDEXP 0.099 0.000 1.000 0.000 0.300
STAOWN 0.000 0.000 0.000 0.000 0.000
FIN 0.209 0.000 1.000 0.000 0.409
Panel C: t-Tests for Differences
t-Tests for Differences t-stat. p-value
H0: AAJONF ¼ AAJONM 0.191 0.849
H0: AADDF ¼ AADDM 0.285 0.776
H0: SIZEF ¼ SIZEM 2.045 0.044
(continued on next page)
Female Auditors and Accruals Quality 215
Accounting Horizons
June 2013
statistically insignificant. Finally, we observe that FAUDITOR and FRATIO are significantly
negatively correlated with SIZE, INVREC, BIG4, AAGE, AINDEXP, and FIN, and positively
correlated with LOSS.
Regression Results
Table 3 reports the estimation results of our panel regressions with absolute abnormal accruals
as the dependent variable. We estimate four alternative regressions with different abnormal accruals
(AAJON or AADD) and different female auditor variables (FAUDITOR or FRATIO). All model
specifications have the same set of control variables. In Models 1 and 2, we use the abnormal
accruals based on the modified Jones (1991) model as the dependent variable. As can be noted from
Table 3, the adjusted R2s of these models are around 20 percent, and the F-statistics are significant
TABLE 2 (continued)
t-Tests for Differences t-stat. p-value
H0: LEVF ¼ LEVM 0.005 0.996
H0: ROAF ¼ ROAM 2.695 0.008
H0: LOSSF ¼ LOSSM 0.734 0.465
H0: OCFF ¼ OCFM 0.105 0.917
H0: INVRECF ¼ INVRECM 1.176 0.243
H0: FOROPRF ¼ FOROPRM 0.220 0.826
H0: SGRTHF ¼ SGRTHM 9.038 0.000
H0: MBF ¼ MBM 0.940 0.349
H0: ACF ¼ ACM 0.239 0.812
H0: BIG4F ¼ BIG4M 0.805 0.422
H0: AAGEF ¼ AAGEM 0.783 0.436
H0: AINDEXP ¼ AINDEXP 1.308 0.194
H0: STAOWNF ¼ STAOWNM 2.144 0.034
H0: FINF ¼ FINM 1.573 0.119
Panel A reports the summary statistics for the sample of Finnish and Swedish firms listed in the NASDAQ OMX Nordic
exchange. Financial institutions (SIC codes 6000–6900) and firms with inadequate data are excluded. Panel B reports the
statistics for the firms with female audit engagement partners. Panel C presents the t-statistics for the null hypothesis that
there is no difference between firms with female and male audit engagement partners. Subscripts F and M in Panel C
refer to firms with female and male audit engagement partners, respectively. Swedish Kronas (SEK) are converted into
Euros using fiscal year-end exchange rates.
Variable Definitions:
AAJON ¼ absolute value of the abnormal accruals based on the modified Jones (1991) model;
AADD ¼ absolute value of the abnormal accruals based on the modified Dechow and Dichev (2002) model;
SIZE ¼ natural logarithm of total assets;
LEV ¼ percentage of total debt to total assets;
ROA ¼ return on assets;
LOSS ¼ dummy variable that equals 1 for firms with negative net income;
OCF ¼ cash flow from operations;
INVREC ¼ inventory and receivables to total assets;
FOROPR ¼ proportion of foreign sales of total sales;
SGRTH ¼ sales growth rate;
MB ¼ ratio of market value of equity to book value of equity;
AC ¼ 1 if the firm has an audit committee;
BIG4 ¼ 1 if the firm has a Big 4 auditor;
AAGE ¼ logarithm of auditor’s age;
AINDEXP ¼ dummy variable that equals 1 if the auditor has over 10 percent market share in a particular industry;
STAOWN ¼ dummy variable equaling 1 if the Finnish or the Swedish government holds shares in the firm; and
FIN ¼ dummy variable for Finnish firms.
216 Ittonen, Va¨ha¨maa, and Va¨ha¨maa
Accounting Horizons
June 2013
TABLE 3
Absolute Abnormal Accruals
Variable
jAAJONj jAADDj
Model 1 Model 2 Model 3 Model 4
Constant 0.049 0.060 0.044 0.035
(1.07) (1.29) (0.69) (0.55)
Female Auditor Variables
FAUDITOR 0.007 0.016***
(0.76) (3.25)
FRATIO 0.015* 0.014***
(1.81) (2.42)
Control Variables
SIZE 0.009*** 0.009*** 0.011*** 0.012***
(5.25) (5.44) (5.23)*** (5.25)
LEV 0.007 0.006 0.042 0.041***
(0.28) (0.26) (2.86) (2.85)
ROA 0.000 0.000 0.001* 0.001**
(0.75) (0.75) (1.95) (2.00)
LOSS 0.014 0.014 0.010 0.009
(0.11) (1.56) (0.91) (0.86)
OCF 0.109** 0.108** 0.015 0.014
(2.20) (2.20) (0.53) (0.48)
INVREC 0.007 0.007 0.000 0.001
(1.14) (0.95) (0.01) (0.10)
FOROPR 0.000** 0.000** 0.000 0.000
(2.33) (2.06) (0.23) (0.40)
SGRTH 0.000** 0.000** 0.000 0.000
(2.45) (2.45) (0.83) (0.90)
MB 0.003** 0.003** 0.005*** 0.005***
(2.14) (2.15) (3.11) (3.09)
AC 0.014** 0.013** 0.002 0.003
(2.56) (2.49) (0.32) (0.34)
BIG4 0.014* 0.014* 0.001 0.000
(1.79) (1.83) (0.09) (0.04)
AAGE 0.014 0.012 0.014 0.017
(1.18) (0.99) (0.85) (0.99)
AINDEXP 0.002 0.003 0.010*** 0.010**
(0.66) (0.62) (2.62) (2.47)
STAOWN 0.006 0.005 0.012 0.012
(0.69) (0.60) (1.08) (1.09)
FIN 0.001 0.001 0.005 0.005
(0.83) (1.23) (1.50) (1.54)
Industry Fixed Effects Yes Yes Yes Yes
Year Fixed Effects Yes Yes Yes Yes
Adjusted R2 0.197 0.199 0.194 0.193
F-stat. 8.906*** 8.947*** 8.680*** 8.596***
(continued on next page)
Female Auditors and Accruals Quality 217
Accounting Horizons
June 2013
at the 1 percent level. The estimated coefficients for the female auditor variables are negative in
both regression specifications, and marginally significant in Model 2 with FRATIO, the proportion
of females of all audit engagement partners, as the test variable. Hence, these estimates provide
support for our research hypothesis that the client firms of female auditors are associated with better
accruals quality. The coefficient estimates for our control variables SIZE, OCF, and AC are negative
and statistically highly significant, suggesting that accruals quality is generally higher in large,
profitable firms that have established an audit committee. Interestingly, our estimates also indicate
that the level of abnormal accruals is higher in firms that are audited by a Big 4 audit firm.13
In Models 3 and 4, the abnormal accruals are extracted from the modified Dechow and Dichev
(2002) model. The adjusted R2s of the regressions are around 19.5 percent, and the F-statistics are
significant at the 1 percent level. Regarding the variables of interest, it can be noted from Table 3
that the coefficients for FAUDITOR and FRATIO are negative and statistically significant at the 1
percent level in both regressions. These estimates indicate that the level of abnormal accruals is
lower for firms with female audit engagement partners and, therefore, consistent with our research
hypothesis, suggest that the quality of financial reporting is better in the client firms of female
TABLE 3 (continued)
*, **, *** Denote significance at the 0.10, 0.05, and 0.01 levels, respectively.
The table reports the estimates of four alternative versions of the following regression model:
jAAi;tj ¼ a0 þ b1FEMALEi;t þ b2SIZEi;t þ b3LEVi;t þ b4ROAi;t þ b5LOSSi;t þ b6OCFi;t þ b7INVRECi;t
þ b8FOROPRi;t þ b9SGRTHi;t þ b10MBi;t þ b11ACi;t þ b12BIG4i;t þ b13AAGEi;t þ b14AINDEXPi;t
þ b15STAOWNi;t þ b16FINi;t þX
n1
k¼1
akSICki
þ X
2007
y¼2006
xyYEARy
i þ e;
where jAAi,tj denotes the absolute value of abnormal accruals for firm i in year t, based either on the Jones (1991)
(AAJON) or the Dechow and Dichev (2002) (AADD) model. The test variable in our regression specification is
FEMALE, which is either (1) FAUDITOR, a female auditor dummy equaling 1 if the firm has appointed a female
audit partner, or (2) FRATIO, which is defined as the proportion of female audit engagement partners to all audit
engagement partners (designed to control for audits with two engagement partners). The t-statistics (reported in
parentheses) are based on standard errors corrected for clustering at the firm level.
Variable Definitions:
SIZE ¼ natural logarithm of total assets;
LEV ¼ percentage of total debt to total assets;
ROA ¼ return on assets;
LOSS ¼ dummy variable that equals 1 for firms with negative net income;
OCF ¼ cash flow from operations;
INVREC ¼ inventory and receivables to total assets;
FOROPR ¼ proportion of foreign sales of total sales;
SGRTH ¼ sales growth rate;
MB ¼ ratio market value of equity to book value of equity;
AC ¼ 1 if the firm has an audit committee;
BIG4 ¼ 1 if the firm has a Big 4 auditor;
AAGE ¼ logarithm of auditor’s age;
AINDEXP ¼ dummy variable that equals 1 if the auditor has over 10 percent market shares in a particular industry;
STAOWN ¼ dummy variable equaling 1 if the Finnish or the Swedish government holds shares in the firm;
FIN ¼ dummy variable for Finnish firms;
SIC ¼ dummy variable according to industry classification codes; and
YEAR ¼ dummy for fiscal years.
13 The high concentration of the Nordic audit market may partially explain this somewhat counterintuitive finding.
In our sample, over 90 percent of the firms are audited by a Big 4 auditor.
218 Ittonen, Va¨ha¨maa, and Va¨ha¨maa
Accounting Horizons
June 2013
auditors. The coefficient estimates for our control variables indicate that the magnitude of abnormal
accruals decreases with increasing firm size, leverage, and profitability, and increases with growth
prospects. While the coefficients for other auditor attributes appear insignificant in Models 3 and 4,
the coefficient estimates for the industry-specialist dummy variables are positive and significant,
suggesting that the involvement of industry-specialist audit engagement partners does not improve
the quality of accruals.
So far, we have measured the extent of earnings management by the absolute value of
abnormal accruals. To further examine the effects of auditor gender on accruals quality, we next
differentiate between income-decreasing and income-increasing earnings management by splitting
the sample based on the sign of the abnormal accruals. There are 51 observations with female
auditors in the income-increasing subsample and 40 in the income-decreasing subsample. The
estimation results of the alternative regression specifications with positive and absolute negative
abnormal accruals as the dependent variables are reported in Table 4. As can be noted from the
table, the adjusted R2s of the regressions vary between 10.8 and 45.9 percent, and the F-statistics
are statistically highly significant in each model. In general, the estimation results provide support
for our research hypothesis. The estimated coefficient for FAUDITOR is negative and statistically
significant in the model with positive Dechow and Dichev (2002) accruals as the dependent
variable, and the coefficients for both FAUDITOR and FRATIO are negative and marginally
significant in the models with absolute negative Jones (1991) accruals as the dependent variable.
Hence, these estimates suggest that the client firms of female audit engagement partners are
associated with more restrained use of both income-increasing and income-decreasing accruals.
Regarding the control variables, it can be noted from Table 4 that firm size and the alternative
performance measures are the most important factors for explaining the level of positive and
negative abnormal accruals.
Overall, the regression estimates reported in Tables 3 and 4 indicate that the gender of the audit
engagement partner is associated with the quality of financial reporting. In particular, our
estimations provide considerable evidence to suggest that firms with female audit partners are
associated with smaller abnormal accruals, thereby implying that female auditors may constrain
excessive earnings management practices in their client firms. Furthermore, we also find evidence
indicating that the client firms of female auditors have smaller income-increasing, as well as
income-decreasing, accruals. These findings are generally consistent with the experimental results
of Chung and Monroe (2001), O’Donnell and Johnson (2001), and Gold et al. (2009), who
document that female auditors may be more efficient and less influenced by their clients.
Furthermore, our empirical findings are consistent with the cognitive psychology and behavioral
economics literature that have documented significant gender-based differences in diligence,
conservatism, and risk tolerance (see, e.g., Feingold 1994; Byrnes et al. 1999; Costa et al. 2001;
Dwyer et al. 2002; Eckel and Grossman 2002; Beu et al. 2003; Nettle 2007; Schmitt et al. 2008;
Pierce and Sweeney 2010).
Additional Tests
We conduct several additional tests to examine the robustness of our empirical findings. First,
given that the current accruals may be more susceptible to management’s judgmental or estimation
bias compared to the non-current component of accruals, we follow the approach of Ashbaugh et al.
(2003) and use the modified Jones (1991) model to estimate abnormal current accruals. We then
reestimate Equation (3) with the abnormal current accruals as the dependent variable. The
estimation results based on absolute abnormal current accruals (not tabulated) are qualitatively
similar to the results reported in Table 3. The negative coefficients for the female auditor variables,
however, are insignificant at the conventional significance levels. The estimates based on positive
Female Auditors and Accruals Quality 219
Accounting Horizons
June 2013
TABLE 4
Positive and Negative Abnormal Accruals
Variable
Positive Accruals Absolute Negative Accruals
AAJON AAJON AADD AADD jAAJONj jAAJONj jAADDj jAADDj
Constant 4.041*** 4.099*** 2.725*** 3.320*** 5.839*** 5.815** 2.412 2.380
(6.69) (6.63) (2.80) (3.26) (2.78) (2.57) (1.28) (1.25)
Female Auditor Variables:
FAUDITOR 0.000 0.236** 0.274* 0.101
(0.00) (1.98) (1.89) (0.71)
FRATIO 0.067 0.027 0.297* 0.099
(0.45) (0.16) (1.94) (0.46)
Control Variables:
SIZE 0.022 0.023 0.187*** 0.190*** 0.255*** 0.256*** 0.134*** 0.137***
(0.75) (0.81) (2.95) (2.97) (8.78) (9.01) (23.48) (34.56)
LEV 0.323 0.320 0.600 0.563 0.025 0.018 0.875*** 0.865***
(0.74) (0.73) (1.02) (0.98) (0.10) (0.07) (4.17) (3.97)
ROA 0.063*** 0.063*** 0.006 0.006 0.060*** 0.059*** 0.006 0.006
(3.53) (3.54) (1.14) (1.09) (9.23) (9.55) (1.44) (1.46)
LOSS 0.684*** 0.687*** 0.418* 0.428* 0.187** 0.183** 0.403** 0.397**
(2.83) (2.84) (1.76) (1.80) (2.23) (2.24) (2.26) (2.16)
OCF 8.726*** 8.728*** 1.642*** 1.746*** 6.853*** 6.847*** 1.557*** 1.551***
(7.81) (7.86) (2.76) (2.85) (11.65) (11.83) (3.28) (3.25)
INVREC 0.532** 0.539** 0.222 0.294 0.720*** 0.696*** 0.666 0.672
(2.44) (2.45) (0.49) (0.65) (2.86) (2.97) (1.59) (1.61)
FOROPR 0.004** 0.004** 0.001 0.001 0.006*** 0.006*** 0.001 0.001
(2.50) (2.49) (0.27) (0.18) (4.37) (4.57) (1.04) (0.86)
SGRTH 0.003*** 0.003*** 0.001* 0.001** 0.002 0.002 0.000 0.000
(2.79) (2.84) (1.84) (2.31) (1.44) (1.49) (0.23) (0.23)
MB 0.036*** 0.036*** 0.043** 0.042* 0.027 0.027 0.039** 0.038**
(4.49) (4.39) (2.02) (1.95) (0.83) (0.84) (2.35) (2.43)
AC 0.045 0.050 0.214 0.213 0.044 0.032 0.201 0.207
(0.48) (0.53) (1.22) (1.16) (0.28) (0.19) (1.35) (1.38)
(continued on next page)
220 Ittonen, Va¨ha¨maa, and Va¨ha¨maa
Accounting Horizons
June 2013
TABLE 4 (continued)
Variable
Positive Accruals Absolute Negative Accruals
AAJON AAJON AADD AADD jAAJONj jAAJONj jAADDj jAADDj
BIG4 0.223** 0.228** 0.090 0.115 0.024 0.030 0.038 0.035
(2.21) (2.27) (0.72) (1.00) (0.16) (0.20) (0.36) (0.33)
AAGE 0.116 0.128 0.056 0.073 0.771 0.769 0.115 0.121
(0.64) (0.71) (0.20) (0.25) (1.51) (1.41) (0.23) (0.24)
AINDEXP 0.181 0.180 0.576*** 0.581*** 0.211** 0.207** 0.068 0.067
(1.52) (1.54) (5.09) (5.32) (2.37) (2.11) (0.76) (0.73)
STAOWN 0.172 0.168 0.497 0.478 0.671 0.682 0.076 0.077
(1.15) (1.15) (1.52) (1.44) (1.51) (1.50) (0.52) (0.56)
FIN 0.117 0.119 0.074 0.092 0.037 0.032 0.177** 0.175**
(1.51) (1.54) (1.02) (1.23) (0.49) (0.43) (2.15) (2.07)
Industry Fixed Effects Yes Yes Yes Yes Yes Yes Yes Yes
Year Fixed Effects Yes Yes Yes Yes Yes Yes Yes Yes
Adjusted R2 0.386 0.386 0.111 0.108 0.459 0.459 0.241 0.240
F-stat. 13.764*** 13.732*** 3.173*** 3.102*** 11.002*** 10.952*** 6.039*** 6.021***
***, *** Denote significance at the 0.10, 0.05, and 0.01 levels, respectively.
The table reports the estimates of eight alternative versions of the following regression model:
lnjAAi;tj ¼ a0 þ b1FEMALEi;t þ b2SIZEi;t þ b3LEVi;t þ b4ROAi;t þ b5LOSSi;t þ b6OCFi;t þ b7INVRECi;t þ b8FOROPRi;t þ b9SGRTHi;t þ b10MBi;t þ b11ACi;t
þ b12BIG4i;t þ b13AAGEi;t þ b14AINDEXPi;t þ b15STAOWNi;t þ b16FINi;t þX
n1
k¼1
akSICki
þ X
2007
y¼2006
xyYEARy
i þ ei;t ;
where jAAi,tj denotes either positive or absolute negative abnormal accruals for firm i in year t, based either on the Jones (1991) (AAJON) or the Dechow and Dichev (2002)
(AADD) model. The test variable in our regression specification is FEMALE, which is either (1) FAUDITOR, a female auditor dummy equaling 1 if the firm has appointed a female
audit partner, or (2) FRATIO, which is defined as the proportion of female audit engagement partners to all audit engagement partners (designed to control for audits with two
engagement partners). The t-statistics (reported in parentheses) are based on standard errors corrected for clustering at the firm level.
Variable Definitions:
SIZE ¼ natural logarithm of total assets;
LEV ¼ percentage of total debt to total assets;
ROA ¼ return on assets;
LOSS ¼ dummy variable that equals 1 for firms with negative net income;
(continued on next page)
Female Auditors and Accruals Quality 221
Accounting Horizons
June 2013
TABLE 4 (continued)
OCF ¼ cash flow from operations;
INVREC ¼ inventory and receivables to total assets;
FOROPR ¼ proportion of foreign sales of total sales;
SGRTH ¼ sales growth rate;
MB ¼ ratio market value of equity to book value of equity;
AC ¼ 1 if the firm has an audit committee;
BIG4 ¼ 1 if the firm has a Big 4 auditor;
AAGE ¼ logarithm of auditor’s age;
AINDEXP ¼ dummy variable that equals 1 if the auditor has over 10 percent market share in a particular industry;
STAOWN ¼ dummy variable equaling 1 if the Finnish or the Swedish government holds shares in the firm;
FIN ¼ dummy variable for Finnish firms;
SIC ¼ dummy variable according to industry classification codes; and
YEAR ¼ a dummy for fiscal years.
222 Ittonen, Va¨ha¨maa, and Va¨ha¨maa
Accounting Horizons
June 2013
and absolute negative abnormal accruals are consistent with the results reported in Table 4, and the
coefficients for FAUDITOR and FRATIO appear negative and marginally significant in the models
with absolute negative accruals as the dependent variable.
Second, we attempt to ascertain that our findings are not affected by potentially redundant
variables by estimating a more parsimonious model. In particular, we reestimate all regressions with
SIZE, ROA, and MB as the only control variables. The estimation results of these regressions (not
tabulated) are qualitatively similar to the results reported in Tables 3 and 4. Most importantly, our
additional estimations indicate that absolute abnormal accruals are statistically significantly lower
for firms with female audit engagement partners. Furthermore, in order to ensure that our results are
not affected by multicollinearity, we reestimate all regressions with ROA as the only performance
control. Again, the estimates of these regressions (not tabulated) suggest that the client firms of
female auditors are associated with better accruals quality. Hence, we conclude that our empirical
findings are robust to alternative model specifications.
The results of Barua et al. (2010) and Peni and Va¨ha¨maa (2010) suggest that firms with female
CFOs are associated with more conservative financial reporting practices. Because the gender of the
firm’s CFO may potentially be correlated with the gender of the firm’s audit engagement partner,
we reestimate the regressions with a CFO gender dummy as an additional control variable. For this
purpose, we manually collect data on CFO gender for all Swedish firms from their annual reports.14
The estimation results (not tabulated) based on this subsample of Swedish firms are consistent with
our main findings, and indicate that female auditors are significantly negatively associated with
abnormal accruals also after controlling for the gender of the firm’s CFO.
Finally, it should be noted that our empirical analysis may suffer from endogeneity problems. It
is possible that firms with better accruals quality may strategically select female audit engagement
partners or, alternatively, that female auditors self-select into less-risky firms with better financial
reporting quality due to the gender differences in risk tolerance. Ultimately, the effect of female
auditors on earnings management could be tested by examining whether the level of abnormal
accruals actually decreases after the appointment of a female audit engagement partner. Hence, we
identify changes in auditor gender during our sample period and reestimate the regressions with two
additional dummy variables that indicate male-to-female and female-to-male audit partner changes.
The auditor-change dummy variables equal 1 for the transition year and the subsequent year.
During our sample period, there are 12 changes from a male auditor to a female auditor, and eight
changes from a female auditor to a male auditor. The estimated coefficient for the male-to-female
dummy variable is negative and statistically significant at the 10 percent level in the regression with
absolute Jones (1991) accruals as the dependent variable, thereby suggesting that the appointment
of a female audit engagement partner after a male auditor may improve accruals quality. The
coefficient estimates for the female-to-male dummy variable appear insignificant in the regressions.
Nevertheless, given the very small number of changes in auditor gender, these estimates are
constrained by a lack of statistical power.
Limitations
We acknowledge several limitations in our analysis. First, it should be noted that the general
theoretical background for the analysis is mostly drawn from psychology and behavioral economics
literature and, consequently, our empirical findings have to be regarded as somewhat exploratory.
Second, our sample is relatively small, and includes only 91 firm-year observations with female
14 We are able to obtain data on CFO gender for 476 firm-year observations. In this subsample of Swedish firms, the
correlation coefficient between female audit engagement partners and female CFOs is 0.069 and statistically
significant at the 10 percent level.
Female Auditors and Accruals Quality 223
Accounting Horizons
June 2013
audit engagement partners. The small number of firms with female auditors obviously limits the
statistical power of our tests. Third, the sample consists of the NASDAQ OMX-listed firms from
Finland and Sweden. It is uncertain to what extent our empirical findings apply in other countries
and in other institutional settings. It should also be noted that Finland and Sweden are among the
leading countries in the world in terms of gender equality and, therefore, the argumentation based
on higher competence due to the glass ceiling phenomenon may be less valid in our sample.15
Nevertheless, some degree of gender discrimination is still likely to exist in Finland and Sweden
and, therefore, we presume that females may have to demonstrate extra competence in order to
attain positions as audit engagement partners.
Furthermore, we acknowledge that the abnormal accruals used in the empirical analysis are
subject to measurement errors, and may not provide perfect estimates of the extent of earnings
management. As can be noted from Tables 3 and 4, there are some inconsistencies between the
regressions with the Jones (1991) and the Dechow and Dichev (2002) abnormal accruals as the
dependent variable. An obvious extension of our analysis would be to use alternative measures of
earnings management, for instance, such as the timeliness of loss recognition (see Ball and
Shivakumar 2005) and the likelihood of meeting or beating analysts’ earnings forecasts (see, e.g.,
Matsumoto 2002; Koh et al. 2008).
Finally, it is important to recognize that our empirical analysis may suffer from endogeneity
problems. We have controlled for a set of firm-specific factors that, according to the prior literature,
affect accruals quality. However, we may have omitted some correlated variables, and moreover,
some firm characteristics may simultaneously affect the appointment of a female audit engagement
partner and earnings management. It is possible that firms with better earnings quality may
strategically select female auditors for their allegedly higher competence or, alternatively, that
female auditors self-select into less-risky firms with higher earnings quality due to gender-based
differences in risk tolerance. In order to address these endogeneity problems, we examined whether
the quality of financial reporting changes after the appointment of a female auditor. Given that we
are unable to completely rule out the possibility that the appointment of a female auditor is
endogenously determined, any causal interpretations of our findings should be made with caution.
CONCLUSIONS
In this paper, we empirically examine the association between accruals quality and the gender
of the firm’s audit engagement partner. Motivated by the behavioral differences between women
and men, we hypothesize that gender-based differences in diligence, conservatism, and risk
tolerance may have important implications for the audit process and auditor judgments and, thus,
ultimately, for the quality of audited financial information. In our empirical analysis, we utilize the
unique institutional setting of the Finnish and Swedish NASDAQ OMX-listed firms in which the
disclosure of the identity of the audit engagement partner(s) is mandatory in audit reports. We use
the modified Jones (1991) model and the modified Dechow and Dichev (2002) model to measure
accruals quality in the client firms, and run several alternative panel regressions of abnormal
accruals on female audit partner variables and firm-specific controls.
Our empirical findings demonstrate that the gender of the audit engagement partner is
associated with the quality of financial reporting. In particular, our findings suggest that the client
firms of female audit engagement partners may have smaller absolute abnormal accruals in their
reported earnings. We also document that female auditors may constrain the use of both
income-increasing and income-decreasing accruals. Thus, consistent with the prior experimental
15 Finland and Sweden are ranked third and fourth, respectively, in the gender equality ranking of the World
Economic Forum (WEF) (see WEF 2010).
224 Ittonen, Va¨ha¨maa, and Va¨ha¨maa
Accounting Horizons
June 2013
studies on auditor gender effects, the empirical findings reported in this paper suggest that gender
differences may affect the audit process. Moreover, the results are broadly consistent with the vast
body of literature on gender differences in diligence, conservatism, and risk tolerance. In general,
our findings suggest that the behavioral differences between women and men may have important
implications for the quality of auditing and financial reporting.
REFERENCES
Ali, A., T-Y. Chen, and S. Radhakrishnan. 2007. Corporate disclosures by family firms. Journal of
Accounting and Economics 44: 238–286.
Ashbaugh, H., R. LaFond, and B. Mayhew. 2003. Do nonaudit services compromise auditor independence?
Further evidence. The Accounting Review 78: 611–639.
Baldry, J. 1987. Income tax evasion and the tax schedule: Some experimental results. Public Finance 42:
357–383.
Ball, R., and L. Shivakumar. 2005. Earnings quality in U.K. private firms: Comparative loss recognition.
Journal of Accounting and Economics 39: 83–128.
Balsam, S., J. Krishnan, and J. S. Yang. 2003. Auditor industry specialization and earnings quality.
Auditing: A Journal of Practice & Theory 22: 71–97.
Barber, B., and T. Odean. 2001. Boys will be boys: Gender, overconfidence, and common stock investment.
Quarterly Journal of Economics 116: 261–292.
Barnett, T., K. Bass, and G. Brown. 1994. The ethical judgments of college students regarding business
issues. Journal of Education for Business 69: 333–338.
Bartov, E., F. Gul, and J. Tsui. 2001. Discretionary-accruals models and audit qualifications. Journal of
Accounting and Economics 30: 421–452.
Barua, A., L. Davidson, D. Rama, and S. Thiruvadi. 2010. CFO gender and accruals quality. Accounting
Horizons 24: 25–39.
Becker, L., M. DeFond, J. Jiambalvo, and K. Subramanyam. 1998. The effect of audit quality on earnings
management. Contemporary Accounting Research 15 (1): 1–24.
Bernardi, R., and D. Arnold. 1997. An examination of moral development within public accounting by
gender, staff level and firm. Contemporary Accounting Research 14: 653–668.
Beu, D., M. Buckley, and M. Harvey. 2003. Ethical decision-making: A multidimensional construct.
Business Ethics: A European Review 12: 88–107.
Byrnes, J. P., D. C. Miller, and W. D. Schafer. 1999. Gender differences in risk taking: A meta-analysis.
Psychological Bulletin 125: 367–383.
Caramanis, C., and C. Lennox. 2008. Audit effort and earnings management. Journal of Accounting and
Economics 45: 116–138.
Chi, H., and C. Chin. 2011. Firm versus partner measures of auditor industry expertise and effects on
auditor quality. Auditing: A Journal of Practice & Theory 30: 201–229.
Christie, A. A., and J. L. Zimmerman. 1994. Efficient and opportunistic choices of accounting procedures:
Corporate control contests. The Accounting Review 69 (4): 539–566.
Chung, J., and G. Monroe. 2001. A research note on the effects of gender and task complexity on an audit
judgment. Behavioral Research in Accounting 13: 111–125.
Cohen, J., G. Krishnamoorthy, and A. Wright. 2004. The corporate governance mosaic and financial
reporting quality. Journal of Accounting Literature 23: 87–152.
Costa, P., A. Terracciano, and R. McCrae. 2001. Gender differences in personality traits across cultures:
Robust and surprising findings. Journal of Personality and Social Psychology 81: 322–331.
Croson, R., and U. Gneezy. 2009. Gender differences in preferences. Journal of Economic Literature 47: 1–27.
DeAngelo, H., L. DeAngelo, and D. Skinner. 1994. Accounting choice in troubled companies. Journal of
Accounting and Economics 17: 113–143.
Dechow, P., and I. D. Dichev. 2002. The quality of accruals and earnings: The role of accrual estimation
errors. The Accounting Review 77: 35–59.
Female Auditors and Accruals Quality 225
Accounting Horizons
June 2013
Dechow, P., R. Sloan, and A. Sweeney. 1995. Detecting earnings management. The Accounting Review 70:
193–225.
DeFond, M., and J. Jiambalvo. 1993. Factors related to auditor-client disagreements over income-increasing
accounting methods. Contemporary Accounting Research 9: 415–431.
Dwyer, P., J. Gilkeson, and J. List. 2002. Gender differences in revealed risk taking: Evidence from mutual
fund investors. Economics Letters 76: 151–158.
Eckel, C., and P. Grossman. 2002. Sex differences and statistical stereotyping in attitudes toward financial
risk. Evolution and Human Behavior 23: 281–295.
Fallan, L. 1999. Gender, exposure to tax knowledge, and attitudes towards taxation: An experimental
approach. Journal of Business Ethics 18: 173–184.
Feingold, A. 1994. Gender differences in personality: A meta-analysis. Psychological Bulletin 116: 429–
456.
Fondas, N., and S. Sassalos. 2000. A different voice in the boardroom: How the presence of women
directors affects board influence over management. Global Focus 12: 13–22.
Francis, J., E. Maydew, and H. Sparks. 1999. The role of Big 6 auditors in the credible reporting of accruals.
Auditing: A Journal of Practice & Theory 18: 17–34.
Francis, J., P. Michas, and S. Seavey. 2012. Does audit market concentration harm the quality of audited
earnings? Evidence from audit markets in 42 countries. Contemporary Accounting Research 30: 325–
355.
Francis, J., R. LaFond, P. Olsson, and K. Schipper. 2005. The market pricing of accruals quality. Journal of
Accounting and Economics 39: 295–327.
Frankel, R., M. Johnson, and K. Nelson. 2002. The relation between auditors’ fees for nonaudit services and
earnings management. The Accounting Review 77: 71–105.
Geiger, M., and D. North. 2006. Does hiring a new CFO change things? An investigation of changes in
discretionary accruals. The Accounting Review 81: 781–809.
Ghosh, A., and D. Moon. 2005. Auditor tenure and perceptions of audit quality. The Accounting Review 80:
585–612.
Gold, A., J. Hunton, and M. Gomaa. 2009. The impact of client and auditor gender on auditors’ judgments.
Accounting Horizons 23: 1–18.
Green, C., N. Jegadeesh, and Y. Tang. 2009. Gender and job performance: Evidence from Wall Street.
Financial Analysts Journal 65: 65–78.
Gul, F., S. Fung, and B. Jaggi. 2009. Earnings quality: Some evidence on the role of auditor tenure and
auditors’ industry expertise. Journal of Accounting and Economics 47: 265–287.
Holthausen, R. 1990. Accounting method choice: Opportunistic behavior, efficient contracting, and
information perspectives. Journal of Accounting and Economics 12: 207–218.
Ittonen, K., and E. Peni. 2012. Auditor’s gender and audit fees. International Journal of Auditing 16: 1–18.
Jenkins, D., and U. Velury. 2008. Does auditor tenure influence the reporting of conservative earnings?
Journal of Accounting and Public Policy 27: 115–132.
Jenkins, D., G. D. Kane, and U. Velury. 2006. Earnings quality decline and the effect of industry specialist
auditors: An analysis of the late 1990s. Journal of Accounting and Public Policy 25: 71–90.
Jianakoplos, N., and A. Bernasek. 1998. Are women more risk averse? Economic Inquiry 36: 620–630.
Johnson, J., and P. Powell. 1994. Decision making, risk and gender: Are managers different? British
Journal of Management 5: 123–138.
Johnson, V. E., I. K. Khurana, and J. K. Reynolds. 2002. Audit-firm tenure and the quality of financial
reports. Contemporary Accounting Research 19: 637–660.
Jones, J. 1991. Earnings management during import relief investigations. Journal of Accounting Research
29: 193–228.
Jones, K., G. Krishnan, and K. Melendrez. 2008. Do models of discretionary accruals detect actual cases of
fraudulent and restated earnings? An empirical analysis. Contemporary Accounting Research 25:
499–531.
Kim, J-B., R. Chung, and M. Firth. 2003. Auditor conservatism, asymmetric monitoring and earnings
management. Contemporary Accounting Research 20: 323–359.
226 Ittonen, Va¨ha¨maa, and Va¨ha¨maa
Accounting Horizons
June 2013
Kinney, W. R., and R. D. Martin. 1994. Does auditing reduce bias in financial reporting? A review of auditrelated
adjustment studies. Auditing: A Journal of Practice & Theory 13: 151–156.
Koh, K., D. Matsumoto, and S. Rajgopal. 2008. Meeting or beating analyst expectations in the postscandals
world: Changes in stock market rewards and managerial actions. Contemporary Accounting
Research 25: 1067–1098.
Kothari, S., A. Leone, and C. Wasley. 2005. Performance matched discretionary accrual measures. Journal
of Accounting and Economics 39: 163–197.
Krishnan, G. 2003. Audit quality and the pricing of discretionary accruals. Auditing: A Journal of Practice
& Theory 22: 109–126.
Kumar, A. 2010. Self-selection and the forecasting abilities of female equity analysts. Journal of
Accounting Research 48: 393–435.
La Porta, R., F. Lopez-de-Silanes, A. Shleifer, and R. Vishny. 1998. Law and finance. Journal of Political
Economy 106: 1113–1155.
Levin, I., M. Snyder, and D. Chapman. 1988. The interaction of experimental and situational factors and
gender in a simulated risky decision-making task. Journal of Psychology 122: 173–181.
Lin, J., and M. Hwang. 2010. Audit quality, corporate governance, and earnings management: A metaanalysis.
International Journal of Auditing 14: 57–77.
Matsumoto, D. 2002. Management’s incentives to avoid negative earnings surprises. The Accounting
Review 77: 483–514.
McNichols, M. 2002. Discussion of the quality of accruals and earnings: The role of accrual estimation
errors. The Accounting Review 77: 61–69.
Meek, G., R. Rao, and C. Skousen. 2007. Evidence on factors affecting the relationship between CEO stock
option compensation and earnings management. Review of Accounting and Finance 6: 304–323.
Myers, J., L. Myers, and T. Omer. 2003. Exploring the term of the auditor-client relationship and the quality
of earnings: A case for mandatory auditor rotation? The Accounting Review 78: 779–799.
Nagy, A. 2005. Mandatory audit firm turnover, financial reporting quality, and client bargaining power: The
case of Arthur Andersen. Accounting Horizons 19: 51–68.
Nelson, M., J. Elliott, and R. Tarpley. 2002. Evidence from auditors about managers’ and auditors’ earnings
management decisions. The Accounting Review 77: 175–202.
Nettle, D. 2007. Empathizing and systemizing: What are they, and what do they contribute to our
understanding of psychological sex differences? British Journal of Psychology 98: 237–255.
Niskanen, J., J. Karjalainen, M. Niskanen, and J. Karjalainen. 2011. Auditor gender and corporate earnings
management behavior in private Finnish firms. Managerial Auditing Journal 26: 778–793.
O’Donnell, E., and E. Johnson. 2001. The effects of auditor gender and task complexity on information
processing efficiency. International Journal of Auditing 5: 91–105.
Peni, E., and S. Va¨ha¨maa. 2010. Female executives and earnings management. Managerial Finance 35:
629–645.
Pierce, B., and B. Sweeney. 2010. The relationship between demographic variables and ethical decision
making of trainee accountants. International Journal of Auditing 14: 79–99.
Piot, C., and R. Janin. 2007. External auditors, audit committees and earnings management in France.
European Accounting Review 16: 429–454.
Powell, M., and D. Ansic. 1997. Gender differences in risk behaviour in financial decision making: An
experimental analysis. Journal of Economic Psychology 18: 605–627.
Public Company Accounting Oversight Board (PCAOB). 2009. Concept Release on Requiring the
Engagement Partner to Sign the Audit Report. PCAOB Release No. 2009–005. Washington, DC:
PCAOB.
Schmitt, D., A. Realo, M. Voracek, and J. Allik. 2008. Why can’t a man be more like a woman? Sex
differences in big five personality traits across 55 cultures. Journal of Personality and Social
Psychology 94: 168–182.
Sharma, V., D. Sharma, and U. Ananthanarayanan. 2011. Client importance and earnings management: The
moderating role of audit committees. Auditing: A Journal of Practice & Theory 30: 125–156.
Female Auditors and Accruals Quality 227
Accounting Horizons
June 2013
Sinani, E., A. Stafsudd, S. Thomsen, C. Edling, and T. Randoy. 2008. Corporate governance in
Scandinavia: Comparing networks and formal institutions. European Management Review 5: 27–40.
Srinidhi, B., and F. Gul. 2007. The differential effects of auditors’ nonaudit and audit fees on accrual
quality. Contemporary Accounting Research 24: 596–629.
Watkins, A., W. Hillison, and S. Morecroft. 2004. Audit quality: A synthesis of theory and empirical
evidence. Journal of Accounting Literature 23: 153–193.
Watson, J., and M. McNaughton. 2007. Gender differences in risk aversion and expected retirement
benefits. Financial Analysts Journal 63 (4): 52–62.
World Economic Forum (WEF). 2010. The Global Gender Gap Report 2010. Geneva, Switzerland: World
Economic Forum.
228 Ittonen, Va¨ha¨maa, and Va¨ha¨maa
Accounting Horizons
June 2013
Copyright of Accounting Horizons is the property of American Accounting Association and
its content may not be copied or emailed to multiple sites or posted to a listserv without the
copyright holder’s express written permission. However, users may print, download, or email
articles for individual use.