Do Audit e Fforts Increase The Future Equity Value of Client Firm?

The current issue and full text archive of this journal is available on Emerald Insight at:
https://www.emerald.com/insight/0268-6902.htm
Future equity
Do audit efforts increase the value of client
future equity value of client firm? firm
Sang Ho Lee
Korea Capital Market Institute, Seoul, Republic of Korea
Seung Uk Choi
Department of Accounting and Taxation, Kyung Hee University, Received 15 November 2019
Revised 1 August 2020
Seoul, Republic of Korea, and 24 October 2020
Accepted 7 December 2020
Ji Yeon Ryu
Division of Business Administration, Yonsei University, Wonju, Republic of Korea
Abstract
Purpose – The purpose of this study is to examine the association between additional audit efforts and
clients’ future equity value. The study hypothesizes that auditors’ additional audit efforts directly increase
clients’ stock return performance. Additionally, this study expects that the additional audit effort lowers the
likelihood of audit failure and improves accounting information quality, thereby indirectly increasing clients’
future equity return performance.
Design/methodology/approach – The regression and portfolio return tests are conducted using
observations from 2003 to 2016. This study uses the abnormal audit hours as a proxy for additional audit
effort using mandatorily disclosed audit hour data from Korean listed firms. The study also conducts
mediation analyses to examine the causal intermediate steps that link audit effort to client equity return
performance.
Findings – The paper documents a significant and positive association between abnormal audit hours and
clients’ subsequent years’ stock return performance and Tobin’s Q. This finding is accentuated for clients
audited by Big N auditors or with greater demand for superior audit service. This finding is robust after
controlling for various proxies of accounting quality. The portfolio return tests also find evidence that
investors cannot fully perceive the value of audit efforts. A battery of additional tests does not alter the main
findings.
Practical implications – The results provide implications for investors and policymakers by
emphasizing the importance of audit efforts in value-creation. Moreover, this study’s findings suggest that
auditors’ assurance, insurance and information roles are all the important drivers of this value-creation.
Originality/value – This study highlights a prominent feature of audit effort that enhances the value of
auditees.
Keywords Market anomaly, Audit effort, Abnormal audit hour, Equity value,
Hedge portfolio returns
Paper type Research paper
1. Introduction
Prior research documents the positive impact of audit effort on the client’s information
environment (Caramanis and Lennox, 2008; Chen et al., 2011; Pittman and Fortin, 2004).
However, only a few have investigated the direct effect of audit effort on auditees’ value-
creation [1]. Filling this void, we examine whether the auditors’ additional efforts could
Managerial Auditing Journal
© Emerald Publishing Limited
0268-6902
JEL classification – G14, M42 DOI 10.1108/MAJ-11-2019-2473
MAJ affect audit clients’ future equity value by providing assurance, insurance and information
values. We then divide audit determinants into audit supply side and demand side and test
whether these determinants influence the association. Finally, we examine whether
investors perceive auditors’ additional effort in their decision-making process using a zero-
cost investment portfolio approach.
Our research expects extensive audit effort to be positively related to the audited clients’
future equity value for the following reasons [2]. First, auditors can reduce stock price crash
risk and prevent loss if they spend an appropriate amount of audit hours, recognizing client-
specific risks (Callen and Fang, 2017; Hogan and Wilkins, 2008). Second, audit effort reduces
the implied cost of equity capital by lowering clients’ information uncertainty (Chen et al.,
2011; Pittman and Fortin, 2004). Third, abnormal audit hours are indirectly related to
positive stock returns by enhancing financial reporting (Becker et al., 1998; Francis and
Krishnan, 1999; Krishnan, 2003a; Lambert et al., 2007). Specifically, extensive audit
procedures could deter managers from reporting low-persistent and high-accrual earnings.
Thus, we conduct a cross-sectional regression analysis after controlling for various proxies
of financial reporting quality and firm-specific risk characteristics identified from prior
studies to examine whether abnormal audit hours have a direct positive impact on future
stock return performance.
Auditors’ traditional roles provide further insight into the relationship between audit
hours and clients’ equity values. Auditor’s widely accepted roles are assurance, information
(signaling) and insurance (liability). We posit that additional audit hours affect clients’
future equity values through all three channels of auditors’ roles.
First, the existing studies document that external auditors’ assurance services affect the
credibility of disclosed financial statements (Becker et al., 1998; Blackwell et al., 1998) and
lower the costs of debt and equity (Francis et al., 2005). Therefore, assurance may affect the
relationship between audit hours and client equity return performance as extensive audit
efforts increase clients’ equity value through improvements in the financial reporting
quality. Second, the information role of audit service implies that auditors could help
address information asymmetry between audit clients and capital providers (Wallace, 2004).
The existing study finds that clients experience a lower cost of debt and equity (Mansi,
Maxwell, and Miller, 2004), lower underpricing (Chang et al., 2008) and higher earnings
response coefficients (Ghosh and Moon, 2005) by being audited by Big N auditors.
Therefore, hiring auditors with expertise and capacity to spend sufficient audit effort
signals to the market, contributing to clients’ equity value. Third, the insurance (or liability)
role suggests that large auditors provide greater insurance coverage in the event of audit
failure and securities litigation. Additionally, as large auditors may spend a greater amount
of audit hours (Caramanis and Lennox, 2008), the market perceives the insurance value of
these firms audited by Big N auditors differently over those audited by non-Big N auditors,
therefore, having a different effect on clients’ future equity value. Overall, we examine how
different roles of audit service affect our main results: increased audit hours enhance clients’
stock return performance [3].
Our empirical analysis adopts 13,859 firm-year observations from firms listed in the
Korean capital market from 2003 to 2016. The setting is ideal for our research because the
Act on External Audit requires all listed firms in Korea to disclose audit hours in their
annual reports mandatorily. The result shows a positive and significant correlation between
abnormal audit hours and one-year ahead monthly stock returns, reflecting a direct
influence of abnormal audit effort on clients’ future equity value even after controlling for
various reporting quality effects.
Next, we identify the audit determinants that may affect the relationship between Future equity
abnormal audit hours and future stock performance. From the audit-supply perspective, Big value of client
N auditors have greater capacity and resources to increase audit efforts than non-Big N
auditors (Hogan and Wilkins, 2008). Therefore, an hour of Big N auditor has greater audit
firm
value than a non-Big N auditor. Big N auditors also have greater motivation to protect their
reputation and avoid expensive litigation (DeAngelo, 1981a, 1981b; Dye, 1993; Lennox and
Pittman, 2010). Based on this rationale, we find that the future stock return performances of
clients audited by Big N auditors are greater than those audited by Non-Big N auditors.
We also consider the audit demand perspective. International Financial Reporting
Standard (IFRS) adoption and high financial statement comparability are known to improve
accounting information quality (De Franco et al., 2011; Landsman et al., 2012). We find that
the association between abnormal audit hours and future equity value weakens firms with
high financial statement comparability under IFRS. These firms have a smaller room for
improvement. Overall, these findings indicate that both supply- and demand-side subjects
incentivized by high-quality audit services are instrumental in enhancing firms’ equity
value.
Next, we perform causal mediation analyses using the accounting quality proxy and
sanctioned firms from the regulatory review as mediators. We find that better accounting
quality and a lower likelihood of audit failure are the causal chains linking audit hours to
equity value. The results show that these intermediate steps explain 9.12%–25.80% of audit
hours, resulting in greater client firm equity value.
Finally, the hedge portfolio taking a long position in the top decile and a short position in
the bottom decile of the abnormal audit hour portfolio earns a monthly average excess
return of 0.36%. Such a significant and positive arbitrage chance indicates that investors do
not fully perceive the efforts of abnormal audit information in their investment decisions.
We also perform various robustness tests, such as excluding extreme performing firms,
applying alternative criteria to measure abnormal audit effort and constructing portfolios,
controlling for accrual anomaly, and using alternative firm value proxies, such as Tobin’s Q.
Findings remain consistent with the main results.
The contributions of this study are as follows. First, to the best of the authors’
knowledge, this is the first study to examine the relationship between abnormal audit hours
and firms’ future equity value. While prior studies focused on audit efforts to improve
financial reporting quality (Caramanis and Lennox, 2008), this study extends the literature
by proving that strong audit efforts can create the firm value incremental to improve
accounting quality. Moreover, we disentangle audit effort from audit fee premiums by using
audit hours (DeFond and Zhang, 2014), thereby extending previous findings using audit fees
as a proxy of audit effort. Second, prior studies find that investors do not efficiently perceive
going concern audit opinions in making investment decisions (Kausar et al., 2009; Taffler
et al., 2004). Our study expands the field of study by finding that investors do not fully
perceive audit effort information, even though it is publicly available and economically
meaningful. Therefore, we suggest that regulators consider effective methods to adequately
disclose audit hours to enhance markets’ understanding of audit effort information.
The remainder of this paper is organized as follows. Section 2 presents the prior literature
and proposes our hypotheses. Section 3 describes our research design and sample. Section 4
provides the empirical results, and Section 5 concludes the paper.
2. Prior literature and hypothesis development

This study examines the significant association between abnormal audit hours and audit
clients’ future equity value based on the following rationale. First, high-quality audit effort
MAJ could prevent clients’ stock price crashes when auditors recognize the client firms’ various
risks that could affect stock price and equity value decline (Callen and Fang, 2017).
Second, auditors’ additional effort could reduce investors’ demand for the risk premium,
lowering the required discount rate. Newly listed companies audited by Big N audit firms
have a lower cost of debt capital than those not audited by Big N auditors (Pittman and
Fortin, 2004; Chen et al., 2011).
Third, previous studies predict that audit effort toward client firms’ various aspects
impacts clients’ values. For instance, Bronson et al. (2017) document that auditors assign
additional efforts to foreign firms cross-listed in the US market than domestic US firms,
lowering their litigation risks. Cho et al. (2019) show that auditors exert more effort when
their clients’ annual reports have lower readability. Asante-Appiah (2020) reports that
auditors’ efforts are positively associated with firms’ negative reputation in environmental,
social and governance practices. Overall, these empirical results suggest that auditors’
efforts may affect the litigation, reputation and credibility of accounting information,
increasing clients’ equity values.
While a direct positive effect of abnormal audit hours on clients’ equity value is expected,
there could also be an indirect effect; extensive audit effort influences clients’ stock prices by
enhancing financial reporting quality (Lambert et al., 2007). High audit quality is closely
related to the credibility of financial reporting and reduces investors’ expected risk (Khurana
and Raman, 2004) [4]. Therefore, an increase in audit hours improves accounting
information reliability (Caramanis and Lennox, 2008). Thus, we control variables known to
affect financial reporting quality, as suggested by prior studies, and examine the correlation
between abnormal audit hours and future abnormal returns. This leads us to investigate the
direct effect of additional audit effort over accounting quality on clients’ market value. We
propose our first hypothesis as follows:
H1. Abnormal audit hours have a positive association with one-year-ahead stock
performance after controlling for accounting quality.
We examine the association from two different perspectives: the audit-supply and audit-
demand side. First, we focus on the supplier of audit services: auditors. Auditors have
different capacities and intentions that could influence audit efforts. Therefore, we use the
determinants known to affect abnormal audit hours by analyzing auditor characteristics
related to abnormal audit hour input.
First, prior literature finds that audit firm size is positively related to audit quality
(DeAngelo, 1981a). Davidson and Neu (1993) state that larger auditing firms, the Big N audit
firms, provide higher quality audit services than smaller ones. Greater management error
forecast supports this statement as managers’ incentive to minimize the difference between
forecasted and reported income is reduced through high audit quality by Big N auditors.
Second, Big N audit firms have greater expertise, larger human resources and capital. Thus,
they increase audit input using their large pool of human resources for the same level of risk
recognized during audit procedures than non-Big N auditors (Hogan and Wilkins, 2008).
Additionally, their competence and expertise enhance accounting information quality,
positively affecting audit clients’ equity value (Krishnan, 2003b). Jiang et al. (2019) highlight
that audit quality improves when non-Big N auditors are acquired by Big N audit firms,
whereas mergers and acquisitions among non-Big N audit firms have an insignificant effect
on audit quality. Overall, Big N auditors can exert greater audit effort and provide higher
audit value because of greater expertise compared to a non-Big N auditor. Additionally, their
sufficient human resources and greater audit effort input positively benefit clients’ equity
values. If more Big N audit hours are better than more non-Big N hours, because of their
expertise and resources, the positive association between abnormal audit hours and future Future equity
stock performance, stated in the first hypothesis, would be stronger for the Big N auditor value of client
client firm sample:
firm
H2a. The positive association between abnormal audit hours and future stock
performance expected in H1 would be more pronounced for Big N auditors’
clients.
Second, we examine the association from the perspective of audit clients: audit demand side.
We expect the positive correlation between abnormal audit fees and firms’ equity value to be
weaker when the demand for high-quality audits is low, such as audit clients with a high-
quality information environment. Particularly, we examine the effect of IFRS adoption and
financial statement comparability on our main findings.
Although there are conflicting arguments, Landsman et al. (2012) provide evidence of
increased accounting informativeness in countries with mandatory IFRS adoption [5].
Owing to IFRS adoption, a stronger information environment indicates that there is little
room for information quality improvement by abnormal audit effort. Therefore, the
influence over future stock return by abnormal audit hours would be limited for client firms
following IFRS. Additionally, financial statement comparability could weaken the positive
association between abnormal audit hours and future stock returns, as shown in H1. Higher
comparability of financial statements reduces stakeholders’ information search costs and
benefits their investment decision-making process (De Franco et al., 2011). The effect of
abnormal audit hours on increasing future stock returns expected in H1 weakens because
investors use information efficiently from highly comparable financial statements. In sum,
the abnormal audit hour effect in improving accounting information quality is marginal for
clients under IFRS or high financial statement comparability. Therefore, we formulate the
following hypotheses:
H2b. The positive association between abnormal audit hours and future stock
performance expected in H1 would weaken after IFRS adoption.
H2c. The positive association between abnormal audit hours and future stock
performance expected in H1 would weaken client firms with high financial
statement comparability.
3. Research design
3.1 Estimation of abnormal audit hour
This study uses abnormal audit hours as a proxy for audit efforts. Although audit fee is
another proxy of audit efforts, prior studies have highlighted that audit fee reflects auditors’
economic bonding (Frankel et al., 2002; Myers et al., 2003; Reynolds and Francis, 2000),
which could damage audit quality. Meanwhile, actual audit hours imply firm-specific risk,
especially audit risk, creating an endogenous relationship between audit hours and equity
value [6]. Therefore, we measure abnormal audit hours by controlling various client-specific
risks, audit complexity and auditor characteristics.
First, we estimate the normal level of audit effort. We collect the actual audit hour data of
non-financial firms from 2002 to 2014. It is estimated by the year-by-year Fama and
MacBeth (1973) regressions using equation (1), following prior studies (Choi et al., 2010; Choi
and Wong, 2007; Craswell and Francis, 1999; Francis, 1984; Francis and Simon, 1987; Hribar
et al., 2014; O’Keefe et al., 1994; Palmrose, 1989):
MAJ lnAHit ¼ b 0 þ b 1 lnTAit þ b 2 LEVit þ b 3 BMit þ b 4 ISSUEit þ b 5 BIG4it
þ b 6 LARGEOWNit þ b 7 FOROWNit þ b 8 CONit þ b 9 CONFM
þ b 10 INVRECit þ b 11 ROAit þ b 12 CFOTAit þ b 13 LOSSit þ b 14 OPNit
þ b 15 SPVISEit þ b 16 FIRSTit þ b 17 KOSPIit þ b 18 lnAFit þ RIND þ « it
(1)
Here lnAHit is the natural log of the total audit hour for firm i in fiscal year t, and the
remaining variables are defined in Appendix 1.
We considered the following variables as the determinants of normal audit hours,
following prior studies. Audit efforts may increase with client size, leading to a positive
association between them. To control for client size, we incorporate lnTA, CON and CONFM
(Francis and Simon, 1987; Palmrose, 1989). Auditors may spend more time for clients with
more complicated business operations, resulting in a positive relationship between audit
complexity and audit hours. Variables INVREC and ISSUE are controlled for audit
complexity (Hribar et al., 2014). LEV, ROA, CFOTA and LOSS are used to proxy for the
client’s operating risk (Francis, 1984; O’Keefe et al., 1994). The coefficients of these variables
are expected to be positive. Greater audit efforts are required for high growth firms rather
than low-growth firms, and we include BM, expected to be negative on audit hours (Choi et
al., 2010). OPN and SPVISE capture unobservable audit risk (Francis and Simon, 1987;
Palmrose, 1989). BIG4 is controlled to capture the effect of different auditor characteristics
on audit effort. A positive sign of BIG4 indicates that Big 4 auditors exert a greater level of
audit effort. LARGEOWN and FOROWN are proxies for ownership structure and corporate
governance effects on audit effort (Choi and Wong, 2007). FIRST is used to reflect initial
engagement effects on audit efforts, leading to a positive association with audit hours
(Craswell and Francis, 1999). lnAF is controlled to consider the demand for audit services.
The demand for audit services is positively correlated with the level of audit fees. We
include KOSPI to alleviate potential bias from listed market characteristics. Industry fixed
effects (RIND) observe time-invariant heterogeneity among industries [7]. Newey–West
robust standard errors are used to correct for heteroskedasticity and serial correlation
(Newey and West, 1987).
Appendix 2 provides the estimated normal audit hours using 15,054 firm-year
observations over the 13 years, 2002–2014. Using the estimated coefficients from the year-
by-year Fama-MacBeth (1973) regressions of equation (1), we obtain the fitted value of audit
hours (lnAH ^ it ) as the normal level of audit hours. The adjusted R2 of the model is 67.20%,
and the overall goodness-of-fit is significant at the 1% level. Finally, we measure abnormal
audit hours (ABLNAHit) as the difference between the natural log of actual audit hours
(lnAHit) and the normal level of audit hours.
To test the robustness of the results, we recalculated equation (1) with various alterations.
First, we consider the actual audit hours as a scaling variable to construct a percentage measure.
Second, we incorporate a simplified model that considers only three or four controls regarding
client size and audit complexity (e.g. lnTA, BM, BIG4 and CONFM). Untabulated for brevity, the
results using alternative methods are qualitatively similar; thus, we cannot alter our inferences.
3.2 Cross-sectional regression model

This study uses equation (2) to examine the association between abnormal audit hours and
future abnormal returns [8]:
abRETitþ1 ¼ b 0 þ b 1 ABLNAHit þ b 2 ACCQualityVARit þ Controls Future equity
þ RYEAR þ RIND þ « it (2) value of client
firm
Here abRETitþ1 is the one-year ahead annual stock return, ABLNAHit is the abnormal audit
hours and ACCQualityVARit is a set of accounting quality variables for firm i in year t. We
consider five different model specifications to control for various accounting quality proxies
to analyze whether abnormal audit hours contain incremental informativeness after
controlling for accounting qualities known to affect stock returns. The following accounting
quality variables are controlled one variable at a time: accrual-based earnings management
(AEM), financial statement opaqueness (OPAQUE), real earnings management (REM),
profitability (ROA) and accounting conservatism (CONSERV) [9]. ACCQualityVAR is null
for the BASIC model.
The period of abRET starts four months after the fiscal year-end and ends in March of
the subsequent year to give the market adequate time to perceive the financial statement
information. To control for risk factors known to predict cross-sectional stock returns
effectively, our baseline model (BASIC) contains market capitalization (lnSIZE), book-to-
market ratio (lnBM), market beta (BETA), idiosyncratic volatility (IDVOL) and momentum
(MOM). BIG4 is considered to control audit quality. Prior research documents that
governance characteristics such as the number of directors or directors’ expertise are related
to a firm’s market value (DeFond et al., 2005; Black and Kim, 2012). To control for the
governance effect on market returns, we include the size of the board (lnBOD), the
accounting and legal expertise of audit committee members (lnSPBOA), board independence
(RoutBOD), CEO-chairman duality (CEODuality) and foreign investors’ ownership
(FOROWN). Year-fixed effects (RYEAR) are included to control for the time-series variance
of our variables. Industry-fixed effects (RIND) are included to observe time-invariant
heterogeneity among industries based on K-SIC2 [10].
3.3 Sample
The sample includes firms listed on the Korea Stock Exchange. Table 1 lists the selection
criteria. Firms satisfying the following criteria are excluded from the sample:
firms belonging to the financial industry;
fiscal year does not end in December;
Sample selection criteria Firm-years
Initial sample
All firm-year observations listed in Korea stock exchange (KOSPI and KOSDAQ) 20,834
during the fiscal year of 2002–2014 and available to get one-year ahead stock
returns during the period of 2003.04 to 2016.03
(Less)
Observations with financial firms (959)
Observations with non-December fiscal month (895)
Observations with impaired capital and non-unqualified opinion (137)
Observations with penny stocks (609)
Observations with unavailable to measure abnormal audit hours (3,515)
Observations with unavailable to measure other control variables (860) Table 1.
Final sample for the cross-sectional regression analysis 13,859 Sample distribution
MAJ impaired capital;
non-unqualified opinion;
penny stock compared to previous fiscal period [11]; and
missing data for measuring variables.
The final sample consists of 13,859 firm-year observations from 2003 to 2016.
The audit hour data are collected from TS2000, distributed by the Korea Listed Companies
Association. The financial statement data is obtained from KISVALUE provided by the NICE
Information Service, stock returns and market capitalization data are acquired from DataGuide
5.0, provided by the FnGuide Incorporate. Monetary Stabilization Bonds yield data, a proxy for
the risk-free rate, is collected from the Economic Statistics System of the Bank of Korea.
4. Empirical results
4.1 Descriptive statistics
Table 2 reports the descriptive statistics and correlation matrix for the variables. All
continuous variables for cross-sectional regressions are winsorized at the 1% and 99%
levels, respectively. Panel A shows descriptive statistics for the sample used for cross-
sectional regression. The mean (median) value of abRET is 0.231 (0.091) and has a positively
skewed distribution similar to prior studies. The mean (median) of ABLNAH is 0.001 (0.023),
and the distribution is skewed to the left. The negatively skewed distribution of ABLNAH
indicates the likelihood of extremely insufficient audit effort. The mean (median) of jPMDAj
is 0.105 (0.074). Other accounting quality proxies show similar descriptive statistics from the
previous studies. For instance, the mean values of REM3yr and C-Score are 0.007 and 0.430,
respectively. The mean (median) value of BETA is 0.990 (0.961), and we expect the market
risk characteristic to be similar to that of the market universe. Descriptive statistics for other
control variables are similar to those reported in prior literature.
According to Panel B of Table 2, the Pearson correlation between ABLNAH and abRET
is 0.029, a significantly positive correlation. Next, abRET is negatively correlated with
jPMDAj, jPMDA3yrj and REM3yr. These correlations support the hypothesized positive
association between audit effort and future stock returns and highlight the need to control
for variables related to accounting quality.
4.2 Cross-sectional regressions of future stock returns on abnormal audit hours (H1)
In this section, we examine whether ABLNAH and future abnormal returns have a positive
cross-sectional correlation. Table 3 presents the OLS regression results over the period
2003–2016. The BASIC model in Column (1) is our baseline specification from equation (2),
which controls the various well-known determinants of expected stock returns. From
Column (2) of the AEM model to Column (6) of the CONSERV model, we incorporate each
proxy of accounting qualities to verify that ABLNAH has incremental predictability on the
expected stock returns [12]. Including these accounting quality proxies also enables us to
check whether auditors’ insurance and/or information roles affect the client’s equity value
after controlling for its assurance role. In every specification of Table 3, the coefficient of
ABLNAH is significantly positive at the 5% or higher level. These results further show that
the positive association between the ABLNAH and future abnormal returns is not
subsumed by other proxies of accounting qualities that predict stock returns. These effects
are also economically meaningful. Based on our baseline result of Column (1), a one-
standard-deviation increase in ABLNAH is related to a 5.17% {=(0.030 0.398)/0.231}
Panel A: Descriptive statistics for cross-sectional regressions
Variables Obs. Mean Std dev P1 P5 P10 Q1 Median Q3 P90 P95 P99
abRET 13,859 0.231 0.580 0.602 0.395 0.291 0.124 0.091 0.412 0.901 1.338 2.604
TobinQ 13,859 1.185 0.890 0.416 0.547 0.627 0.767 0.966 1.306 1.909 2.459 4.557
ABLNAH 13,859 0.001 0.398 1.410 0.655 0.425 0.186 0.023 0.232 0.448 0.587 0.855
jPMDAj 13,859 0.105 0.104 0.001 0.006 0.013 0.033 0.074 0.142 0.235 0.315 0.503
jPMDA3yrj 13,859 0.330 0.215 0.045 0.084 0.113 0.177 0.279 0.424 0.617 0.766 1.074
REM3yr 13,859 0.007 0.159 0.576 0.265 0.162 0.057 0.018 0.093 0.179 0.241 0.367
ROA 13,859 0.023 0.098 0.361 0.168 0.082 0.002 0.033 0.073 0.118 0.150 0.226
C-SCORE 13,859 0.430 6.586 24.106 3.422 1.279 0.223 0.171 0.634 1.850 3.935 30.314
lnSIZE 13,859 11.379 1.502 8.844 9.420 9.756 10.345 11.112 12.080 13.378 14.554 16.070
lnBM 13,859 6.846 0.759 4.859 5.494 5.832 6.364 6.894 7.381 7.788 8.014 8.419
BETA 13,859 0.990 0.556 0.279 0.141 0.320 0.621 0.961 1.326 1.694 1.978 2.550
IDVOL 13,859 0.142 0.069 0.051 0.065 0.075 0.096 0.125 0.168 0.228 0.277 0.412
MOM 13,859 0.216 0.634 0.664 0.492 0.383 0.182 0.065 0.432 0.971 1.441 2.778
BIG4 13,859 0.544 0.498 0.000 0.000 0.000 0.000 1.000 1.000 1.000 1.000 1.000
lnBOD 13,859 0.798 0.513 0.000 0.000 0.000 0.693 0.693 1.099 1.386 1.609 1.946
lnSPBOA 13,859 0.402 0.430 0.000 0.000 0.000 0.000 0.693 0.693 1.099 1.099 1.386
RoutBOD 13,859 0.143 0.111 0.000 0.000 0.000 0.063 0.143 0.200 0.286 0.333 0.500
CEODuality 13,859 0.912 0.284 0.000 0.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
FOROWN 13,859 0.070 0.120 0.000 0.000 0.000 0.001 0.012 0.081 0.235 0.350 0.542
(continued)
matrix
Descriptive statistics
value of client
and correlation
Table 2.
Future equity
firm
MAJ
Table 2.
Panel B: Correlation matrix for cross-sectional regressions (Pearson/Spearman)
Variables (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19)
(1) abRET 0.353 0.007 0.035 0.061 0.067 0.050 0.021 0.0560.055 0.012 0.053 0.035 0.001 0.004 0.010 0.002 0.002 0.019
(2) TobinQ 0.221 0.027 0.091 0.130 0.104 0.104 0.041 0.278
0.797 0.095 0.317 0.187 0.021 0.025 0.094 0.048 0.107 0.083
(3) ABLNAH 0.029 0.016 0.003 0.016 0.020 0.012 0.000 0.031
0.015 0.018 0.042 0.005 0.014 0.026 0.038 0.012 0.009 0.013
(4) jPMDAj 0.022 0.043 0.005 0.574 0.064 0.028 0.011 0.081
0.103 0.072 0.165 0.046 0.036 0.065 0.015 0.018 0.046 0.061
(5) jPMDA3yrj 0.028 0.074 0.001 0.618 0.088 0.065 0.026 0.125
0.155 0.109 0.238 0.092 0.060 0.102 0.019 0.028 0.073 0.095
(6) REM3yr 0.047 0.057 0.021 0.057 0.066 0.295 0.050 0.1690.088 0.040 0.141 0.054 0.093 0.025 0.070 0.053 0.024 0.172
(7) ROA 0.005 0.037 0.005 0.160 0.174 0.260 0.075 0.261
0.131 0.072 0.241 0.280 0.065 0.011 0.033 0.044 0.018 0.236
(8) C-SCORE 0.017 0.049 0.006 0.001 0.003 0.014 0.006 0.1580.046 0.019 0.051 0.022 0.058 0.070 0.043 0.004 0.027 0.113
(9) lnSIZE 0.073 0.103 0.008 0.082 0.120 0.180 0.243 0.004 0.407 0.063 0.248 0.200 0.335 0.394 0.226 0.005 0.153 0.620
(10) lnBM 0.027 0.378 0.013 0.124 0.168 0.107 0.020 0.028 0.384 0.114 0.268 0.226 0.033 0.009 0.103 0.043 0.105 0.151
(11) BETA 0.055 0.033 0.010 0.069 0.088 0.051 0.056 0.025 0.047 0.110 0.273 0.006 0.013 0.039 0.006 0.085 0.006 0.004
(12) IDVOL 0.177 0.171 0.045 0.173 0.220 0.118 0.292 0.004 0.249 0.271 0.229 0.042 0.163 0.212 0.082 0.026 0.109 0.285
(13) MOM 0.008 0.123 0.012 0.014 0.045 0.042 0.197 0.004 0.152 0.306 0.058 0.178 0.002 0.025 0.007 0.000 0.029 0.052
(14) BIG4 0.017 0.025 0.000 0.035 0.057 0.092 0.079 0.000 0.350 0.024 0.007 0.161 0.018 0.194 0.131 0.017 0.098 0.259
(15) lnBOD 0.033 0.022 0.034 0.071 0.109 0.046 0.041 0.008 0.454 0.009 0.031 0.191 0.007 0.192 0.276 0.678 0.171 0.316
(16) lnSPBOA 0.012 0.042 0.042 0.016 0.022 0.081 0.021 0.003 0.293 0.101 0.007 0.085 0.007 0.143 0.296 0.090 0.049 0.162
(17) RoutBOD 0.010 0.012 0.003 0.012 0.026 0.007 0.031 0.004 0.019 0.026 0.076 0.032 0.014 0.003 0.681 0.124 0.010 0.002
(18) CEODuality 0.016 0.040 0.000 0.050 0.070 0.020 0.038 0.012 0.151 0.102 0.004 0.090 0.010 0.098 0.171 0.057 0.010 0.136
(19) FOROWN 0.027 0.030 0.002 0.050 0.075 0.196 0.208 0.009 0.540 0.121 0.066 0.231 0.009 0.244 0.301 0.177 0.002 0.103
Notes: All continuous variables are winsorised at extreme 1% in Panel A and Panel B. In Panel B, italic numbers indicate the significance at 1%. Refer to
variable definitions in Appendix 1
Pred. (1) (2) (3) (4) (5) (6)
Future equity
Variables Sign BASIC AEM OPAQUE REM ROA CONSERV value of client
ABLNAH þ 0.030** 0.029** 0.029** 0.026** 0.029** 0.030**
firm
(2.24) (2.21) (2.23) (2.00) (2.20) (2.25)
jPMDAj 0.220***
(4.28)
jPMDA3yrj 0.148***
(5.18)
REM3yr 0.321***
(8.48)
ROA þ 0.390***
(5.77)
C-SCORE þ 0.002**
(2.25)
lnSIZE 0.008 0.009 0.010* 0.010* 0.015*** 0.008
(1.44) (1.60) (1.74) (1.83) (2.62) (1.43)
lnBM þ 0.085*** 0.081*** 0.077*** 0.093*** 0.079*** 0.086***
(7.67) (7.29) (6.85) (8.31) (7.21) (7.69)
BETA þ 0.029*** 0.030*** 0.030*** 0.031*** 0.030*** 0.029***
(2.63) (2.70) (2.73) (2.80) (2.72) (2.58)
IDVOL þ 1.815*** 1.840*** 1.855*** 1.883*** 1.962*** 1.817***
(14.90) (15.17) (15.36) (15.53) (16.09) (14.92)
MOM þ 0.015 0.017* 0.021** 0.017* 0.031*** 0.015
(1.56) (1.77) (2.16) (1.75) (3.06) (1.58)
BIG4 þ 0.024** 0.024** 0.023** 0.020* 0.025** 0.024**
(2.27) (2.26) (2.20) (1.95) (2.44) (2.26)
lnBOD 6 0.008 0.008 0.008 0.017 0.023 0.008
(0.45) (0.46) (0.46) (1.01) (1.35) (0.44)
lnSPBOA 6 0.003 0.003 0.003 0.002 0.005 0.003
(0.24) (0.25) (0.26) (0.18) (0.39) (0.25)
RoutBOD 6 0.077 0.074 0.074 0.099 0.115* 0.076
(1.10) (1.07) (1.06) (1.44) (1.69) (1.09)
CEODuality 6 0.024 0.023 0.022 0.026 0.023 0.024
(1.52) (1.49) (1.42) (1.63) (1.44) (1.52)
FOROWN þ 0.163*** 0.161*** 0.157*** 0.116** 0.140*** 0.162***
(3.06) (3.02) (2.95) (2.19) (2.68) (3.03)
Constant 0.598*** 0.540*** 0.476*** 0.646*** 0.516*** 0.594***
(4.27) (3.87) (3.39) (4.61) (3.73) (4.25)
Year FE Included Included Included Included Included Included
Industry FE Included Included Included Included Included Included Table 3.
Observations 13,859 13,859 13,859 13,859 13,859 13,859 Cross-sectional
R-squared 0.057 0.058 0.059 0.063 0.060 0.057 regression results
between abnormal
Notes: This table reports cross-sectional regression results of one-year-ahead abnormal stock returns (abRET) on
abnormal audit hours (ABLNAH). Refer to Appendix 1 for variable definitions; *, ** and *** denote the significance at audit hours and one-
10%, 5% and 1% level, respectively; t-statistics in parentheses are corrected for robust standard errors clustering at the year-ahead monthly
firm level stock returns (H1)
increase in annual return performance, relative to its mean. Overall, these results support
H1.
The results of the control variables are consistent with our ex ante predictions. The
coefficients of accounting proxies are generally negative and significant, except for ROA.
Next, LnBM, IDVOL and BIG4 are positively significant, whereas MOM is negatively
significant.
MAJ 4.3 Audit supply-side sub-samples: Big4 auditor versus non-Big4 auditor (H2a)
In this section, we further investigate the relationship between abnormal audit efforts and
firms’ equity value from the perspective of the audit supply side. We expect that the positive
association between abnormal audit hours and future stock returns would be stronger for
the Big4 auditor sub-sample.
Panel A of Table 4 shows the regression results with the Big4 sub-samples. In every
model of Panel A in Table 4, the coefficient of the ABLNAH is positive. The coefficients
are all significant, except in Column (4). Column (1) shows that the effects yield a 5.25%
{=(0.031 0.376)/0.222} increase in annual return performance relative to its mean for a
one-standard-deviation increase in ABLNAH. However, in every specification of Panel B
in Table 4, the coefficient of ABLNAH is statistically insignificant. This result suggests
that the direct and indirect effects of audit quality on abnormal stock returns are limited
without sufficient audit effort. These results support H2a that the positive association
observed between abnormal audit hours and future equity value is stronger for firms
audited by Big N auditors. As previous studies document that clients hire Big 4 auditors
to give a positive signal to the market and to guarantee insurance coverage, the results in
Table 4 further suggest that the audit effort on the auditee’s equity value is partially
driven by auditors’ insurance and or information roles.
4.4 Audit demand side sub-samples

This section examines the relationship between abnormal audit efforts and future returns,
considering audit demand-side characteristics. Particularly, we focus on the effect of IFRS
adoption and accounting comparability. The indirect influence of abnormal audit effort is
expected to be marginal for firms with high accounting quality and have low demand for
high-quality audits.
4.4.1 Pre-IFRS period versus post-IFRS period (H2b). Panel A of Table 5 presents the
regression results with a pre-IFRS adoption sample. In every specification of Panel A in Table 5,
the coefficient of ABLNAH is significantly positive at 10% or higher level. The economic
significance is meaningful as a one-standard-deviation increase in ABLNAH amplifies the
annual return performance relative to its mean by 5.66% {=(0.032 0.430)/0.243}, as shown in
Column (1). However, for all models of Panel B in Table 5, the coefficient of ABLNAH is
statistically insignificant. These results are consistent with prior literature that mandatory IFRS
adoption enhances accounting transparency and decreases stock price crash (DeFond et al.,
2015; Landsman et al., 2012). Therefore, the results indicate that the indirect influence on firms’
future equity value would be marginal after the mandatory adoption of IFRS by improved
accounting quality through abnormal audit efforts. Overall, these results support H2b.
4.4.2 Lower accounting comparability versus higher accounting comparability (H2c).
Panel A of Table 6 presents the regression result with accounting comparability below the
median sample. In every specification of Panel A in Table 6, the coefficient ABLNAH is
significantly positive at the 5% level. The BASIC model of Column (1) shows that the
monthly return performance relative to its mean increases by 9.19% {=(0.054 0.395)/
0.232} for a one-standard-deviation increase in ABLNAH. However, all coefficients of
ABLNAH in Panel B of Table 6 are statistically insignificant. These results suggest that
abnormal audit hours’ effect in reducing information asymmetry would be marginal for
clients with high financial statement comparability. Overall, the results suggest that the
positive association between abnormal audit hours and future abnormal returns is weaker
for firms with high financial statement comparability, consistent with H2c.
Pred. (1) (2) (3) (4) (5) (6)
Variables Sign BASIC AEM OPAQUE REM ROA CONSERV
Panel A: Big4 sample

ABLNAH þ 0.031* 0.029* 0.029* 0.025 0.032* 0.031*
(1.80) (1.71) (1.69) (1.43) (1.88) (1.79)
jPMDAj 0.246***
(3.60)
jPMDA3yrj 0.129***
(3.38)
REM3yr 0.354***
(7.21)
ROA þ 0.394***
(4.41)
C-SCORE þ 0.001
(1.06)
lnSIZE 0.003 0.002 0.002 0.002 0.003 0.003
(0.40) (0.33) (0.25) (0.27) (0.40) (0.40)
lnBM þ 0.081*** 0.077*** 0.074*** 0.095*** 0.079*** 0.081***
(5.60) (5.34) (5.12) (6.45) (5.53) (5.60)
BETA þ 0.012 0.011 0.010 0.007 0.010 0.012
(0.79) (0.73) (0.71) (0.50) (0.70) (0.81)
IDVOL þ 1.727*** 1.764*** 1.766*** 1.800*** 1.858*** 1.728***
(9.46) (9.65) (9.62) (9.89) (9.94) (9.47)
MOM þ 0.010 0.008 0.005 0.010 0.003 0.010
(0.75) (0.60) (0.39) (0.70) (0.25) (0.73)
lnBOD 6 0.003 0.003 0.004 0.010 0.017 0.003
(0.16) (0.16) (0.19) (0.49) (0.84) (0.16)
lnSPBOA 6 0.007 0.006 0.006 0.000 0.008 0.007
(0.45) (0.42) (0.40) (0.01) (0.50) (0.45)
RoutBOD 6 0.019 0.012 0.016 0.037 0.054 0.019
(0.22) (0.14) (0.18) (0.45) (0.65) (0.22)
CEODuality 6 0.015 0.014 0.013 0.018 0.015 0.015
(0.80) (0.73) (0.68) (0.96) (0.82) (0.80)
FOROWN þ 0.105* 0.102* 0.098 0.054 0.078 0.104*
(1.72) (1.69) (1.63) (0.88) (1.31) (1.71)
(continued)
depending on Big4/
value of client
regression results
Cross-sectional
Table 4.
Future equity
non-Big4 sub-
stock returns
year-ahead monthly
firm
samples (H2a)
audit hours and one-
between abnormal
MAJ
Table 4.
Pred. (1) (2) (3) (4) (5) (6)
Constant 0.578*** 0.525*** 0.478*** 0.705*** 0.539*** 0.576***

(3.19) (2.91) (2.65) (3.84) (3.04) (3.18)
Industry FE Included Included Included Included Included Included
Observations 7,540 7,540 7,540 7,540 7,540 7,540
R-squared 0.052 0.054 0.054 0.060 0.055 0.052
Panel B: Non-Big4 sample
ABLNAH þ 0.024 0.024 0.026 0.024 0.021 0.025
(1.15) (1.16) (1.22) (1.11) (1.03) (1.17)
jPMDAj 0.196**
(2.53)
jPMDA3yrj 0.170***
(4.13)
REM3yr 0.279***
(4.95)
ROA þ 0.407***
(4.06)
C-SCORE þ 0.003*
(1.90)
lnSIZE 0.038*** 0.039*** 0.041*** 0.041*** 0.049*** 0.037***
(3.35) (3.49) (3.65) (3.64) (4.13) (3.34)
lnBM þ 0.091*** 0.086*** 0.080*** 0.092*** 0.079*** 0.092***
(5.31) (5.01) (4.62) (5.37) (4.57) (5.35)
BETA þ 0.071*** 0.071*** 0.072*** 0.070*** 0.070*** 0.070***
(4.19) (4.23) (4.26) (4.17) (4.22) (4.17)
IDVOL þ 1.877*** 1.894*** 1.918*** 1.936*** 2.032*** 1.881***
(10.65) (10.79) (11.00) (11.08) (11.57) (10.67)
MOM þ 0.035*** 0.037*** 0.041*** 0.038*** 0.052*** 0.035***
(2.63) (2.75) (3.06) (2.82) (3.79) (2.61)
lnBOD 6 0.010 0.010 0.013 0.003 0.005 0.010
(0.29) (0.29) (0.38) (0.10) (0.14) (0.30)
lnSPBOA 6 0.011 0.011 0.010 0.014 0.009 0.011
(0.57) (0.56) (0.52) (0.71) (0.44) (0.57)
(continued)
Pred. (1) (2) (3) (4) (5) (6)
RoutBOD 6 0.060 0.061 0.050 0.089 0.089 0.057

(0.46) (0.46) (0.38) (0.68) (0.68) (0.43)
CEODuality 6 0.028 0.028 0.027 0.031 0.023 0.026
(0.92) (0.94) (0.89) (0.98) (0.75) (0.87)
FOROWN þ 0.177* 0.176* 0.172* 0.145 0.157 0.176*
(1.79) (1.78) (1.73) (1.50) (1.61) (1.78)
Constant 0.399* 0.334 0.240 0.381 0.236 0.396*
(1.72) (1.43) (1.03) (1.63) (1.00) (1.71)
Observations 6,319 6,319 6,319 6,319 6,319 6,319
R-squared 0.079 0.080 0.082 0.083 0.083 0.080
Notes: This table reports cross-sectional regression results of one-year-ahead abnormal stock returns (abRET) on abnormal audit hours (ABLNAH). Panel A
shows the result with Big4 samples. Panel B shows the result with non-Big4 samples. Refer to Appendix 1 for variable definitions; *, ** and ***denote the
significance at 10%, 5% and 1% level, respectively; t-statistics in parentheses are corrected for robust standard errors clustering at the firm level
value of client
Table 4.
Future equity
firm
MAJ
Table 5.
stock returns
Cross-sectional
regression results
depending on pre-
between abnormal
sub-samples (H2b)
year-ahead monthly
IFRS and post-IFRS

Pred. (1) (2) (3) (4) (5) (6)
Panel A: Pre-IFRS sample (2002–2010)

ABLNAH þ 0.032** 0.031* 0.032** 0.029* 0.031* 0.032**
(1.97) (1.95) (2.00) (1.80) (1.95) (1.98)
jPMDAj 0.241***
(4.08)
jPMDA3yrj 0.163***
(4.95)
REM3yr 0.321***
(6.41)
ROA þ 0.394***
(4.73)
C-SCORE þ 0.001
(1.19)
lnSIZE 0.017** 0.016** 0.015** 0.014* 0.010 0.017**
(2.22) (2.13) (1.99) (1.88) (1.30) (2.22)
lnBM þ 0.116*** 0.111*** 0.105*** 0.123*** 0.109*** 0.116***
(8.22) (7.85) (7.42) (8.69) (7.87) (8.24)
BETA þ 0.038** 0.040** 0.040** 0.041** 0.037** 0.038**
(2.35) (2.43) (2.42) (2.51) (2.32) (2.33)
IDVOL þ 1.818*** 1.845*** 1.860*** 1.891*** 1.972*** 1.820***
(11.70) (11.91) (12.07) (12.21) (12.67) (11.70)
MOM þ 0.018 0.020* 0.025** 0.018 0.033*** 0.018
(1.47) (1.68) (2.09) (1.52) (2.70) (1.48)
BIG4 þ 0.033** 0.033** 0.032** 0.030** 0.035** 0.033**
(2.36) (2.35) (2.27) (2.16) (2.49) (2.36)
lnBOD 6 0.008 0.008 0.007 0.016 0.024 0.008
(0.35) (0.33) (0.32) (0.70) (1.02) (0.34)
lnSPBOA 6 0.005 0.005 0.005 0.001 0.008 0.005
(0.29) (0.27) (0.31) (0.04) (0.49) (0.28)
RoutBOD 6 0.067 0.062 0.060 0.080 0.105 0.065
(0.69) (0.64) (0.63) (0.84) (1.10) (0.67)
CEODuality 6 0.026 0.025 0.023 0.028 0.024 0.026
(1.24) (1.21) (1.12) (1.32) (1.17) (1.24)
(continued)
Pred. (1) (2) (3) (4) (5) (6)
FOROWN þ 0.063 0.060 0.056 0.020 0.036 0.062

(0.96) (0.92) (0.85) (0.31) (0.56) (0.95)
Constant 1.129*** 1.065*** 0.986*** 1.169*** 1.049*** 1.128***
(6.42) (6.07) (5.61) (6.64) (6.05) (6.41)
Observations 8,253 8,253 8,253 8,253 8,253 8,253
R-squared 0.059 0.061 0.063 0.065 0.063 0.060
Panel B: Post-IFRS sample (20122014)
ABLNAH þ 0.003 0.003 0.002 0.002 0.002 0.003
(0.12) (0.13) (0.09) (0.08) (0.09) (0.11)
jPMDAj 0.185
(1.44)
jPMDA3yrj 0.082
(1.30)
REM3yr 0.327***
(4.57)
ROA þ 0.292**
(2.27)
C-SCORE þ 0.006***
(2.83)
lnSIZE 0.052*** 0.053*** 0.053*** 0.054*** 0.058*** 0.051***
(5.27) (5.42) (5.40) (5.54) (5.79) (5.21)
lnBM þ 0.016 0.013 0.012 0.024 0.010 0.015
(0.84) (0.68) (0.64) (1.28) (0.56) (0.83)
BETA þ 0.027* 0.027* 0.027* 0.028* 0.029* 0.025
(1.66) (1.67) (1.69) (1.73) (1.77) (1.56)
IDVOL þ 2.193*** 2.215*** 2.215*** 2.243*** 2.294*** 2.190***
(8.19) (8.27) (8.23) (8.37) (8.28) (8.19)
MOM þ 0.055*** 0.056*** 0.057*** 0.059*** 0.067*** 0.054***
(3.04) (3.08) (3.12) (3.30) (3.56) (2.95)
(continued)
value of client
Table 5.
Future equity
firm
MAJ
Table 5.
Pred. (1) (2) (3) (4) (5) (6)
BIG4 þ 0.020 0.020 0.020 0.016 0.021 0.019

(1.05) (1.06) (1.06) (0.87) (1.12) (1.02)
lnBOD 6 0.029 0.030 0.030 0.041 0.041 0.030
(0.99) (1.04) (1.04) (1.45) (1.42) (1.03)
lnSPBOA 6 0.005 0.005 0.005 0.000 0.005 0.006
(0.26) (0.27) (0.25) (0.01) (0.24) (0.31)
RoutBOD 6 0.203* 0.203* 0.202* 0.231** 0.233** 0.206*
(1.82) (1.82) (1.81) (2.10) (2.09) (1.84)
CEODuality 6 0.021 0.021 0.021 0.022 0.020 0.021
(0.81) (0.80) (0.80) (0.83) (0.74) (0.80)
FOROWN þ 0.254*** 0.252*** 0.249*** 0.197** 0.242*** 0.249***
(2.93) (2.91) (2.87) (2.32) (2.81) (2.87)
Constant 0.353 0.395* 0.403* 0.316 0.437* 0.347
(1.57) (1.75) (1.78) (1.40) (1.96) (1.55)
Observations 4,300 4,300 4,300 4,300 4,300 4,300
R-squared 0.107 0.107 0.107 0.113 0.108 0.108
shows the result with pre-IFRS samples. Panel B shows the result with post-IFRS samples. Refer to Appendix 1 for variable definitions; *, ** and ***denote the
Pred. (1) (2) (3) (4) (5) (6)
Panel A: Lower accounting comparability sample

ABLNAH þ 0.054** 0.051** 0.053** 0.052** 0.049** 0.054**
(2.53) (2.42) (2.50) (2.47) (2.38) (2.55)
jPMDAj 0.338***
(4.20)
jPMDA3yrj 0.146***
(3.37)
REM3yr 0.223***
(3.98)
ROA þ 0.407***
(4.06)
C-SCORE þ 0.002*
(1.92)
lnSIZE 0.029*** 0.030*** 0.030*** 0.031*** 0.038*** 0.029***
(2.92) (3.06) (3.07) (3.11) (3.75) (2.90)
lnBM þ 0.097*** 0.089*** 0.087*** 0.097*** 0.083*** 0.097***
(5.39) (4.96) (4.80) (5.38) (4.59) (5.38)
BETA þ 0.041** 0.043** 0.042** 0.043** 0.041** 0.040**
(2.36) (2.45) (2.42) (2.45) (2.37) (2.27)
IDVOL þ 1.966*** 1.989*** 1.991*** 1.993*** 2.097*** 1.967***
(10.04) (10.21) (10.25) (10.14) (10.58) (10.05)
MOM þ 0.038*** 0.040*** 0.043*** 0.040*** 0.056*** 0.038***
(2.77) (3.00) (3.18) (3.00) (3.90) (2.80)
BIG4 þ 0.032* 0.032* 0.032* 0.030* 0.033* 0.032*
(1.84) (1.86) (1.83) (1.70) (1.93) (1.85)
lnBOD 6 0.024 0.021 0.020 0.029 0.039 0.024
(0.78) (0.69) (0.66) (0.94) (1.25) (0.76)
lnSPBOA 6 0.005 0.006 0.005 0.006 0.001 0.005
(0.24) (0.29) (0.27) (0.28) (0.04) (0.24)
RoutBOD 6 0.258** 0.242** 0.237* 0.272** 0.301** 0.256**
(2.09) (1.96) (1.91) (2.22) (2.49) (2.08)
CEODuality 6 0.002 0.001 0.000 0.001 0.002 0.001
(0.07) (0.05) (0.01) (0.02) (0.08) (0.05)
(continued)
accounting
value of client
depending on
regression results
Cross-sectional
Table 6.
Future equity
firm
high sub-samples
comparability low/
stock returns
year-ahead monthly
(H2c)
between abnormal
MAJ
Table 6.
Pred. (1) (2) (3) (4) (5) (6)
FOROWN þ 0.082 0.077 0.072 0.039 0.070 0.078

(0.90) (0.85) (0.80) (0.42) (0.80) (0.86)
Constant 0.495** 0.399* 0.372 0.474** 0.331 0.479**
(2.05) (1.67) (1.54) (1.97) (1.37) (1.98)
Observations 5,358 5,358 5,358 5,358 5,358 5,358
R-squared 0.077 0.080 0.079 0.080 0.081 0.078
Panel B: Higher accounting comparability sample
ABLNAH þ 0.018 0.017 0.017 0.012 0.019 0.019
(0.90) (0.85) (0.85) (0.59) (0.92) (0.93)
jPMDAj 0.175**
(2.14)
jPMDA3yrj 0.143***
(3.29)
REM3yr 0.366***
(5.74)
ROA þ 0.375**
(2.03)
C-SCORE þ 0.003**
(1.99)
lnSIZE 0.005 0.005 0.005 0.004 0.003 0.005
(0.62) (0.62) (0.55) (0.43) (0.29) (0.60)
lnBM þ 0.060*** 0.059*** 0.054*** 0.076*** 0.064*** 0.061***
(3.24) (3.14) (2.90) (3.97) (3.37) (3.28)
BETA þ 0.009 0.009 0.010 0.011 0.011 0.009
(0.48) (0.51) (0.56) (0.61) (0.61) (0.48)
IDVOL þ 2.370*** 2.385*** 2.389*** 2.431*** 2.440*** 2.376***
(10.23) (10.25) (10.31) (10.63) (10.44) (10.28)
MOM þ 0.008 0.008 0.012 0.007 0.016 0.007
(0.45) (0.49) (0.69) (0.44) (0.90) (0.41)
(continued)
Pred. (1) (2) (3) (4) (5) (6)
BIG4 þ 0.030* 0.030* 0.029* 0.027* 0.032* 0.030*

(1.83) (1.81) (1.78) (1.67) (1.95) (1.81)
lnBOD 6 0.034 0.034 0.032 0.027 0.026 0.034
(1.34) (1.34) (1.28) (1.08) (1.02) (1.36)
lnSPBOA 6 0.029 0.029 0.028 0.021 0.030* 0.029
(1.61) (1.59) (1.55) (1.21) (1.68) (1.63)
RoutBOD 6 0.068 0.072 0.066 0.054 0.042 0.070
(0.61) (0.64) (0.59) (0.49) (0.38) (0.63)
CEODuality 6 0.017 0.017 0.016 0.020 0.015 0.016
(0.79) (0.79) (0.73) (0.93) (0.69) (0.77)
FOROWN þ 0.157** 0.158** 0.156** 0.131* 0.133* 0.157**
(2.19) (2.20) (2.17) (1.85) (1.86) (2.18)
Constant 0.563** 0.539** 0.475** 0.694*** 0.589** 0.560**
(2.45) (2.34) (2.05) (2.97) (2.54) (2.43)
Observations 5,391 5,391 5,391 5,391 5,391 5,391
R-squared 0.072 0.073 0.074 0.080 0.074 0.073
(Panel B) shows the result with the below-median (above-median) comparability sample. Accounting comparability is measured as the average score of top four
firms within the two-digit SIC industry following De Franco et al. (2011). Refer to Appendix 1 for other variable definitions; *, ** and ***denote the significance
at 10%, 5% and 1% level, respectively; t-statistics in parentheses are corrected for robust standard errors clustering at the firm level
value of client
Table 6.
Future equity
firm
MAJ 4.5 Hedge portfolio analyses
In this section, we examine whether investors perceive abnormal audit hours when
making investment decisions. We perform additional analyses to test whether
abnormal profit could be achieved through risk-free arbitrage using abnormal audit
hours’ information, buying the high and selling the low of abnormal audit hours’
portfolio.
If investors recognize the positive influence of abnormal audit hours on future stock
returns and use the information in the decision-making process, the current stock price
would already reflect the abnormal audit hour effect, making abnormal returns
unlikely. However, if investors fail to understand the effect of abnormal audit hours on
increasing future equity value, the current stock price would not reflect this effect,
allowing future abnormal profit by forming a hedge portfolio using abnormal audit
hours.
Table 7 reports the results of the monthly time-series regressions of the abnormal
audit hour portfolios. The first part of Panel A in Table 7 reports the equal-weighted
average raw returns. The rest of Panel A in Table 7 presents the risk-adjusted returns
using the capital asset pricing model (CAPM) (Lintner, 1965; Sharpe, 1964), the Fama–
French three-factor model (Fama and French, 1993) and the Carhart four-factor model
(Carhart, 1997). The magnitude and the significance of each factor loadings are similar,
as shown in prior studies [13]. Overall, the decile of firms with the highest abnormal
audit hours (High) significantly outperforms the decile with the lowest abnormal audit
hours (Low). The High–Low hedge portfolio that takes a long position in high abnormal
audit hour firms and a short position in low abnormal audit hour firms earns an average
equal-weighted monthly return of 0.48% (t-stat: 2.51), significant at 5%. The excess
returns of CAPM a, three-factor a and four-factor a for the High–Low hedge portfolio
are 0.48%, 0.36% and 0.36%, respectively (t-stat: 2.45, 2.74 and 2.70) and significant at
5% or 1%. These economically and significantly positive hedge returns indicate that
investors do not fully reflect abnormal audit efforts information in their investment
decision.
4.6 Causal mediation analyses

Higher audit effort increases clients’ future equity return performance, perceived as
greater audit effort, directly enhancing clients’ firm value. Alternatively, it is argued
that intermediate steps, such as a lower likelihood of audit failure or better accounting
quality because of higher audit effort, indirectly increase equity value. This section
checks these causal chains by conducting causal mediation analyses following Hicks
and Tingley (2011).
Additionally, the analysis identifies the effects of auditors’ assurance and insurance
roles. We conjecture that greater audit effort enhances audit quality and results in equity
value creation regarding the assurance mechanism. We use audit quality as the mediator to
address the assurance mechanism. Regarding the insurance mechanism, the insurance value
of audit service declines as firms are sanctioned by the regulatory agency and are exposed to
higher litigation risk. Therefore, we use litigation risk as the mediator.
The association between audit hours and stock return performance is examined after
considering both mediators. In the first stage, the mediator is set as the dependent variable,
whereas the audit hour is set as an independent variable to estimate the mediation effects.
We examine the association between the estimated audit hours effects in the first-stage
regression and future stock return performance in the second stage.
ABLNAH (Low) (Mid) (Mid)
Decile 1 2 3 4 5 6 7
Average Raw Return

a (%) 1.79 1.97 2.15 2.17 2.14 2.11 2.17
(3.66) (3.83) (3.74) (3.88) (3.94) (3.91) (4.31)
CAPM: Rpt Rft ¼ a þ b ðRmt Rft Þ þ « pt
a (%) 0.84 1.00 1.18 1.21 1.17 1.14 1.21
(2.88) (3.03) (3.34) (1.21) (3.75) (3.43) (3.97)
Rm Rf 0.95 0.98 0.98 0.97 0.98 0.98 0.98
(15.03) (13.52) (12.87) (14.03) (17.87) (12.36) (14.25)
R2 0.592 0.630 0.627 0.619 0.603 0.625 0.605
3 Factor Model: Rpt Rft ¼ a þ b 1 ðRmt Rft Þ þ b 2 SMBt þ b 3 HMLt þ « pt
a (%) 0.60 0.54 0.87 0.77 0.79 0.66 0.81
(4.07) (3.99) (5.03) (3.66) (5.03) (5.58) (4.91)
Rm Rf 1.01 1.03 1.03 1.01 1.03 1.03 1.03
(40.29) (34.78) (25.93) (28.87) (37.87) (41.94) (32.01)
SMB 0.85 0.83 0.81 0.79 0.86 0.86 0.87
(27.32) (17.67) (27.57) (17.31) (19.79) (41.16) (25.63)
HML 0.03 0.20 0.10 0.19 0.14 0.21 0.14
(0.57) (3.63) (1.79) (2.40) (2.63) (4.12) (2.65)
R2 0.909 0.915 0.906 0.877 0.896 0.927 0.914
4 Factor Model: Rpt Rft ¼ a þ b 1 ðRmt Rft Þ þ b 2 SMBt þ b 3 HMLt þ b 4 UMDt þ « pt
a (%) 0.62 0.56 0.87 0.77 0.81 0.66 0.86
(4.50) (3.96) (4.90) (3.54) (5.26) (5.94) (5.23)
Rm Rf 1.01 1.03 1.03 1.01 1.03 1.03 1.03
(39.49) (35.32) (25.08) (28.53) (38.09) (42.41) (33.76)
SMB 0.85 0.83 0.81 0.79 0.85 0.86 0.87
(27.42) (19.12) (29.30) (17.30) (20.94) (45.39) (29.67)
HML 0.04 0.20 0.10 0.19 0.15 0.21 0.16
(0.62) (3.69) (1.66) (2.48) (2.68) (4.44) (2.57)
UMD 0.03 0.03 0.01 0.01 0.04 0.00 0.08
(0.55) (0.70) (0.13) (0.19) (1.26) (0.02) (2.94)
R2 0.909 0.915 0.906 0.877 0.897 0.927 0.916
Notes: This table represents the equal-weighted average (excess) return (in %) for abnormal audit hour portfolios from April 2003 to March 2016, with 156 monthly observations. Stocks are allocated
into decile groups of fiscal year t abnormal audit hour (ABLNAHt) at each April of year t þ 1. CAPM a is monthly excess return on market factor (Lintner, 1965; Sharpe, 1964). Three-factor a is
monthly excess return on market, size and value factor (Fama and French, 1993). Four-factor a is monthly excess return on market, size, value and momentum factor (Carhart, 1997). Dividend yields
are reflected in stock returns and assumed to be re-invested. Mid–Low indicates the difference of one-year-ahead monthly excess return performance between firms in the mean of 5th and 6th decile
(Mid) and in the bottom decile(Low) of ABLNAHt. High–Low indicates the difference of one-year-ahead monthly excess return performance between firms in the top decile(High) and in the bottom
decile(Low) of ABLNAHt; *, ** and ***denote the significance at 10%, 5% and 1% level, respectively; t-statistics in parentheses are corrected for Newey–West heteroskedasticity and autocorrelation
consistent standard error
(continued)
audit hour
One-year-ahead
value of client
Table 7.
Future equity
portfolio returns
firm
sorted on abnormal
MAJ
Table 7.
ABLNAH (High)
Decile 8 9 10 Mid–Low High–Mid High–Low
Average Raw Return

a (%) 2.14 2.05 2.28 0.33 0.15 0.48
(4.09) (3.9) (3.86) (2.69)*** (1.02) (2.51)**
CAPM: Rpt Rft ¼ a þ b ðRmt Rft Þ þ « pt
a (%) 1.17 1.06 1.32 0.31 0.16 0.48
(3.50) (3.98) (3.47) (2.53)** (1.12) (2.45)**
Rm Rf 0.97 1.01 0.96 0.03 0.02 0.01
(12.79) (13.84) (12.99) (1.46) (0.61) (0.48)
R2 0.617 0.676 0.587 0.007 0.003 0.001
3 Factor Model: Rpt Rft ¼ a þ b 1 ðRmt Rft Þ þ b 2 SMBt þ b 3 HMLt þ « pt
a (%) 0.91 0.80 0.97 0.12 0.25** 0.36***
(5.51) (5.36) (6.11) (1.13) (2.04) (2.74)
Rm Rf 1.03 1.06 1.02 0.02 0.01 0.01
(42.00) (38.73) (27.98) (1.50) (0.49) (0.36)
SMB 0.84 0.74 0.88 0.00 0.02 0.02
(24.24) (20.68) (15.18) (0.18) (0.47) (0.51)
HML 0.05 0.07 0.11 0.14*** 0.07 0.08
(1.41) (2.00) (1.07) (3.05) (0.90) (0.88)
R2 0.916 0.910 0.903 0.066 0.022 0.013
4 Factor Model: Rpt Rft ¼ a þ b 1 ðRmt Rft Þ þ b 2 SMBt þ b 3 HMLt þ b 4 UMDt þ « pt
a (%) 0.93 0.82 0.98 0.11 0.25** 0.36***
(5.37) (5.00) (6.31) (1.22) (2.03) (2.70)
Rm Rf 1.03 1.06 1.02 0.02 0.01 0.01
(44.26) (42.57) (29.70) (1.50) (0.51) (0.37)
SMB 0.83 0.74 0.88 0.00 0.02 0.02
(24.93) (21.94) (15.33) (0.21) (0.45) (0.50)
HML 0.06 0.07 0.11 0.14*** 0.06 0.08
(1.53) (2.24) (1.12) (3.17) (0.89) (0.97)
UMD 0.03 (0.92) 0.03 (0.60) 0.02 0.01 0.01 0.00
(0.50) (0.15) (0.22) (0.01)
R2 0.916 0.910 0.903 0.066 0.022 0.013
Panel A of Table 8 presents the result of causal mediation analysis for audit quality on the Future equity
relationship between audit hours and clients’ stock return performance in the subsequent year. value of client
We find that both audit hours and mediators for audit quality have a significant
association with one-year-ahead stock return performance (0.030, 0.325; t-stat: 2.26, 5.79,
firm
respectively). The mean value of the total audit hour effect is 0.033, whereas its direct effect
is 0.030. However, the indirect effect of audit effort is 0.003 and explains 9.12% of the total
audit hour effect on the client’s equity value creation.
Panel B presents the audit effort mechanism on the client’s stock return performance from
the audit service’s insurance value. In the first stage of logit regression, we set litigation risk as
the mediator, captured by whether the client was sanctioned by the Financial Supervisory
Service (equivalent to SEC). As shown in Panel B, clients experience negative stock return
performance as litigation risk increases with the likelihood of audit failure from lack of audit
Panel A: Mediation effects of audit quality (negative value of jPMDAj)

Variable Pred. sign Coefficient t-Stat.
(1st OLS) MediatorðAuditQÞ ¼ a þ b 1 lnAH þ Controls þ «
lnAH þ 0.010 *** (2.95)
(2nd OLS) abRET ðþ1Þ ¼ a þ b 1 lnAH þ b 2 Mediatorð AuditQÞ þ Controls þ «
lnAH þ 0.030 ** (2.26)
AuditQ þ 0.325 *** (5.79)
Observations 13,859 R-squared 0.054
Effect Mean 95% Confidence interval
Low High
Average casual mediation 0.033 0.008 0.060
effect
Direct effect of lnAH 0.030 0.005 0.056
Total effect of lnAH 0.003 0.001 0.006
Total effects mediated 9.12%
Panel B: Mediation effects of high litigation risk (sanctioned firm from the regulatory review)
Variable Pred. sign Coefficient t-[z-]
stat.
(1st Logit) MediatorðHighRiskÞ ¼ a þ b 1 lnAH þ Controls þ «
lnAH þ 0.625 ** [2.54]
(2nd OLS) abRET ðþ1Þ ¼ a þ b 1 lnAH þ b 2 MediatorðHighRiskÞ þ Controls þ «
lnAH þ 0.032 ** (2.03)
HighRisk 0.106 ** (2.14)
Observations 8,096 R-squared 0.053
Effect Mean 95% Confidence interval
Low High
Average casual mediation 0.043 0.008 0.082
effect
Direct effect of lnAH 0.032 0.002 0.063
Total effect of lnAH 0.011 0.000 0.029
Total effects mediated 25.80% Table 8.
Causal mediation
Notes: This table reports the casual mediation analyses of audit quality and litigation risk on the effects of audit
relationship between audit hours (lnAH) and one-year-ahead abnormal stock returns (abRET). Panel A and quality and litigation
Panel B show the average casual mediation effects of audit quality (AuditQ) and high litigation risk
risk on the
(HighRisk), respectively. AuditQ is measured as the negative value of jPMDAj. HighRisk is a dummy
variable which has value of 1 if the firm is regulatory reviewed with enforcement action, otherwise is 0. The relationship between
mediation effects are estimated following Hicks and Tingley (2011). Refer to Appendix 1 for other variable audit hours and one-
definitions: *, ** and ***denote the significance at 10%, 5% and 1% level, respectively; t-statistics (z- year-ahead abnormal
statistics) in parentheses (brackets) are corrected for robust standard errors clustering at the firm level returns
MAJ effort (0.106; t-stat: 2.14). The client’s future equity value declines as the insurance value of
audit service deteriorates. The insurance role explains 25.80% of the total audit hour’s effects
on the client’s future equity value. Overall, we find that both assurance and insurance roles
affect the association between audit hours and the client’s future equity value.
4.7 Robustness tests

We conduct a battery of tests to verify the robustness of our findings. Robustness test
results using regression and portfolio approaches, controlling for accrual anomaly, and
various other sensitivity test results are provided throughout this section.
4.7.1 Using alternative proxies of client firm value. In this section, we use Tobin’s Q as an
alternative proxy for future firm value. Although this study documents the role of audit
effort on market-return based firm value, Tobin’s Q is also a widely used proxy in the
literature. The results replacing the return to Tobin’s Q are shown in Table 9. Results using
the full sample are given in Panel A. Consistent with the previous findings, the coefficients
of ABLNAH are all positive and significant. Additionally, ABLNAH is positive and
significant in the Big4 sample (Panel B), pre-IFRS subsample (Panel D) and low accounting
comparability sample (Panel F). However, the coefficients of ABLNAH are all insignificant
in the other counterparts. Overall, the results using Tobin’s Q support our hypotheses.
4.7.2 Role of information intermediates. Prior research considers information intermediates
to control for the information role of auditors (Mansi et al., 2004). Following Mansi et al. (2004),
we include credit ratings and analyst followings in our regression model. The untabulated
results show that the coefficient of ABLNAH is still significant even after controlling for credit
ratings (lnRATING) and analyst followings (lnANALYST). As the model also controls
accounting quality proxies representing the assurance level, we suggest that the remaining
insurance role has explanatory power to increase clients’ stock return performance.
4.7.3 Using alternative proxies of audit efforts. For robustness checks, major regressions
are re-performed using abnormal audit fees (ABLNAF) as an alternative proxy for abnormal
audit effort. The normal audit fee model is set following prior studies (Choi et al., 2010;
Hribar et al., 2014). The untabulated results show that client equity value increases as
abnormal audit fees increase, consistent with previous findings.
We also use total audit hours (lnAH) as an alternative proxy for abnormal audit effort.
Investors may perceive total audit hours information in investment decision-making.
However, they may not fully understand the abnormal audit efforts because of the model’s
complexity. Therefore, to eliminate this concern, we reanalyze the main regressions using
the natural log of naïve audit hours (lnAH). The untabulated results are similar to the main
results. Total audit hours information is not perceived in the investment decision-making
process. However, as mentioned, there could be a potential endogeneity problem between
total audit hours and returns, requiring caution while inferring the results.
Next, we control audit fees and abnormal audit fees and check whether abnormal audit
hours still positively affect client equity. The untabulated results show that the coefficients
of ABLNAH are positive and significant in both abRET and TobinQ models, even after
controlling for audit fees and abnormal audit fees.
4.7.4 Other sensitivity checks. Although untabulated for brevity, we performed various
sensitivity analyses and reexamined the main findings; the results did not alter our inferences.
First, in an extreme performing firm, such ephemeral performance could be
misunderstood as a market anomaly (Kraft et al., 2006). Consequently, we remove
observations with a yearly buy-and-hold return greater than 300% and smaller than 50%.
The untabulated findings suggest that extreme firm performance may not drive the positive
association between abnormal audit hours and future abnormal returns. Second, we re-
Pred. (1) (2) (3) (4) (5) (6)
Panel A: All sample

ABLNAH þ 0.035** 0.035** 0.035** 0.034** 0.035** 0.036**
(2.16) (2.16) (2.14) (2.14) (2.17) (2.23)
Controls Included Included Included Included Included Included
Year and industry FE Included Included Included Included Included Included
Observations 13,859 13,859 13,859 13,859 13,859 13,859
R-squared 0.667 0.667 0.667 0.668 0.667 0.668
Panel B: Big4 sample
ABLNAH þ 0.052** 0.052** 0.051** 0.051** 0.051** 0.051**
(2.50) (2.50) (2.48) (2.48) (2.47) (2.49)
Observations 7,540 7,540 7,540 7,540 7,540 7,540
R-squared 0.789 0.789 0.789 0.789 0.789 0.789
Panel C: Non-Big4 sample
ABLNAH þ 0.018 0.018 0.018 0.018 0.019 0.021
(0.63) (0.62) (0.62) (0.63) (0.68) (0.74)
Observations 6,319 6,319 6,319 6,319 6,319 6,319
R-squared 0.645 0.645 0.645 0.645 0.645 0.646
Panel D: Pre-IFRS sample (2002–2010)
ABLNAH þ 0.044* 0.044* 0.043* 0.044* 0.045* 0.044*
(1.77) (1.76) (1.74) (1.76) (1.75) (1.78)
Observations 8,253 8,253 8,253 8,253 8,253 8,253
R-squared 0.639 0.639 0.640 0.639 0.639 0.641
Panel E: Post-IFRS sample (2012–2014)
ABLNAH þ 0.010 0.010 0.010 0.011 0.009 0.010
(0.27) (0.26) (0.25) (0.27) (0.23) (0.27)
(continued)
value of client
using Tobin’s Q
Robustness checks
Table 9.
Future equity
firm
MAJ
Table 9.
Pred. (1) (2) (3) (4) (5) (6)
Observations 4,300 4,300 4,300 4,300 4,300 4,300

R-squared 0.899 0.899 0.899 0.899 0.899 0.899
Panel F: Low accounting comparability sample
ABLNAH þ 0.066*** 0.066*** 0.066*** 0.066*** 0.069*** 0.066***
(3.05) (3.04) (3.07) (3.06) (3.16) (3.06)
Observations 5,358 5,358 5,358 5,358 5,358 5,358
R-squared 0.747 0.747 0.747 0.747 0.748 0.748
Panel G: High accounting comparability sample
ABLNAH þ 0.038 0.038 0.039 0.038 0.038 0.033
(1.19) (1.19) (1.21) (1.19) (1.19) (1.07)
Observations 5,391 5,391 5,391 5,391 5,391 5,391
R-squared 0.651 0.651 0.652 0.651 0.651 0.653
Notes: This table reports cross-sectional regression results of Tobin’s Q ratio (TobinQ) on abnormal audit hours (ABLNAH). TobinQ is the sum of market
capitalization and total liabilities divided by total assets (Tobin and Brainard 1977). Refer to Appendix 1 for other variable definitions; *, ** and ***denote the
perform hedge portfolio return performance in other forms, such as using a value-weighted Future equity
portfolio return, a portfolio by terciles, quartiles, quintiles and septiles and find similar value of client
returns. Finally, we alter the abnormal audit hour estimation model as mentioned in
equation (1) in Section 3, and the results remain consistent.
firm
5. Conclusion
This study analyzes the association between abnormal audit effort and firms’ future equity
value. We provide empirical evidence of additional audit effort, enhancing clients’ equity
value, especially by reducing the likelihood of audit failure and improving the quality of
financial reporting. Additionally, our study finds that audit effort information is not fully
reflected in the equity market, suggesting that a zero-cost investment strategy using
abnormal audit hours’ information is significantly effective. We also perform various
robustness tests: causal mediation analyzes using sanctioned firms and accounting quality,
alternative abnormal audit effort measurements, and alternative portfolio construction were
used and additionally controlled for information intermediates and audit fees. In all
additional robustness tests, the results were robust.
We expand the horizon of abnormal audit effort literature by documenting how auditors’
extra effort could benefit clients’ equity value by moderating possible audit failure and improving
financial reporting quality. Moreover, the results of this study highlight the benefits of additional
audit effort. Given that the audit hour information is publicly available in Korea and has an
economically meaningful impact on firms’ equity value, our study provides practical implications
for investors who do not fully perceive audit hour data while making investment decisions.
Further, our findings suggest that regulators and policymakers should contemplate the system
and regulation on disclosing audit hours information to be beneficial to market participants.
Despite various additional robustness tests, our study has the following limitations.
First, the potential endogeneity concern embedded in the audit hour model is not ruled out
completely. Second, it is difficult to isolate the role of audit service from others. Although all
three auditing attributes have a significant influence, we cannot compare the magnitude of
their impacts on client value creations. We suggest that future studies separate the role of
auditing and investigate the magnitude of each role in creating the client’s equity value.
Notes
1. It is because of limited data availability. For instance, as a proxy for audit effort, prior studies
have generally used the audit fee. However, using audit fee cannot disentangle audit effort from
the risk premia (DeFond and Zhang, 2014).
2. We use abnormal audit hour to incorporate auditors’ incremental steps on detecting and
correcting material misstatements. Also, using abnormal audit hour is one way to mitigate
endogeneity concern existing in the association between audit hour and client firm outputs. We
discuss this issue more in research design section.
3. It is theoretically and empirically difficult to distinguish one role from the others (Knechel et al.,
2013; Jacob et al., 2019). Therein, we take the following steps to investigate how traditional
auditors’ roles would affect the association between audit effort and clients’ stock return
performance. First, we control well-known accounting quality proxies, such as discretionary
accruals, real activity management and accounting conservatism, which capture the expected
assurance levels (Li et al., 2009). If abnormal audit hours are positively related to future stock
returns even after controlling for accounting quality, then we can assume that insurance and/or
information roles contribute to our findings. Second, we divide our sample into Big 4 and non-Big
4 auditor clients and test whether our results are similar in both subsamples. If we document
significant results only in the Big 4 clients’ subsample, we could conclude that auditors’
MAJ insurance and/or information roles contribute to clients’ equity values. Third, to isolate the effect
of assurance and insurance role, we conduct causal mediation analyses. Specifically, we examine
the two casual chains – the likelihood of audit failure and better accounting quality – which are
related to the expected insurance and assurance levels, respectively. Fourth, to control for the
information role, we include credit ratings and analyst followings in our regression model (Mansi
et al., 2004). If the results are still consistent even after controlling these variables, then we
suggest that audit hours’ assurance and/or insurance aspects are influencing clients’ equity
values.
4. Firms with low audit quality suffer from low accrual credibility (Francis et al., 1999), and therefore have
a low financial reporting quality. However, firms with high audit quality are related to high accrual
quality (Krishnan, 2003a); thus, information users could rely on accounting information in making more
accurate judgment. In other words, low quality of accounting information has a negative effect on stock
price because of its lack of credibility.
5. There are conflicting evidences regarding the effect of IFRS adoption on accounting information
quality. Atwood et al. (2011) state reported earnings from firms following US GAAP have higher
relation to future earnings and cash flows compared to the firms under IFRS. Doukakis (2014) uses 22
IFRS-adopted European countries and finds IFRS adoption does not have a significant association with
accrual-based earnings management nor real activity-based earnings management.
6. See Caramanis and Lennox (2008) for more issues of endogeneity in audit hour.
7. Industry is divided according to Korea Standard Industry Classification 2-digit (hereafter, K-SIC2).
8. We perform the annual pooled regression; however, the results are similar when we use month-
by-month Fama-MacBeth (1973) cross-sectional regression, and Newey–West robust standard
errors to correct for heteroskedasticity and serial-correlation (Newey and West, 1987).
9. To avoid a potential multi-collinearity problem among accounting quality variables, we do not
include them collectively.
10. Definitions of all the other variables are shown in Appendix 1.
11. Results remain similar when (1)–(5) firms are included in the sample.
12. These specifications are important for the contribution of the research. Even if abnormal audit effort and
stock return have a significant association, contribution of the study would be limited if the incremental
explanatory power is weak compared to the other proxy known as financial reporting quality.
13. The value of coefficient of momentum factor (UMD) is negative, inconsistent with result using US
sample. The different direction of results is because Korean stock market results in full reversal
of return even in short period of time (Lee and Ohk, 2015).
References
Asante-Appiah, B. (2020), “Does the severity of a client’s negative environmental, social and
governance reputation affect audit effort and audit quality?”, Journal of Accounting and Public
Policy, Vol. 39 No. 3.
Atwood, T.J., Drake, M.S., Myers, J.N. and Myers, L.A. (2011), “Do earnings reported under IFRS tell us
more about future earnings and cash flows?”, Journal of Accounting and Public Policy, Vol. 30
No. 2, pp. 103-121.
Becker, C.L., Defond, M.L., Jiambalvo, J. and Subramanyam, K.R. (1998), “The effect of audit quality on
earnings management”, Contemporary Accounting Research, Vol. 15 No. 1, pp. 1-24.
Black, B. and Kim, W. (2012), “The effect of board structure on firm value: a multiple
identification strategies approach using korean data”, Journal of Financial Economics,
Vol. 104 No. 1, pp. 203-226.
Blackwell, D.W., Noland, T.R. and Winters, D.B. (1998), “The value of auditor assurance: Evidence from Future equity
loan pricing”, Journal of Accounting Research, Vol. 36 No. 1, pp. 57-70.
value of client
Bronson, S.N., Ghosh, A. and Hogan, C.E. (2017), “Audit fee differential, audit effort, and
litigation risk: an examination of ADR firms”, Contemporary Accounting Research, Vol. 34 firm
No. 1, pp. 83-117.
Callen, J.L. and Fang, X. (2017), “Crash risk and the auditor–client relationship”, Contemporary
Accounting Research, Vol. 34 No. 3, pp. 1715-1750.
Caramanis, C. and Lennox, C. (2008), “Audit effort and earnings management”, Journal of Accounting
and Economics, Vol. 45 No. 1, pp. 116-138.
Carhart, M.M. (1997), “On persistence in mutual fund performance”, The Journal of Finance, Vol. 52
No. 1, pp. 57-82.
Chang, X., Gygax, A.F., Oon, E. and Zhang, H.F. (2008), “Audit quality, auditor compensation and
initial public offering underpricing”, Accounting and Finance, Vol. 48 No. 3, pp. 391-416.
Chen, H., Chen, J.Z., Lobo, G.J. and Wang, Y. (2011), “Effects of audit quality on earnings management
and cost of equity capital: evidence from China”, Contemporary Accounting Research, Vol. 28 No.
3, pp. 892-925.
Cho, M., Hyeon, J., Jung, T. and Lee, W.J. (2019), “Audit pricing of hard-to-read annual reports”, Asia-
Pacific Journal of Accounting and Economics, pp. 1-26, doi: 10.1080/16081625.2019.1600418.
Choi, J.H. and Wong, T.J. (2007), “Auditors’ governance functions and legal environments: an
international investigation”, Contemporary Accounting Research, Vol. 24 No. 1, pp. 13-46.
Choi, J.H., Kim, J.B. and Zang, Y. (2010), “Do abnormally high audit fees impair audit quality?”,
AUDITING: A Journal of Practice and Theory, Vol. 29 No. 2, pp. 115-140.
Cohen, D.A. and Zarowin, P. (2010), “Accrual-based and real earnings management activities around
seasoned equity offerings”, Journal of Accounting and Economics, Vol. 50 No. 1, pp. 2-19, doi:
10.1016/j.jacceco.2010.01.002.
Craswell, A.T. and Francis, J.R. (1999), “Pricing initial audit engagements: a test of competing theories”,
The Accounting Review, Vol. 74 No. 2, pp. 201-216.
Davidson, R.A. and Neu, D. (1993), “A note on the association between audit firm size and audit
quality”, Contemporary Accounting Research, Vol. 9 No. 2, pp. 479-488.
DeAngelo, L.E. (1981a), “Auditor size and audit fees”, Journal of Accounting and Economics, Vol. 3
No. 3, pp. 183-199.
DeAngelo, L.E. (1981b), “Auditor independence, ‘low balling’, and disclosure regulation”, Journal of
Accounting and Economics, Vol. 3 No. 2, pp. 113-127.
DeFond, M. and Zhang, J. (2014), “A review of archival auditing research”, Journal of Accounting and
Economics, Vol. 58 No. 2/3, pp. 275-326, doi: 10.1016/j.jacceco.2014.09.002.
DeFond, M.L., Hann, R.N. and Hu, X. (2005), “Does the market value financial expertise on audit
committees of boards of directors?”, Journal of Accounting Research, Vol. 43 No. 2, pp. 153-193.
DeFond, M.L., Hung, M., Li, S. and Li, Y. (2015), “Does mandatory IFRS adoption affect crash risk?”,
De Franco, G., Kothari, S.P. and Verdi, R.S. (2011), “The benefits of financial statement comparability”,
Journal of Accounting Research, , Vol. 49 No. 4, pp. 895-931.
Doukakis, L.C. (2014), “The effect of mandatory IFRS adoption on real and accrual-based earnings
management activities”, Journal of Accounting and Public Policy, Vol. 33 No. 6, pp. 551-572.
Elsevier.
Dye, R.A. (1993), “Auditing standards, legal liability, and auditor wealth”, Journal of Political Economy,
Vol. 101 No. 5, pp. 887-914.
Fama, E.F. and French, K.R. (1993), “Common risk factors in the returns on stocks and bonds”, Journal
of Financial Economics, Vol. 33 No. 1, pp. 3-56.
MAJ Fama, E.F. and MacBeth, J.D. (1973), “Risk, return, and equilibrium: empirical tests”, Journal of Political
Economy, Vol. 81 No. 3, pp. 607-636, doi: 10.1086/260061.
Francis, J.R. (1984), “The effect of audit firm size on audit prices: a study of the australian market”,
Journal of Accounting and Economics, Vol. 6 No. 2, pp. 133-151.
Francis, J.R. and Krishnan, J. (1999), “Accounting accruals and auditor reporting conservatism”,
Contemporary Accounting Research, Vol. 16 No. 1, pp. 135-165.
Francis, J.R. and Simon, D. (1987), “A test of auditing pricing in the small-client segment of the US audit
market”, The Accounting Review, Vol. 62 No. 1, pp. 145-157.
Francis, J.R., Maydew, E.L. and Sparks, H.C. (1999), “The role of big 6 auditors in the credible reporting
of accruals”, AUDITING: A Journal of Practice and Theory, Vol. 18 No. 2, pp. 17-34.
Francis, J., LaFond, R., Olsson, P. and Schipper, K. (2005), “The market pricing of accruals quality”,
Frankel, R.M., Johnson, M.F. and Nelson, K.K. (2002), “The relation between auditors’ fees for nonaudit
services and earnings management”, The Accounting Review, Vol. 77 No. S1, pp. 71-105.
Ghosh, A. and Moon, D. (2005), “Auditor tenure and perceptions of audit quality”, The Accounting
Review, Vol. 80 No. 2, pp. 585-612.
Hogan, C.E. and Wilkins, M.S. (2008), “Evidence on the audit risk model: do auditors increase audit fees
in the presence of internal control deficiencies?”, Contemporary Accounting Research, Vol. 25
No. 1, pp. 219-242.
Hicks, R. and Tingley, D. (2011), “Causal mediation analysis”, The Stata Journal: Promoting
Communications on Statistics and Stata, Vol. 11 No. 4, pp. 605-619.
Hribar, P., Kravet, T. and Wilson, R. (2014), “A new measure of accounting quality”, Review of
Accounting Studies, Vol. 19 No. 1, pp. 506-538.
Hutton, A.P., Marcus, A.J. and Tehranian, H. (2009), “Opaque financial reports, R2, and crash risk”,
Journal of Financial Economics, Vol. 94 No. 1, pp. 67-86, doi: 10.1016/j.jfineco.2008.10.003.
Jacob, J., Desai, N. and Agarwalla, S.K. (2019), “An examination of factors driving big 4 audit fee
premiums: evidence from india’s audit market”, Accounting Horizons, Vol. 33 No. 2, pp. 43-58.
Jiang, J., Wang, I.Y., P. and Wang, K. (2019), “Big N auditors and audit quality: new evidence from
quasi-experiments”, The Accounting Review, Vol. 94 No. 1, pp. 205-227.
Kausar, A., Taffler, R.J. and Tan, C. (2009), “The going-concern market anomaly”, Journal of Accounting
Research, Vol. 47 No. 1, pp. 213-239.
Khan, M. and Watts, R.L. (2009), “Estimation and empirical properties of a firm-year measure of
accounting conservatism”, Journal of Accounting and Economics, Vol. 48 Nos 2/3, pp. 132-150,
doi: 10.1016/j.jacceco.2009.08.002.
Khurana, I.K., Raman, K.K. (2004), “Litigation risk and the financial reporting credibility of big 4 . . .”,
Kothari, S.P., Leone, A.J. and Wasley, C.E. (2005), “Performance matched discretionary accrual
measures”, Journal of Accounting and Economics, Vol. 39 No. 1, pp. 163-197, doi: 10.1016/j.
jacceco.2004.11.002.
Kraft, A., Leone, A.J. and Wasley, C. (2006), “An analysis of the theories and explanations offered for
the mispricing of accruals and accrual components”, Journal of Accounting Research, Vol. 44
No. 2, pp. 297-339.
Krishnan, G.V. (2003a), “Audit quality and the pricing of discretionary accruals”, AUDITING: A
Journal of Practice and Theory, Vol. 22 No. 1, pp. 109-126.
Krishnan, G.V. (2003b), “Does big 6 auditor industry expertise constrain earnings management?”,
Accounting Horizons, Vol. 17 No. s-1, pp. 1-16. No. SUPPL.,
Lambert, R., Leuz, C. and Verrecchia, R.E. (2007), “Accounting information, disclosure, and the cost of
capital”, Journal of Accounting Research, Vol. 45 No. 2, pp. 385-420.
Landsman, W.R., Maydew, E.L. and Thornock, J.R. (2012), “The information content of annual earnings Future equity
announcements and mandatory adoption of IFRS”, Journal of Accounting and Economics,
Vol. 53 No. 1-2, pp. 34-54. Nos value of client
Lee, M. and Ohk, K.Y. (2015), “Market anomalies and multifactor models: comparison between firm
the FF model and the CNZ model”, Journal of Korean Securities Association, Vol. 44 No. 5,
pp. 855-885.
Lennox, C. and Pittman, J.A. (2010), “Big five audits and accounting fraud”, Contemporary Accounting
Research, Vol. 27 No. 1, pp. 209-247.
Li, Y., Stokes, D.J., Taylor, S.L. and Wong, L. (2009), “Audit quality, accounting attributes and the cost
of equity Capital”, Working PaperOctober 2, available at: https://papers.ssrn.com/sol3/papers.
cfm?abstract_id=1481823
Lintner, J. (1965), “Security prices, risk, and maximal gains from diversification”, The Journal of
Finance, Vol. 20 No. 4, pp. 587-615.
Mansi, S.A., Maxwell, W.F. and Miller, D.P. (2004), “Does auditor quality and tenure matter to investors?
Evidence from the bond market”, Journal of Accounting Research, Vol. 42 No. 4, pp. 755-793.
Myers, J.N., Myers, L.A. and Omer, T.C. (2003), “Exploring the term of the auditor-client relationship
and the quality of earnings: a case for mandatory auditor rotation?”, The Accounting Review,
Vol. 78 No. 3, pp. 779-799.
Newey, W.K. and West, K.D. (1987), “Hypothesis testing with efficient method of moments estimation”,
International Economic Review, Vol. 28 No. 3, pp. 777
O’Keefe, T.B., Simunic, D. A. and Stein, M.T. (1994), “The production of audit services: evidence from a
major public accounting firm”, Journal of Accounting Research, Vol. 32 No. 2, pp. 241-261.
Palmrose, Z.-V. (1989), “The relation of audit contract type to audit fees and hours”, The Accounting
Review, Vol. 64 No. 3, pp. 488-499.
Pittman, J.A. and Fortin, S. (2004), “Auditor choice and the cost of debt capital for newly public firms”,
Reynolds, J.K. and Francis, J.R. (2000), “Does size matter? The influence of large clients on office-level
auditor reporting decisions”, Journal of Accounting and Economics, Vol. 30 No. 3, pp. 375-400.
Roychowdhury, S. (2006), “Earnings management through real activities manipulation”, Journal of
Accounting and Economics, Vol. 42 No. 3, pp. 335-370, doi: 10.1016/j.jacceco.2006.01.002.
Sharpe, W.F. (1964), “Capital asset prices: a theory of market equilibrium under conditions of risk”, The
Journal of Finance, Vol. 19 No. 3, pp. 425-442.
Taffler, R.J., Lu, J. and Kausar, A. (2004), “In denial? Stock market underreaction to going-concern audit
report disclosures”, Journal of Accounting and Economics, Vol. 38 Nos 1/3, pp. 263-296.
Tobin, J. and Brainard, W.C. (1977), “Asset markets and the cost of capital, economic progress private
values and public policies: essays in honor of William Fellner”.
Wallace, W.A. (2004), “The economic role of the audit in free and regulated markets: a look back and a
look forward”, Research in Accounting Regulation, Vol. 17, pp. 267-298.
Corresponding author
Seung Uk Choi can be contacted at: suchoi@khu.ac.kr
MAJ Appendix 1
Dependent variable: stock return performance

abRET = One-year-ahead yearly stock returns adjusted by the returns from size and book-to-
market benchmark portfolios. The return period starts on the first day of April after the
fiscal year-end
TobinQ = The sum of market capitalization and total liabilities divided by total assets
Interest variable: abnormal audit hours
ABLNAH = Abnormal audit hours measured as the difference between natural log of actual audit
hours and estimated normal audit hours
Control variables (1): Accounting quality characteristics
jPMDAj = Absolute value of performance-matched discretionary accruals. Discretionary accruals
are measured based on modified Jones model and matching is conducted with peer
firm’s ROA performance on the industry-year basis (Kothari et al., 2005)
jPMDA3yrj = Opaqueness in financial reporting measured as sum of jPMDAj for recent three years
(Hutton et al., 2009)
REM3yr = Real earnings management activities for recent three years calculated by the mean of
abnormal cash flows in operation and discretionary expenses (Roychowdhury, 2006;
Cohen and Zarowin, 2010)
ROA = Net income divided by beginning total assets;
C-SCORE = Firm-specific conservatism score (Khan and Watts, 2009)
Control variables (2): firm-specific characteristics
lnSIZE = Natural log of the beginning market capitalization
lnBM = Natural log of the beginning book-to-market ratio
BETA = Systematic risk estimated by the market model using monthly stock returns over at
least 24-month rolling windows up to 36-month rolling windows
IDVOL = Idiosyncratic volatility measured as the standard deviation of residuals from the
regressions of BETA estimation
MOM = Stock return performance past six months ending one month prior the portfolio
formulation
BIG4 = Dummy variable which has value of 1 if the auditor is one of the Big4 accounting firms,
otherwise 0
lnBOD = Natural log of the total number of board directors
lnSPBOA = Natural log of the total number of audit committee members with accounting and legal
expertise
RoutBOD = Ratio of outside directors to total directors
Table A1. CEODuality = Dummy variable which has value of 1 if the CEO also serves as board chair, otherwise 0
Variable definitions FOROWN = Percentage of equity ownership held by foreign investors
Appendix 2 Future equity
value of client
firm
Variables Pred. sign Coefficient t-statistic
Intercept þ/2 26.243*** (232.66)

lnTAt þ 0.104*** (4.12)
LEVt þ 20.033* (22.15)
BMt 2 20.014* (22.15)
ISSUEt þ 0.011 (0.80)
BIG4t þ 0.255*** (5.85)
LARGEOWNt þ/2 20.103*** (27.89)
FOROWNt þ/2 0.012 (0.30)
CONt þ 20.080 (21.52)
CONFMt þ 0.113*** (4.57)
INVRECt þ 20.017 (20.90)
ROAt 2 20.122 (21.49)
CFOTAt þ 0.056** (2.53)
LOSSt þ 0.006 (0.19)
OPNt þ 20.250 (21.42)
SPVISEt þ 20.011 (20.33)
FIRSTt þ 0.030 (1.23)
KOSPIt þ 0.037*** (5.67)
lnAFt þ 0.561*** (23.94)
Industry Dummies Included

R-squared 0.696
Adjusted R-squared 0.672
Firm-years 15,054
Notes: This table presents yearly regression results of audit hour model during the fiscal year of 2002–
2014:
lnAHit ¼ b 0 þ b 1 lnTAit þ b 2 LEVit þ b 3 BMit þ b 4 ISSUEit þ b 5 BIG4it
þ b 6 LARGEOWNit þ b 7 FOROWNit þ b 8 CONit þ b 9 CONFM þ b 10 INVRECit
þ b 11 ROAit þ b 12 CFOTAit þ b 13 LOSSit þ b 14 OPNit þ b 15 SPVISEit þ
b 16 FIRSTit þ b 17 KOSPIit þ b 18 lnAFit þ RIND þ « it
We define the abnormal audit hour(ABLNAHt) as the difference between natural log of actual audit hour(lnAHt)
and estimated normal audit hour. Variable definitions are as follows: lnAH is the natural log of total audit hour;
lnTA is the natural log of total asset; LEV is the total debt divided by total assets; BM is the book value of equity
divided by market value of equity; ISSUE takes 1 if the firm has issued long-term debt or equity capital, otherwise
0; BIG4 takes 1 if the auditor is one of the Big4 accounting firms, otherwise 0; LARGEOWN is the percentage of
equity ownership held by the largest shareholders; FOROWN is the percentage of equity ownership held by
foreign investors; CON takes 1 if the firm has consolidated subsidiaries, otherwise 0; CONFM is the number of
consolidated subsidiaries; INVREC is the sum of inventory and accounting receivable divided by beginning total
assets; ROA is the net income divided by beginning total assets; CFOTA is the operating cash-flow divided by
beginning total assets; LOSS takes 1 if the firm reports loss, otherwise 0; OPN takes 1 if the firm has non-
unqualified audit opinion, otherwise 0; SPVISE takes 1 if the firm has issues for administration, otherwise 0;
FIRST takes 1 if the firm is audited for the first time, otherwise 0; KOSPI takes 1 if the firm is listed in KOSPI, 0 if
listed in KOSDAQ; lnAF is the natural log of total audit fee; *, ** and ***denote the significance at 10%, 5% and Table A2.
1% level, respectively; t-statistics in parentheses are corrected for Newey–West heteroskedasticity and Estimating abnormal
autocorrelation consistent standard error audit hour

Do Audit e Fforts Increase The Future Equity Value of Client Firm?

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

Do Audit e Fforts Increase The Future Equity Value of Client Firm?

Uploaded by

Copyright:

Available Formats

The current issue and full text archive of this journal is available on Emerald Insight at:

2. Prior literature and hypothesis development

3.2 Cross-sectional regression model

Sample selection criteria Firm-years

4.4 Audit demand side sub-samples

Panel A: Big4 sample

Constant 0.578*** 0.525*** 0.478*** 0.705*** 0.539*** 0.576***

RoutBOD 6 0.060 0.061 0.050 0.089 0.089 0.057

IFRS and post-IFRS

Panel A: Pre-IFRS sample (2002–2010)

FOROWN þ 0.063 0.060 0.056 0.020 0.036 0.062

BIG4 þ 0.020 0.020 0.020 0.016 0.021 0.019

Panel A: Lower accounting comparability sample

FOROWN þ 0.082 0.077 0.072 0.039 0.070 0.078

BIG4 þ 0.030* 0.030* 0.029* 0.027* 0.032* 0.030*

4.6 Causal mediation analyses

Average Raw Return

Average Raw Return

Panel A: Mediation effects of audit quality (negative value of jPMDAj)

4.7 Robustness tests

Panel A: All sample

Observations 4,300 4,300 4,300 4,300 4,300 4,300

Dependent variable: stock return performance

Intercept þ/2 26.243*** (232.66)

Industry Dummies Included

You might also like

Constant 0.578* 0.525* 0.478* 0.705* 0.539* 0.576*