Contemporary Accounting Research

Vol. 26 No. 2 (Summer 2009) pp. 393–402 © CAAA
doi:10.1506/car.26.2.3

Discussion of “Mandatory Audit Partner Rotation,
Audit Quality, and Market Perception:
Evidence from Taiwan”*

E. MICHAEL BAMBER,

University of Georgia

LINDA SMITH BAMBER,

University of Georgia

1. Introduction

Chi, Huang, Liao, and Xie (2009) examine the effect of mandatory audit partner
rotation on audit quality and the market’s perception of audit quality, measured
using abnormal accruals and earnings response coefficients (ERCs). The effect of
audit partner rotation on audit quality is an important question. Audit partner rota-
tion is costly for auditing firms, especially with the increased frequency of rotation
mandated by the Sarbanes-Oxley Act. Moreover, as DeFond and Francis (2005,
26) note, “there is a greater need than ever for objective scientific evidence to guide
public policy-making in auditing now that it is explicitly controlled by a govern-
ment agency [the Public Company Accounting Oversight Board] rather than the
active product of competitive market forces”.
To date, there has been little empirical evidence on the costs or benefits of

Canadian Institute of
Chartered Accountants

, the

Certified General Accountants of Ontario

, the

Certified Man-
agement Accountants of Ontario

, and the

Institute of Chartered Accountants of Ontario

. We
thank Michael Willenborg (associate editor) for giving us the opportunity to discuss this paper.
We have benefited from helpful comments by Orie Barron, Jeremy Griffin, and Isabel Yanyan
Wang.

394 Contemporary Accounting Research

CAR

Vol. 26 No. 2 (Summer 2009)

will lead to higher-quality audits. As Chi et al. (2009) recognize, however, auditor
rotation has costs as well as benefits. Given the loss of client-specific knowledge
that potentially impairs the effectiveness and quality of the audit, there is a real


Auditor rotation can be accomplished by changing either the audit firm or the audit
partner. Most of the evidence on the effects of auditor rotation on audit quality is
based on audit firm rotation, although to date regulators have yet to require audit
firm rotation. Instead, regulators in the United States, United Kingdom, Taiwan,
Australia, and many other countries around the globe effectively require audit part-
ner rotation.
Because audit firm rotation is not mandatory, extant research explores the
effect of voluntary audit firm (non)rotation, by documenting the relation between
audit firm tenure and measures of audit quality. The general result from this litera-
ture is that longer audit firm tenure is associated with higher-quality financial
reporting. For example, Johnson, Khurana, and Reynolds (2002) and Myers,
Myers, and Omer (2003) conclude that longer audit firm tenure constrains extreme
absolute abnormal accruals. Mansi, Maxwell, and Miller (2004) find that longer
audit firm tenure is associated with higher bond ratings and a lower cost of debt,
and Ghosh and Moon (2005) find that longer audit firm tenure is associated with
greater value-relevance of reported earnings and higher Standard & Poors’ ratings
of the client’s shares. Using a field-based analysis, Bamber and Iyer (2007) find
longer audit firm tenure mitigates acquiescence to the client’s preferences.

2

These
results are consistent with critics’ arguments that the costs of audit firm rotation (in

Discussion of “Mandatory Audit Partner Rotation and Audit Quality” 395

CAR

Vol. 26 No. 2 (Summer 2009)

effects, increasing the power of the empirical tests becomes more important.
Chi et al. (2009) add to the emerging literature on the effects of audit partner
rotation. This literature is more limited than the audit firm rotation literature
because of limited availability of data on the identity of lead audit partners running
specific engagements. Australia and Taiwan are exceptions where the lead part-
ner(s) can be identified. On the basis of an analysis of a sample of clients in 1995
before audit partner rotation became mandatory in Australia, Carey and Simnett
(2006) conclude that audit quality suffers when the audit partner’s tenure exceeds
seven years. Specifically, when the partner’s tenure exceeds seven years, finan-
cially distressed clients are less likely to receive a going-concern qualification, and
there is also some evidence that the client is less likely to report earnings that just
miss breakeven. (They also conclude that this deterioration in audit quality is
attributable to non–Big 6 auditors.) However, Carey and Simnett (2006) find no
evidence that signed or absolute abnormal accruals are associated with audit partner
tenure. In contrast, Chen et al. (2008) conclude that audit quality improves with
audit partner tenure. In the 1990 – 2001 period, before audit partner rotation
became mandatory in Taiwan, Chen et al. (2008) conclude that both absolute
performance-matched abnormal accruals and income-increasing performance-
matched abnormal accruals decrease with audit partner tenure.

4

396 Contemporary Accounting Research

CAR

Vol. 26 No. 2 (Summer 2009)

Chi et al. (2009) contribute by examining the effect of



p-

value of 0.05 or better. Lindsay (1994) provides
evidence that editorial “bias against the null” occurs in the accounting discipline.
Bamber, Christensen, and Gaver (2000, 124) further argue that in combination
with the bias against publishing replications (which is more extreme in the field of
accounting than in hard sciences, where replication is the norm), editorial bias
against the null “can lead to a situation where the first published studies are more
likely to reject the null, and these initial studies have a disproportionate effect on
subsequent research due to the bias against publishing replications”. For these rea-
sons, we argue that it is vitally important to publish “no-results” studies.
That said, it is equally important for researchers to make the case that the
study’s empirical tests are powerful enough to detect an economically material
effect, should one exist. Above, we have explained why any effect of audit partner
rotation is likely to be modest. Detecting an effect of modest size requires powerful
tests. Two inescapable features of the Chi et al. 2009 setting likely reduce the
power of the study’s tests.
First, focusing on the effects of

mandatory

audit partner rotation (which is,
after all, the auditor rotation question of greatest practical import) necessarily limits
variation in the audit partner tenure variable. Chen et al. (2008) consider tenure of
5 years or less to be short, and tenure exceeding 10 years to be long. Carey and

Discussion of “Mandatory Audit Partner Rotation and Audit Quality” 397

CAR

Audit quality is unobservable. Most of the literature draws inferences about audit
quality based on traditional, noisy measures of earnings quality. Although the two
are related, they are by no means isomorphic. A high-integrity client can produce
high-quality earnings whether or not the audit is of high quality. Conversely, even a
high-quality audit cannot be expected to identify or adjust all of a client’s low-quality
reporting choices, so a high-quality audit does not guarantee high-quality earnings.

6

Two of the most popular measures of “audit quality” are abnormal accruals
(e.g., Johnson et al. 2002; Myers et al. 2003; Carey and Simnett 2006; Chen et al.
2008), and ERCs (e.g., Ghosh and Moon 2005). However, the popularity of such
measures does not mean that they are good proxies for earnings quality, much less
audit quality. To their credit, Chi et al. (2009) forthrightly acknowledge some of
the limitations of these proxies. In hopes of motivating future researchers to develop
sharper and more refined proxies for audit quality, we expand on their discussion.
Abnormal accruals are residuals from models of nondiscretionary (or normal)
accruals. The explanatory power of these models rarely exceeds 30–35 percent.

Is
it plausible that most of the variation in aggregate accruals is due to inappropriate
earnings management that auditors could have detected, but failed to eliminate?

That is the implication of using abnormal accruals to evaluate audit quality. Given
the modest explanatory power of these models and other well-known concerns
reviewed in McNichols 2000, it is not surprising that Jones, Krishnan, and Melendrez
(2008) find that most measures of abnormal accruals (including performance-
matched abnormal accruals such as those in Chi et al. 2009) do not have any
incremental explanatory power (beyond total accruals) in explaining extremely


Chi et al. (2009) identify a pressing question of real practical import: Does manda-
tory audit partner rotation improve audit quality? Because audit partners are not
identified in North American audit reports, the authors capitalize on available
evidence from Taiwan. This is an excellent use of international data to address a ques-
tion of real importance in North America, but one that we cannot answer in a North
American context. The empirical work is consistent with current standard method-
ology. Although we take issue with the paper’s proxies for audit quality, this concern
applies to the voluminous stream of archival research on audit quality, of which
Chi et al. 2009 is simply one example.
Turning to directions for future research, both the audit firm rotation literature
and the audit partner rotation literature focus on the net effects of rotation, rather
than separately identifying the costs and benefits of rotation. Future research could
contribute by providing more insight into the specific costs and benefits. Because
rotation creates significant costs, it would be useful to know, for example, whether
and to what extent audit partner rotation leads to any measurable benefits at all.
With the benefit of hindsight, it seems apparent that the magnitude of any
effects of mandatory audit partner rotation would be modest. Particularly given the
limited cross-sectional variation in audit partner tenure and the noisy proxies for
audit quality, it is not surprising that the study does not find much empirical evi-
dence of an effect.
Future researchers who find themselves with a project where the results fail to
reject the null would do well to heed Greenwald’s 1975 advice on gracefully failing
to reject the null hypothesis. For example, he discusses the importance of provid-
ing convincing evidence that the empirical proxies have construct validity, and he
also suggests providing evidence on the size of an effect that the empirical analysis
would be able to detect. The idea is to demonstrate that the study fails to reject the

Discussion of “Mandatory Audit Partner Rotation and Audit Quality” 399

CAR

tional features of the audit environment. For example, rather than focusing on
aggregate accruals, researchers could focus on specific accruals that (a) are eco-
nomically significant, (b) are not well explained in accompanying disclosures,
(c) are susceptible to manipulation, and (d) whose manipulation auditors are likely
to be able to detect. Examples include loan loss provisions of financial institutions
and insurers’ loss reserves. Field studies such as the Nelson, Elliott, and Tarpley
2002 analysis of managers’ decisions about how to attempt earnings management
and auditors’ decisions whether to respond by requiring adjustments to the finan-
cial statements may be helpful in identifying specific powerful contexts.
DeAngelo (1981, 186) defines the quality of audit services as the “market
assessed joint probability that a given auditor will

both

(a) discover a breach in the
client’s accounting system, and (b) report the breach”. Researchers might incorporate
factors related to auditors’ ability to detect questionable reporting practices, such
as auditor industry specialization, auditor or audit firm experience, industry-specific
characteristics, client complexity, or audit firm alumni in client firm top manage-
ment. Factors related to auditors’ willingness to report or correct a questionable
reporting practice, such as the importance of the client, the effect of the practice on

400 Contemporary Accounting Research

CAR

Vol. 26 No. 2 (Summer 2009)

reaching an earnings target, the quality of the client’s corporate governance, or
other measures of independence, should also affect audit quality.

audit partner must remain rotated off for five years. In their empirical analysis, Chi et
al. (2009) appropriately exclude clients where one audit partner rotated off in 2003 and
rotated back on in 2004. The proportion of clients with this type of superficial rotation
is about 10 percent of their preliminary sample, which is, however, much lower than
the proportion Chen et al. (2008) report for 2003 and 2004. It appears that the
voluntary rotation benchmark group could include superficial rotation, if audit firms
engaged in superficial partner rotation in anticipation of the requirements for
mandatory rotation.
4. Chen et al. (2008, 419) recognize that the differences between their inferences and
Carey and Simnett’s 2006 inferences could be due to differences in research design
(e.g., 1-year versus 12-year sample periods) or to institutional differences between
Australia and Taiwan. Chen et al. also find that longer audit firm tenure is associated
with higher earnings quality, similar to results found in studies using U.S. data. Carey
and Simnett (2006) do not investigate the relation between audit firm tenure and
earnings quality. However, Chen et al. (2008) argue that because audit partner tenure is
strongly associated with audit firm tenure, Carey and Simnett’s (2006) evidence that
longer audit partner tenure is associated with lower quality earnings suggests that

Discussion of “Mandatory Audit Partner Rotation and Audit Quality” 401

CAR

Vol. 26 No. 2 (Summer 2009)
longer audit firm tenure may also be associated with lower quality earnings in
Australia, contrary to results based on U.S. data.
5. Part of the reason for the modest difference in audit partner tenure arises because each
engagement has two audit partners. Because the data do not distinguish the lead
partner, the authors take the average of the two partners’ tenure. As a result, partner
tenure can actually be shorter in the benchmark samples where rotation is not required
or where there was voluntary rotation. For example, client A has two partners who have

Barron, O., O. Kim, S. Lim, and D. Stevens. 1998. Using analysts’ forecasts to measure
properties of analysts’ information environment.

The Accounting Review

73 (4): 421–33.
Botosan, C. 1997. Disclosure level and the cost of equity capital.

The Accounting Review72 (3): 323–49.
Carey, P., and R. Simnett. 2006. Audit partner tenure and audit quality.

The Accounting
Review

81 (3): 653–76.
Chen, C., C. Lin, and Y. Lin. 2008. Audit partner tenure, audit firm tenure, and discretionary
accruals: Does long auditor tenure impair earnings quality?

Contemporary Accounting
Research

25 (2): 415–45.
Chi, W., H. Huang, Y. Liao, and H. Xie. 2009. Mandatory audit partner rotation, audit
quality, and market perception: Evidence from Taiwan.

Contemporary Accounting
Research

of Practice & Theory

24 (Supplement): 5–30.
Deis, D., and G. Giroux. 1992. Determinants of audit quality in the public sector.

The
Accounting Review

67 (3): 462–79.
Ghosh, A., and D. Moon. 2005. Auditor tenure and perceptions of audit quality.

The
Accounting Review

80 (2): 585–612.
Greenwald, A. 1975. Consequences of prejudice against the null hypothesis.

Psychological
Bulletin

82 (1): 1–20.
Johnson, V., I. Khurana, and J. Reynolds. 2002. Audit-firm tenure and the quality of
financial reports.

Contemporary Accounting Research

19 (4): 637–60.
Jones, K., G. Krishnan, and K. Melendrez. 2008. Do models of discretionary accruals detect
actual cases of fraudulent and restated earnings? An empirical analysis.


investors? Evidence from the bond market.

Journal of Accounting Research

42 (4):
755–93.
Matsumoto, D. 2002. Management’s incentives to avoid negative earnings surprises.

The
Accounting Review

77 (3): 483–514.
McNichols, M. 2000. Research design issues in earnings management studies.

Journal of
Accounting and Public Policy

19 (4–5): 313–45.
Myers, J., L. Myers, and T. Omer. 2003. Exploring the term of the auditor-client relationship
and the quality of earnings: A case for mandatory auditor rotation?

The Accounting
Review

78 (3): 779–99.
Nelson, M., J. Elliott, and R. Tarpley. 2002. Evidence from auditors about managers’ and
auditors’ earnings management decisions.

The Accounting Review