Auditor Size and the Cost of Equity Capital over Auditor Tenure

• DOI: http://doi.org/10.1111/ijau.12071
• Author: J. Kenneth Reynolds,David B. Bryan
• Date: 01 November 2016
• Published date: 01 November 2016

Auditor Size and the Cost of Equity Capital over Auditor Tenure

David B. Bryan

1

and J. Kenneth Reynolds

2

1

The University of North Fl orida, United States

2

Florida State University, UnitedStates

This study examines whether the difference in the cost of equity capital between clients of large and small auditors

(‘the cost of equity capital disparity’) varies with auditor tenure. We find that the cost of equity capital disparity

increases over auditor tenure, suggesting a decline in the perceived audit quality of small auditors relative to large

auditors as tenure lengthens. The results suggest that this occurs because there is a deterioration in perceived audit

quality over tenure for small auditors combined with no change in perceived audit quality over tenure for large

auditors. Furthermore, we find that the cost of equity capital disparity does not increase over auditor tenure for

industry specialist auditors, implying that auditor specialization may act as a substitute for auditor size in reducing

investor concerns about audit quality declining over tenure for small auditors.

Key words: Audit quality, auditor tenure, auditor industry specialization, cost of equity capital

INTRODUCTION

Enhancing the credibility of financial reports is an

important function of high quality audits (Simunic & Stein,

1987; Slovin, Sushka & Hudson, 1990; Datar, Feltham &

Hughes, 1991; Khurana & Raman, 2004, 2006; Boone,

Khurana & Raman, 2008). In fact, regulators argue that

the perception of high audit quality is critical for ensuring

investor faith in the audit profession (e.g., Levitt, 2000).

For example, Arthur Levitt, the former C hairman of the

Securities and Exchange Commission, noted ‘It is not

enough that the accountant on an engagement act

independently. For investors to have confidence in the

quality of the audit the public must perceive the accountant

as independent’(Levitt, 2000). In this study, we integrate

the lines of literature that examine whether auditor tenure

and auditor size are associated with perceptions of audit

quality. Specifically, we investigate whether the difference

in perceived audit quality between large and small auditors

changes over auditor tenure. Furthermore, we also

incorporate auditor industry specialization by examining

whether specialization affects how the difference in

perceived audit quality between large and small auditors

changes over tenure.

Audit quality can affect investor perceptions of

information risk because higher quality audits provide

more rigorous verification of financial information, and

thus stronger monitoring (Khurana & Raman, 2004, 2006;

Fan & Wong, 2005; Boone et al., 2008; Choi & Lee, 2014).

As audit quality increases, uncertainty surrounding the

quality of the reported financial information decreases,

implying a lower degree of information risk and a lower

cost of equity capital (e.g., Khurana & Raman, 2004, 2006;

Boone et al., 2008). Accordingly, prior research commonly

utilizes the cost of equity capital as a proxy for investor

perception of audit quality (e.g., Khurana & Raman, 2004,

2006; Boone et al., 2008; Boone, Khurana & Raman, 2010;

Cassell et al., 2013). Using the cost of equity capital as a

measure of perceived audit quality, this study examines

whether the difference in the costof equity capital between

clients of large and small auditors (‘the cost of equity

capital disparity’) varies with auditor tenure.

Regulators have often expressed concern that longer

auditor tenure can lead to lower audit quality because of

the potential for impaired auditor independence (e.g.,

Securities and Exchange Commission (SEC), 1994;

Government Accountability Office (GAO), 2003; Public

Company Accounting Oversight Board (PCAOB), 2011).

On the other hand, information asymmetry between the

auditor and client is reduced as tenure lengthens,

potentially leading to higher quality audits. We argue that

the associationbetween auditor tenure and perceivedaudit

quality is likely to vary based on auditor size due to the

effects of auditor independence, learning, and audit

technology. Using a sample from the period 2003–2010,

we find that the cost of equity capital disparity increases

over auditor tenure, suggesting a deterioration in

perceived audit quality for small auditors relative to large

auditors as tenure lengthens. Further analysis implies that

this occurs because there is a decline in perceived audit

quality over tenure for small auditors combined with no

change in perceived audit quality over tenure for large

auditors. We also find that the cost of equity capital

disparity does not increase over tenure for industry

specialist auditors, consistent with the notion that

specialization may act as a substitute for auditor size in

attenuating investor concerns about audit quality

deteriorating over tenure for small auditors.

This study should be of interest to regulators, investors,

and academics. First, regulators should be interested in

this study because the results contribute to the debate

about mandatory auditor rotation. Regulators have

expressed concerns about audit quality declining over

tenure (e.g., Securities and Exchange Commission (SEC),

1994) and one proposed solution is to require auditor

rotation (e.g., Government Accountability Office (GAO),

2003). In fact, in 2011 the PCAOB issued Concept Release

2011-006, which solicits comments on mandatory audit

firm rotation (Public Company Accounting Oversight

Board (PCAOB), 2011). Our results imply that auditor

rotation would only improve perceptions of audit quality

for small auditors who are not industry specialists,

suggesting that mandatory auditor rotation would impose

a broad cost on capital market participants, with the

benefits only being realized by firms that are audited by

small non-specialist auditors. Hence, we add to the

growing body of literature cautioning regulators that

mandatory auditor rotation may be counterproductive.

Correspondenceto: David B. Bryan, Departmentof Accounting & Finance,

University of North Florida, 1 UNF Drive, Jacksonville, FL 32224, USA.

Email: david.bryan@unf.edu

International Journal of Auditing doi: 10.1111/ijau.12071

Int. J. Audit. 20:278–294 (2016)

©2016 John Wiley& Sons Ltd ISSN 1090-6738

Investors should be interested in our study because our

results indicate that the interaction of auditor size and

auditor tenure affects information risk and the cost of

equity capital. We show that there is only a positive

association between auditor tenure and the cost of equity

capital for firms audited by a small auditor. Therefore,

investors concerned about changes in information risk

attributable to auditor tenure should consider auditor

size when making assessments of information risk.

Furthermore, we also show that the positive association

between auditor tenure and the cost of equity capital

for firms audited by small auditors is eliminated for

industry specialist auditors. Thus, investors that are

concerned about changes in information risk over auditor

tenure should also consider whether the auditor is an

industry specialist when assessing information risk.

Lastly, this study should alsobe of interest to academics.

A considerable amount of prior research has investigated

perceptions of audit quality. Our study contributes to the

streams of research that examine whether auditor size

and auditor tenure are associated with perceived audit

quality. We also contribute to the auditor specialization

literatureby finding that the cost of equity capital disparity

does not increase over tenure for industry specialist

auditors, implying that specialization may act as a

substitute for auditor size in reducing investor concerns

about audit quality declining over tenure for small

auditors. At the international level, academics should be

interested in this study because many countries are either

considering mandatory auditor rotation or already require

it. For example, mandatory auditor rotation has been

imposed in Brazil, Italy, Singapore, South Korea, and

Indonesia. The European Union also recently approved

mandatory auditorrotation. Although our study is limited

to a sample from theUnited States, the results suggestthat

auditor rotation will only improve perceptions of audit

quality for small auditors who are not industry specialists,

which calls into question the wisdom of requiring auditor

rotation. Future research can explore whether these results

generalize to other nations.

BACKGROUND AND PRIOR LITERATURE

Auditor size and audit quality

Prior research has found that large auditors are associated

with higher quality audits than small auditors. For

example, prior research finds that large auditors are

associated with higher earnings quality (e.g., Becker et al.,

1998; Francis, Maydew & Sparks, 1999; Krishnan, 2003;

Carver, Hollingsworth & Stanley, 2011; DeFond, Erkens &

Zhang, 2014; Comprix & Huang, 2015), a lower chance of

fraud (e.g., Lennox& Pittman, 2010), a lower likelihood of

restatements (e.g., DeFond et al. 2014; Eshleman & Guo,

2014),a greaterpropensity to issue a going-concernopinion

(e.g., Chan & Wu, 2011; DeFond et al., 2014), and more

accurate analyst forecasts (e.g., Behn, Choi & Kang, 2008).

Prior research also suggests that large auditors are

perceived as providinghigher quality audits. For example,

Khurana and Raman (2004) find that large auditors are

associated with a lower cost of equity capital than small

auditors, and Teoh and Wong (1993) find that large

auditors are associated with a higher earnings response

coefficient compared to small auditors. Also, Mansi,

Maxwell and Miller (2004) as well as Pittman and Fortin

(2004) find that large auditors are associated with a lower

cost of debt than small auditors.

1

In sum, prior research

generally indicates that (1) large auditors provide higher

audit quality than small auditors and (2) perceptions of

audit quality are reflective of this difference.

2

Auditor tenure and audit quality

Existing research frequently investigates the association

between auditor tenure and audit quality using abnormal

accruals to proxy for audit quality. Johnson, Khurana and

Reynolds (2002) and Myers, Myers and Omer (2003) find

a negative association between auditor tenure and

abnormal accruals. Relatedly, Gul, Fung and Jaggi (2009)

find that earnings quality, measured using abnormal

accruals, is higher in short tenure for clients of industry

specialist auditors than it is for clients of non-specialists,

while Lim and Tan (2010) find that accrua ls quality is

higher over tenure for clients of specialist auditors

compared to clients of non-specialists. Examining partner-

level data, Chen, Lin and Lin (2008) and Chi et al. (2009)

find a negative association between tenure and abnormal

accruals, and that thoseabnormal accruals are even higher

in the year after a change in partners.

3

Other common proxies for audit quality used in testing

the tenure association include earnings benchmarks (e.g.,

Carey & Simnett, 2006; Davis, Soo & Trompeter, 2009),

conservatism (e.g., Jenkins & Velury, 2008; Li, 2010),

restatements (e.g., Stanley & DeZoort, 2007), fraud (e.g.,

Carcello & Nagy, 2004), and the propensity to issue a

going-concern opinion (e.g., Geiger & Raghunandan,

2002; Carey & Simnett, 2006). Results from this research

are more mixed than the abnormal accrual results.

4

More directlyrelated to our study,prior research has also

examined therelation between auditortenure and perceived

audit quality. Ghosh and Moon (2005) find that with

increases in tenure, the earnings response coefficient

increases, earnings have a greater effect on stock rankings,

and earnings have a greater effect on one-year-ahead

analyst forecasts of earnings. Boone et al.(2008)findthat

auditor tenure andthe cost of equity capital are negatively

related, suggesting that investors view tenure favorably.

However, the authors also findthat perceived audit quality

declines after approximately 13years, as reflected in an

increasing cost of equity capital (Boone et al., 2008). Mansi

et al. (2004) find that both auditor tenure and auditor size

are negatively associated with the cost of debt. However,

they do not examinethe interaction between auditortenure

and auditor size. Dao, Mishra and Raghunandan (2008)

find a positive association between votes against auditor

ratification and auditor tenure.

Azizkhani, Monroe and Shailer (2013) use a sample of

Australian firms to investigate the association between

the cost of equity capital and both audit firm and audit

partner tenure. With respect to audit firm tenure, the

authors find a nonlinear association between auditor

tenure and the costof equity capital for non-Big 4 auditors,

wherebyperceptions of audit qualityinitially improve over

tenure and then begin to deteriorate after approximately

10.7years (Azizkhani et al., 2013). The authors find no

association between audit firm tenure and the cost of

equity capital for Big 4 auditors. Our study differs from

Azizkhani et al. (2013) in several ways. First, Azizkhani

et al. (2013) utilizeAustralian data and examinea relatively

small sampleof observations during a time periodthat was

primarilybefore the accounting scandalsin the early 2000s.

We utilize a much larger sample of observations from the

United States and our sample period extends from 2003 to

2010. Second, Azizkhani et al. (2013) find that perceptions

Auditor Sizeand the Cost of Equity Capital overAuditor Tenure 279

©2016 John Wiley & Sons Ltd Int. J. Audit. 20:278–294 (2016)