Audit Quality Effects of an Individual Audit Engagement Partner Signature Mandate

• Author: Allen D. Blay,Adrian Valencia,Caren Schelleman,Matthew Notbohm
• Date: 01 November 2014
• DOI: http://doi.org/10.1111/ijau.12022
• Published date: 01 November 2014

Audit Quality Effects of an Individual Audit Engagement Partner

Signature Mandate

Allen D. Blay,1Matthew Notbohm,2Caren Schelleman3and Adrian Valencia4

1Florida State University

2University of North Dakota

3Maastricht University

4Florida Gulf Coast University

The European Union recently mandated that audit engagement partners sign their name to the audit opinion.

However, there is little evidence demonstrating whether such a mandate had any influence on audit quality. We

examine the audit quality effects of a partner signature mandate by comparing multiple measures of audit quality

in years after adoption of an audit partner signature mandate in the Netherlands, a country with no tradition of

partner signatures, to the audit quality in years prior to adoption, as well as to the United Kingdom, a country with

a similar legal and accounting environment, but which did not adopt partner-level signatures at the same time. We

detect no substantial change in audit quality using multiple measures and sample selection criteria. Although a

null result cannot demonstrate that no effect exists, power tests suggest the lack of an economically significant

effect. Our results support the arguments of practitioners who believe that mandatory partner signatures will not

influence audit quality.

Key words: Audit quality, mandatory partner-level signatures, accruals, regulation

INTRODUCTION

Over the past decade, many countries (Australia, Taiwan,

and China), in addition to the countries of the European

Union (EU), have mandated that individual audit

engagement partners must sign their name to the audit

opinion for publicly traded companies (Advisory

Committee on the Auditing Profession, 2008; Chen,Lin &

Lin, 2008; Fargher, Lee & Mande, 2008; Su, 2010; Gold

et al., 2012). The primary argument made by regulators

for a mandatory partner signature is the suggestion that

audit quality would improve if partners are required to

sign the audit opinion. The argument is based on the

assumption of increased accountability to individual

audit partners (Institute of Chartered Accountants in

England and Wales, 2005), without any additional legal

liability. Extant research in both accounting and

psychology indicates that accountability reduces decision

biases (Kennedy, 1993; Brazel, Agoglia & Hatfield, 2004),

leads to increases in an individual’s conformity with

established behavior standards (Pryor et al., 1977; Diener,

1979; Silvia, 2002), and increases individualresponsibility

for performance (Schlenker et al., 1994). Experimental

studies also find that a personal signature increases

honesty and integrity (Davidson & Stevens, 2013), and

can lead to increased moral reasoning and lower

misreporting of private information (Blay et al. , 2013).

However, it is unclear whether an engagement

partner’s signature would actually increase audit quality

in practice. First, an assumption that the engagement

partner’s signature would increase accountability

inherently assumes that current accountability levels are

insufficient (Bierstaker et al., 2009). In addition, economic

predictions of audit quality effects rely on an increase in

legal liability (e.g., Basu, 1997).1Furthermore, practitioner

auditors also argue that there is no reason to believe that

audit quality would improve as a result of an auditor’s

signature. Deloitte strongly stated in a comment letter,

when referring to a partner signature proposal in the US,

‘We do not believe this requirement has any close nexus

to audit quality’ (Deloitte, 2008). Ernst & Young (2009),

KPMG (2009), and PricewaterhouseCoopers (2009),

among many other international firms, also expressed

objections to the proposed rule. Some members of the

working party of the Institute of Chartered Accountants

in England and Wales suggested that such a requirement

would provide no different incentives for partners to

keep audit quality high (Institute of Chartered

Accountants in England and Wales, 2005). Therefore, the

audit quality effect of a mandatory partner signature

policy is an empirical question.

The effects of mandatory adoption of an engagement

partner signature requirement are difficult to analyze for

several reasons. First, many countries (e.g., Australia)

had a tradition of partner-level signatures prior to

enacting a mandatory policy. In addition, other

countries adopted mandatory partner signing in

conjunction with other broad changes that make

inferences difficult (e.g., adoption of International

Financial Reporting Standards, IFRS). Finally, because

the adoption was mandatory in all EU member states,

substantially all firms adopt at the same time and there

are no early or late adoption control groups, making

empirical comparison unreliable.

We attempt to gain some insight into the audit quality

effects of the EU’s policy of mandatory partner-level

signatures by using the novel approach of a multi-year

analysis of audit quality in two countries with

reasonably similar accounting and legal environments,

but where the timing of mandatory partner-level

signature adoption is different, and IFRS adoption was

not uniform. Substantially all firms in the Netherlands

adopted partner-level signatures for fiscal years ending

on or after December 31, 2005. Prior to adoption, there

was no history of partner-level signatures in the

Netherlands.2In addition, this adoption date was

several years prior to other EU countries with no prior

Correspondence to: Allen D. Blay, Florida State University, P.O. Box

3061110, Tallahassee, FL 32306-1110, USA. Email: ablay@cob.fsu.edu

International Journal of Auditing doi:10.1111/ijau.12022

Int. J. Audit. 18: 172–192 (2014)

© 2014 John Wiley & Sons Ltd ISSN 1090-6738

history of partner-level signatures. Thus, we attempt to

control for country-specific effects using a pre- and

post-period within-country analysis, and we control for

the systematic effects of other concurrent shocks using

UK firms as a control sample (Francis, Nanda & Wang,

2006).3

To test the audit quality effects of this policy

implementation we use several estimates of client

earnings properties (abnormal accruals, magnitude of

accruals relative to cash flows, and earnings benchmark

tests). We include extensive controls for other country,

firm, auditor and year-specific factors that could

influence our findings. We predict that if a partner-level

signature improves audit quality, we will detect an

increase in audit quality in the Netherlands in the years

a firm’s audit report contains an engagement partner

signature, when compared to earlier periods or when

compared to the UK in the same time period (when

partner-level signatures were not yet required). This

finding would provide support for the psychological

explanation set forth by standards setters. Alternatively,

failing to document inter-temporal and/or cross-

country audit quality differences would provide some

support for the economic argument set forth by

practitioners.

We find no evidence that audit quality in the

Netherlands is improved by a mandatory engagement

partner-level signature requirement relative to both

prior years and a control sample of UK firms. Although

we acknowledge that a null finding is a joint conclusion

about the research question and the power of the test,

these lack of findings are robust to multiple models of

accruals, and to two measures of earnings benchmarks,

as well as to implementation-year tests, an extended

sample of firms, extended sample periods, matched pair

testing, and auditor/firm characteristic partitioning.

Further, we provide some evidence that our tests would

have the power to detect a reasonably-sized effect. Thus,

we believe our study provides preliminary evidence

that archival measures of audit quality are unable to

detect any effects of the partner-level signature

adoption.

These findings should be interesting to several parties.

Members of the working party of the Institute of

Chartered Accountants in England and Wales (2005)

suggested, ‘Signing the name of the firm provides just as

much motivation to a partner to get the opinion right as

he/she is signing as an individual on behalf of the other

partners within the firm.’ Similarly, members of the

Public Company Accounting Oversight Board (PCAOB),

during recent discussions of a similar US partner

signature proposal, expressed doubts about the audit

quality improvingeffects of such a policy (Defelice, 2009).

Our inability to reject the null of no audit quality impact

from an engagement partner’s signature adds some

empirical support to their suspicions.4Additionally,

investors, who rely on qualityinformation in the financial

statements, may be very interested in evidence about the

audit quality effects of this policy. Lastly, academics may

find this information interesting because it can contribute

to the empirical evidence related to both auditquality and

engagement partner signatures.

The remainder of this study is organized as follows.

The next section provides background information and

presents our research question. The sample and research

design is then discussed, followed by our results and a

final concluding section.

BACKGROUND AND RESEARCH QUESTION

Australian partners have been signing audit opinions for

many years, with the legal requirement to do so being

included in the Australian Corporations Act of 2001,

section 324(10) (Ryken, Radich & Fargher, 2007; Fargher

et al., 2008). Additionally, Taiwan requires that both the

engagement partner and the review partneron each audit

sign their own names to the audit opinion (Chen et al.,

2008). American regulators at the PCAOB had been

discussing a partner signature mandate, but are now

weighing an alternativepartner “disclosure” requirement

(PCAOB, 2009b, 2011a, 2011b). Thus, the global financial

community appears to be moving towards mandated

audit partner signatures.

The finalversion of the EU’s 8th Directive was issued in

May 2006 (European Commission, 2006). The directive

sets forth several new requirements for the auditor of

public interest companies including adherence to

IAASB’s International Auditing Standards,

implementation of a quality assurance (quality control)

system, external audits of internalcontrols, limitations on

auditor provided non-audit services, rotation of audit

partners every 7 years, among other requirements

(Braiotta, 2005). Additionally, article 28 states, ‘Where an

audit firm carries out the statutory audit, the audit report

shall be signed by at least the statutoryauditor(s) carrying

out the statutory audit on behalf of the audit firm’

(European Commission, 2006).

Member states were required to adopt the directive’s

partner signature mandate into their local laws at or

before June 2008 (Gold et al., 2012). Atthat time, many EU

member states had already adopted the partner signature

requirements. Germany, for example, had required

engagement partner signatures since 1985 (Gold et al.,

2012). The Netherlands implemented a partner signature

policy in 2006 via article 29 of the Dutch Law on Audit

Firm Supervision (Wet toezicht accountantsorganisaties,

2006). Most recently, the UK implemented this

requirement as part of the UK CompaniesAct of 2006 (UK

Companies Act, 2006), effective for 2009 fiscal years.

During discussions of the EU’s audit partner signature

mandate, some members argued that an individual audit

partner signature requirement would improve audit

quality. Similarly, members of the PCAOB, during recent

discussions of a similar US proposal, suggested a

psychological effect of signing one’s own name, causing

an individual to feel a ‘greater sense of personal

accountability’, and they briefly mentioned academic

literature suggesting such a psychological effect of

individuals signing their names (PCAOB, 2009a).

However, some involved in these discussions

questioned whether such a requirement would in fact

increase audit quality. Some members of a working party

of the Institute of Chartered Accountants in England and

Wales (2005) questioned whether an audit partner’s

signature would have any effect on incentives to audit.

Similarly, some PCAOB members, and Deloitte in a

comment letter, argued that the audit team makes the

decision, not the individual audit partner. Therefore,

because the audit partner’s signature would be limited to

the engagement partner’s work, it would be unlikely to

have an effect on the quality of an audit team’s decision

(Deloitte, 2008; PCAOB, 2009a). Additionally, some

considered whether an individual partner’s signature

might cause him/her to view the audit as only his/her

responsibility,driving the partner to decide not to consult

Audit Quality Effects of an Individual Audit Engagement Partner Signature Mandate 173

Int. J. Audit. 18: 172–192 (2014)© 2014 John Wiley & Sons Ltd