How Is the Audit Market Affected by Characteristics of the Nonaudit Services Market?

• Published date: 01 June 2021
• Author: HENRY L. FRIEDMAN,LUCAS MAHIEUX
• Date: 01 June 2021
• DOI: http://doi.org/10.1111/1475-679X.12347

DOI: 10.1111/1475-679X.12347

Journal of Accounting Research

Vol. 59 No. 3 June 2021

Printed in U.S.A.

How Is the Audit Market Affected

by Characteristics of the Nonaudit

Services Market?

HENRY L. FRIEDMAN∗AND LUCAS MAHIEUX†

Received 31 August 2019; accepted 28 December 2020

ABSTRACT

How can features of the markets for audit and nonaudit services (NAS) af-

fect an audit f‌irm’s incentives to invest in audit quality, average audit quality,

and social welfare? We address these questions in a model focusing on com-

petition in both audit and NAS markets. We show that, when audit and NAS

demand are positively correlated, prohibiting auditors from providing NAS

to audit clients leads to higher investments in audit quality, but can decrease

average audit quality if marginal clients switch to lower quality auditors. The

effect on social welfare can be positive or negative, depending on the dis-

tribution of clients’ service demands. General bans on auditor provision of

NAS can, via similar channels, increase or decrease audit quality and social

welfare. Overall, our f‌indings suggest a more nuanced view of how regulating

an auditor’s provision of NAS might affect audit quality and social welfare,

∗Anderson School of Management, University of California, Los Angeles; †Tilburg School

of Economics and Management, Tilburg University

Accepted by Regina Wittenberg Moerman. We thank the associate editor,two anonymous

reviewers, Christopher Bleibtreu (discussant), Judson Caskey, Jenny Chu, Ilan Guttman, Gilad

Livne, Xiaojing Meng, Michael Minnis, Joshua Ronen, Brett Trueman, Raghu Venugopalan

(discussant), Gaoqing Zhang (discussant), and conference/workshop participants at Bocconi

University, Boston College, Columbia University,New York University, Tilburg University, the

2019 AAA Annual Meeting, the 2019 ARW at the University of Zurich, the 2018 CFEA at

TulaneUniversity, and the 2019 FARS Midyear Meeting for helpful comments. Mahieux grate-

fully acknowledges funding from the European Research Council (Grant Agreement Number

669217). Any remaining errors are ours.

[The copyright line for this article was changed on 17 February 2021 after original online

publication.]

959

© 2021 The Authors. Journal of Accounting Research published by Wiley Periodicals LLC on behalf of The

Accounting Research Center at the University of Chicago Booth School of Business

This is an open access article under the terms of the

CreativeCommonsAttribution-NonCommercial-NoDerivs License, which permits use and distribution in

any medium, provided the original work is properly cited, the use is non-commercial and no modif‌ications

or adaptations are made.

960 h. l. friedman and l. mahieux

and are driven by the effects of multimarket competition on the auditor’s

incentives to invest in audit quality, rather than previously identif‌ied auditor

independence or knowledge spillover channels.

JEL codes: L11, L13, L51, M41, M42

Keywords: audit quality; auditing services; nonaudit services; competition

1. Introduction

Researchers, regulators, and practitioners have long-standing concerns

about the negative effects of nonaudit services (NAS) on audit quality (see,

e.g., DeAngelo [1981a], Che, Langli, and Svanström [2018], Kowaleski,

Mayhew, and Tegeler [2018]). Payments for NAS by client f‌irms can impair

auditor independence and motivate auditors to ignore client def‌iciencies.

Simunic [1984] and Bouwens [2018], in contrast, suggest positive spillovers

from NAS provision to audit quality, because of, for instance, knowledge

transfer across audit and NAS employees at the same f‌irm. These arguments

are generally constructed in a single-client setting, in which one auditor

provides services to one client. Such a setting precludes a number of inter-

esting economic forces from operating, including the potential for auditors

to target audit and NAS client segments via quality and pricing decisions.1

In this paper, we focus on interactions between audit and NAS using an

industrial organization perspective. We use a model to examine how com-

petition in the market for NAS can affect the audit f‌irm’s choices related to

f‌irm-wide quality investments, audit pricing, and the implications of regula-

tions prohibiting auditors from offering NAS. Our analysis provides a novel

perspective on how economic forces related to NAS can affect audit out-

comes through channels involving the industrial organization of the audit

and NAS markets.

Our model features three types of economic agents and two services mar-

kets. A set of heterogeneous client f‌irms demand professional services from

auditors and consultants. Auditors provide assurance about f‌inancial re-

ports, and consultants provide NAS to help f‌irms gain value. Our primary

interest lies in the choices made by a high-quality audit f‌irm that can make

audit-improving investments and offers both audit and NAS to clients. A

low-quality audit f‌irm offers a base-level audit and competes with the high-

quality auditor in the audit market, whereas an NAS-only consultant com-

petes with the high-quality auditor in the NAS market. Our setup follows

from our primary focus on how features of the NAS market can inf‌luence

the high-quality auditor’s offerings.2

1A single-client model also cannot distinguish between auditor-wide and engagement-

specif‌ic investments. The Public Company Accounting Oversight Board has highlighted f‌irm-

wide investments in human capital and infrastructure as important aspects contributing to

audit quality PCAOB [2015].

2In our main analysis, we abstract away from features such as audit independence, knowl-

edge spillovers, audit standards, and costs of audit failure studied in previous research,

non-audit services’ effects on the audit market961

Audit services for each client f‌irm are modeled as in the standard single-

client models of Dye [1995] and Laux and Newman [2010]. Each client

f‌irm has a project that can be either good or bad, with good projects pro-

viding higher cash f‌lows in expectation. We refer to the probability of hav-

ing a bad project as the business risk of a client f‌irm, and allow this risk to

vary across client f‌irms. Each client f‌irm’s investors must choose whether

to make a f‌ixed investment in the f‌irms’ project. In order to value each

client f‌irm’s project and subsequently make an informed investment, the

investors require each client f‌irm’s manager to issue an accounting report.

Absent an audit, each client f‌irm’s manager would report that the f‌irm’s

project is good. The audit probabilistically detects misreporting in this ac-

counting report, and therefore exposes bad projects with a probability that

depends on the quality of the audit purchased by the f‌irm.

Before offering its services to clients, the high-quality auditor can invest

in f‌irm-wide audit quality, with higher audit quality increasing the probabil-

ity that its audits successfully detect misreporting.3The cost of audit quality

is increasing both in the quality delivered and the number of clients it is

delivered to, which captures costs of hiring and training high-quality em-

ployees to deliver services. The low-quality auditor’s audit quality is f‌ixed

(e.g., at the quality required by auditing standards). The high-quality au-

dit f‌irm’s investment in quality is observable to client f‌irms, ref‌lecting, for

instance, the auditor’s reputation. This reputation affects client f‌irms’ will-

ingness to pay for audit services, and it is this audit revenue channel that

motivates investment in audit quality.4

Client f‌irms also can purchase NAS, which increase client f‌irm value, ei-

ther from the high-quality auditor or from the NAS competitor, that is, a

pure-play consulting f‌irm. In the baseline model, we focus on the case in

which client valuation of audit and NAS are positively correlated. Indeed,

positive demand correlation seems the most relevant case as auditors would

then like to sell audit services and NAS to the same clients and NAS re-

strictions would prevent this. Both the high-quality auditor and the NAS

competitor offer equivalent-quality NAS, and neither can use NAS quality

discussed further in section 1.1. We introduce additional competition in the NAS market in

section 4.2 and show that our results still hold as long as there is some degree of differenti-

ation. We present results related to changes in audit standards in section 4.3. In section 4.4

and in appendix B, we show that our main results are robust to formally separating preparers

from users of f‌inancial statements. In appendix B, we also allow for the low-quality auditor to

choose its audit quality, allow the NAS competitor to choose the quality of its services, and

model mandates that impose penalties or costs on f‌irms that are not audited.

3Audit quality in our model is f‌irm wide, applying to all of the auditor’s engagements.

This captures constructs such as audit f‌irm culture, internal controls, or reputation, rather

than engagement-level auditor effort as captured by many prior single-client models (e.g.,

Dye [1995]). Such f‌irm-wide audit quality can also be interpreted as capturing the probability

that any given engagement partner or staff member acts in good faith and avoids shirking.

4The revenue-based motive does not conf‌lict with previously studied motives related to

auditor liability or audit standards, which are discussed further in section 1.1.