This blog has been written by Lucas Mahieux about his recently accepted paper “Auditors’ Incentives and Audit Quality: Non-audit Services versus Contingent Audit Fees”.
What are the incentive effects of the provision of non-audit services (NAS) on auditors? This important
policy question is of concern to regulators given the debate that has been raging for years on whether
audit firms should provide NAS to audit clients. This debate is often reduced to a simple cost and
benefit tradeoff. On the one side, joint NAS and audit services provision is likely to be more efficient
in terms of production costs because of knowledge spillovers (Simunic, 1984). On the other side, NAS
may threaten auditor independence because it creates an economic bond between the auditor and the
client (DeAngelo, 1981). In the absence of a clear sense of the incentive effects, the conventional wisdom
that “providing both NAS and audit services to the same client threatens auditor independence and may
affect audit quality” seems to prevail (Causholli et al., 2014). Nevertheless, Ewert (2004) emphasizes that
“the incentive problems of combining NAS and auditing are still an open issue.” More recently, auditor
independence rules related to NAS appear to be headed in opposite directions in the US and in the UK,
which indicates a disagreement among regulators (CFO, 2020).
In my recent paper, I depart from the conventional wisdom and analyze a novel tradeoff related
to the provision of NAS by auditors to audit clients. I also investigate the joint implications of NAS
and contingent audit fees (CAF) for audit quality. Specifically, I build a framework that highlights
a positive incentive effect of NAS: the possibility of providing NAS contingent on detecting financial
misstatements may increase the auditor’s effort to detect those misstatements. This represents a benefit
of the provision of NAS by auditors. Nonetheless, my analysis also underlines that the provision of NAS
reduces auditor independence and may decrease audit quality. When considering a single firm, there is no
role for regulation as investors trade off the cost and benefit of using NAS to provide effort incentives to
the auditor. However, in the presence of multiple client firms, I highlight the negative externalities caused
by the decrease in audit quality when investors rely on peer-firms’ financial statements to make investment
decisions. Restricting auditors from providing NAS may then be desirable. Thus, regulators face a tradeoff
between the ex ante positive incentive effect and the ex post decrease in auditor independence. Removing
the current restrictions on CAF for unfavorable audit opinions may offset the ex post decrease in audit
quality while preserving the ex ante incentives.
The results of the paper contribute to both the regulatory and academic communities. Regulators
and practitioners fear that auditors would be unwilling to challenge a client if a negative audit opinion
would mean losing future NAS contracts (e.g., Bell et al., 2015). However, regulators seem to focus on the
ex post lack of auditor independence without considering the ex ante incentive effects. My findings shed
some light on the desirability of the provision of NAS to audit clients, and suggest regulators may need
to investigate more carefully the tradeoff between ex ante incentives and an ex post decrease in auditor
independence. In particular, I show that having fully independent auditors may not always be optimal.
Moreover, regulators have banned CAF in most jurisdictions. However, CAF could be used to provide
incentives to auditors to exert audit effort in environments in which market-based incentive forces are
not sufficient. My results show that removing the current restrictions on contingent audit fees jointly
with banning non-audit services may increase auditor independence while preserving the ex ante effort
incentives.
Finally, my model also generates several testable empirical predictions. The main empirical takeaway
is that a ban on the provision of NAS to audit clients could increase auditor independence but decrease
audit quality. This result is important given that many empirical studies use proxies of audit quality to
assess auditor independence. My analysis suggests that the relation between those two key variables may
be more complex. A direct consequence is that a positive relationship between NAS fees and financial
misstatements does not imply that NAS negatively affect audit effort and audit quality. On the contrary,
I show that a positive relationship may indicate that incentives are provided to auditors using NAS and
that this mechanism improves audit effort and ex ante audit quality.
This paper has recently been accepted at the European Accounting Review, and its permanent link will be
https://doi.org/10.1080/09638180.2022.2066011
References
Bell, T. B., Causholli, M. and Knechel, W. R. (2015), ‘Audit Firm Tenure, Non-Audit Services, and Internal Assessments
of Audit Quality’, Journal of Accounting Research 53(3), 461–509.
Causholli, M., Chambers, D. J. and Payne, J. L. (2014), ‘Future Nonaudit Service fees and Audit Quality’, Contemporary
Accounting Research 31(3), 681–712.
CFO (2020), ‘SEC Planning to Loosen Auditor Independence Rules’.
DeAngelo, L. E. (1981), ‘Auditor Independence,‘Low Balling’, and Disclosure Regulation’, Journal of Accounting and Economics
3(2), 113–127.
Ewert, R. (2004), ‘Audit Regulation, Audit Quality, and Audit Research in the Post-Enron Era: an Analysis of Nonaudit
Services’, The Econonomics and Politics of Accounting. Oxford University Press, Oxford pp. 239–264.
Simunic, D. A. (1984), ‘Auditing, Consulting, and Auditor Independence’, Journal of Accounting Research pp. 679–702.
