Two Studies Raise Questions About Audit Partner Rotation
In 2002, the Sarbanes-Oxley Act required audit partner rotation on a five-year cycle. The theory behind this requirement was that long audit partner tenures could lead to the partner becoming too closely associated with the client and that a “fresh look” by a newly-involved partner might turn up financial reporting issues that the prior partner had ignored or missed. The supposed benefits of a fresh look also lead the PCAOB to float the idea of mandatory audit firm rotation in a 2011 concept release. The Board dropped the idea after the House of Representatives passed a bi-partisan bill in 2013 prohibiting the PCAOB from requiring companies to automatically change auditors.
Two recent academic studies examine the merits of the underlying rationale for partner rotation. These studies find no evidence that audit quality declines as engagement partner tenure increases – within the constraint of the five year tenure limit imposed by SOX – and no evidence that the incoming engagement partner’s fresh look uncovers reporting issues that his or her predecessor missed. On the other hand, these studies provide some support for the argument against rotation – that the new partner on an engagement is more likely to miss issues during the first year because he or she is still coming up the learning curve regarding the client business and financial reporting challenges.
Mandatory Audit Partner Rotations and Audit Quality in the United States
Mandatory Audit Partner Rotations and Audit Quality in the United States (available here for purchase) appears in the August 2020 issue of Auditing: A Journal of Practice and Theory (AJPT). The authors, Huan Kuang, Huimin Li, Matthew G. Sherwood, and Robert L. Whited, examined SEC filings and comment letters between 2003 and 2019 to identify partner changes that occurred after five years of partner tenure. They state that they “do not find evidence consistent with rotation materially improving audit quality.” Specific findings of their study include:
No evidence that abnormal accruals (a measure of the likelihood of financial reporting manipulation) differ in financial reporting periods before and after mandatory partner rotation.
Limited, but statistically significant, evidence that financial statements audited during the initial years after a mandatory change in audit partners are more likely to be subsequently restated than are those audited in the final years of the outgoing engagement partner’s term. This is particularly likely when the audit firm’s tenure with the client is short. While the sample size was small and material restatements are infrequent, the authors observe that “the weight of our evidence is not consistent with partner rotation yielding material improvements to the audit process due to the ‘fresh look’ of the new audit partner.”
No evidence that audit firms materially increase or decrease audit fees following partner rotation. The authors note that this “could be because changing partners does not influence the number of audit hours or that the new partner does increase audit hours but does not pass these costs on to the client. * * * If there is a learning curve for the new partner who performs additional audit procedures when partner rotation occurs, then the audit firm may suffer financially in the year of partner rotation if profit margins are squeezed.”
On the Economics of Mandatory Audit-Partner Rotations and Tenure: Evidence from PCAOB Data
In On the Economics of Mandatory Audit-Partner Rotations and Tenure: Evidence from PCAOB Data, Brandon Gipper, Luzi Hail, and Christian Leuz reach conclusions similar to those of Kuang, Li, Sherwood, and Whited regarding the audit quality implications of partner rotation. Gipper, et al. state that they “find no evidence for audit quality declines over the tenure cycle and, consistent with this result, little support for fresh-look benefits after five-year mandatory rotations.” However, they reach a different conclusion regarding the economic impact, finding that audit fees decline immediately after rotation, but subsequently rise. The results of this research will appear in a future issue of The Accounting Review and are currently available on SSRN, an online repository of scholarly research in the social sciences and humanities.
Gipper, Hail, and Leuz based their work on a non-public PCAOB dataset that covers audits from 2008 to 2014 of over 3,300 clients of the six largest U.S. audit firms. This dataset includes 2,385 mandatory engagement-partner rotations and contains engagement information, including audit hours, partner hours, billing realization, review partner names and hours, and internal client-risk ratings. The findings of Gipper and his colleagues touch on the audit quality impact of rotation, firm management of partner rotation, and the economic impact of rotation.
Audit quality. Like the authors of the study published in AJPT, Gipper, Hail, and Leuz find no evidence of a decline in audit quality over the five-year rotation cycle or of any fresh-look benefits from mandatory partner rotation. Moreover, like the other study, they find some evidence that audit quality declines in the first years after rotation. “We show that, for the average engagement of a Big-6 firm, audit quality is unrelated to partner tenure, except for announcements of restatements, which are more frequent in the first two years after rotation.”
Rotation quality management. The study finds that audit firms manage partner rotation to reduce the likelihood of audit failures. In particular, audit firms employ “shadowing” – a practice in which the incoming partner observes the work of the outgoing partner during the last year of his or her tenure. However, the incidence of shadowing varies with the market and the experience level of the incoming partner. “[E]vidence of partner shadowing is present only in less concentrated local audit markets suggesting that audit firms dedicate more resources to minimize disruption and prevent client attrition in competitive environments where clients have more outside options. In addition, we find more evidence of shadowing for experienced partners, plausibly due to the importance or complexity of the clients assigned to them.”
Economic effects. Unlike the AJPT study, the Gipper-Hail-Leuz paper finds that rotation does impact audit fees. Specifically, audit fees decline immediately after the new partner takes over, but subsequently decline. Conversely, audit hours, total partner hours and review partner hours rise significantly in the first year of the incoming partner’s tenure but fall in the following years. They state:
“When a new partner begins a five-year cycle, she needs additional time familiarizing herself with the client and the audit procedures in place. She may also consider updating procedures or ask audit staff to help her with the transition. As the new partner does not yet have a relationship with the client, it could be harder for her to ask for fee increases. Moreover, the client could use partner rotation as a way to mount fee pressures.”
Like shadowing, these economic effects of rotation differ depending on the competitiveness of the local audit market, client size, and partner experience. For example, auditors charge lower audit fees in more competitive audit market and encounter greater fee pressure. This results in lower fees in the first year after partner rotation. However, audit fees increase more steeply in these more competitive markets during the five-year cycle. Further, “newer, less experienced partners spend more time on the engagement in the first year of the rotation cycle, but also exhibit greater declines in hours in years four and five. This pattern is consistent with having a steeper learning curve than experienced partners.”
Comment: These findings do not in themselves either demonstrate that partner rotation has no benefits or, if it has benefits, that five years is the optimum rotation period. Gipper, Hail, Leuz observe that a “plausible explanation” for the lack of evidence that audit quality is related to partner tenure “is that in the relatively robust U.S. reporting and audit environment the five-year rotation mandate is sufficiently short to prevent, at least on average, major declines in audit quality over the partner cycle.” Further, their research “does not speak to the relation between partner tenure and audit quality beyond the five-year period. It could well be that a longer rotation term would yield similar results, while being less costly to auditors and clients.”
From an audit committee perspective, five-year partner rotation is a regulatory requirement and therefore unavoidable. Audit committees might however want to take note of the evidence in both studies that audit quality (as measured by the need to subsequently restate financial statements) declines in the first years after a new partner takes charge. “Shadowing” during the final year of the out-going partner’s tenure appears to be a logical step to reduce the risk that the incoming partner will miss a material error while still learning about the company and its financial reporting issues. But whether the audit firm employs shadowing seems to depend on factors like the competitiveness of the audit market, the experience of the incoming partner, and the size of the client. Audit committees facing a partner rotation should ask what steps the audit firm will take to ensure a smooth transition to the new partner and whether the new partner will shadow the old during his or her final year. If the firm does not plan on using shadowing, the committee should ask why not.