A perennially contentious issue, auditor rotation has led regulators in both the United States and Europe to require public companies to change auditors periodically—in the European Union by requiring firms to invite bids from other audit firms after ten years and in the U.S. by mandating rotation after five years of the engagement partner overseeing audits of a corporate client (but not demanding rotation of the partner’s audit firm itself).
With the U.S. mandate dating from the Sarbanes-Oxley act of 2002 (SOX) and the E.U. requirement adopted twelve years later, a lingering question has been whether the U.S. will follow suit by broadening its statute to require periodic audit-firm rotation, a step that would likely provoke strong opposition from the accounting industry.
Some new research that probes the two reasons most frequently advanced for mandating auditor rotations: 1) that personal ties developed over time between auditors and clients can compromise the accountants’ professional independence and, thus, the quality of financial reporting; and 2) that mandating rotations brings fresh looks to audits which likely enhance the quality of reporting.
Studies in two peer-reviewed journals of the American Accounting Association, Auditing: A Journal of Practice and Theory and The Accounting Review, find no significant fall-off in reporting quality over the course of partners’ five-year tenures, as limited by SOX, and little or no evidence that the fresh looks mandated by SOX make for improved audits. Some evidence even emerges in the AJPT study of audit-quality decline with a new engagement partner at the helm, perhaps reflecting a fall-off in knowledge about the client.
Comments Robert L. Whited of North Carolina State University, a co-author of the AJPT paper: “I think this study has implications both for current partner-rotation requirements and for potential audit-firm rotations. Some have argued in favor of rotation that fresh eyes of a new audit partner help improve audit quality, and a similar argument is made by those favoring mandatory firm rotation. However, in our sample, we find no evidence of an improvement in the first year or two following rotation.”
Drawing on publicly available data, the AJPT study employs a novel method to focus on the years immediately preceding and following mandated partner rotations. Joining Prof. Whited as co-authors in the AJPT study are Huan Kuang and Matthew G. Sherwood of the University of Massachusetts Amherst and Huimin Li of the University of New Hampshire.
The Accounting Review study, in contrast, draws on massive proprietary data made available to the researchers by the Public Company Accounting Oversight Board (PCAOB). A joint effort of Brandon Gipper of Stanford University, Luzi Hail of the University of Pennsylvania, and Christian Leuz of the University of Chicago, their study represents the first partner-tenure and mandatory-rotation analysis for a large cross-section of publicly traded U.S. companies. The sample consists of 3,300 corporate clients of the country’s six largest audit firms and covering 2,385 mandatory engagement-partner rotations over the seven-year period 2008 through 2014.
The study, “On the Economics of Mandatory Audit-Partner Rotations and Tenure: Evidence from PCAOB Data,” focuses particularly on the clients of BDO USA, Deloitte & Touche, Ernst & Young, Grant Thornton, KPMG, and PricewaterhouseCoopers. The corporate clients in the sample constitute about 46% of SEC registrants in the U.S. and account for about 85% of aggregate market capitalization.
The professors found that, on average, quality over the five-year mandatory rotation cycle is unrelated to the length of partners’ tenures with clients, except for announcements of restatements, which are more frequent in the first two years after rotation. While this increase in announcements of prior misstatements suggests a benefit of fresh looks, other important indicators of audit quality do not: for example, financial misstatements are no less likely to occur in those two years, and accrual levels do not decline, accruals being non-cash accounting items that are often subjective (such as estimates of inventory value or of bad debts) and are therefore considered particularly subject to management manipulation.
The researchers conclude, then, that “for the average client engagement mandatory rotation appears to be short enough or the U.S. audit environment robust enough to prevent auditor capture or complacency. At the same time, we find only limited evidence of fresh-look benefits.”
Adds Prof. Gipper: “Our findings also suggest a likely reason for this apparent lack of fresh-look benefits—namely, that audit firms anticipate and invest resources to reduce potential disruption arising from mandatory partner rotations.”
The AJPT study, “Mandatory Audit-Partner Rotations and Audit Quality in the United States,” covers the 17-year period 2003 through 2019. Prof. Whited and his colleagues overcame the absence of engagement-partner identification through most of that time with a novel method: they scanned 1.3 million corporate filings with the SEC to identify instances where companies disclosed a change in engagement partner to comply with the SOX-mandated five-year limit. These rare disclosures, combined with a smaller number obtained through SEC comment letters, resulted in a sample of 171 engagement-partner switches due specifically to the SOX mandate and not some other cause.
The professors compared the quality of firms’ financial reporting in the year and two years before and after the switch of engagement partners, using such standard indicators as financial misstatements and abnormal levels of accruals. Within the five-year tenure limit enforced by SOX, they write, they found neither “an impairment of audit quality as tenure increases,” nor “a material improvement in audit quality following rotation (i.e., fresh look). On the contrary, we find some evidence that material misstatements are more likely to occur in the first year or years following a mandatory audit-partner rotation, though we urge caution given the limited sample and the low frequency of misstatements.”
The AJPT study appears in the quarterly’s current issue (August/October), and the Accounting Review study has been accepted for a future issue of the periodical, which is published six times yearly.