Research on rotation

By Janet Morrill

Studies show that audit-firm or audit partner rotation doesn't necessarily improve audit quality

Many of us have participated in discussions about whether to renew a current auditor or adopt a rotation policy. While publicly traded firms must rotate their audit partners, many anxious to follow best practices have considered going further and rotating the audit firm. Private companies have no rotation requirements but nevertheless periodically consider replacing the auditor.

The requirement to rotate auditors, or at least audit partners, is an attempt to reduce the threats to auditor independence largely arising from familiarity. The concern is that over the long term, the auditor becomes too close to the client and loses his or her “honest disinterestedness”.¹ For example, auditor independence can decrease as friendships develop, the auditor becomes too closely identified with the interests of client management, the audit plan becomes stale, or the auditor becomes reluctant to make decisions that indicate that past decisions were incorrect.

The threats posed by a long-term auditor-client relationship must be balanced against the costs of switching auditors and the risks to audit quality posed by a new auditor. Many contend that new auditors may have less knowledge of the client’s business, operations and risks. New auditors, particularly those who set a low audit fee, may also be anxious to recoup their initial audit costs and may be too accommodating.

The arguments for and against auditor rotation seem persuasive. What is needed is an understanding of which effects appear to prevail over a large number of companies, a large number of auditors and a long time span. Research in this area indicates that on balance audit-firm rotation does not improve audit quality.

Rotation of audit firms by public companies
Several recent studies have found that audit quality of public firms is generally lower in the early years of audit-firm tenure, arguing against audit-firm rotation.

Audit quality is difficult to quantify, so it is useful to understand the various measures researchers use and to ensure that the conclusions hold under different measures. Geiger and Raghunandan (2002)² measured audit quality as whether the auditor had issued a going-concern qualification in the prior year for US clients that declared bankruptcy. This is a joint measure of whether the auditor was able to detect a going-concern problem and willing to issue the qualification. They found that longer-term auditors were more likely to qualify their opinions under these circumstances.

Other studies have used accruals to measure audit quality. Total accruals are simply the difference between net income and cash flow from operations and include such items as amortization and changes in working capital accounts. Researchers often focus on discretionary, sometimes referred to as “abnormal,” accruals. To measure these, researchers determine a normal level of accrual based on firm size and fixed assets and deduct these normal accruals from total accruals to arrive at discretionary accruals. Discretionary accruals are seen to be where management has room to play. After all, a manager could never convince his or her auditor that there should be zero amortization, no allowance for doubtful accounts or no accounts payable. However, within some reasonable zone, managers may be able to shade their accruals to move income toward a desired target. When these discretionary accruals are high in absolute value, this is taken as evidence of greater earnings management and lower earnings quality.

Johnson et al (2002)³ and Myers et al (2003)⁴ found that the absolute values of total accruals and discretionary accruals were higher when the auditor was new, indicating that reporting quality appears to suffer when the auditor changes. A study of firms audited by Arthur Andersen that hired new auditors following Andersen’s collapse also found that audit quality did not improve following the rotation (Blouin, Grein and Rountree, 2007).⁵

Johnson et al also looked at the persistence of accruals as a measure of audit quality. There are two principal schools of thought about accruals. The first says managers use accruals opportunistically, enriching themselves by manipulating the stock price and/or increasing their bonuses. The second says managers use accruals efficiently to enable users to better predict future cash flows and earnings. Researchers can distinguish between these two motivations by seeing if the managed income is a better predictor of future earnings than unmanaged income is. If so, the manager is behaving efficiently, earnings quality is higher and the auditor is justified in allowing those accruals. Alternatively, less persistent accruals suggest the auditor should have limited the accruals. Johnson et al find the accruals component of earnings is less persistent for firms with shorter-term auditors, suggesting lower audit quality for these firms.

Finally, Carcello and Nagy (2004)⁶ looked at firms accused of fraudulent financial reporting and compared them with similar firms where such allegations had not been made. After controlling for corporate governance characteristics, the alleged fraudulent financial reporting firms were more likely to be audited by new auditors than the firms in the nonfraud group.

Rotation of audit partners by public companies
Rotating audit partners rather than the entire audit firm is an interesting compromise position. While there are new personalities and fresh eyes at the partner level, there is still the same audit firm methodology, prior year working papers and in many cases the same staff. The costs of switching, if any, are lower than those that would be incurred if the entire audit firm were replaced and the predecessor partner can still provide knowledge of the business as a resource to the successor partner.

Research here is sparse as the identity of the partner in charge of the audit is typically not disclosed. However, Carey and Simnett (2006)⁷ were able to study the effects of audit partner rotation in Australia where this information is required disclosure, although audit-partner rotation was not mandatory at the time of their sample. They found no evidence of deterioration of audit quality with long-term auditors using abnormal accruals. However, they found evidence of deterioration as long-tenure auditors were less likely to issue going-concern opinions for troubled companies and evidence of deterioration using another measure — firms just beating earnings benchmarks.

The logic of beating earnings benchmarks as a measure of audit quality is that managers are generally rewarded for meeting or just beating benchmarks and are often punished for missing them. Given that meeting or beating benchmarks is so desirable, management perhaps steers hard to ensure that the target is achieved and audit quality must be low as the auditor did not detect the earnings management or did not stop it. Various bench-marks are used by researchers, such as analyst forecasts, management forecasts, breakeven profits or prior year earnings. Using the latter two measures, Carey and Simnett found companies with long-tenure auditors seemed to beat breakeven profits (defined as 2% of total assets) or report earnings just slightly better than the prior years more often than one would expect. In general, then, this study found some evidence to support rotation of the audit partner, but the preponderance of evidence suggests that rotation impairs audit quality.

Rotation of audit firms by private companies
All the research cited studied public companies, primarily as they are the ones for which data is available. Can this evidence be generalized to a private company’s auditor renewal decision?

The public company evidence suggests the cost of reduced business knowledge on the part of the new auditor appears to be greater than the familiarity costs associated with the incumbent auditor. While this is also likely to be the case with private companies, there are some differences to consider.

Smaller private companies typically have less complex operations, so even a new auditor may have sufficient knowledge of the business to perform an effective audit, making rotation more feasible. However, as smaller companies have less bargaining power with their auditors (typically seen as more closely related to a self-interest threat for the auditor), perhaps incumbent auditors are less vulnerable to familiarity threats, so rotation is less necessary.

To date, there is only one published research study on private companies in Belgium (Knechel and Vanstraelen, 2007).⁸ That study found rotating auditors appeared to have no impact on audit quality as measured by the likelihood of giving a going-concern opinion to a distressed firm. Long tenure auditors did not improve audit quality but did not reduce it either. Again, if the costs of switching auditors are considered, rotating auditors to improve audit quality does not appear to make sense.

Indirect costs of auditor retention: independence perceptions
It is important to consider how financial statement users perceive a long-tenure auditor. If they believe that auditor rotation improves audit quality, the nonswitching firm’s financial statements might lack credibility. In that case, a company may save money by not switching auditors but may incur indirect costs through a higher cost of capital.

In an experiment with MBA and law students, Gates, Lowe and Reckers (2007)⁹ found that confidence in the financial statements increased when the audit firm was rotated. Rotating the audit partner had no effect. However, it does not appear that markets value auditor rotation. Ghosh and Moon (2005)¹⁰ examined stock market reactions to announcements of earnings audited by long-term auditors versus new auditors. They found that the market reacts more strongly to earnings audited by long-term auditors, suggesting that auditor retention improves the perceived credibility of earnings.

Similarly, Mansi et al (2004)¹¹ looked at the connection between auditor tenure and the cost of debt. Controlling for other factors, they found that longer tenure auditors were associated with lower debt costs, consistent with lenders attaching more credibility to financial information audited by long-term auditors.

Conclusion
The research on audit tenure fairly consistently indicates that audit-firm rotation does not improve audit quality (please see the table on page 63) and there is only one study providing mixed evidence that audit partner rotation may be beneficial. In fact, in many cases audit quality seems to deteriorate when the auditor is replaced. Taken together with the cost of switching auditors and the trust that many financial statement users seem to place in audits performed by long-term auditors (please see table on page 64), it appears that companies should not feel compelled to change auditors. Furthermore, standard-setters and regulators should also consider whether mandatory ro-tation of either audit partners or audit firms is in the best interests of society.

Janet Morrill, CA, CGA, PhD, is with the I. H. Asper School of Business, University of Manitoba

Technical editor: Christine Wiedman, PhD, FCA, School of Accounting and Finance, University of Waterloo

References

Mautz, R. K. and H. A. Sharaf. 1961. "The Philosophy of Auditing." American Accounting Association: USA
Geiger, M. A. and K. Raghunandan. 2002. "Auditor Tenure and Audit Reporting Failures." Auditing 21 (1): 67-78
Johnson, V. E., I. K. Khurana and J. K. Reynolds. 2002. "Audit-Firm Tenure and the Quality of Financial Reports." Contemporary Accounting Research 19 (4): 637-660
Myers, J. N., L. A. Myers and T. C. Omer. 2003. "Exploring the Term of the Auditor-Client Relationship and the Quality of Earnings: A Case for Mandatory Auditor Rotation?" The Accounting Review 78 (3): 779-799
Blouin, J., B. M. Grein and B. R. Rountree. 2007. "An Analysis of Forced Auditor Change: The Case of Former Andersen Clients." The Accounting Review 82 (3): 621-650
Carcello, J. V. and A. L. Nagy. 2004. "Audit Firm Tenure and Fraudulent Financial Reporting." Auditing 23 (2): 55-70
Carey, P. and R. Simnett. 2006. "Audit Partner Tenure and Audit Quality." The Accounting Review. 81 (3): 653-676
Knechel, R. W. and A. Vanstraelen. 2007. "The Relationship between Auditor Tenure and Audit Quality Implied by Going Concern Opinions." Auditing 26 (1): 113-131
Gates, S. K., D. J. Lowe and P. M. J. Reckers. 2007. "Restoring Public Confidence in Capital Markets through Auditor Rotation." Managerial Auditing Journal 22 (1) : 5- 17
Ghosh, A. and D. Moon. 2005. "Auditor Tenure and Perceptions of Audit Quality." The Accounting Review 80 (2): 585-612
Mansi, S. A., W. F. Maxwell and D. P. Miller. 2004. "Does Auditor Quality and Tenure Matter to Investors? Evidence from the Bond Market." Journal of Accounting Research 42 (4): 755-793