Auditor Resignations and the Importance of Monitoring Client Acceptance Risk Free

Prior research suggests that auditor resignations from clients reflect auditors' efforts to manage their risks. However, when predecessor auditors drop clients, successor auditors are needed to fill the void. For years, accounting firms have been acutely aware of the fact that accepting another firm's clients must be done with caution. In fact, both the Public Company Accounting Oversight Board (PCAOB) and the American Institute of Certified Public Accountants' (AICPA) Auditing Standards Board have quality control standards that require accounting firms to assess the risks associated with accepting a new client or continuing to serve an existing client (PCAOB 2003; AICPA 2007). While firms must comply with these quality control standards, recent research demonstrates the importance of this requirement for managing engagement risk.

Our recently published study in Accounting Horizons, “An Ex Post Examination of Auditor Resignations” (Catanach et al. 2011), examines client acceptance patterns and client outcomes following auditor resignations. More specifically, this study uses a sample of auditor resignations to examine two issues: (1) why accounting firms assume the role of successor auditor on these presumably risky engagements, and (2) the future outcomes of clients accepted by these successor auditors.¹ Our analyses reveal three key findings. First, smaller accounting firms accept the successor auditor role for resigned clients at a considerably greater rate than do larger firms. Second, resigned clients accepted by smaller firms are riskier on several dimensions than those accepted by larger firms. Third, resigned clients accepted by smaller firms are associated with weaker long-term financial ratios, shorter survival tenures, and a greater likelihood of adverse outcomes relative to those accepted by larger firms.

This study highlights two issues that reiterate the need for successor auditors to proceed with care when accepting clients from which the predecessor auditor resigned. First, it is important for auditors to conduct a thorough client acceptance and continuance process that also is proactive and adaptive in identifying and incorporating key risk factors. This is especially vital for clients from which the prior accounting firm has walked away from the audit. Second, accounting firm leadership, from the partners serving in a risk management capacity to partners at the engagement level, should consider routinely reviewing past client acceptance and continuance decisions to determine how well their processes are working. Whereas Type I errors in the client acceptance process (in which an accounting firm resigns from a “good” client) are decisions that a firm presumably regrets, Type II errors in the client acceptance process (in which an accounting firm becomes the successor auditor on a “bad” client) potentially can be ruinous for the firm.

Most auditors probably recall that Ernst & Whinney resigned from the ZZZZ Best engagement before the audit blew up. What they may not recall is that, as recently as 2007, Barry Minkow, the former CEO of ZZZZ Best, stated publicly that the worst thing that can happen to a company is for the auditor to resign (Foy 2007). Findings from prior research are consistent with Barry Minkow's observation (e.g., DeFond et al. 1997; Krishnan and Krishnan 1997). These studies find that clients from whom the predecessor auditor resigned were more likely to be associated with certain indicators of risk (e.g., declining financial health, prior going-concern opinions) than clients from which the auditor was dismissed. Prior studies also have examined the extent to which the size of the accounting firm plays a role in the successor auditor's decision to accept a resigned client, and find that larger accounting firms are less likely to serve as the successor auditor when the predecessor auditor has resigned (Raghunandan and Rama 1999).

While auditor change studies conclude that clients from which accounting firms resign are risky, little is known about what happens in the periods after these predecessor auditors have resigned from the engagement. Our primary analyses are based on a sample of auditor resignations that occurred between 1994 and 2000, and they reveal several interesting insights about the successor auditor's decision to accept a resigned client.²

We first document a dramatic shift in resigned clients from larger accounting firms to smaller firms within the auditor resignations sample. We develop a basic ratio to use in analyzing changes in a firm's client portfolio. This “turnover ratio” is based on how many clients an accounting firm added (i.e., assumed the successor auditor role) divided by how many the same firm dropped (i.e., were the predecessor auditor). A ratio greater than (less than) one indicates that a firm added more (less) clients than it dropped. A high turnover ratio implies that an accounting firm either intentionally or unintentionally overlooked key risk factors during its client acceptance process. In the former case, a firm may have been pursuing a growth strategy or attempting to offset the effect of lost clients, which resulted in adding some clients with riskier profiles to its portfolio. In the latter case, a firm's client screening process may not have effectively assessed the client's risk profile, which led the firm to accept an increased number of resigned clients.

We find substantial differences in turnover ratios between the larger and smaller accounting firms. Table 1 shows that, within the sample of clients experiencing auditor resignations, larger firms are the predecessor auditor for 73 percent of the sample (171 of 235), yet the successor auditor for only 38 percent of the sample (89 of 235). In contrast, smaller firms assume the successor auditor role on the 82 net clients from which the larger firms resigned (add 146 clients, drop 64 clients).³ Overall, Table 1 is consistent with the interpretation that larger firms exercise caution when assuming the successor auditor role, and smaller firms become the primary supplier of audit services to resigned clients.

We next investigate factors associated with successor auditors' decisions to accept resigned clients. This is an interesting question because, presumably, accounting firms know that resigned clients are risky, yet they still choose to accept the engagement. We found that, when larger accounting firms choose to conduct the audit for resigned clients, several factors are critical. Client size is important because not all accounting firms have the resources to audit multinational corporations. Additionally, firms screen on several dimensions of client risk. For example, larger firms are more likely to avoid clients with prior going-concern opinions, high-risk comments from Form 8-K filings, and complex operations. Larger firms also are more likely to avoid clients from which the prior auditor resigned after the client's year-end (i.e., when the audit likely is underway). It is one thing to resign, but to resign in the middle of the audit is a signal that should not be ignored. Further, larger firms accept the successor auditor role on resigned clients when the firm believes it has the industry expertise to help mitigate the risks associated with the client.

Finally, we analyze future outcomes associated with these client acceptance decisions. We classify clients that continue to operate as active enterprises during the entire ex post period as “good” future outcomes, and clients that file for bankruptcy or have their securities delisted during the ex post period as “bad” outcomes. As reported in Table 2, there are notable differences in outcomes associated with resigned clients accepted by larger and smaller accounting firms. Most significantly, clients file for bankruptcy or have their securities delisted in 27 percent of engagements where larger firms assume the successor auditor role; however, this likelihood almost doubles, to 51 percent, when the successor auditor is a smaller firm. Moreover, comparing the predecessor auditor bankruptcy and delisting percentages with the successor auditor bankruptcy and delisting percentages, the larger accounting firm percentage declines by 14 percent (from 41 percent to 27 percent), while the smaller firm percentage increases by 8 percent (from 43 percent to 51 percent).⁴ Together, the findings are consistent with the previous discussion in that clients with observable risk factors at the time of an auditor resignation commonly give rise to negative future outcomes. However, we are unable to distinguish whether the fact that larger accounting firms' clients experience significantly fewer negative future outcomes is due to a comparative advantage that larger firms have (i.e., a more effective client acceptance process and/or less pressure to retain clients) or to larger firms having different risk preferences (i.e., being more risk-averse).

The client acceptance and continuance process is one of the most important priorities for the leadership of any accounting firm; it is a classic case of risk management. In our study, we observe clear differences in client acceptance decisions for a sample of clients from which the predecessor auditor resigned. By substantially reducing their proportion of resigned clients, larger accounting firms significantly reduce their exposure to negative future client outcomes. As noted previously, it is an open question whether the portfolio rebalancing successes achieved by larger accounting firms are attributable to a comparative advantage in their client acceptance process, a difference in their risk tolerance, or some combination of the two. We leave this question for future research.

In most cases, accounting firms cannot completely mitigate engagement risk by increasing the scope of testing or by charging a fee premium. Consequently, all accounting firms should strive to enhance their client selection models and should revisit those models on a regular basis. Of equal importance, firms should routinely review and adapt their client acceptance processes in order to prospectively avoid high-risk clients. Considering the size of audit fees relative to lawsuits, accounting firms must screen resigned clients very carefully before accepting them.