Successor Auditors Research Articles

An Empirical Analysis of Voluntarily Supplied Client-Auditor Realignment Reasons

In this paper we investigate whether voluntarily disclosed reasons for auditor-client realignments (as encouraged by the SEC) have information content for investors. After classifying realignment reasons into two types - verifiable and non-verifiable, with the latter representing disclosures about the auditor-client relationship not evident from alternative sources, we find that, as predicted by the good news precept of theoretical signaling models, non-verifiable realignment reasons are positively associated at the time of their announcement with abnormal returns. We also investigate whether voluntarily disclosed realignment reasons are associated with the relative size of the predecessor or successor auditor. We find that clients are more likely to cite service-related (non-verifiable) reasons when dismissing large predecessor auditors, and are more likely to cite fee-related (non-verifiable) reasons when choosing small successor auditors. These findings are consistent with auditors competing for the clients of large auditors by offering better or broader services, and with smaller auditors competing based upon price. All of our findings are robust to controlling for mandatory auditor change disclosures (auditor client disagreements, reportable events, and going-concern opinions), and operating, financing and investing activities found in prior research to be associated with auditor changes.

Going Concern Decisions and Auditor Switches: An Economic Perspective

This research examines the effects of economic incentives on the strategic interaction between an auditor and client when a potential going concern situation exists. This experimental economics investigation has the following game structure. The auditor first obtains information relevant in predicting the client's viability and then conveys to the client an intention to express a clean or going concern opinion. A client can switch auditors in an attempt to avoid a going concern opinion and the potential, self-fulfilling nature of the auditor's prediction of business termination. This incentive to switch auditors poses a credible threat to the incumbent auditor. Conducting the test in the laboratory enables us to vary 1) the likelihood the successor auditor will issue a clean opinion and 2) the presence of the self-fulfilling prophecy effect. By varying the levels of these treatments, we are able to test the predicted effect on the audit partner's reporting decision and the client's switching decision. The results strongly indicate that treatments 1) and 2) both had a significant effect on subject behavior. However, contrary to the equilibrium predictions, auditors did not sacrifice their independence as 1) and 2) increased. Clients, on the other hand, did respond opportunistically by switching auditors more frequently as 1) and 2) increased. The effect was strongly significant and was directionally consistent with the analytical predictions.

An Empirical Analysis of Voluntarily Supplied Client-Auditor Realignment Reasons

In this paper we investigate whether voluntarily disclosed reasons for auditor-client realignments (as encouraged by the SEC) have information content for investors. After classifying realignment reasons into two types—verifiable and non-verifiable, with the latter representing disclosures about the auditor-client relationship not evident from alternative sources—we find that, as predicted by the “good news” precept of theoretical signaling models, non-verifiable realignment reasons are positively associated at the time of their announcement with abnormal returns. We also investigate whether voluntarily disclosed realignment reasons are associated with the relative size of the predecessor or successor auditor. We find that clients are more likely to cite service-related (non-verifiable) reasons when dismissing large predecessor auditors, and are more likely to cite fee-related (non-verifiable) reasons when choosing small successor auditors. These findings are consistent with auditors competing for the clients of large auditors by offering better or broader services, and with smaller auditors competing based upon price. All of our findings are robust to controlling for mandatory auditor change disclosures (auditor-client disagreements, reportable events, and goingconcern opinions), and operating, financing, and investing activities found in prior research to be associated with auditor changes.

Quality Levels of Predecessor and Successor Auditors: Evidence from the Reasons Reported by Managers for Changing Auditors

In this study we investigate relations between the quality levels of successor and predecessor auditors for a sample of client-initiated auditor switches. We use reasons reported by managers for changing auditors to investigate whether those reasons are differentially related to changes in the quality levels of successor and predecessor auditors. Based on prior evidence on auditor realignments, we classify the reasons from managers into three categories that appear to summarize the motivating factors surrounding client-initiated changes. The three categories relate to structural changes of the company, audit fees, and accounting disagreements. In this study we introduce a three-level auditor quality proxy that suggests and allows future researchers, to consider auditor quality as other than a binary variable (i.e., Big 6 versus non-Big 6). We find that the quality level of the successor auditor is related to the quality level of the predecessor auditor irrespective of the reason for the change. Our results suggest that auditor quality levels are, on average, unchanged following a client-initiated change. However, an investigation of any changes in auditor quality subsequent to our classification of client-initiated changes into the three categories suggests that managers' reasons are differentially related to the change in quality of the auditor. Managers who cite structure reasons move up in quality, confirming prior literature. Managers who cite fees as a reason tend to move down in quality, which is different from findings in prior papers. Finally, we find that clients who cite accounting disagreement as the reason for changing also move down in quality, which is consistent with managers searching for less conservative auditors. Our evidence offers additional insights into possible reasons that managers change auditors and may suggest the need for financial reporting regulators to consider increased or improved disclosures related to the reasons for changing auditors particularly in the case of client-initiated changes.

Venlafaxine versus fluoxetine in the treatment of depressed patients with concomitant anxiety

Auditor switching preceded by the receipt of a qualified opinion has received considerable attention in recent years. The concern that such switching may be motivated by opinion shopping (i.e., shopping for an improved audit opinion from a new auditor) has resulted in the issuance of policy statements aimed at controlling potential opinion shopping. Two studies (Chow and Rice 1982; Smith 1986) compared the audit opinions of predecessor and succesor auditors but did not find empirical evidence consistent with opinion shopping. Krishnan (1994) examined the audit opinion decisions for clients who switched and found that auditors treated prospective switchers more conservatively than non-switchers in the year prior to a switch.Our study extends Krishnan (1994) by comparing the audit opinion decisions of the predecessor and successor auditors for clients who switched, relative to auditor's treatment of non-switching clients. Following Krishnan (1994), an ordered probit audit opinion model was estimated. Controlling for clients' financial conditions, the model provides threshold values for auditors' qualification decisions. Differential threshold estimates for switching and non-switching clients were assumed to be attributable to auditors' differential treatment of different clients. The results suggest no difference in the treatment of clients who switched in the pre-switch and post-switch years relative to non-switching clients. In fact, clients who switched were treated relatively conservatively by both predecessor and successor auditors, suggesting either the absence of successful opinion shopping or that the increased switching is not motivated by opinion shopping.

Auditor Resignations and the Market for Audit Services

This paper examines the market for audit services when the incumbent auditor of an SEC registrant has resigned from the engagement. While many previous studies of auditor changes have examined auditor dismissals by the client, only a few studies have specifically focused on auditor resignations. Auditor resignations constitute a unique setting because they may indicate increased likelihood of possible future losses to the new auditor, and provide an opportunity to test the demand-and-supply side incentives in the market for audit services. Results from analyses of 156 auditor resignations and a control sample of 375 auditor dismissals indicate that Big 6 firms were less likely to serve as the successor auditor when the predecessor has resigned, after controlling for three other factors identified as proxies for litigation risk to the auditor (client's financial stress, industry membership and proportion of total assets in receivables and inventory). The effects were especially pronounced for the subset of resignees in financial stress. These results support suggestions that the implications of auditor resignations are different from auditor dismissals, and provide supporting evidence for the suggestions that supply-side incentives should be considered in examining the market for audit services.

A Theory of Auditor Resignation

In this paper we propose an economic rationale for auditors resigning from the engagements of risky clients and use this rationale to examine claims that increases in expected auditor liability are giving rise to an increase in auditor resignations.1 Given the relatively inelastic demand by publicly traded firms for audits, it seems puzzling that auditors would resign rather than simply risk-adjust their bids for the increase in expected liability. In addition, it is not clear why a successor auditor, who faces not only the threat of legal liability but also the information disadvantage accompanying a first-year audit, would accept the rejected client. Our theory explains how an incumbent auditor rationally resigns an engagement because any attempt at risk-adjusted pricing leaves him with only unprofitable clients, and how successor auditors who know less about the client than the incumbent auditor can profitably accept the engagement of a client whose auditor has rationally resigned. Several auditing models have specifically addressed why client firms switch auditors (Magee and Tseng [1990], Dye [1991], Teoh [1992], Kanodia and Mukherji [1994], and Gigler and Penno [1995]). The interest in auditor switches arises partly from regulatory concerns that

Evidence on opinion shopping from audit opinion conservatism

Auditor switching preceded by the receipt of a qualified opinion has received considerable attention in recent years. The concern that such switching may be motivated by opinion shopping (i.e., shopping for an improved audit opinion from a new auditor) has resulted in the issuance of policy statements aimed at controlling potential opinion shopping. Two studies (Chow and Rice 1982; Smith 1986) compared the audit opinions of predecessor and succesor auditors but did not find empirical evidence consistent with opinion shopping. Krishnan (1994) examined the audit opinion decisions for clients who switched and found that auditors treated prospective switchers more conservatively than non-switchers in the year prior to a switch. Our study extends Krishnan (1994) by comparing the audit opinion decisions of the predecessor and successor auditors for clients who switched, relative to auditor's treatment of non-switching clients. Following Krishnan (1994), an ordered probit audit opinion model was estimated. Controlling for clients' financial conditions, the model provides threshold values for auditors' qualification decisions. Differential threshold estimates for switching and non-switching clients were assumed to be attributable to auditors' differential treatment of different clients. The results suggest no difference in the treatment of clients who switched in the pre-switch and post-switch years relative to non-switching clients. In fact, clients who switched were treated relatively conservatively by both predecessor and successor auditors, suggesting either the absence of successful opinion shopping or that the increased switching is not motivated by opinion shopping.