감사판단의 사후평가시 기대차이, 직무책임 및 사후편견의 영향

Regulatory reviews on audit reports by the Financial Supervisory Service (the “FSS”) are generally believed to have improved the overall accounting and audit quality in Korea. Under the FSS's review program, the FSS staffs select samples of public companies listed in the Korea Stock Exchange, and evaluate the financial statements for any potential misstatements. However, there have been ever growing concerns among the accounting profession over the fairness and objectivity of the FSS's disciplinary decisions. In particular, one important issue with respect to the review procedure is a potential bias in the ex post evaluation on audit judgment. Despite the concerns about such bias in the accounting profession, little evidence that either refutes or validates the allegation is currently available. We attempt to fill this void. Prior studies report a potential bias in the third party's evaluation of audit results. For instance, Anderson et al. (1993) report that one such a bias is so-called hindsight bias. Hindsight bias exists when individuals overestimate the extent to which outcomes could have been anticipated prior to their occurrence. That is, individuals judge audit quality (e.g., the likelihood of auditor negligence) in light of ex post adverse outcomes (e.g., unfavorable subsequent consequences or undetected material misstatements of earnings). One common and legitimate concern therefore is that adverse outcomes could bias individuals, causing overly harsh conclusions about the likelihood of auditor negligence. Also, Anderson et al. (1993) argue that the evaluation may be biased by a factor such as expectation gap. However, while hindsight bias may contribute to the professional legal disputes, this analysis alone may not resolve all issues since many studies attend on other biases by expectation gaps deriving from prior belief on estimated error occurrence (Peecher and Piercey 2008), or accountability difference by respondent's responsibility or roles (Kennedy 1995). Previous studies report that adverse outcome tends to cause lower evaluation of audit quality due to hindsight bias. Also, the higher the expectation for auditors' responsibility or audit standards to be accomplished is, the more stringent evaluations on audit opinions tend to be. Similarly, more accountable respondents are more likely to be assigned to tough evaluations on audit results. In addition, Peecher and Piercey (2008) find that the higher prior belief on error occurrence is positively associated with the lower level of hindsight bias of the jury ("the reverse hindsight bias"). Several prior studies attempt to design debiasing methods that could be implemented in the legal liability context, given that judicial litigants may be unable to ignore the knowledge about outcomes. We investigate whether such a bias actually affects the professional evaluation on the auditor opinion in the course of the FSS’s review procedures. Specifically, based on the findings in the prior literature, we posit that bias arises from several factors: outcome knowledge (hindsight bias), expectation gap (prior belief on estimated occurrence of error), and accountability difference (i.e. regulatory reviewer's challenging positions vs. auditor's defending positions). We conduct an experimental field survey using 44 staff reviewers of the FSS and 107 auditors of various accounting firms, who are the counterparts involved in the FSS's review procedures. Our results indicate that the FSS reviewers’ evaluation on the auditors’ performance is significantly lower than that of auditors, which is reflective of an accountability difference. Moreover, the higher prior belief on error is, the lower evaluation of audit quality is, which is also reflective of an expectation gap. Perhaps more importantly, our results reveal that evaluative judgments of auditors’ performance are subject to hindsight bias. These results together indicate that professionals including both of auditors and the FSS reviewers are affected by hindsight bias, expectation gap, and accountability in the third party evaluation of audit results. In addition, we find that higher prior belief on estimated error occurrence is positively associated with lower hindsight bias (the reverse hindsight bias). However, we do not find the reverse hindsight bias in the FSS reviewer's evaluation, which indicates that FSS reviewers tend to evaluate audit results lower than auditors, and their level of hindsight bias is similar with that of the auditor. Additional analyses also show that longer experience in the years of audit or regulatory review tends to mitigate the bias affecting the ex post evaluation of audit results. This suggests that longer experience can be used as debiasing method to mitigate the bias. Our study has several contributions. First, most of prior studies on hindsight and other bias are based on the surveys or experimental studies using jury, judge, or auditors. In contrast, our study is based on a field survey on auditors and the FSS reviewers who are the actual counterpart of the FSS review procedures. Second, we find that the hindsight bias, expectation gap and accountability difference together actually affect the judgment of the professionals participating in the FSS review process. Lastly, we find that the audit or the FSS review experience reduces the level of such bias, and therefore can be useful as a debiasing method that mitigates biases.