The Mitigation of Hindsight Bias in Judges' Evaluation of Auditor Decisions
Anderson, John C., Jennings, Marianne M., Lowe, D. Jordan, Reckers, Philip M. J., Auditing: A Journal of Practice & Theory
Key Words: Hindsight bias, Legal liability, Debiasing.
Data Availability: The data upon which this paper is based may be obtained from the authors on request.
Hindsight bias refers to an individual's overestimation of the extent to which they would have foreseen the apparent "inevitability" of an outcome (Hawkins and Hastie 1990; Schkade and Kilbourne 1991). Individuals project knowledge of an outcome into the past and are unaware of the effect that this knowledge has on their perceptions (Fischhoff 1975). As a result, hindsight bias has been found to influence evaluations of the competence of the decision maker (Baron and Hershey 1988). Individuals evaluating decision makers may assume (in hindsight) that certain events were potentially foreseeable and reflect that the decision maker (in foresight) should have been able to anticipate an outcome that became clear only retrospectively (Baron and Hershey 1988; Fischhoff 1975; Mitchell and Kalb 1981). The public accounting profession could be significantly and negatively affected by the hindsight bias phenomenon (Berton and Schiff 1990; Kinney 1993, 1994). For instance, in a classic case, hindsight bias could have influenced the Securities and Exchange Commission (SEC) when it censured Arthur Andersen & Co. (see SEC (1981), Accounting Series Release No. AS-292) for alleged audit failure in the audits of Mattel. Arthur Andersen & Co. was charged with failure to discover that the amount of adjustment for obsolete inventory was significantly understated. A description of the SEC's determination of audit deficiencies, as stated in ASR 292, was determined with the benefit of hindsight (Buchman 1985).
Auditors make decisions from an ex ante position relative to an outcome. However, auditor performance is evaluated ex post because individuals have knowledge of an event outcome (e.g., bankruptcy, management fraud, etc.) (Kennedy 1995). Judicial litigants must determine if the auditor exercised "due professional care" in the conduct of an audit. From a hindsight perspective, judges and jurors may be more apt to believe that the auditor breached the standard of care expected of members of the accounting profession and required by professional standards (Kinney 1993). If hindsight bias is found to exist in the legal system and cannot be mitigated, it suggests to the public accounting profession that tort reform should be undertaken to overcome this bias and thereby achieve a more objective evaluation of auditor performance. Some measure of tort reform was enacted by the U.S. Congress in 1995.(1) That legislation, however, was complex and unlikely to be wholly adequate given the magnitude of the problem (Boyle and Knopf 1996; Goldwasser 1997).
The purpose of this research is to examine the effectiveness of two debiasing methods that potentially could mitigate hindsight bias with professionally trained judges. An experiment was conducted in which we established the existence of hindsight bias with judges and then attempted to mitigate it with two individual debiasing methods. Our results indicate that the debiasing method that redirected attention away from the plaintiff (to other stakeholders) was completely effective in mitigating hindsight bias. However, the debiasing method that encouraged judges to consider alternative outcomes was not effective. These findings lead to two main conclusions. First, we provide evidence that judges' hindsight bias can be mitigated in an audit legal liability context. While prior research has established that judges are susceptible to hindsight bias, this study is the first to document its mitigation with these experienced judicial subjects. This is important given the influence of hindsight bias on auditor legal liability and the fact that it is not easily mitigated. Second, the effectiveness of debiasing methods may be dependent on judicial subject group (judges vs. jurors). Although the alternative outcomes debiasing method was effective in mitigating hindsight bias with jurors (see Lowe and Reckers 1994), it was ineffective with judges in our study. …