When using tools that they perceive to be more sophisticated than others, auditors are more likely to bias their judgments in favor of what they believe their supervisors want, in contrast to when they are not using these solutions, according to a recent study published in the Journal of Emerging Technologies in Accounting.
Researchers studied a pool of 84 external financial statement auditors recruited through the professional network of one of the authors. Their task was to make a preliminary determination of how reasonable the fair value of a client’s business unit was as part of a goodwill impairment test. While the case materials indicated that management had determined that the fair value of the business unit supported the valuation of goodwill without impairment, the case contained seeded errors and inconsistencies among certain assumptions that implied that the analysis was not comprehensive, resulting in an overstated fair value.
Auditors were told that the engagement partner thought previous audits were either overly skeptical or not skeptical enough. They were also told that they would be using an analytics tool that was either more or less sophisticated than others on the market.
When auditors understood that their supervisor thought previous audits were overly skeptical, and they were using a tool that they believed was more sophisticated than the rest, the study found that they were more likely to overweigh evidence in support of engagement partner preferences (in this case, a more gentle, less skeptical audit). This effect, however, did not appear when auditors were using the same tool but were told it was less sophisticated than others.
“We show that the impact of knowledge of the preferences on auditor’s judgments regarding complex estimates is amplified by perceived tool sophistication. That is, when a supervisor expresses a concern that auditors have been overly sensitive to evidence suggesting impairment of a complex estimate, the availability of a more sophisticated tool will lead auditors to believe that the supervisor will evaluate them more favorably if they reach a client-preferred conclusion. … However, when a supervisor expresses concern that auditors have been insufficiently sensitive to evidence suggesting that a misstatement exists, auditors’ sensitivity to loss will dominate, resulting in no impact of perceived tool sophistication on auditor judgments,” said the study’s conclusion.
The researchers believe the experiment showed the “halo effect” at play, where positive impressions of a person, company, brand or product in one area positively influence one’s opinion or feelings in other areas that may or may not be related. With this in mind, said the researchers, while sophisticated data analytic tools have the potential to improve audit quality, the results suggest that perceptions of sophistication may also unintentionally bias auditor judgments, which may undermine the benefits of these tools.
“Thus, firms should carefully consider how they promote the sophistication of their new technology tools, as well as provide additional training to their auditors to help them distinguish characteristics of a tool that are relevant to audit judgments versus those that are not,” said the paper.