ndependence and objectivity are hallmarks of the auditing profession. However, auditors are subject to economic incentives to please clients that could lead them away from perfect independence and objectivity and potentially result in undesirable conduct, even if they comply with professional standards.
Auditors face two potential risks—failing to detect a material misstatement (a “type II” error) and incorrectly concluding that there is a material misstatement (a “type I” error). Under current standards, the auditor is required to achieve a low level of risk of a type II error, but there is no similar prescription for a type I error. In the study, researchers developed an economic model based on the basic assumptions that even though auditors aim to please clients, they also strive to maintain the costs of an audit as planned and to comply with standards. The study stated that the type II error related to an overstatement of company value, while the type I error related to an understatement.
The analysis showed that if the auditor concluded there was a misstatement (type I error) but the client knew or believed that there was none, then the client would encourage the auditor to do more work so that it was possible to render an unqualified opinion. Simply put, no client wants its company’s value understated. As a result, the auditor worked to eliminate the understatement. However, in order to stay within the budget as planned, the auditor let the overstatement drift as high as possible based on the level considered tolerable by professional standards for a type II error. While this finding appeared counter intuitive, the model confirmed it.
Normally, risks of misstatements below materiality might not be of concern. However, the model showed that for investment portfolios of companies in similar industries these misstatements were not averaged out, as investors might expect. Further, the misstatements persisted even when investors diversified portfolios. This is because the misstatements were all likely to be in favor of the client—overstating the company’s value—making the overall misstatement in the portfolios potentially very large. The researchers suggested that by prescribing the relationship between type I and type II errors in auditing standards, the misstatement can be adjusted. They also advocated specifying the level for type I error rates to encourage auditors to avoid the pitfalls revealed in the research.
For the full text of the research paper, see Auditing: A Journal of Practice & Theory, vol. 18, Supplement, 1999.
J. EFRIM BORITZ, PhD, FCA, CISA, is the Ernst & Young Professor of Accounting and Director of the Center for Information System Assurance, School of Accountancy, University of Waterloo, Ontario, Canada. His e-mail address is firstname.lastname@example.org . PING ZHANG, PhD, is associate professor, Rotman School of Management, University of Toronto, Canada.