Analytical procedures and audit planning decisions.

Unexpected fluctuations that influence auditors to revise their audit plans.

Auditors perform analytical procedures in planning the nature, timing and extent of testing. Significant fluctuations between the current year's unaudited data and expected results signal an increased risk of material error and help auditors focus their planned tests on high-risk areas. This study examined auditors' decisions to revise preliminary audit plans after analytical procedures performed during planning revealed significant, unexpected fluctuations. Specifically we examined whether the extent of corroboration for management's explanation for the fluctuation and the presence of an explicit incentive for management to misstate the financial statements influenced auditors' decisions to revise their audit plans.

Auditing standards and theory indicate that when management's explanation for a significant unexpected fluctuation is not corroborated during planning, auditors need to revise their audit plan in order to address the increased risk signaled by the unexpected fluctuation. However, based on prior research related to judgment and decision-making theory, we expected auditors would not revise plans unless management had an explicit incentive to misstate financial results.

Sixty-seven practicing auditors participated in an experiment in which they were asked to make audit planning decisions for a hypothetical client. When analytical procedures revealed an unexpected and significant fluctuation, the auditors had the opportunity to revise their preliminary audit plans. The incentive for management to misstate the financial results was high or low, and the extent of corroboration for management's explanation of the fluctuation was either extensive or minimal, resulting in four conditions--high incentive/extensive corroboration; high incentive/minimal corroboration; low incentive/extensive corroboration; low incentive/minimal corroboration. Auditors were randomly assigned to one of these conditions. The results, as expected, were that auditors were more likely to plan for increased testing when there was little corroboration of management's explanation and management had an explicit incentive to misstate financial results.

In the minimal-corroboration condition, when there was no explicit incentive for management to misstate financial results, only 25% of the auditors revised their plans. When management had an explicit incentive and only minimal corroboration was available, only 56% increased planned tests.

The fact that more than 40% of the auditors did not extend tests, given an unexpected fluctuation and little corroboration for management's explanation, raises concerns about whether underauditing may occur even in a relatively high-risk setting. This finding is consistent with concerns expressed by the AICPA in its 1996-1997 Audit Risk Alert. In its discussion of audit problem areas identified through litigation and audit failures, the AICPA indicated that two major concerns were auditors' (1) willingness to accept management's representations without corroboration, and (2) failure to identify risky situations or to apply professional skepticism and revise audit procedures appropriately.

For the full text of the published work, see Auditing: A Journal of Practice & Theory, Fall 2000, vol. 19, no. 2.

STEVEN GLOVER, PhD, is an associate professor at Brigham Young University, Provo, Utah. His e-mail address is [e-mail]. JAMES JIAMBALVO, PhD, is PriceWaterhouseCoopers and Alumni professor at the University of Washington, Seattle. His e-mail address is [e-mail]. JANE KENNEDY, PhD, is an associate professor and Gregory Fellow at the University of Washington, Seattle. Her e-mail address is [e-mail].