Determinants of the Mix of Audit Procedures: Key Factors that Cause Auditors to Change What They Do

Abstract

Our research objective is to identify characteristics of clients and of audit engagements that are associated with changes in the mix of audit procedures performed by auditors. The conceptual basis for the research is simply that profit maximizing auditors should vary the nature of procedures performed and time spent in performing procedures as marginal benefits and/or costs change. However, prior research has found that audit programs generally tend to be quite rigid. In this archival study, we examine the determinants of the number of audit hours devoted to the performance of 34 common audit procedures over four audit phases - planning, risk assessment, substantive testing, and completion - for a cross-section sample of 113 audits of Dutch companies in 1998/99 by 14 public accounting firms, including a cluster of 56 audits performed by two of the (then) international Big 5 firms. We find that the length of auditor tenure with a client increases audit hours, while a high risk of management fraud weakly increases audit hours - but without changing the mix of procedures. A high level of client pressure on audit time and fees decreases audit hours, particularly in the areas of planning and risk assessment. Use of modern, strategic systems (business risk) based audit approaches, considerably changes the mix of procedures, but these approaches are applied quite differently by Big 5 vs. non-Big 5 firms. Likewise, the degree of auditor-reliance on clients' internal controls changes the mix of procedures, but the effects are quite different for Big 5 vs. non-Big 5 firms. The joint performance of management advisory services (MAS) for an audit client results in relatively less audit time in planning and risk assessment, but does not change total audit hours, suggesting that any benefits from 'knowledge spillovers' from MAS are retained by auditors, and hence may not be detectable in total hours or in audit fees. Finally, our results are consistent with the notion that the details of audit production by Big 5 firms are both more reasonable and predictable than audits by non-Big 5 firms. This can explain why Big 5 audits are more highly valued in the marketplace.