A Modified Audit Production Framework: Evaluating the Relative Efficiency of Audit Engagements

Abstract

We develop a model of audit production based on Data Envelopment Analysis (DEA) using labor cost as input and hours spent on evidence-gathering activities that determine the level of assurance as output. Client characteristics are considered exogenous factors that affect audit production as a whole. We apply the model to a sample of U.S.-based engagements from an international accounting firm. Results indicate that a constrained DEA model using variable returns to scale is appropriate for modeling audit production. We find that audits are more efficient for clients that are larger, have a December year-end, and are highly automated. Audits are less efficient when the auditor relies on internal control, tax services are provided, and the client has subsidiaries. We also find that a well-specified regression-based production model can control for factors that influence auditor efficiency. Finally, we find that inefficiencies are impounded in fees for some industries and firm offices