An Analysis Of Firm Level Auditor Tenure On Audit Hours

Abstract

This study provides further evidence with respect to findings reported by Stein et al. (1994) and O’Keefe et al. (1994) on the demand for the various grades of professional labor audit hours. These studies found that demand for audit hours was a function of three sets of client factors: size, complexity, and risk. However, contrary to what might be expected, they did not find that auditor learning had a significant effect. Using a data set consisting of 75 audits of publicly-held companies obtained from four international accounting firms, we test the generalizability of their findings to our more diverse data set and further examine the effects of auditor learning on audit hours. We find the results reported by Stein et al. and O’Keefe et al. generalizable to our data set, indicating their findings to be robust and our samples comparable. Although perhaps not of economic significance, the data also tends to suggest differential auditor learning across various staff levels in a complex relation. That is, the ability to observe learning is dependent on auditee size. There appears to be some learning for partners, managers, and seniors in audits of smaller auditees. However, given the limitations of our data set and consideration of economic significance versus statistical significance, the evidence is not convincingly strong in support of auditor learning on repeat engagements. We recommend further investigation as to whether both an auditor tenure indicator variable and a auditor tenure by auditee size interaction variable simultaneously considered in the model are an appropriate specification of the audit production function to capture auditor learning in a cross-sectional analysis.