Further Evidence on Auditor Selection Bias and the Big 4 Premium

Abstract

In recent years, the competitiveness of the corporate audit market has received a great deal of attention from policy makers and academic researchers alike. Among the main issues of concern is whether large auditors command a premium when setting fees for statutory audit services, and whether this is symptomatic of a lack of competition in the market for audit services or results from differences in the quality of the product offered by the big 4. A large number of academic studies based on independent data sets find a positive OLS coefficient on a large auditor binary variable in audit fee regressions and interpret this as evidence of a premium. However, recent research on UK private companies suggests that the large auditor premium is explained by auditor self-selection bias and that when this is controlled for using a two-stage Heckman procedure, the premium vanishes. In this paper we examine some of the difficulties in properly specifying the audit fee equation and discuss potential sensitivity of the estimates provided by the two-step model. We re-estimate audit fee equations for over 36,000 UK private companies employing a relatively new development in the applied econometrics literature – propensity score matching. In addition, we employ formal decomposition methods, which have not been used in the audit literature to date, to provide a more comprehensive analysis of big 4 premiums. Our results suggest that evidence of the large auditor premium vanishing when selection bias is controlled for do not seem to generalise and that the Heckman two-step procedure is highly sensitive to model specification. Matching results suggest that auditees of similar size, risk and complexity pay significantly higher fees to big 4 auditors.