Audit pricing and the Big4 fee premium: evidence from Belgium

Purpose - The purpose of this paper is to study audit pricing and the Big4 fee premium in Belgium. While a number of studies have already explored these issues, the Belgian audit market provides an interesting setting to gain an additional insight into the pricing of audit services for many reasons (e.g. audit market concentration in Belgium is much lower than in other countries, the Belgian audit market mainly consists of non‐listed firms, etc.). Design/methodology/approach - Besides the traditional audit fee model, based on seminal work by Simunic, the paper also estimates regression models in which the author allows coefficients to vary across Big4 and non‐Big4 auditors and control for self‐selection (based on a two‐stage procedure). Findings - Using the traditional audit fee model, results suggest that Big4 auditors receive (or are able to charge) a fee premium compared to non‐Big4 auditors. Nevertheless, when the author allows regression coefficients to vary across Big4 and non‐Big4 auditors and control for self‐selection, the aforementioned finding does no longer hold. The results reveal differences in fee structures between Big4 and non‐Big4 auditors and suggest that Big4 auditors consider a richer set of variables when setting their fees. Research limitations/implications - Since Belgian firms are only required to disclose audit fees as from 2007 onwards, the analyses are based on data for one year only. Practical implications - An important implication, at least from an academic point of view, is that the results clearly illustrate and corroborate the need to control for self‐selection when modelling audit fees (while this issue has been ignored by recent audit fee studies). The findings also have implications for the (Belgian) auditing profession. For example, the fact that significant differences are observed in audit pricing between the Big4 and non‐Big4 firms may have an impact on the (Belgian) audit services market (e.g. it might influence the competitive nature of the tendering process). Originality/value - Using a two‐stage procedure, the results corroborate the need to control for self‐selection in modeling audit fees (an issue that has been largely ignored in the audit fee literature). In addition, the results reveal that Big4 and non‐Big4 auditors have different fee structures and that it is therefore important to allow the regression coefficients (and not only the intercept) to vary across both groups. Finally, the findings add to the very scarce evidence on audit pricing for non‐listed firms.