Auditor Tax Advisory Services and Clients’ Tax Avoidance: Do Auditors Draw a Line in the Sand for Tax Advisory Services?

• In Event: Research in Taxation

Abstract

A resurgence of consultancy as a major area of revenue growth for audit firms as well as incidents of audit firms involvement in clients’ tax aggressive activities have reignited concerns about auditors’ provision of non-audit services. Audit industry regulators have taken or are contemplating actions to limit the provision of non-audit services (e.g., E.U. 2014 Directive on audit service). However, prior research shows that non-audit services result in beneficial knowledge spillover that enhances audit quality. A prohibition or sever restrictions on non-audit services can have negative consequences for audit quality and clients’ financial performance. Current PCAOB regulations restrict audit firms from providing tax aggressive strategies to their publicly-listed clients. In our study, we provide large-sample, empirical evidence that current regulations constrains the association between auditor tax advisory services and clients’ level of tax avoidance. Using non-linear regression models, we document a negative but declining association between tax fees paid and clients’ effective tax rates (ETR). Further, we fail to find a significant association between APTS and effective tax rates for the most tax aggressive clients. Based on the coefficients from our models, we estimate the relation does not extend below the eighteen percent ETR line. These results are consistent with auditors finding it increasingly challenging to design tax strategies that lowers a client’s effective tax rate while remaining non-aggressive. Eventually, the audit firm exhausts all available non-aggressive tax strategies and stops providing additional tax advisory services, either for regulatory, reputational, or litigation concerns. Our findings are robust to alternative research designs including long-run measures of tax avoidance, segmenting the sample into quartiles, and quantile regressions. The results of our study should be informative to regulators and other stakeholders in their debate on the benefits and costs of non-audit services.

Authors

• Joanna Shaw, Michigan State University

• Wayne L Nesbitt, Michigan State University

• Anh V Persson, Michigan State University