Separation of the tax and audit function: Motivation and consequences

This study examines companies choosing to separate the tax and audit function prior to and subsequent to the passage of the Sarbanes-Oxley Act ( the Act ). While the provision of tax services is not currently prohibited by the Act, some companies have separated these two functions. An examination of the relationship between tax function separation and audit quality, financial statement visibility and tax minimization proxies is performed for two different samples: (1) a matched pair design for companies engaging their audit firm for tax work in 2001 (2 years before the Sarbanes-Oxley Act was placed in effect) and then separating the tax and audit function by 2003 (the effective year of the Sarbanes-Oxley Act), and (2) a logistic regression analysis on all companies in the Russell 2000, a major market index, for 2002-2004. The data used in this study is compiled from proxy statements filed with the Securities and Exchange Commission and commercially available information from I/B/E/S, COMPUSTAT and AuditAnalytics. After controls for size, industry and growth are added to the model, the results reveal that in some years separation companies pay significantly less for their audit and, as a consequence, have higher instances of financial statement restatement. These results support the idea that the integration of audit and tax provide a better quality audit, supporting prior empirical work of Kinney et al. (2004) and anecdotal evidence from accounting firms. Some evidence for 2004 exists that separation companies are actually becoming less visible in the capital markets. Given that audit quality, analyst following, share turnover and management ownership are decreasing, this implies that these companies are less concerned with financial reporting quality. Overall, these findings suggest that tax service separation may be motivated by factors beyond merely shopping for the best tax provider