Do nonaudit services compromise auditor independence? Further evidence

This paper challenges the findings of Frankel et al. (2002) (FJN). The results of our discretionary accruals tests differ from FJN's when we adjust discretionary current accruals for firm performance. In our earnings benchmark tests, in contrast to FJN we find no statistically significant association between firms meeting analyst forecasts and auditor fees. Our market reaction tests also provide different results than those reported by FJN. Overall, our study indicates that FJN's results are sensitive to research design choices, and we find no systematic evidence supporting their claim that auditors violate their independence as a result of clients purchasing relatively more nonaudit services.