Predicting Bankruptcy After The Sarbanes-Oxley Act Using Logit Analysis

Abstract

Our study proposes firm bankruptcy prediction using logit analysis after the passage of the Sarbanes-Oxley (SOX) Act using 2008-2009 U.S. data. The results of our logit analysis show an 80% (90% with one year before bankruptcy data) prediction accuracy rate using financial and other data from the 10-K report in the post-SOX period. This prediction rate is comparable to other data mining tools. Overall, our results show that, as compared to the prediction rates documented by other bankruptcy studies before SOX, firm bankruptcy prediction rates have improved since the passage of SOX. Our findings shed light on the benefits of SOX by providing evidence that legislation makes the financial reporting more informative. This study is important for regulators to implement public policy. Investors may be interested in our findings to better assess company risk when making portfolio decisions.