CEO Director Connections and Corporate Fraud

Abstract

We study the propensity of firms to commit financial fraud using a sample of SEC enforcement actions from 2000 to 2006. Controlling for several rm characteristics as well as year and Fama-French 48-industry effects, we find a significant relation between fraud probability and CEO-board connectedness. The nature of this relation depends on the institutional origin of the connection. While nonprofessional connectedness due to shared educational and non-business antecedents increase fraud probability, professional connections formed due to common prior employment decrease fraud. The positive effects of professional connectedness are pronounced only when individuals share service as executives. The impact of professional-connections persists after the 2002 Sarbanes-Oxley Act while nonprofessional connections lose significance after SOX. Our results suggest that social ties matter and they can have very different effects depending on the institutional context in which the ties are formed. The results suggest heterogeneity in monitoring even within the set of independent directors and support a complementary collaborative board perspective in which directors not only monitor but also provide advice and counsel to CEOs.