Financial crime and punishment : a meta-analysis

We provide the first quantitative synthesis of the literature on how financial markets react to the disclosure of financial crimes committed by listed firms. While consensus expects negative stock price returns, the exact size of the effect is far from clear. We survey 111 studies published over three decades, from which we collect 480 estimates from event studies. Then, we perform a thorough meta-analysis based on the most recent available techniques. We show that the negative abnormal returns found in the literature seem to be exaggerated by more than three times. Hence, the "punishment" effect, including a reputational penalty, suffers from a serious publication bias. After controlling for this bias, negative abnormal returns suggest the existence of an informational effect. We also document that accounting frauds, crimes committed in common-law countries such as the United States, and allegations are particularly severely sanctioned by financial markets, while the information channels and types of procedures do not influence market reactions.