DO INSIDERS ACT AS ARBITRAGEURS?

This paper presents new evidence supporting the hypothesis that a large part of the returns generated by corporate trading (insider trading, share repurchases, and SEOs) can be attributed to public information rather than to private information. Corpo-rate trades are contrarian and the magnitudes of returns prior to and following trades increase with idiosyncratic risk. These patterns cannot be explained by proxies for future cash-flow news and are not mitigated by increased regulatory oversight of in-sider trading. Furthermore, arbitrage forces are weak at high-idiosyncratic risk firms: institutions are slow to recognize the valuation implications of these trades and returns following the trades firms drift for several months. Overall, our results suggest tha