Essays on the trading behavior of institutional investors and stock return anomalies

In the first chapter, I examine the effect of information networks on the trading behavior of mutual funds and on stock returns. An information or stock ownership linkage between two mutual funds is defined by large positions in the same stock. Mutual funds trade together with other funds in their information network after controlling for the overall trading behavior of the mutual fund sector. The effect is robust and cannot be explained by style investing or geographic location. The paper also examines the effect of the structure of information networks on stock returns and stock volatility. Using network density as a measure for the speed of information diffusion in a network of investors, I find that stocks with a lower network density demonstrate stronger return momentum at medium horizons and stronger return reversal in the long run. The evidence is consistent with the gradual information diffusion model of Hong and Stein (1999). Finally, I provide empirical evidence in support of recent theoretical models that study the asset pricing implications of social networks. I show that centralized information networks lead to a higher volatility of individual stocks in the cross-section and also explain the variation in average stock idiosyncratic volatility over time.;The second chapter focuses on the effect of the firm's information network on underreaction to earnings news. Using size-adjusted network density as a measure for the speed of information diffusion, I find that the underreaction to earnings news increases with decrease in network density. This association between network density and underreaction to earnings news becomes stronger with increasing information uncertainty and is insignificant for the stocks with low information uncertainty. The findings provide evidence for a gradual information diffusion based explanation for underreaction to earnings news with an assumption that flow of private information is required to ascertain the effect of new public information on the firm's value.;In the third chapter (co-authored with Martijn Cremers), we examine the effect of institutional investors' investment duration on the efficiency of stock prices. Using a new duration measure based on quarterly institutional investors' portfolio holdings, the presence of short-term institutional investors can help explain many of the best-known stock return anomalies, possibly because these investors are affected by behavioral biases like overconfidence. Specifically, we find that both momentum returns and subsequent returns reversal are much stronger for stocks with greater proportions of short-term institutional investors. The accruals and share issuance anomalies are also stronger for stocks held primarily by short-term institutional investors. Finally, short-term institutional investors do not seem to recognize the benefits of significant R&D increases, as they tend to under-react to these increases.