Beta reversal and expected security returns

This dissertation contains two parts that study the behavior of security returns in financial markets. The first part shows that the failure of the capital asset pricing model (CAPM) beta to predict individual stocks' expected returns, documented by Fama and French (1992), is largely driven by a small group of stocks with large betas and high idiosyncratic volatility. These stocks' betas tend to reverse in addition to beta instability which is embedded in most stocks. Therefore, even when the CAPM holds period-by-period, the cross-sectional evidence on market betas is weak at best due to the confounding effect of beta instability and reversal. The second part further shows that such a beta reversal is partly predictable by idiosyncratic volatility. As a result, the current beta estimates of individual stocks can signicantly explain the cross-sectional difference in future returns when allowing for an interaction between the current idiosyncratic volatility and beta. More importantly, our simple model can predict the historical expected return well. All results are robust with respect to different measures of beta and idiosyncratic volatility as well as different subsamples.;The second part investigates whether or not the under-performance of actively managed mutual funds is attributed to these funds' investment preferences toward stocks with high idiosyncratic volatility. It finds that these funds prefer to hold stocks with higher idiosyncratic volatility. Their portfolios, however, remain exposed to higher undiversified idiosyncratic volatility. The level of undiversification is proportional to these portfolios' aggregate idiosyncratic volatility. Undiversified idiosyncratic volatility has a significantly negative effect on the funds' performance. I conclude that the under-performance of actively managed mutual funds results from fund managers' self-interested actions: gambling to speculate the extremely abnormal return in financial markets and conducting window dressing to improve the appearance of quarterly reports.