Content area
Full text
Abstract
I develop an agency model where returns-chasing behavior by mutual fund investors causes beta not to be priced to the degree predicted by the standard CAPM. Mutual fund investors chase returns through time, precipitating unusually large aggregate cash inflows into mutual funds just after dramatic market runups. Mutual fund investors also chase returns cross-sectionally across funds so that the highest-performing funds capture the largest fraction of the aggregate inflows into the mutual fund sector. The interaction of these two flow-performance relationships induces an asymmetry in payoffs to mutual funds where fund managers care most about outperforming peers during bull markets. Since high-beta stocks tend to outperform in up markets, active fund managers tilt their portfolios toward high-beta stocks, reducing the beta risk premium in equilibrium. To support the model's time-series flow-performance assumption, I show empirically that market returns have a large economic impact on subsequent aggregate mutual fund flows. In addition, data on mutual fund holdings suggest that the aggregate stock portfolio held by equity funds is overweighted in high-beta stocks relative to the overall market, though this does not include the cash held by mutual funds. Fama-MacBeth tests indicate that the equity premium falls only slightly as the relative size of mutual funds increases, and the relation is not statistically significant.
I. Introduction
One of the most interesting puzzles in investments is the lack of empirical support for the Capital Asset Pricing Model (CAPM). According to the CAPM, the relationship between risk and return is captured by a linear function where risk is measured by beta. In addition, the slope of this security market line should equal the market risk premium. Black, Jensen, and Scholes (1972) find that the slope of the empirical security market line is flatter than r^sub m^ - r^sub f^. Using more recent data, Fama and French (1992) conclude that after controlling for size, "the relation between market beta and average return is flat," prompting the financial press to announce that beta is dead. Chan and Lakonishok (1993) find that if beta has in fact died, it did so around 1983. Beta's ineffectiveness led Fama and French (1993) to augment the traditional single-factor model with two additional priced factors based on the book-to-market effect and the size...