Aggregate Mutual Fund Flows and Cross-Sectional Anomalies

Abstract

I examine the explanatory power of aggregate mutual fund flows for the profitability of price-based (i.e., momentum and 52-week high) and non-price-based (i.e., earnings surprises, profitability, share issuance, accrual and asset growth) anomalies in the cross-section of returns. I find that the flow-based trading of mutual funds contributes to mispricing as measured by the profits to price-based anomalies, especially at times when market-wide funding costs are high. The effect also exists for non-price-based anomalies, but only through the dependence of their profits on momentum. My findings support the view of Lou (2012) and Vayanos and Woolley (2013) that mutual funds’ trading on flows creates feedback that strengthens price-based anomalies, as high-performing funds buy additional shares of high-performing stocks and poorly performing funds sell shares of poorly performing stocks. However, the explanatory power of aggregate mutual fund flows for price-based anomaly returns is only partly attenuated by fund-level variables designed to capture the feedback effect. The flow-induced trading by mutual funds appears to contribute to mispricing for reasons beyond the feedback effect.