The Persistence of Risk-Adjusted Mutual Fund Performance

Abstract

This paper examines mutual fund predictability for common stock funds, using a sample free of survivorship bias, andmeasures performance using risk-adjusted returns. We reconfirm the hot hands result that high return can predict high return in the short run. Like Hendricks, Patel and Zeckhauser, we find that past performance is predictive of future risk-adjusted performance in both the short run and longer run. Furthermore, when we utilize modern portfolio theory (MPT) techniques to allocate capital among funds, we can construct a portfolio of funds based on prior data that significantly outperforms a rule based on past rank alone and that produces a positive risk-adjusted excess return. In addition, we demonstrate the improvement in performance using MPT by selecting a combination of actively managed portfolios that has the same risk as a portfolio of index funds but has higher mean return. While consistent with past studies, our study finds that expenses account for only part of the differences in performance across funds. We find that there is still predictability even after the major impacts of expenses have been removed. Throughout our study we are able to construct portfolios of funds that have small but statistically significant positive risk-adjusted returns during a period where mutual funds in general had negative risk-adjusted returns.

JEL Classification: D89