Jensen’s Alpha and the Market Timing Puzzle

Abstract

Theory predicts that market timing in managed portfolios biases Jensen’s alpha. However, empirical studies have failed to find evidence this bias actually exists. We tackle this puzzle by showing via a nested model approach and various simulations that, for the bias to become economically relevant, its components, the extent of timing activity and market conditions, must be extreme. Empirically, however, such conditions rarely occur, explaining why the bias does not appear in the data. In a comprehensive empirical analysis of US mutual funds, we find that measures of total performance that allow for timing activities are virtually identical to Jensen’s alpha. Hence, the key takeaway of this paper is that Jensen’s alpha is a sufficient measure of total performance, even in the presence of timing.

Appendix available at https://ssrn.com/abstract=2877173

Keywords: Mutual fund performance; stock selection; market timing; total performance; conditional performance

JEL Classification: G11, G23