Mutual Fund Flows and Performance in Rational Markets

TLDR

The absence of persistent active‑management returns does not mean managers lack ability, that performance information is useless, or that chasing performance is futile. The authors develop a simple rational model to benchmark the relationship between returns and fund flows. The model shows that active managers cannot beat passive benchmarks due to competitive capital markets and decreasing returns to scale, and that a strong link between past performance and fund flows eliminates predictability. The model explains many perceived anomalies, shows past performance cannot predict future returns or skill, and finds that about 80 % of active managers have sufficient skill to cover their fees.

Abstract

We develop a simple rational model of active portfolio management that provides a natural benchmark against which to evaluate observed relationship between returns and fund flows. Many effects widely regarded as anomalous are consistent with this simple explanation. In the model, investments with active managers do not outperform passive benchmarks because of the competitive market for capital provision, combined with decreasing returns to scale in active portfolio management. Consequently, past performance cannot be used to predict future returns, or to infer the average skill level of active managers. The lack of persistence in actively managed returns does not imply that differential ability across managers is nonexistent or unrewarded, that gathering information about performance is socially wasteful, or that chasing performance is pointless. A strong relationship between past performance and the flow of funds exists in our model: Indeed, this is the market mechanism that ensures that no predictability in performance exists. Choosing parameters to match the flow-performance relationship and survivorship rates, we find these features of the data are consistent with the vast majority (80%) of active managers having at least enough skill to make back their fees.