Implied Belief Extrapolation (Contrarianism) : Evidence from Mutual Fund Holdings

Behavioral finance lacks direct evidence linking agents’ beliefs to their behaviors. Motivated by revealed preference theories, I provide such evidence using mutual fund stock holdings. Assuming that investors' expectations are consistent with their investment decisions, I relate fund managers’ stock holdings to past factor returns and document four facts about belief formation: 1. In contrast to belief extrapolation, a substantial fraction of mutual fund managers act as contrarian investors who expect lower factor returns after a good factor performance; 2. Distinguished from fund style investment strategy, fund catering strategy and fund risk preference, the two trading patterns imply fund manager expectation biases, which are either extrapolative belief or contrarian belief. Both extrapolator and contrarian fund managers overweight more recent factor returns then distant factor returns when forming their expectations; 3. The degree of contrarianism is highly correlated with future fund performance. The top (contrarian) managers outperform the bottom (extrapolative) managers by a return of 3.4% per annum after adjusted by FFC4 factor models; and 4. Exploring fund characteristics, I find contrarian managers are more experienced, charge higher expense ratios, and manage smaller US equity funds. Overall, my finding provides evidence for belief contrarianism---the opposite of belief extrapolation---as a general trait that is associated with superior investment management performance