John Beshears, James Choi, David Laibson, and Brigitte Madrian
Several experiments have found that subjects are more willing to invest in risky assets with positive expected returns if only aggregated returns are reported to them, rather than the individual component returns. Information aggregation along various dimensions produces this effect: reporting subjects’ portfolio return over the last n ≥ 2 periods once every n periods rather than reporting single-period returns each period, reporting subjects’ portfolio-level returns rather than returns for each individual asset separately, or reporting historical long-horizon return distributions of asset classes rather than historical one-year return distributions of asset classes. These results are consistent with subjects suffering from myopic loss aversion, which is the combination of loss aversion and mental accounting.
The strength and consistency of the experimental results constitute compelling evidence that myopic loss aversion is a real psychological phenomenon that responds to aggregation manipulations. In this project, we conducted an experiment to try to answer a related but separate question: would a financial institution increase the portfolio risk-taking of its clients if it started disclosing returns at a more aggregated level outside the laboratory? We recruited 600 subjects from the general US adult population to participate in a year-long study. Each subject was allocated $325 among four real mutual funds. We paid each subject whatever the $325 would have been worth at the end of the year if the money had been invested according to his or her choices.
Beshears, John, James Choi, David Laibson, and Brigitte Madrian. Does Aggregated Returns Disclosure Increase Portfolio Risk-Taking? 2012.