How Do Prior Gains and Losses Affect Subsequent Risk Taking? New Evidence from Individual-Level Horse Race Bets

Abstract

Many empirical studies focusing on how prior outcomes affect subsequent risk taking report conclusions that appear mutually contradictory. While some studies document increased risk taking after gains and risk aversion after losses, others report the opposite. This paper contributes to this literature by investigating horse race bettors’ sequential risky decisions using individual-level data. A horse race betting market is a laboratory-like environment and can be used to analyze behavioral patterns associated with decision making under risk. We find evidence for (i) the “house money effect” as bettors take riskier wagers after gains and mostly spend the money they have won; (ii) risk aversion after prior losses, which we label a “playing safe effect”; and (iii) a preference for breakeven. Contrary to the widely held conception in the empirical literature, our findings suggest that the “break-even effect” does not necessarily imply an increased preference for riskier bets because bettors may seek to break even with less risky wagers. The online appendix is available at https://doi.org/10.1287/mnsc.2016.2679 . This paper was accepted by Uri Gneezy, behavioral economics.