Helping the meaner, hurting the nicer: The contribution versus distribution game

Wide experimental evidence shows that people do care about their opponents' payoff during social interaction. Our research aims to shed light on the relative importance of different motives in non-selfish choices highlighted in the recent literature. After a standard public-good game, one player is given the possibility to increase or decrease his opponent's payoff. While our baseline treatment replicates the tendency to hurt richer but lower-contributing players and help poorer but higher-contributing players, if we add exogenous assignments we find substantial willingness to hurt the rich, even if they have contributed more, and to help the poor, even if they have contributed less. These results show a greater focus on correcting inequality than on punishing or rewarding particular behavior. Moreover, we also find that subjects disregard efficiency, in terms of the overall "pie" to be shared. Overall, our data support inequality aversion as a more robust phenomenon than reciprocity and efficiency considerations.