Estimating Grouped Patterns of Heterogeneity in Repeated Public Goods Experiments

We revisit the role of social preferences and beliefs in voluntary cooperation. It is well documented that people exhibit heterogeneous patterns of behavior in public goods experiments. We estimate grouped patterns of heterogeneity in two such experiments by making use of recently developed methods to detect unobserved parameter heterogeneity in panel data. Our results suggest that, in the repeated game, a rather large proportion of players is willing to invest strongly in cooperation, matching their beliefs essentially one to one. Interestingly, this group of "in-game cooperators" also includes some individuals which are classified as free riders by traditional player typologies.