An experimental manipulation of divisibility in the public good game

We present the results from two framed field experiments in which members of real communities that share naturally occurring public goods can voluntarily contribute to a common pool which, after being increased by the experimenter, is either divided in individual vouchers (as in standard public good games) to be used for buying private goods, or paid out in group vouchers to be used to buy public goods for the community. We find that contributions are larger in this second case, and when individuals belong to the same community. Furthermore, contributions in our naturally occurring communities tend to grow over time. Finally, contributions in the standard public good game are strongly correlated with contributions to the naturally occurring public good. These results add to the growing literature that investigates the external validity of public good games.