Résumé
We run a series of experiments in which subjects have to choose their level of contribution to a pure public good. The design differs from the standard public good game with respect to the decision procedure. Instead of deciding simultaneously in each round, subjects are randomly ordered in a sequence which differs from round to round. We compare sessions in which subjects can observe the exact contributions from earlier decisions ("sequential treatment with information") to sessions in which subjects decide sequentially but cannot observe earlier contributions ("sequential treatment without information"). The results indicate that sequentiality increases the level of contribution to the public good when subjects are informed about the contribution levels of lower ranked subjects while sequentiality alone has no effect on contributions. Moreover, we observe that earlier players try to influence positively the contributions of subsequent decision makers in the sequence, by making a large contribution. Such behaviour is motivated by the belief that subsequent players will reciprocate by also making a large contribution. We also discuss the effect of group size on aggregate contributions. Finally, we conceptualize a model where agents’ preferences incorporate a “weak” moral motivation element. The moral motivation is “weak” in the sense that contributors update their morally ideal level of contribution according to observed behaviours. This suggested qualification of rational contributors fits well with the patterns observed in the lab.