Altruistic Versus Rational Behavior in a Public Good Game

This paper analyses an evolutionary version of the Public Good game of Eshel, Samuelson, and Shaked (1998) in which agents can choose between imitation and best-reply decision rules. We describe conditions under which altruistic and spiteful (maximizing) behavior arise: these conditions are established for any number of neighbors and any total number of agents in the population. Given mistake-free play, (short-run) outcomes are identical whether agents are constrained to employ an imitation rule only; or they can choose between imitation and best-reply rules. Given the possibility of mistakes, (long-run) outcomes vary across these two scenarios. The paper suggests how to provide public goods and gives an explanation of why we observe seemingly irrational cooperation - altruistic behavior - in the rational world.