Free Riders and Cooperators in Public Goods Experiments: Can Evolutionary Dynamics Explain their Coexistence?

An oft-cited and robust result from Public Goods Game experiments is that, when subjects start playing, the aggregate level of contributions is significantly different from zero. At the same time, a sizeable proportion of players free ride from the outset. Behavioural economics has persuasively shown that these laboratory findings are compatible with the presence of motivationally heterogeneous agents, displaying both standard, self-centred preferences and non-standard, interdependent preferences. However, at the theoretical level, economists would prefer to account for motivational heterogeneity endogenously, instead of simply assuming it from the outset. Our work provides such endogenisation, by assuming that social evolution is driven by material payoffs only. By separately focusing on different types of ‘experimentally salient’ pro-social players (such as Reciprocators, Strong Reciprocators and Altruists), we are able to shed light – to our knowledge, for the first time, within the public good framework – on the evolutionary stability of two-type populations consisting of positive proportions of both ‘nice’ and ‘mean’ guys.