Systematic Misperceptions About Pro-Sociality : Experimental Evidence

We comprehensively investigate misperceptions about pro-social behavior and the extent to which punishment and rewarding is under- or overestimated in laboratory experiments. Our games and treatments cover a large range of decision environments, in which there are different (i) trade-offs (e.g., Pareto vs. non Pareto dominated, socially efficient vs. not efficient), (ii) motives for non-selfish behavior (e.g., inequity aversion, reciprocity), (iii) costs for punishing and rewarding, (iv) incentives for belief accuracy, and (v) number of allocations in a given game. Our analysis shows a systematic underestimation of rewards, both at the aggregate and individual level suggesting that people have pessimistic perceptions or ‘mental models’ about human rewarding behavior, irrespective of the decision environment and incentives. In contrast, beliefs about punishment are more accurate suggesting that people anticipate altruistic punishment to enforce social norms. We also observe that women, economists, and more religious subjects underestimate rewards even more than males, noneconomists, and less religious people