No evidence of inequality aversion in the investment game.

We report experimental evidence on second-movers' behavior in the investment game (also known as the trust game) when there exists endowment heterogeneity. Using a within-subject analysis, we investigate whether or not second-movers exhibit some taste for inequality aversion by returning a larger (smaller) share of the available funds to first-movers who are initially endowed with a lesser (larger) endowment, respectively. Our data suggest that second-movers do not take into consideration the level of endowments when making their decisions as their behavior is consistent across distribution of endowments; i.e., they return the same proportion of the available funds regardless of the endowments. We indeed find that some second-movers have a tendency to return what they have received from first-movers. In our data, there is also a substantial proportion of second-movers who are selfish and return nothing. (JEL Codes: C72, C91, D3, D63).

The emergence of altruism as a social norm.

Expectations, exerting influence through social norms, are a very strong candidate to explain how complex societies function. In the Dictator game (DG), people expect generous behavior from others even if they cannot enforce any sharing of the pie. Here we assume that people donate following their expectations, and that they update their expectations after playing a DG by reinforcement learning to construct a model that explains the main experimental results in the DG. Full agreement with the experimental results is reached when some degree of mismatch between expectations and donations is added into the model. These results are robust against the presence of envious agents, but affected if we introduce selfish agents that do not update their expectations. Our results point to social norms being on the basis of the generous behavior observed in the DG and also to the wide applicability of reinforcement learning to explain many strategic interactions.

Humans expect generosity.

Mechanisms supporting human ultra-cooperativeness are very much subject to debate. One psychological feature likely to be relevant is the formation of expectations, particularly about receiving cooperative or generous behavior from others. Without such expectations, social life will be seriously impeded and, in turn, expectations leading to satisfactory interactions can become norms and institutionalize cooperation. In this paper, we assess people's expectations of generosity in a series of controlled experiments using the dictator game. Despite differences in respective roles, involvement in the game, degree of social distance or variation of stakes, the results are conclusive: subjects seldom predict that dictators will behave selfishly (by choosing the Nash equilibrium action, namely giving nothing). The majority of subjects expect that dictators will choose the equal split. This implies that generous behavior is not only observed in the lab, but also expected by subjects. In addition, expectations are accurate, matching closely the donations observed and showing that as a society we have a good grasp of how we interact. Finally, correlation between expectations and actual behavior suggests that expectations can be an important ingredient of generous or cooperative behavior.

Social preferences and cognitive reflection: evidence from a dictator game experiment.

This paper provides experimental evidence on the relationship between social preferences and cognitive abilities, which we measure using the Cognitive Reflection Test (CRT). We elicit social preferences by way of 24 dictatorial situations, in which the Dictator's choice sets include (i) standard Dictator games, where increasing the Dictator's payoff yields a loss for the Recipient, (ii) efficient Dictator games, where increasing the Dictator's payoff also increases that the Recipient's; as well as other situations in which (iii) either the Dictator's or (iv) the Recipient's monetary payoff is held constant. We partition our subject pool into three groups: reflective (scoring 2 or more in the CRT), impulsive (opting twice or more for the "intuitive" but wrong answers in the CRT) and the remainder. We find that impulsive Dictators show a marked inequity aversion attitude, especially in standard Dictator Games. By contrast, reflective Dictators show lower distributional concerns, except for the situations in which the Dictators' payoff is held constant. In this case, reflective Dictators give significantly more.