عضویت در گروه، اولویت های تیم، و انتظارات

Group membership is a powerful determinant of social behaviour in a variety of experimental games. Its effect may be channelled primarily via the beliefs of group members, or directly change their social preferences. We report an experiment with a prisoner's dilemma with multiple actions, in which we manipulate players’ beliefs and show that group identity has a consistent positive effect on cooperation only when there is common knowledge of group affiliation. We also test the robustness of the minimal group effect using three different manipulations: one manipulation fails to induce group identity, and we observe an unsystematic effect of group membership when knowledge of affiliation is asymmetric.

There is currently a revival of interest among economists in the effect of group membership on individual decision-making.1 It is well known that people tend to behave more pro-socially when they interact with members of their own group, but become less generous, less trusting, and less cooperative towards individuals who belong to different groups. However, there is less agreement about why this happens, and in which conditions group membership has a significant effect. The experiment described in this paper extends research on group membership in two directions. (1) Using a two-person public goods game (or multiple-action prisoner's dilemma), it tries to discriminate between two alternative explanations of group identity effects. Does group membership change people's goals (by, for example, modifying the argument of their utility function) or does it change people's expectations concerning what other individuals will do? (2) The experiment probes the robustness of group effects comparing three different versions of the classic minimal group paradigm (Tajfel et al., 1971). While the answer to the first question appears rather straightforward – the effect of group membership is channelled mainly through people's expectations – the results on the second front are less univocal. Minimal group manipulations appear to be fragile, and have unsystematic effects when knowledge of group membership is asymmetric. In some sessions group identity increases transfers to fellow group members, in some it decreases transfers, and in others it has no effect at all. The paper is organised as follows: Section 2 sketches the theoretical background and briefly reviews the experimental literature. The design of the experiment is illustrated in Section 3, while Section 4 describes and discusses critically the main results. Section 5 concludes with a summary and general comments.

When they choose their contribution level to a public good, subjects tend to contribute more if they are matched with an in-group member than if they are matched with an out-group member. Thus, pro-social cooperative behaviour is affected by group affiliation. Systematic discrimination however takes place only if both subjects have access to information about the group membership of the other player. When knowledge is asymmetric, fully informed participants do not cooperate more with in-group than with out-group subjects. Overall their average level of contribution is statistically indistinguishable from that of partially informed subjects. The message is that the minimal group paradigm acts primarily on individual beliefs, and through this channel modifies behaviour in games of cooperation. All theories that postulate a direct link between group identity and preferences – unmediated by mutual beliefs – are therefore refuted by the evidence. The results of this experiment however raise important questions regarding the robustness of minimal group effects. Out of three attempted manipulations, one certainly failed to induce significant discrimination and must be taken as a warning against attempts to ground group identity on mere random labelling. The other two manipulations successfully replicated the classic in–out effect with common knowledge of group membership, but produced unsystematic effects in the asymmetric conditions. This variance is probably due to subjects’ attempts to apply in an unusual context a behavioural norm that is appropriate for situations where group affiliation is common knowledge among players.