An Experimental Study of Individual Risk and Altruism Within a Collective

• In Event: Experiments in Lab-Style Games

Abstract

How does the risk of punishment affect individuals' decisions as to whether or not to participate in costly collective actions, such as protests or social movements? Participation in a collective action is often viewed through the lens of conflict: one's individual interest in not paying the costs of participation faces off against the benefit to the collective that accrues if enough people participate. Since the benefits of collective action, such as policy or even regime change, tend to be public and non-excludable, free-riding is attractive: one can gain the benefits of the collective action without paying the costs. In order to explain the incidence of particularly costly revolutionary protests, such as in the Arab Spring and the 1989 Soviet Union Revolutions, literatures across disciplines have largely focused on explanations that address variation in the benefits of collective action. Variation in the costs of acting, however, have seen less attention. In particular, variation in the costs of repression have not been analyzed with the same attention to individual beliefs and behavior.

This paper aims to add to our understanding of repression and collective action via a behavioral game-theoretic model and an incentivized laboratory experiment designed to test the conclusions of the model. The model incorporates both collective action and punishment/repression, disaggregates costs from punishment into those affecting only oneself and those affecting others in one's group, and makes use of pro-social preferences. The incentivized experiment tests nearly all the comparative statics of the model, and additionally explores differences in the form of information on punishment risks passed to experimental participants.

The combination of model and experiment allows identification of how variation in individuals' perceptions of individual versus collective risk and altruistic tendencies translate into their likelihood of participation in a costly collective action, as well as the likelihood that the collective action will be successful in its goals. Those questions are of central substantive and intellectual importance: successful collective actions can lead to substantial policy or even regime changes, and at times may be required for such changes. Yet, while we might expect higher expected costs of repression to reduce participation, we don't have a strong theoretical understanding of, or strong quantitative empirical support regarding, the manner by which people respond to variation in the degree to which they or others are vulnerable to repression. That prevents us from answering important questions. For instance, do individuals care about repression that would be applied to other members of their community, even if they themselves would be unaffected? Does the existence of others unaffected by repression alter one's own decision to participate?

Authors

• Gloria Cheung, Duke University

• David A. Siegel, Duke University