A Theory of Leadership Selection in Small Groups - With Evidence from Ugandan Farmer Associations∗

Small groups, such as farmer associations, micro-lending groups, or civic organizations, play an important role in economic, political and social life in all manner of countries and circumstances. This paper studies factors that affect the ability and effort exerted by leaders of these groups, and hence the effectiveness of the groups in providing public goods to their members. We argue that small groups differ from larger political units in a number of important ways, and offer a model adapted to the small group setting. The model suggests that, under certain conditions, groups face a tension when choosing the level of effort to demand from their leaders. If groups demand too little effort, they obtain high ability leaders, but these leaders exert little effort. Increasing effort demands increases the effort exerted by leaders, but may cause high ability members to self-select out of the candidate pool. The overall result is an inverted U-shaped relationship between groups’ effort demands and the value of the public good that their leaders produce. Whether this trade-off exists depends crucially on the level of private income opportunities available to group members. These predictions are tested using data gathered for the purpose from a sample of Ugandan farmer associations. The data support the predictions of the model and suggest that variation in the value of the group public good produced by the leader can have a meaningful impact on group member’s welfare. ∗We are grateful to Alex Barnard, Eliana Horn, and Vivian Lu for their research assistance, to MassimoMorelli, Matt Winters, Lucie Gadenne, Marcos Nakaguma and participants at the Columbia Political Economy Breakfast for helpful comments, and to our local team of research assistants. G.G. gratefully acknowledges support from the NSF Doctoral Dissertation Improvement Grant SES-0921204.