Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Farmer groups are the cornerstone of many agricultural projects in low income countries. The success of such projects crucially depends on the ability of group members to cooperate. We conducted a series of public goods experiments to study within-group cooperation in Malawian farmer groups. We combine results from these experiments with survey data and qualitative interviews. Our results shed light on the heterogeneous capacity of groups to cooperate. We find that democratically run groups, in particular those with close social ties, are more cooperative compared to groups with leader-driven decision-making. Focus groups indicate that this democracy is deliberative in nature, characterized by open discussion that aggregates preferences, increases common knowledge, and creates goodwill. A second set of experiments in which we experimentally vary the decision-making processes yields quantitatively similar results in arbitrary groupings of farmers and null results in pre-existing groups with established decision-making procedures, demonstrating the stickiness of institutional rules. Our results imply that group formation and functioning needs to be included in the design phase of agricultural projects in low income countries.