Theoretical, Empirical, and Experimental Research on Collective Action Problems

This research investigates collective action problems to better understand the extent to which groups can solve them. Society faces many challenging collective action problems. A broad goal of the research is ultimately to help society by contributing to the improvement of existing institutions for solving these problems.

Many fundamental theoretical and empirical questions remain unresolved about how to efficiently aggregate preferences and information in the face of individual free riding incentives, especially when side payments are not possible. There are many open questions about actual human decision making and behavior under different institutional environments.

The theoretical arm of this research is aimed at better understanding the efficient aggregation of preferences and information, using an organizational design approach. The experiments provide information on actual behavior in these institutions, pointing to how they may be modified to improve decision making; and how the theory might be modified in light of behavioral findings from the experiments.

The empirical research is aimed at better understanding the forces underlying voter turnout in mass elections.

More specifically, the theoretical research addresses specific questions about the relative performance of different organizational structures, or mechanisms, where measures of performance may include the probability of group success at overcoming the free rider problems, or, more generally, economic efficiency. The research team first investigates these questions taking a mechanism design perspective that differs from the usual approaches to study “public goods.” The new perspective studies honest and obedient communication mechanisms, which coordinate the actions of group members without commitment, coercion or monetary transfers.

These “Volunteer Based Organizations” (VBOs) require only very simple binary communication in which members express a willingness to either volunteer or free ride. The research goes on to explore: (1) the efficiency properties of VBO mechanisms; (2) a “mechanism equilibrium” when there are multiple competing groups, each of which solves collective action problems which can crowd each other out; and (3) an application of VBO mechanisms to a group-based model of voter turnout.

The application to voter turnout also involves an empirical project to structurally estimate the VBO model of group turnout with a new voter-turnout dataset. The researchers then go beyond this static setting and study the dynamics of collective action problems, where individual volunteering or free riding decisions evolve sequentially over time. In many applications, for example international environmental agreements (IEAs), grass root campaigns or uprisings against authoritarian regimes, collective action evolves over time, starting with a small core of activists/volunteers.

Individuals who are most willing to volunteer (lowest cost) commit earlier, which in turn can create a bandwagon effect inducing higher cost individuals to follow. This fosters coordination and improves efficiency, allowing individuals in the group to be screened sequentially. Behavior in this environment can be studied as a dynamic participation game; or as a dynamic mechanism design problem if the group can design an honest and obedient mechanism that takes advantage of the dynamic structure.

The research studies both approaches and then proceeds to investigate two further extensions: (1) the effect of size asymmetries across the group members; and (2) the effect of non-stationarities, reflecting changes in the value of group success or costs of participation as time unfolds. The third arm of the research is experimental and involves four laboratory experiments and one large-scale online experiment.

Three of the laboratory experiments are directly motivated by the static model of VBOs; the fourth is motivated by the dynamic theory. The large-scale online experiment explores group sizes an order of magnitude large than is possible in the laboratory.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.