Essays in Dynamic Cooperation: Bargaining and Contribution Games

Abstract

In many economic and political settings, agents with non-aligned interests must work together to enact mutually beneficial change. My dissertation consists of three chapters that analyze problems that combine features of bargaining and contribution games. In particular, I examine how institutional rules, the size of teams, and the attitudes of agents affect dynamic cooperation. In Chapter 1, I analyze a repeated bargaining environment where the previously negotiated agreement becomes the present status quo. Agreements’ efficiency decays in the absence of revision, giving both parties an incentive to re-negotiate. I study how limited foresight— or the inability of agents to account for the full dynamic linkage from one agreement to the next— affects behavior. I show that limited foresight is a key driver of gridlock. Moreover, I demonstrate that while limited foresight and inter-temporal discounting both overweight imminent outcomes, these have the opposite effects on equilibrium. I solve the model in the case of legislative bargaining and relate its findings to observed phenomena in policy-making. In Chapter 2, Matias Iaryczower, Santiago Oliveros, and I study sequential contributions to public goods. We consider a model where agents with different valuations for the good are selected at random to contribute in each period. Agents face classical free-riding incentives in preferring to offload the cost of contributing to other players. We show that if the project is sufficiently large, the unique equilibrium of the model displays endogenous contribution cycles, in which agents of different types alternate making gradual contributions towards completion of the project. Finally, in Chapter 3, I explore a setting that combines features of both bargaining and contribution games. Two players cooperate to complete a joint project, wherein each of several periods,one agent— the project leader— is responsible for advancing the project at personal cost. Project leaders are choose not only the extent of progress they enact but also the direction in which the project moves, over which their preferences diverge with their counterparts. I show that increasing the incumbency of an efficient project leader decreases her productivity while the opposite is true for an inefficient contributor.