Electronic Thesis and Dissertation Repository

Thesis Format

Integrated Article

Degree

Doctor of Philosophy

Program

Economics

Supervisor

Zheng, Charles Z.

Abstract

An important barrier to conflict resolution is asymmetric information. That is adversaries have private information about their objectives, resources, and strengths during the conflict and have incentives to misrepresent this information during the negotiations. Third-party institutions, like mediators, can help adversaries to reach an agreement by making a peace proposal. In this thesis, I explore the implication of asymmetric information of players to the design of an optimal peace proposal by a mediator.

Chapter 2, co-authored with Charles Zheng, studies a problem of conflict mediation where a mediator proposes a split of a good between two ex-ante identical contestants thereby preempting a conflict if and only if both accept the proposal. A contestant worries that accepting the proposal may signal weakness that may be exploited in the event of conflict. Thus, when conflict cannot be fully preempted, the socially optimal proposal offers to one of the contestants a minimally larger share of the good than what it offers to the other party so that the former contestant always accepts it without fearing any part of its private information being revealed.

Chapter 3 examines the participation decisions of the players in a mediation process. The action of participating in the mediation conveys information to the opponent. The mediator wants to minimize the probability of conflict subject to incentive compatibility and full participation constraint. I find that despite ex-ante identical players, a certain class of biased proposals augmented by a randomization device, that incentivizes the favored player to always accept, satisfy full participation. The equal proposal cannot satisfy full participation even with randomization. These results are true when the type distribution is binary or a continuum of types.

Chapter 4 studies a conflict model where adversaries can renege on an accepted agreement. A mediator whose objective is to maximize welfare subject to renege-proof constraint proposes a peaceful split of a contested prize between two players. Despite ex-ante identical players, the renege-proof optimal proposal is a biased proposal that the favored player always accepts. This proposal is even more biased than the lopsided proposal that maximizes welfare when players have full commitment.

Summary for Lay Audience

An important barrier to a peaceful settlement between adversaries is asymmetric information, i.e., each involved party is uncertain about the objective or resources of the opponent. The difficulty of assessing the cost of conflict can lead to the collapse of negotiations. Mediators can help adversaries to reach an agreement by making peace proposals. In this thesis, I examine the design of optimal peace proposals by a mediator.

Chapter 2, co-authored with Charles Zheng, studies conflict mediation problems where a mediator proposes a split of a prize between two players that if they both accept, they avoid a conflict. Players have private information about their cost of exerting effort at conflict. I find that, when conflict cannot be fully avoided, despite equal players, the optimal proposal that maximizes social welfare, is a specific biased proposal. It is the smallest proposal that the favored player always accepts, without worrying to signal any weakness by accepting the proposal.

Chapter 3 considers a conflict mediation problem where adversaries can decide to participate in a mediation procedure or trigger conflict immediately. The act of participation signals information about players' private information. A mediator, who can provide economic incentives, offers a peace surplus of an agreement to two players to prevent a conflict if both accept the proposal. The mediator wants to design proposals that guarantee the full participation of the players. I find that despite equal players the peace proposals that guarantee the full participation of players are a class of biased proposals, augmented with a fair coin. When the type distribution is binary, the lopsided proposals also minimize the probability of conflict subject to the full participation constraint.

Chapter 4 studies conflict mediation problems where players can renege on an accepted proposal. Players can learn about each other by observing each other's decisions in the mediation. They can use this information to reassess the cost of conflict and renege on the previously accepted proposal. I find that a mediator who is interested in the design of renege-proof welfare-maximizing proposals should propose even more biased proposals compared to the environments where players have full commitment.

Share

COinS