Семинары

International conflicts rarely erupt instantaneously. They almost always result from a process of escalation. The potential costs and benefits arising from these escalations are often most uncertain to the parties involved. The purpose of this paper is to study conflict situations by focusing on the role this uncertainty plays in the prevalence of conflict escalation and the prospects of ending it peacefully.

The authors using formal models to study escalating conflicts, notably Nalebuff and Fearon, rely on Schelling's early insight of "dancing on the cliff edge". While this literature is providing important insights on conflict situations, it is still rather thin in the analysis on the role of information in such conflicts: in real life players face a major uncertainty regarding their own costs and benefits and our model will not assume this uncertainty away.

Our investigation will proceed in two directions: In the first one we will focus on the effect of learning on the intensity of escalation. We introduce a model of gradual learning in which sides of the conflict acquire information about their own cost (from escalating) and benefits (from prevailing in the conflict). The second direction will involve the study of effects of an unexpected shocks to the sides' payoffs, here we provide comparative static results regarding the distribution of these costs and benefits. Thus we depart from the existing literature on escalating conflicts that assumes players to have a perfect knowledge of their own payoffs (and regards the payoffs of opponents to be the main source of uncertainty).