Theoretical Implications of Empirical Models: An Application to Conflict Studies

• In Event: Theoretical Implications of Empirical Models

Abstract

Establishing causal relations with the data that is typically available is an effort that is plagued by endogeneity concerns in many areas of substantive interest. Confronting these challenges has stimulated a growing literature, drawing on methods from statistics, bid data, and formal theory. The most important concern in causal studies is the articulation of a clear mechanism at work in the data, and how that mechanism can be isolated empiricallly. Given this importance, a natural connection arises with a branch of formal theory that prioritizes the articulation of mechanisms. To study how this process should unfold we consider an application to conflict studies and generalize the canonical bargaining model of conflict focusing specifically on the different kinds of uncertainty that facilitate the incidence and outcome of conflict, and discuss the kinds of quasi-experiments that emerge from these sources of uncertainty. We identify two distinct mechanisms that may introduce exogenous variation into potential conflicts. First, incidence risk is the uncertainty regarding whether a crisis escalates into violence, and arises from private information between potential combatants. Second, result risk arises from imperfect information about the ultimate outcome of conflict, which implies that the initiation of conflict essentially triggers a lottery. We discuss the kinds of substantive questions that can be addressed using each, with particular emphasis on highlighting the theoretical underpinnings of different empirical strategies.

Authors

• Scott Tyson, University of Rochester

• Jessica Sun, Emory University