Leaders, States, and Reputations (with Scott Wolford)
Reputational concerns are ubiquitous explanations for war, yet consistent evidence of reputations is elusive. We attribute this to two issues. First, extant literature searches for the payment of reputational costs, yet strategic censoring should systematically bias observational data against revealing them. Second, the locus of reputation is often ambiguous, yet the choice of leader or state as unit of observation has inferential consequences. In contrast, our research design (a) focuses on observable implications of reputational theories in an appropriate sample and (b) considers two competing sources of the private information that creates reputational incentives: changes in national leaders and changes in domestic political institutions. Consistent with our expectations, leadership turnover and regime change are each associated with initially high probabilities that militarized disputes escalate to war, and those probabilities decline over time. Reputations are in evidence, but analysts must look for them in the right place.
Current status: under consideration for special issue of the Journal of Conflict Resolution