Peaceful uncertainty: When power shocks do not create commitment problems

Does a shock to the balance of power cause the advantaged actor to exploit its newfound advantage by initiating conflict? The modeling literature on commitment problems as a source of war makes a central assumption that states know and anticipate power shifts. We relax this assumption such that states must estimate future power shifts by looking at past and present capabilities-both their own and those of their adversaries. We incorporate these estimates, and their attendant uncertainty, into a model of war. We find that commitment problems remain a source of war, but that the existing models overpredict war by ignoring this dynamic. States continuously updating their estimates and accounting for uncertainty promotes peace. It follows that the apparent window of opportunity-in which the power balance becomes suddenly favorable to one side- poses less of a threat to peace than previous theories suggest. This result has applications to nuclear proliferation dynamics and conflict in general. We find empirical support for the model in tests analyzing power shifts and interstate wars.