Evaluating UN Peacekeeping with Matching to Improve Causal Inference

Previous statistical studies of the effects of UN peacekeeping have generally suggested that UN interventions have a positive effect on building a sustainable peace after civil war. However recent methodological developments have called these findings into question because the cases in which the UN intervened were quite different in a variety of factors than those in which they did not. We argue that this problem arises because of the nonrandom assignment of UN operations to civil conflicts. As a result of this problem estimates of the effects of the UN operations were largely a result of the researchers’ assumptions about the proper statistical model to use rather than the data. Therefore we cannot be sure if the estimated causal effect of peacekeeping is due to peacekeeping itself or to the functional form of the model that the researchers chose. We correct for this problem by pre-processing our data using matching techniques on a sample of UN interventions in post-Cold-War conflicts and find that the UN interventions are indeed effective in the sample of post-civil conflict interventions, but that UN interventions while civil wars are still ongoing have no effect.