Peace on quicksand? Challenging the conventional wisdom about economic growth and post-conflict risks

In a widely cited study, Collier, Hoeffler & Soderbom show that economic growth reduces the risk of post-conflict peace collapse – particularly when the UN is present with a peace mission. These findings are encouraging for interventionist international policymakers. We replicate their study using data from the UCDP/PRIO Armed Conflict Database instead of the Correlates of War database. We generate a series of different datasets on the basis of different coding criteria commonly used in the literature, and rerun a simplified version of their model. Our results do not support their findings regarding the risk-reducing effect of economic growth and UN involvement. At best, the results are mixed. Some of the models even suggest that economic growth may increase the risk of post-conflict peace collapse. Overall, we are forced to conclude that the impacts of economic growth and UN involvement on the risk of post-conflict peace collapse are neither clear nor simple. The differences in the results seem to be driven by two sources: the conflicts included in the original datasets and the coding of the start and end dates of the conflicts.