When to expect a coup d’état? An extreme bounds analysis of coup determinants

Abstract

Over the last several decades, both economists and political scientists have shown interest in coups d’e´tat. Numerous studies have been dedicated to understanding the causes of coups. However, model uncertainty still looms large. About one hundred potential determinants of coups have been proposed, but no consensus has emerged on an established baseline model for analyzing coups. We address this problem by testing the sensitivity of inferences to over three million model permutations in an extreme bounds analysis. Overall, we test the robustness of 66 factors proposed in the empirical literature based on a monthly sample of 164 countries that covers the years 1952–2011. We find that slow economic growth rates, previous coup experiences, and other forms of political violence to be particularly conducive to inciting coups.