Democratization After Military Coups d'État

Abstract

Military coups are one of the clearest threats to democracy, often reshaping the political future of a state. This thesis investigates whether the duration of rule following a realized military coup is associated with changes in democratization, and whether that relationship is conditioned by pre-coup military strength, democratic baseline, and natural resource wealth. To test this relationship, this thesis assembles an original dataset of 285 realized military coups across 80 countries from 1945 through early 2026 by integrating the Illinois Cline Center’s Coup d’État Project, Polity V regime scores, Correlates of War military expenditure data, and World Bank resource rent indicators. I estimate a two-stage regression framework with country-clustered standard errors, supplemented by a placebo check comparing successful and failed coup attempts, an extended analysis of single-coup country episodes, and a regional fixed effects robustness check. Initially, the results suggest post-coup duration is positively associated with democratization in simple models, however, this association disappears once controls are introduced. The central finding is that a country’s democratic baseline before a coup, rather than the length of the post-coup interval, is the strongest predictor of post-coup political change. These findings suggest that democratic recovery after coups is shaped more by institutional inheritance than by the passage of time, with implications for how scholars and policymakers should approach democratic recovery after a military coup.