Uncertain times: The causal effects of coups on national income

Abstract

We use doubly robust difference-in-differences models to estimate the causal effect of successful coups on national incomes. We find that real per capita gross domestic product (GDP) decreases by 10%–12% 5 years after a coup and the effect has not begun to diminish at that point. When we investigate the economic and political mechanisms behind this outcome, we find that our result is mostly driven by a fall in investment and in the rule of law, along with an increase in repression. Given the size of the effect, preventing coups can be seen as a significant development issue, and though the international community has taken steps to discourage coups, further consideration of anticoup policies seems well-warranted.