The Changing of the Guard: Succession and Violence Then and Now

The transfer of power from one leader to another constitutes a moment of vulnerability for political systems. A power vacuum risks devolving into violent power struggles between rival claimants for power, potentially resulting in civil war. Different societies have applied different institutional solutions to this “succession problem”. We develop a comprehensive framework, focusing on two key dimensions – the frequency of contested succession and degree of constraints put on the executive post-succession – and discuss common institutional solutions from different periods of history. For instance, monarchies with primogeniture mitigate uncertainty associated with the first dimension, whereas certain institutions in modern democracies address the second dimension. Yet, which type of political system that minimizes the overall risk of succession-related violence remains an open empirical question. We leverage two new datasets to address this question. Our empirical analysis covers a 1000-year period in a European-, and the last two centuries in a global sample of polities. We find that the number of leadership successions has increased dramatically over time, especially over the last 200 years. However, the total level of succession-related violence is only marginally higher in the more recent period. Further, the overall probability of succession-related civil war is actually higher in certain types of relatively democratic systems than in certain types of autocratic ones, mainly due to more frequent leader successions in democracies.