Counterbalancing and Coup Failure

Abstract

This chapter assesses the predictions of the theory using a new dataset that tracks how rulers organize and use the presidential guards, police, and other coercive institutions under their command. The dataset includes 110 countries between 1960 and 2010. Because it includes fine-grained information on features of individual security forces, including the chain of command through which each force reports to the regime and where it is deployed, the dataset allows for the development of more precise measures of counterbalancing than previously possible. It first presents descriptive data on the frequency with which rulers counterbalance and how the use of counterbalancing varies across countries and over time. It then tests statistically the argument that counterbalancing is associated with coup failure. In order to test arguments about the determinants of coup failure, the chapter also compiles new data on the identity of coup plotters in four hundred coup attempts. The chapter considers competing explanations for the negative association between counterbalancing and coup success, which posit that some other factor might explain both counterweights and coup outcomes, but shows that available evidence is inconsistent with them. The finding that counterbalancing makes coups less likely to succeed helps explain why some leaders have successfully prevented military coups, while others have not.