This article analyzes how the selection process for the executive affects the risk of rebellion and insurgencies in sub-Saharan Africa between 1971 and 1995. Four executive recruitment processes are distinguished that are characteristic for the African context: (1) a process without elections; (2) single-candidate elections; (3) single-party, multiple-candidate elections; and (4) multiparty executive elections. The results suggest that single-candidate elections and multiparty elections substantially reduce the risk of insurgencies compared with systems without any kind of executive elections. They further show that during times of political instability, the risk of large-scale violent dissent increases substantially. The article supports findings of the civil war literature that higher levels of income are associated with a lower risk of intrastate violence, while oil-exporting countries are at a higher risk of rebellion. In short, this article further strengthens the need to use more specific measures of elements of political regimes that also take into account regional particularities, in order to paint a more informative picture of how political structures influence the risk of internal violence.
Carey, Sabine C. (2007) Rebellion in Africa: Disaggregating the Effect of Political Regimes, Journal of Peace Research 44 (1): 47–64.