Causes of Civil War in Asia and Sub‐Saharan Africa: A Comparison^*

Abstract

Objectives. This article explores the extent to which economic development, ethnic and religious fractionalization, domestic governance, and international trade openness affect civil war in postcolonial Asia and Sub‐Saharan Africa (SSA) from 1950 through 1992. Methods. We estimate a set of multivariate logit models with the generalized estimating equation (GEE) method for time‐series cross‐sectional (TSCS) data. Results. Both in Asia and in SSA, civil war is less likely with increased economic development and trade openness, while mixed autocratic‐democratic regimes raise the likelihood that states will experience civil war. Although neither ethnic nor religious fractionalization has any statistically significant effect on civil war in SSA, civil war in Asia is more likely with greater ethnic fractionalization. Conclusions. Despite cross‐regional variation in causes of civil war, economic development and trade openness seem to play a consistent role in reducing civil war.