Is There Evidence of a Poverty-Conflict Trap?

Abstract

This paper develops a model to explain why some countries appear to be caught in a poverty-conflict trap whereas others do not. We construct a simple two-period model where consumption and investment decisions are made in the presence of governments who consider initiating diversionary conflict to raise their chances of remaining in power. The resulting theory shows that governments who are less selfish and whose economies provide higher expected returns to capital formation are unlikely to engage in such behavior. In contrast, economies with selfish leaders and lower gains from capital formation may fall prey to engaging in avoidable conflicts thereby lowering investment and hence future growth. We label this the poverty-conflict trap. We then test the theory by estimating a structural model in which conflict (both external and internal) is jointly determined with growth. Using panel data for over 152 countries from 1950-88, we find little support for a poverty-conflict trap. However, after conditioning the information set on the initial conditions of geography, private, public and human capital investment, we find evidence for a conditional poverty-conflict trap-which is broadly consistent with our model. Therefore, on must focus on several factors of development to eliminate the poverty-conflict trap.

JEL Classification: H1, H5, H8