Aiding violence or peace? The impact of foreign aid on the risk of civil conflict in sub-Saharan Africa

This paper considers the impact of foreign aid flows on the risk of civil conflict. We improve on earlier studies on this topic by addressing the problem of the endogenous aid allocation using GDP levels of donor countries as instruments. A more structural addition to the literature is that we efficiently control for unobserved country specific effects in typical conflict onset and conflict continuation models by first differencing. The literature often overlooks the dynamic nature of these types of models, thereby forcing unlikely i.i.d. structures on the error terms implicitly.1 As a consequence, malfunctioning institutions, deep-rooted political grievances, or any other obvious, yet unobserved and time persistent determinants of war are simply assumed away. We find a statistically significant and economically important negative effect of foreign aid flows on the probability of ongoing civil conflicts to continue (the continuation probability), such that increasing aid flows tends to decrease civil conflict duration. We do not find a significant relationship between aid flows and the probability of civil conflicts to start (the onset probability