This paper uses detailed information on the latitude and longitude of conflict events within a set of Sub-Saharan African countries to study the impact of external income shocks on the likelihood of violence.
We consider a number of external demand shocks faced by the countries or the regions within countries — temporary shocks such as changes in the world demand for agricultural commodities, and longer-lasting events such as financial crises in the partner countries or permanent changes in foreign trade policy — and combine these with information reflecting the natural level of trade openness of the location.
We find that (i) the incidence, intensity and onset of conflicts are generally negatively and signicantly correlated with income shocks within locations; (ii) this relationship is signicantly weaker for the most remote locations, i.e those located away from the main seaports, (iii) at the country-level, only large, long-lasting shocks — financial crises, trade policy — have an effect on the outbreak and location of new events.
Altogether, our results therefore suggest that external income shocks are important determinants of the intensity and geography of conflicts within countries, and provide support in favor of the opportunity cost theories of war.
I think there are some weaknesses, and additional work to do (I am a bit suspicious of the ACLED conflict dataset they and others use) but this is a must-read paper for anyone interested in the empirical causes of conflict, and a nice example of how to bridge the divide between sub-national and cross-national analysis.
I’m in the midst of revising my own paper, and will hopefully be posting again in the next week.