Blood for oil

There’s quite a bit of scholarship about oil and civil conflict, but the role oil plays in driving interstate war is a bit less understood. A clever paper by Fracesco Caselli of LSE, Massimo Morelli of Columbia University, and Dominic Rohner of the University of Lausanne attempts to measure this by looking at whether the geographic location of oil reserves effects the likelihood of conflict between neighboring states. 

According to previous research cited in the paper, 86 percent of all full-blown international wars between 1945 and 1987 were between neighboring states, 72 percent of them involving disputes over territory. Nearly half of all changes in international borders between 1816 and 1996 involved some level of international conflict. So if competition over oil is really driving international conflict, one would expect wars to be more likely when the oil is closer to the border, and thus easier to capture — think of Saddam Hussein’s invasion of Kuwait, or the ongoing border conflict between Sudan and South Sudan.    

Using the Correlates of War dataset, the authors pulled out 606 examples of conflicts between 1946 and 2008 where the two sides either shared a border or were seperated by less than 400 miles of water. In cases where the countries’ posessed oil, they used a georeferenced dataset to measure the proximity of oil fields to the border in question. Here’s what they found:

Quantitatively, the effect of geographic location is very sizeable. …As already noted, the average risk of conflict in our sample is 5.7 percent. This drops to 3.1 percent for country pairs in which neither country has oil. In contrast, when one country in the pair has oil, and this oil is right at the border (Distance = 0), the probability of conflict is almost 4 times as large: 11.6 percent. But this greater likelihood of conflict is very sensitive to distance. Indeed when the oil is located at the maximum theoretical value for our distance measure (Distance = 1) the likelihood of conflict is similar to the likelihood when neither country has oil.

The last two bars in the figure look at the case where both countries have oil. In the first instance, asymmetry is maximal: one country has oil right at the border (MinDist=0), the other at the maximum distance (MaxDist=1). The likelihood of conflict is almost three times as large as in the case where neither country has oil, or 8.6 percent. In the second instance, we look at a case of perfect symmetry: both countries have oil at a distance that is one half of the maximum distance (MinDist=MaxDist=0.5).

The likelihood of conflict is a much more modest 4.1 percent.

It’s a compelling finding, though from a non-specialist’s point of view, it also seems to show one of the limitations of relying on datasets for polticial analysis. It seems like a qualitative look at the facts surrounding some of the conflicts in question could pretty easily answer whether or not oil was a major motivating factor for the sides. 

This is a very different take on the issue from Jeff Colgan’s recent book, which I wrote about earlier this month. Colgan argues that oil doesn’t cause interstate wars because states are fighting over the prize, but because of how the presence of oil effects a country’s domestic politics: Oil income lowers the potential costs of instigating armed conflict. 

Caselli, Morelli, and Rohner address the possibility that oil-posessing states are more aggressive or militarily strong, but of course, this wouldn’t explain why the proximity of oil to a border seems to be a factor.

It would be interesting to expand this to wars between non-bordering countries. Anti-war groups often argue that U.S. interventions abroad are motivated by oil. But are assymetric conflicts like this actually any more common when there’s oil in the ground?