Levitt and Dubner on War

by F.

Steven Levitt and Stephen Dubner have an interesting piece in the New York Times on war and weather. “The causes of any war are of course incredibly complex,” they write. “Or are they?”

The economists Edward Miguel, Shanker Satyanath and Ernest Sergenti have written a paper that uses rainfall to explore the issue of civil war in Africa. Twenty-nine of 43 countries in sub-Saharan Africa, they note, experienced some kind of civil war during the 1980’s or 1990’s….

The economists discovered that one of the most reliable predictors of civil war is lack of rain. Using monthly rainfall data from many different African countries (most of which, significantly, are largely agricultural), they found that a shortage of rain in a given growing season led inevitably to a short-term economic decline and that short-term economic declines led all too easily to civil war. The causal effect of a drought, they argue, was frighteningly strong: “a 5-percentage-point negative growth shock” — a drop in the economy, that is — “increases the likelihood of civil war the following year by nearly one-half.”

The piece is here.