The deadly underside of food aid

A great year for American wheat farmers can mean a very bad year for countries torn by civil war.

That’s one finding of a new study by Yale and Harvard economists, who examine an unintended consequence of aid to developing countries: “an increase in US food aid increases the incidence and duration of civil conflicts.”

Aid workers and journalists have long complained about humanitarian aid “actually promoting conflict,” Nathan Nunn of Harvard and Associate Professor of Economics Nancy Qian of Yale write in American Economic Review.

Citing reports that rebels and governments alike steal or otherwise misuse food aid, the researchers crunched 36 years of data to determine “whether these accounts reflect extreme cases or are representative of the average effect of humanitarian aid on conflict.”

Their conclusion: “the concerns of critics are very real.” Sending more food doesn’t cause war, but it really does increase the “incidence and duration” of existing civil conflicts. (There’s no apparent effect on conflicts between countries.)

To analyze the effects of food sent to developing countries between 1971 and 2006, Qian and Nunn used two variables: changes in US wheat production and yearly differences in the amount of aid different countries received. The bigger the wheat crop, the more wheat the United States sent overseas. Since the United States provides more than half of the world’s food aid, and since “the main component of humanitarian aid is food aid,” the consequences are significant.

The economists found that, on average, for every 10 percent increase in food aid, the incidence of civil conflict increased about 4 percent. By that estimate, if the US doubled food shipments to help hungry civilians in a war-torn country, the probability of any conflict would increase by 40 percent.

Qian and Nunn cite numerous ways that food can fuel conflict. Some people estimate that up to 80 percent of food aid is stolen, often along with transport vehicles and equipment. Rebels use stolen food to fortify their fighters, directly or in exchange for weapons. Governments appropriate donations to feed their soldiers, or punish rebellious areas by withholding aid.

The authors are quick to point out that aid has many positive effects. Their results “should not be interpreted in isolation,” they caution, but should be seen as “a small first step toward understanding the costs and benefits of food aid and humanitarian aid policies.”