Is climate change responsible for the conflict in Darfur?

The pundits say yes, but what do the data say?

According to a Democracy Now interview with Columbia professor Mahmood Mamdani, climate change is a main driver of the conflict:

From the late 1970s you have had a significant desertification, and you’ve been having in the north of Darfur basically a situation where people’s simply entire livelihoods are destroyed, and which has been one of the elements, because it has driven the nomadic population in the north down into the south.

Jeff Sachs agrees in this 2005 book interview:

Two things have happened. First, the population has doubled in the last generation, and second, the rainfall has gone down sharply. These are very hungry, crowded people, and now they are killing each other.

Darfur’s climate crisis even finds its way into Al Gore’s Inconvenient Truth:

Focus most of all on this part of Africa just on the edge of the Sahara. Unbelievable tragedies have been unfolding there and there are a lot reasons for it. Darfur and Niger are among those tragedies. One of the factors that has been compounding this is the lack of rainfall and the increasing drought.

But if a thousand pundits say it’s so, does that make it true?

One of my favorite African political economists, Michael Kevane of Santa Clara U, has the novel idea of looking at the rainfall data, and comes to a different conclusion:

Rainfall in Darfur did not decline significantly in the years prior to the eruption of major conflict in 2003; rainfall exhibited a flat trend in the thirty-years preceding the conflict (1972-2002). The claim that climate change explains the conflict rests on the observation that rainfall in Darfur has declined when comparing the present thirty-year period of 1972-2002 with earlier periods. This is strongly evident for El Fasher and El Geneina but less clear for the more southerly rainfall stations. Rainfall is basically stationary over the pre– and post-1972 sub-periods. A theory linking climate change around 1972 with an outbreak of conflict in 2003 has no compelling supporting evidence at present.

Michael argues that elsewhere in the Sahel the same climatic conditions have prevailed, but that regional institutions have kept resource conflicts in check:

While semi-arid conditions of low rainfall with high variability, both temporally and spatially, are tough environments for humans, there is much evidence that Sahelian societies have coped through evolving strategies such as income diversification, agricultural intensification and migration, rather than large-scale violence. Life is not peaceful: there are endemic low-level struggles over resources, particularly between herders and farmers. However, conflict has not erupted into warfare because regional institutions and central governments have by-and-large projected their authority into Sahelian hinterlands and arrested violent situations before they have intensified.

What is different about Darfur, he argues, is simple:

An elite ruling the country from Khartoum that has preferred to exclude peripheral populations from genuine participation in political processes.

Of course, the fact that rainfall shocks do not precede the Darfur crisis do not make rainfall unimportant to conflict. Ted Miguel, Shanker Satyanath and Ernest Sergenti have a paper in the JPE that finds that years of low rainfall are followed by a higher propensity for civil war. Michael’s darfur analysis simply suggests we ought to take a look at the supporting data before we make a conflict the poster child for the rainfall-conflict relationship.

Of course, this entire debate raises an important challenge to Miguel et al’s analysis. They argue that the rainfall leads to conflict because of its impact on incomes, and suggest that this provides evidence of a causal link from low incomes to the outbreak of civil war. The novelty of their paper is that they use rainfall as a statistical instrument to isolate and identify this income-conflict link.

Nearly all of the climate change and conflict discussions, however, focus on the migratory impacts of rainfall–pastoral peoples starting to prey on the lands and livelihoods of more settled peoples, for instance. This was the Darfur story in particular. If so, income is not the only mediator between rainfall and conflict. Statistically speaking, the exclusion restriction on the instrument is potentially violated. This would be an interesting proposition for someone to collect data on and test.

P.S. See some of Michael’s other Darfur opinion pieces here.

P.P.S. The Democracy Now interview with Mamdani is really worth reading since it has a number of very thoughtful and provocative points.