Civil armed conflicts: the impact of the interaction between climate change and agricultural potential

The goal of this paper is to examine the impact of rising world temperatures on the incidence of civil armed conflicts, focusing on a specific mechanism: the interaction between variations in annual temperatures and variations in agricultural potential. We assemble a dataset from various sources for 172 countries from 1946 till 2014. Agricultural potential is based on the Food and Agricultural Organization fs definition of a country land suitability for growing basic crops. Annual temperature data come from the Climate Research Unit of the University of East Anglia. Data on civil armed conflicts is from the Uppsala Conflict Data Project. Using a fixed-effect approach, our identification strategy is akin to a natural experiment where the exogenous interaction between the temporal variation in temperature within a country and the cross-country variation in agricultural potential allows identifying the effect of this interaction on conflict incidence. The findings indicate that temperature and agricultural potential are substitutes and have offsetting effects on conflict incidence. We find that in a country with low agricultural potential a one degree increase in temperature is associated with a 3% increase in conflict incidence. However, when agricultural potential is high, a one degree increase in temperature is associated with a 5% decrease in conflict incidence. The results are tested against various robustness checks.