An econometric analysis on the direct and indirect effects of climate variability on conflict. The case of Kenya.

Highlights

• This study shows that in Kenya climate variability is associated with higher conflict risks. Decreased rainfall is linked to more frequent violent conflicts through an increase in the percentage of stunting children, with each unit increase in below-average rainfall anomalies associated with an 8% increase in foreseen violent conflicts per grid as mediated by malnutrition.
• More specifically, below-average rainfall anomalies computed over 12 months prior to the households’ interviews are associated with an 18.8 percentage point increase in the households with at least one stunted child. Moreover, malnutrition contributes to an increase in violent conflicts, with a one percentage point increase in malnourished households associated to a 0.4% increase in average foreseen violent conflicts.
• Below-average rainfall anomalies computed over 12 months do not directly affect conflict risks, but only indirectly through malnutrition, implying that, when considering long-term periods, nutrition insecurity may become a dominant channel mediating the climate-conflict relationship in Kenya.
• Findings also suggest that climate variability, when considering shorter time periods, is likely to have a significant direct effect on conflict risks through channels other than malnutrition.