Does inequality lead to civil wars? A global long-term study using anthropometric indicators (1816-1999)

We test for the influence of absolute and relative deprivation - proxied by anthropometric methods - on civil war risk. A comprehensive height data set allows us to go back to 1816 for a global sample. We measure absolute deprivation using human stature and we use height inequality within birth cohorts to measure relative deprivation. We take care that selectivity caused by missing values does not bias the results. We find that relative economic deprivation within populations (i.e., inequality) had a strong and consistent impact on the propensity to start civil wars. By contrast, absolute deprivation was significant in most but not all specifications. We also attend to potential endogeneity through instrumental variables.