Resumen 

We explore the effect of historical ethnic borders on contemporary conflict in Africa. In particular, using artificial regions (i.e., grids of 5050 km), we document that both the intensive and extensive margins of contemporary conflict are concentrated in the proximity of historical ethnic borders. To mitigate concerns due to non-random assignment and potential mismeasurement of historical ethnic borders, we follow an instrumental variable approach that exploits variations in potential ethnic borders generated by a plausibly exogenous ethno-spatial partition of Africa. We find that grid-cells with historical ethnic borders have 30 percentage points higher probability to experience conflict events and 13 percentage points higher probability of being the initial location of a conflict. Our results hold across different types of conflict and are robust to accounting for country and ethnicity fixed-effects, a large set of geographical confounders, other sources of conflict, and variations in cell-sizes. Additionally, we find that geographical characteristics that are complementary to border demarcation mitigate the effects of historical borders on contemporary conflict, suggesting that tangibility, observability and immutability of ethnic borders may prevent conflict. Further, we find that population pressure and competition for resources (in particular conflict over land) exacerbate conflict at the proximity of the historical ethnic borders. Finally, we also document that cultural proximity across the borders as well as similarity in economic subsistence increase the probability of conflict at the borders.

Para obtener mayor información sobre el presentador, acceda Aquí.