Capital Cities, Conflict, and Misgovernance: Theory and Evidence

We investigate the links between capital cities, conict, and the quality of governance, starting from the assumption that incumbent elites are constrained by the threat of insurrection, and that this threat is rendered less e_ective by distance from the seat of political power. We develop a model that delivers two key predictions: (i) conict is more likely to emerge (and to dislodge incumbents) closer to the capital, and (ii) isolated capital cities are associated with misgovernance. We show evidence that both patterns hold true robustly in the data, as do other ancillary predictions from the model.