Business as unusual. An explanation of the increase of private economic activity in high-conflict areas in Afghanistan

In this paper we use a unique dataset that combines spatial detailed information on conflict events and on households' activity, to show a positive and significant correlation between violent conflict and entrepreneurship in Afghanistan. We build spatial and IV identifications to estimate the effect of different measures of conflict on the investment in a range of private economic activities of nearby households. The results consistently show that the level of conflict, its impact, and to a lesser extent its frequency, increase the probability that a household engages in self-employment activities with lower capital intensity and in activities related to subsistence agriculture, and reduce the probability of investing in higher capital self-employment. Overall, by increasing entrepreneurship, conflict pushes the country towards a regressive structural change. However, the magnitude of most of the effects is quite small. The paper contributes to a literature that, due to data constraints and identification issues, has not yet delivered conclusive evidence.