Escalation of civil war in Nepal: The role of poverty, inequality and caste polarisation

A growing literature examines effects of poverty, inequality and polarisation on civil war. Few studies examine effects at very local levels despite considerable spatial heterogeneity in many civil wars. We study Nepal's civil war, which escalated sharply from 2001, using geo-coded data on 15,000 conflict deaths. We also use small-area estimation to form poverty and inequality estimates for almost 4000 localities. Contrary to prior findings, it appears that higher local poverty rates reduced the risk of conflict and the number of deaths. This negative association is explained by the shift in strategy by the rebels, to target richer middle class and urban areas so as to access resources as a way to win the war. We also find that local relative wealth inequality is associated with escalation of the civil conflict, suggesting that relative wellbeing affects decisions about rebellion and conflict. Caste polarisation also raises odds of conflict and the number of deaths, especially where the dominant caste groups were larger. In a society where individual identity and alliances are defined by a discriminatory and unequal caste system, the probability of conflict is likely to be higher.