A growing literature studies the causes and consequences of civil conflict. Several of these studies are of Nepal’s civil conflict that lasted for a decade from 1996. Coincident with this conflict was a surge in work-related emigration, affecting education and human capital investments, labour supply, and local economic activity. While a few prior studies look at the effects of Nepal’s conflict on emigration, many key features of the conflict are ignored. In particular, heterogeneity due to the highly localized nature of the conflict tends to be ignored by the type of data and research designs used in prior studies. In this thesis, I contribute to the literature on the causes and consequences of civil conflict by constructing and using very local indicators of driving forces and intensity of Nepal’s civil conflict. Contrary to prior literature that suggests there is more conflict in poorer regions, I find poverty is associated with less conflict risk and conflict intensity, partly due to a change in strategy by the rebels in Nepal to target richer middle class and urban areas to win the war. At the same time, inequality and caste-based polarisation at local levels were major contributors that escalated the civil conflict. In order to study conflict escalation and also to study the impact of conflict intensity on migration, I construct a database of conflict deaths, and of poverty, inequality and caste polarization for almost 4000 localities. In contrast, most previous research on Nepal relies on data aggregated to the district level, suppressing detail as there are only 75 districts and there was far more variation in conflict intensity within districts than between districts. Estimated impacts of conflict on emigration are substantially distorted by using district-level conflict data. The growth in emigration that sees over two million Nepalese working abroad has also given a great boost to remittances, which are now equivalent to about one-third of GDP. The effects of emigration and remittances on human capital investment – particularly child schooling and child labour – remain debated in the literature. I construct a two-wave panel at the locality level to study these issues, with careful attention paid to the endogeneity of emigration and to spatial spillover effects. In localities with a greater increase in emigration, the net enrolment rate in secondary education rises and child labour for older children decreases. A possible channel for this effect is that migration and remittances help credit constrained households to cover extra schooling costs and compensate for the foregone earnings from child labour. In addition to finding an overall positive effect of emigration, I find that the direct (own) effect of emigration increases labour supply of the female labour force but decreases the labour supply in the agriculture sector. There are also positive spatial spillover effects for rural localities, that likely occur through rising local wage rates. Given these generally positive effects of emigration, the fact that local conflict intensity suppressed the growth in emigration suggests that there is a lasting negative effect of the civil conflict in particular parts of Nepal. These effects would not be apparent in the overly aggregated data on conflict that has been used to date. Thus, one contribution of this thesis is to show the benefit of a very local focus on the causes and consequences of civil conflict.