Three Essays on Causes and Consequences of Violence and Conflict

This dissertation revolves around the theme of violence, crime and conflict. It is an attempt at (a) improving the understanding of the causes of violence and crime, and (b) the consequences of violence and conflict. Specifically, the first chapter investigates if air pollution can be a contributor to crime, the second chapter looks at the labor market impacts of terrorism in non-war zones, and the last chapter tries to understand the human capital accumulation decisions of permanently displaced individuals.The first chapter looks at air pollution as a potential contributor to criminal activity. Using the seasonal variation in increase in air pollution --due to increased rice stubble burning-- in the Punjab province of Pakistan, I explore the relationship between air pollution and crime. I combine eight different sources of data and use an instrumental variable approach to estimate the causal impact of air pollution on crime. Air pollution increases both violent and non-violent crimes, but the increase in violent crimes is much more salient. A one standard deviation increase in seasonal variation in air pollution increases violent crime by at least 15 percent. Back of the envelope calculations suggest that the cost of increased crime due to air pollution are at least 5 million US dollars, but maybe as high as 600 million US dollars. One potential mechanism driving the estimates is the reduction in earnings for middle aged male individuals due to high air pollution. The results suggest that the social costs of air pollution are much wider than those previously considered. They also have significant implications for developing countries whose economies rely on agriculture. The second chapter of the dissertation looks at the impact of violence in non-war zones on incomes. A non-war zone does not have an active presence of the military, and it is not characterized by mass migration or shutting down of the economy. I use data from Pakistan on intermittent but sustained terrorist attacks for this purpose. After accounting for the intensity of the attacks, incomes reduce by about 2.5 percent on average due to terrorist attacks. However, the effect is almost twice in the same month in which a terrorist attack takes place in a district. The effects are more severe for low skilled members of the labor force as well as for relatively inexperienced members. The two potential channels driving this impact are the changes in employment compositions across different occupations and a reduction in the number of days worked. From a policy perspective, relief and welfare efforts targeted towards (a) the most vulnerable groups in the labor force and (b) the sectors of the economy directly affected by the violence seem to be the best possible response -instead of general aid aimed at overall rebuilding of the economy which is more relevant for a war or conflict zone. The last chapter considers the event of partition between India and Pakistan in 1947 that induced forced migration and permanent displacement of about 14.5 million people. I compare the educational outcomes of migrants and natives who were still in school going age when they were forced to migrate. I use a differences-in-differences approach. I also address the recent development in related literature on concerns related to identification, power, and bias in a differences-in-differences approach. I find that migrants are more likely to achieve certain educational milestones than their native counterparts. The results may be driven by the choices of older migrants pertaining to location and occupation. They also highlight the importance of an enabling and convenient environment for migrants which plays a crucial role in the pursuit of their educational goals.