My dissertation considers the role of weather on a variety of economic outcomes, including homicide, deforestation, and mortality.
In chapter 1, in order to better understand why weather drives conflict, I focus my analysis on one of the most damaging forms of conflict: homicide. My results suggest that failing to recognize the potential for systematic differences across the homicide rate by race in the United States presents an inaccurate reflection of the true relationship between temperature and homicide. Previous research has found that colder temperatures decrease homicide while warmer temperature increase it. I find that black victims drive much of the negative effect for cold temperatures, while white victims drive the positive effect. While this finding alone is a new addition to the literature, these results are consistent with different mechanisms driving the relationship between temperature and homicide. For white victims, I cannot rule out the hypothesis that heat acts as a physiological stressor leading to violence and aggression. However, for black victims, and especially young black males, this hypothesis has limited support. Instead it appears that for black victims, an interactional mechanism plays a role, as colder temperatures reduce the homicide rate drastically.
In chapter 2, in collaboration with co-authors, we study the relationship between agricultural productivity and deforestation in the Brazilian Amazon. Using annual variation in growing-season temperature, we demonstrate that increased soy yields lead to greater land in agriculture and ultimately more deforestation. We find a delayed effect between the increase in expansion of planted area and deforestation consistent with patterns of indirect land use change documented in the literature. Our findings suggest there may be negative environmental spillovers from policies that increase productivity for highly developed agricultural industry in the tropics.
Finally, in chapter 3, in joint work with a co-author, we study the relationship between temperature extremes and mortality. Although there is consistent evidence that temperature extremes lead to significant reductions in health, there has been limited analysis focused in the developing world. The few existing studies to date document that rural populations drive much of the mortality response to temperature, but offer little insight into what role adaptation plays in urban areas. This chapter attempts to overcome this gap in the literature by considering the relationship between temperature, mortality, and adaptation in Thailand. However, our results raise serious concerns over data quality issues. We find that extreme high temperature leads to decreased mortality rates, which is the opposite effect that has been documented in the literature. We also consider residential energy use as a possible adaptation strategy and model the relationship between temperature and energy use. Here we find that high temperature days relative to moderate days cause residential energy use to increase, which is consistent with the literature.
