Author Description

This dissertation lies at the intersection of environmental economics and development economics. It includes three essays that empirically examine the effects of climate change and extreme weather events on agriculture in both developed and developing countries. The first essay explores how the choice of weather data sets could affect estimates of climate change impacts. A large body of empirical literature finds that high temperatures are detrimental to a wide range of economic outcomes. These effects are often identified from the within-location temporal variation in exposure to the extreme right tail of the temperature distribution. Here, we document large discrepancies in exposure to extreme temperatures across six high-resolution gridded weather data sets in the US, where weather data is considered to be of high quality. We explore and illustrate the consequences of these data discrepancies in the estimation of potential climate change impacts on agriculture. We find that most climate change impacts based on different climate data sets are not statistically different from each other. Yet, the choice of the underlying weather data set can account for up to 48 percent of estimated warming damages on US crop yields. These findings highlight an important, but generally unrecognized, source of uncertainty in estimates of climate change impacts and the need for more systematic intercomparisons of widely used geospatial data sets in environmental social sciences. In the second essay, we estimate the impact of self-reported occurrences of droughts and floods on crop and livestock net income in Sub-Saharan Africa during the period 2009-2016. Based on a pooled data set for five countries, we find robust negative and heterogeneous impacts of droughts and floods across different levels of irrigation, poverty, and agricultural diversification, including reductions of net crop income by 34 percent and 61 percent due to droughts and floods, respectively. The study also confir