We're Not in Kansas Anymore: A Geospatial Analysis of the Impact of Climate Change on the U.S. Agricultural Sector using Discrete Choice Models & Ricardian Comparative Advantage

Abstract

This Senior Thesis investigates the impact of climate change on crop growth and land allocation in a subset of 12 top agricultural producer states in the US. More specifically, it uses a model developed by Costinot, Donaldson, & Smith (2016), rooted in Ricardian comparative advantage, in order to construct counterfactuals of how output patterns would differ if farmers continue to produce with the current land allocation and crop mix, and if they adjust to grow crops according to 2080 land productivity under the most aggressive Hadley A1FI post-climate change scenario. The findings indicate that, on a first-order productivity level, there are mixed gains and losses across counties for different crops. Nevertheless, it was found that farmers gain more and lose less when adjusting their production and crop mix to update new productivity information. The paper then proceeds to note that the welfare gain from production adjustment differs across counties, and that these changes are correlated with the differential level of technology-induced expenses and non-crop outputs such as livestock and poultry.