Essays on climate change and resource economics

Abstract

<p>This dissertation analyzes farmers’ behaviors in response to climate change and technology adoption. The first essay analyzes the adaptive responses of Midwestern farmers to regional climate conditions through land use change and crop insurance purchases. The results of this study can be summarized as follows. First, we find that climate conditions have a significant effect on farmers’ decisions regarding crops to grow, insurance purchases, and land allocation. Second, federal crop insurance mitigates farmers’ incentive to adapt to climate conditions such as intensive rainfall events. Third, federal crop insurance programs have induced Midwest farmers to allocate more acreage to corn and soybeans.</p> <p>The second essay studies economic and environmental implications of genetically modified (GM) corn and information technology adoption by analyzing Midwestern farmers’ corn yield and nutrient management. The findings can be summarized as follows. First, GM corn and its combination with pest scouting increase corn yield and nitrogen use. Second, the effects of GM corn and/or pest scouting adoption on corn yield and nitrogen use are greater for fields having low soil productivity. Third, yield monitor and its combination with pest scouting have positive effects on corn yield and nitrogen use.</p> <p>The third essay examines the effects of uncertainty regarding climate measures on forecasting future land use: variations in projected weather data sets and methods of forming farmers’ expectations regarding weather variables. We analyze decadal land use change over the Midwest based on five general circulation models (GCMs) and six assumptions regarding how to form expected weather conditions. From out-of-sample forecasting tests, we find that the predictive accuracy of models depends on the choice of GCM and methods of forming farmers’ expectations regarding weather variables. However, we find that forecasting results based on models consisting of yearly agronomic variables are more stable and have better predictive accuracy than models consisting of monthly variables. In addition, we estimate forecast land use in 2030 based on the best model and verify that two uncertainties have a significant effect on the forecasting results. Last, the predicted land use over the Midwest in 2030 shows that corn and soybean acreage will expand to the northwest.</p>