Essays in Environmental Economics

Abstract

This dissertation presents research on environmental economics and policy, linking the effects of environmental uncertainty with policy effectiveness. In the first chapter, I study the effects of weather and seasonal forecasts on US agri- culture. The economics discipline has accumulated evidence on the negative impacts of extreme heat on economic activity. Less understood are the actions that economic agents in the expectation of extreme heat. This chapter studies the effect of seasonal climate forecasts on US agriculture. I find that the behavioral responses to forecasts of extreme heat have much larger financial implications than the direct effects of extreme heat. Specifically, I find that an additional day above 29 degrees Celsius in the growing season forecast, causes farm revenues to fall by 1.83% when the occurrence of a day above 29 degrees Celsius reduces revenues by 0.20%. The large drop in revenue is the result of large drops in investment on the part of farmers. These losses in revenue have large and significant consequences for farm profitability. I study some of the underlying changes in investments that farmers make in response to forecasts of extreme heat and show that farmers reduce investments but direct more of their remaining resources towards staple crops like corn, cotton, and soybeans. In the second chapter, I study the effectiveness of short-term pollution policies. This chapter analyzes the effect of an emergency air pollution control policy to address short-run spikes of ozone concentrations in Mexico City. The multi-faceted policy combines a driving restriction with industrial emissions controls to reduce ozone. I use hourly measurements from Mexico City’s pollution monitoring network and leverage variation in the pollution measure that triggers the policy to implement a regression-discontinuity design to calculate causal estimates of the pollution reductions from the policy. I find that the policy signifi- cantly reduces ozone pollution over the entire monitoring network using multiple measures. In particular, the daily 1-hour maximum ozone reading falls by 16.4 ppb, a 17% reduction. The peak ozone reduction benefits of the policy are spatially concentrated, but the spatial distribution of benefits is not associated with high population density, vehicle ownership, or proxies for income. The mortality benefits from reduced ozone are between USD 3.7 million and USD 5.6 million for each advisory. I find no evidence of reductions of CO and NOX, two pollutants closely associated with motor vehicles, suggesting that the benefits of the policy could be maintained without a driving restriction. In my final chapter, I study the interaction between day-ahead markets and the inter- mittent nature of solar energy in California to identify the environmental benefits and costs of policies that would expand solar capacity or require solar producers to commit to lower amounts of production in day-ahead markets to reduce variability and uncertainty in solar generation. I estimate the causal effect of changes in solar generation on emissions from fossil fuel generators taking into account the dynamic nature of decision making in power markets and distinguishing between the effects of predicted solar generation and unexpected deviations from the day-ahead forecast. I find that increasing hourly solar generation in the day-ahead market by 1% leads to a larger reduction in fossil fuel generation from California and imported resources, leading to a drop in CO2 emissions of 16 tons. I also find evidence that local fossil fuel generation provides a flexible resource to respond to deviations from the solar generation forecast. This provision of flexibility has an environmental cost, in- creasing CO2 emissions and implies that policies to reduce uncertainty about real-time solar generation would have environmental benefits.