Author ORCID Identifier

https://orcid.org/0000-0001-9266-2869

Date of Award

8-10-2021

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Economics

First Advisor

Dr. Garth Heutel

Second Advisor

Dr. Andrew Feltenstein

Third Advisor

Dr. Stefano Carattini

Fourth Advisor

Dr. Marc Hafstead

Abstract

The essays in this dissertation discuss modeling techniques for international trade and their application to environmental policy. In addition, I present evidence of the effect of air pollution on worker sick leave.

Chapter 1 presents an application of computable general equilibrium (CGE) modeling in environmental policy evaluation. In most CGE models, researchers assume that goods are differentiated by their origin of production, known as the Armington model of international trade. In this chapter, I consider their application to carbon taxes, carbon leakage, and border carbon adjustments (BCAs). BCAs are designed to address carbon leakage, which is a phenomenon where areas not subject to an emissions tax increase their emissions in response to regulated areas decreasing emissions. I find that the non-Armington model predicts a higher carbon leakage rate compared to typical Armington models. I also find that border rebates are more effective than border tariffs at reducing leakage.

Chapter 2 presents an application of the non-Armington model to the North American Free Trade Agreement (NAFTA). Previous studies of free trade agreements have shown that the Armington model may underpredict changes in trade. In this chapter, I build a non-Armington model that still incorporates observable features of the international trade market. I then simulate the trade effects from NAFTA, and I compare these results to previous studies. I find that the non-Armington model can generate larger changes in trade than the Armington model.

Chapter 3 discusses how pollution might affect worker productivity, specifically the probability of taking sick leave from work. In this chapter, I evaluate these studies using a causal inference technique to quantify the impact of the CAAA on the number of days workers miss due to illness. I discuss the possible issues with using simple hazard rates for illness in making calculations of missed days and I also discuss how paid sick leave may influence results. Using a DiD framework, I find that the CAAA reduced the probability of taking a sick day in a given week by 0.1 percentage points.

DOI

https://doi.org/10.57709/24084865

Download

Share

COinS