Date of Award

Spring 2022

Project Type


Program or Major


Degree Name

Doctor of Philosophy

First Advisor

Bruce Elmslie

Second Advisor

Ju-Chin Huang

Third Advisor

Norman Sedgley


This dissertation examines the impacts of poor air quality and the rise of intra-industry trade on different labor market outcomes like migration, workings hours, and wages. This thesis contributes two distinct strands of literature, literature on air pollution focusing on the economic consequences of poor air quality, and international trade literature. The first two essays examine how a moderate level of air pollution can drive migration decisions and impact working hours in the United States. The third essay investigates the role of the rise in within industry trade on occupational wages in the United States. This dissertation aims to contribute to the above-mentioned two strands of literature by providing pieces of evidence that Americans’ decisions to migrate or to reduce working hours are guided by air quality, and occupational wage movements can be explained by the rise in intra-industry trade.

The first essay examines the impact of air quality on internal migration in the US from 2006 to 2019. A pseudo-gravity model of migration with PM2.5 concentration as an environmental factor of migration is used in the study. Endogeneity and zero values in the dependent variable are the two potential estimation issues in the model. The study uses average thermal inversion strength as an exogenous variation for the PM2.5 concentration. A Poisson model with the thermal inversion strength as an instrument (also known as IV-PPML) is used to study the effect of air pollution on domestic bilateral migration. The baseline results suggest that the “pull” factors of migration at destinations play a crucial role over the “push” factors of migration in the origin. In other words, individuals will choose a cleaner destination when they migrate. The estimated effects of air pollution on migration echo that a 1% reduction in PM2.5 concentration at the destination county leads to an increase in bilateral migration from origin county to destination county by 1.7%.

The second essay attempts to explain the effects of air pollution on male and female working hours in the US from 2006 to 2019. The empirical model in the study suffers from two potential biases, endogeneity, and sample selection bias. A joint IV-Heckman method is used to simultaneously address the potential problems of endogeneity and sample selection introduced by OLS. As in the first chapter, thermal inversion strength is implemented as an instrumental variable. The study finds that the working hours of women, rather than men, are significantly affected by increased exposure to PM2.5. The baseline result suggests that a 10% increase in average annual PM2.5 concentration leads to a 3.5% significant reduction in female working hours and no significant reduction in male working hours. Furthermore, this study finds strong evidence of racial disparity in the impact of air pollution on working hours. The effect of air pollution on working hours is more pronounced for the Black population in general and women in particular. The study also confirms that the working hours of women with children below five years are adversely affected due to heightening air pollution irrespective of their marital status. These findings suggest that air pollution may contribute to the gender difference in working hours in the US.

The third essay examines the effects of within industry trade on relative wage movements in the US from 2002 through 2017. Recent decades have seen a significant increase in the proportion of trade that is within industries relative to trade that is across industries. Classical trade theory and theories of the outsourcing of intermediate goods emphasize the Stolper Samuelson-type effects of relative wages. No such systematic movements in relative wages are expected from the rise in intra-industry trade. We extend a two-stage wage decomposition to the case of imperfect competition and intra-industry trade to investigate the role of rising within industry trade on relative occupational wages over the period from 2002 through 2017. We find that the rise in within industry trade is more significant than other structural variables considered in the study in explaining movements in relative wages over this period.

This research provides considerable evidence that low and moderate air pollution can drive migration decisions in the United States. It is also evident from this work that there exists a gender difference in the impact of air pollution on labor supply. Finally, findings from this research also suggest that a rise in intra-industry trade can drive relative wage movements in the United States