This dissertation is composed of three chapters on the effects of import exposure. For my dissertation I mainly use the variation of import competition across local labor markets to explore its impact on labor market outcomes (e.g., wages and employment status), human capital investment decisions (choice of college major), and education-job mismatch. Chapter one explores the relationship between increasingly intense Chinese import competition and American college students’ choice of major in the 2000s. By employing a modified version of the measure for Chinese import competition from Autor, Dorn, and G. Hanson (2013) and analyzing the relationship between industries and college majors, I find that rising Chinese trade exposure of nineteen industries in the 2000s has a negative effect on American students’ choice of six engineering majors. The magnitudes of the effects range from 0.62 to 0.69 percentage point decreases in the probability of choosing those six engineering majors. I also find that males are more negatively affected by Chinese import competition in terms of the choice of the six engineering majors, whereas no significant results exist if I restrict my sample to females. Chapter two analyzes how increased trade exposure affects students’ choice of STEM major. I first present a simple model to illustrate how trade exposure impacts students’ utility functions through their self-beliefs about labor market outcomes and then use assorted data to show that import competition positively affects the choice of STEM major. I find that increased import exposure in the 2000s leads to 1.05 and 0.72 percentage point increases in the probability of choosing STEM majors for college underclassmen and upperclassmen, respectively. As for labor market outcomes, my results suggest that a rise in import competition leads to a pronounced negative effect on weekly wages, employment status, and full-time employment across STEM and non-STEM occupations from the late 1990s through the 2000s. STEM occupations, however, are less negatively impacted by import competition, which helps explain why a rise in import exposure increases the probability of students choosing STEM majors. Chapter three investigates the impact of import exposure on education-occupation mismatch. I first use the concept of a matching function to explain the connection between mismatch and the supply of and demand for college graduates. Next, I use an input-output table to construct a measure of import exposure that accounts for both direct and indirect trade shocks. Findings show that increased import exposure leads to a rise in education-occupation mismatch from 2011 through 2019. Moreover, for the supply side I present that a rise in import exposure significantly increases the number of bachelor’s degrees awarded in 4-year colleges and in most degree fields. However, for the demand side, I do not observe corresponding increases in occupational employment for most fields of education. The unbalanced demand for and supply of college graduates might potentially explain the rise in education-occupation mismatch.

Manufacturing has historically played a significant role in productivie and R&D. Jorenson (2001) suggests that advances in microprocessors alone were associated with 50 percent of total factor productivity growth in the US and worldwide in the 1990s. This outsided role in R&D and productivity appears to continue today, even with significant changes across the sector in technology and globalization. US manufacturing is a disproportionate source of private R&D spending relative to its share of employment and global value aded (GVA) and has higher an average labor productivity relative to other sectors.

Background: Previous research has shown a significant association between psychological distress (PD) and all-cause mortality. However, there is a dearth of studies quantifying the contributions of sociodemographic and behavioral characteristics to group differences in mortality. In this study, we identify factors of mortality differences by PD. Methods: The Blinder-Oaxaca decomposition analysis was used to quantify the contributions of individual sociodemographic and behavioral characteristics to the observed mortality differences between United States (US) adults with no PD and those with serious psychological distress (SPD), using the pooled data from the 1997-2014 National Health Interview Survey prospectively linked to the 1997-2015 National Death Index (N = 263,825). Results: Low educational level, low household income, and high proportions of current smokers, renters, former drinkers, and adults experiencing marital dissolution contributed to high all-cause mortality among adults with SPD. The relative percentage of all-cause mortality disparity explained by socioeconomic and demographic factors was 38.86%. Approximately 47% of the mortality disparity was attributed to both sociodemographic and behavioral risk factors. Lower educational level (21.13%) was the top contributor to higher all-cause mortality among adults with SPD, followed by smoking status (13.51%), poverty status (11.77%), housing tenure (5.11%), alcohol consumption (4.82%), marital status (3.61%), and nativity/immigrant status (1.95%). Age, sex, and body mass index alleviated all-cause mortality risk among adults with SPD. Conclusions and Global Health Implications: Improved education and higher income levels, and reduced smoking among US adults with SPD might eliminate around half of the all-cause mortality disparity by SPD. Such a policy strategy might lead to reductions in mental health disparities and adverse health impacts both in the US and globally.

I analyze the relationship between immigrant status and an array of economic outcomes with a special focus on undocumented status and housing. I evaluate the role public policy has played in influencing the observed relationships. In the first chapter, I provide evidence that fear of deportation results in search frictions among undocumented immigrants in the rental housing market, leading undocumented immigrants to pay a rent premium and devote a greater share of their income to housing. Making use of a triple differences empirical strategy, I show that sanctuary city policies work to reduce or even eliminate this premium. In the second chapter, using both difference-in-differences and synthetic control methods, I show that undocumented status is a barrier to homeownership. The Deferred Action for Childhood Arrivals policy (DACA) reduced the existing homeownership gap between undocumented and legal resident immigrants. Additionally, I find that a clarification made by the U.S. Treasury Department in 2003 that expanded the availability of certain financial services to undocumented immigrants led to an increase in the relative number of Hispanic home loan applications in counties with the highest concentrations of Hispanic undocumented immigrants, providing evidence that limitations on access to credit have been at least one factor responsible for the homeownership gap. In the final chapter, I make use of a fuzzy differences-in-discontinuities empirical strategy to evaluate the consequences of providing Medicaid to low-income immigrants. I find evidence that provision of public health insurance reduces the uninsured rate among lowincome immigrants but also crowds out spending on private insurance. Importantly, the reduction in private insurance is driven by immigrants substituting away from purchased (non-group) insurance. I do not find evidence of a similar reduction in employer-sponsored health insurance. Consistent with this finding, I also do not detect reductions in labor supply in response to Medicaid.

Introduction: the percentage of people with cognitive difficulty reported in 2000, 2010, and 2019 among residents in New York City. Specifically, residents from the five boroughs in New York City—Bronx, Brooklyn, Manhattan, Queens, and Staten Island—were included in this analysis. Methods: This report uses the American Community Survey PUMS (Public Use Microdata Series) data for all years released by the Census Bureau and reorganized for public use by the Minnesota Population Center, University of Minnesota, IPUMSusa, (https://usa.ipums.org/usa/index.shtml). See Public Use Microdata Series Steven Ruggles, J. Trent Alexander, Katie Genadek, Ronald Goeken, Matthew B. Schroeder, and Matthew Sobek. Integrated Public Use Microdata Series: Version 5.0 [Machine-readable database]. Minneapolis: University of Minnesota, 2020. Discussion: The percentage of people with cognitive difficulty remained relatively stable in New York City at around 5% between 2000 and 2019. Overall, Non-Hispanic whites reported a slightly higher percentage of cognitive difficulty, but by 2019, Asians reported the highest at 5.3%, and non-Hispanic blacks the lowest at 4.5%. By 2019, Non-Hispanic whites and Asians had the highest cognitive difficulty in the Bronx (5.4%), non-Hispanic blacks in Brooklyn (6.5% by 2019), Latinos in Manhattan (13.3% in 2019), and Asians in Queens (5.0%) and Staten Island (6.3%). Significant differences were observed among the five largest Latino subgroups in New York City. Cognitive difficulty among Puerto Ricans consistently increased across the three timepoints, but it remained relatively stable among Mexicans and Colombians, while decreasing among Ecuadorians and Dominicans. By 2019, rates of cognitive difficulty among Puerto Ricans increased to the highest overall at 8.5%, whereas Dominicans (1.7%) and Ecuadorians (0.5%) decreased to the lowest percentages of cognitive difficulty across all timepoints.

: Traditional models relying on standard variables like the U.S. Hispanic unemployment rate fared well in explaining remittances to CAPDR and Mexico during the pre-pandemic period. However, they fail to predict the sustained growth in remittances since June 2020, including the significant increase in the average amount remitted. Using data from over 300 remittances corridors (from 23 U.S. states to 14 Salvadoran departments), we find that this increase is primarily explained by the dynamics of U.S. states real wages, as well as more temporary factors like U.S. unemployment relief (including the extraordinary pandemic support), U.S. states mobility, and COVID-19 infections at home. The paper also analyses what role the change in the modes of transmission of remittances, additional U.S. fiscal stimulus and U.S. labor market developments, especially in the sectors were CAPDR and Mexican migrants preponderantly work, play in explaining aggregate remittances growth.

We document a robust negative relationship between mean annual hours in an occupation and the dispersion of annual hours within that occupation. We study a unified model of occupational choice and labor supply that features heterogeneity across occupations in the return to working additional hours and show that it can match the key features of the data both qualitatively and quantitatively. Occupational choice in our model is shaped both by selection on comparative advantage and selection on tastes for leisure. Our quantitative work finds that the dominant source of differences in hours across occupations is selection on tastes for leisure.

The log odds ratio is a common parameter to measure association between (bi-nary) outcome and exposure variables. Much attention has been paid to its parametric but robust estimation, or its nonparametric estimation as a function of confounders. However, discussion on how to use a summary statistic by averaging the log odds ratio function is surprisingly difficult to find despite the popularity and importance of averaging in other contexts such as estimating the average treatment effect. We propose a couple of efficient double/debiased machine learning (DML) estimators of the average log odds ratio, where the odds ratios are adjusted for observed (potentially high dimensional) confounders and are averaged over them. The estimators are built from two equivalent forms of the efficient influence function. The first estimator uses a prospective probability of the outcome conditional on the exposure and confounders; the second one employs a retrospective probability of the exposure conditional on the outcome and confounders. Our framework encompasses random sampling as well as outcome-based or exposure-based sampling. Finally, we illustrate how to apply the proposed estimators using real data.

We exploit a recent panel of city-level data with rich demographic information to estimate the distributional effects of Department of Defense spending and its effects on a range of social outcomes. The income generated by defense spending accrues predominantly to households without a bachelor's degree. These households as well as Black households tend to disproportionately benefit from this spending. Defense spending also promotes a range of beneficial social outcomes that are often targeted by government programs, including reductions in poverty, divorce rates, disability rates, and mortality rates, as well as increases in homeownership, health insurance rates, and occupational prestige. These effects vary across demographic groups.

Introduction: This report analyzes trends in poverty in New York City over a period spanning from the year 1990 to 2019, including maps of poverty hot spots in the city. Methods: This report uses the American Community Survey PUMS (Public Use Microdata Series) data for all years released by the Census Bureau and reorganized for public use by the Minnesota Population Center, University of Minnesota, IPUMSusa, (https://usa.ipums.org/usa/index.shtml). See Public Use Microdata Series Steven Ruggles, J. Trent Alexander, Katie Genadek, Ronald Goeken, Matthew B. Schroeder, and Matthew Sobek. Integrated Public Use Microdata Series: Version 5.0 [Machine-readable database]. Minneapolis: University of Minnesota, 2020. Discussion: Latinos and non-Hispanic blacks were more affected by poverty (21.4% and 19.4% respectively) compared to non-Hispanic whites and Asians (9.9% and 14.6%). And while the Puerto Rican population had the highest poverty rate (23.2% in 2019), Colombians registered the lowest poverty rates among the largest Latino subgroups in New York (10.8% in the same year). The second main finding is that poverty is not equally distributed across space, and that these patterns persisted despite the general decline of poverty between the years 2000 and 2019. Essentially, poverty hotspots that were mainly located in the Bronx when poverty rates peaked, remained in the same geographical areas even after the general decline in poverty. Additionally, the highest poverty areas are also majority Latino and non-Hispanic black areas. Conversely, areas with low poverty rates are mainly also regions with higher levels of non-Hispanic white population.

Women are still devoting more time than men to child-rearing. That makes women's time spent at work crucial for their fertility decisions. At the same time, the global temperature has been rising, affecting labor supply and productivity. In this study, we investigate whether temperature influences fertility through its effect on the female labor supply (num-ber of hours worked) and productivity. We also test the adaptation hypothesis. According to the latter, couples' fertility in warmer climates may be less affected by higher temperatures. We employ individual-level data from the IPUMS-USA for the period 2002-2019 and temperature data from the World Bank between 1901 and 2001. The obtained results show that temperature affects fertility mainly in the category of women who work full-time in jobs with lower educational demands. We found no effect on fertility for women of either higher or lower education, who work part-time. We also found some evidence in favor of the adaptation hypothesis.

Place-based attachments are important but often overlooked. Place-based attachments can be beneficial but often harm individuals tied to struggling areas. In this address, I discuss my own education and migration experiences and then more generally discuss sense of belonging as a friction to migration. I also present descriptive statistics related to place-based attachments. Most persons born in the U.S. live in their birth state as adults. Birth-state residence has increased over time, especially among the highly educated. I also present evidence that college graduates who reside in their birth state experience a wage penalty that is increasing over time.

This study examined differences across Latine heritage groups (i.e., Mexican, Puerto Rican, Cuban, Dominican, Central American, South American) in rates of US drug overdose mortality. The study utilized 2015-2019 mortality data from the National Center for Health Statistics for 29,137 Hispanic individuals who died of drug overdose. Using population estimates from the American Community Survey, age-standardized drug overdose mortality rates were calculated by specific Latine heritage and sex, nativity, educational attainment, and geographic region. Standardized rate ratios (SRRs), incidence rate ratios (IRRs) from negative binomial regression models, and 95% Confidence Intervals (CIs) were calculated, and multiple imputation was used for missing Latine heritage group in select models. Drug overdose mortality rates in the Puerto Rican heritage population were more than three times as high as in the Mexican heritage population (IRR 3.61 [95% CI 3.02-4.30] in unadjusted model; IRR 3.70 [95% CI 3.31-4.15] in model adjusting for age, sex, nativity, educational attainment, and region; SRR 3.23 [95% CI, 3.15-3.32] in age-standardized model with missing Hispanic heritage imputed). Higher age-standardized rates of drug overdose mortality were observed in males than females across all Latine groups, yet the magnitude of the sex differential varied by Latine heritage. The relationship between drug overdose mortality and nativity differed by Latine heritage; in all groups except Puerto Rican, overdose mortality rates were significantly higher in the US-born than those not US-born. In contrast, overdose mortality rates were significantly lower in US-born Puerto Ricans than in Puerto Ricans who were not US-born (e.g., born in Puerto Rico; SRR, 0.84 [95% CI 0.80-0.88]). The relationship between drug overdose mortality and educational attainment (for ages 25+) also varied between Latine groups. The diverse subgroups comprising the US Latine population vary not only in rates of drug overdose mortality, but also in demographic risk factors for fatal drug overdose.

Scholarship on slavery's legacy and enduring consequences has largely been limited to the South under the assumption that this region continues to be more deeply affected by slavery than other regions, notably the Northeast. Also overlooked is the extent to which slavery's consequences generalize beyond Black disadvantage, harming other racialized ethnic minority groups such as Latinos. Scholarship has paid little attention to the role of the state's institutions of social control-notably, law enforcement-in transmitting slavery's legacy after its formal abolishment. I address these issues using quantitative techniques assessing the relationship between prior slave dependence and contemporary policing practices in U.S. Southern and Northeastern counties. I argue that as a result of slavery's influence on local legal apparatuses and institutions of social control-in areas in the South and Northeast where slave dependence was greater-law enforcement today is less likely to protect minorities, resulting in higher rates of hate crime underreporting by police. Findings reveal slavery's nefarious consequences disadvantage Black populations but also spillover to Latinos, particularly in the South. In both regions, contemporary Black population concentration mediates slavery's relationship with the rate of police underreporting of anti-Black crimes.

Introduction: Persistently high rates of neonatal and maternal mortality have been associated with home births in many low-income and middle-income countries (LMICs). However, causal evidence of the effect of institutional deliveries on neonatal and maternal health outcomes is limited in these settings. Methods: We investigate the effect of institutional deliveries on neonatal mortality and maternal postpartum complications in rural India using data from the 2015-2016 Indian Demographic and Health Survey and an instrumental variable methodology to overcome selection bias issues inherent in observational studies. Specifically, we exploit plausibly exogenous variation in exposure to a road upgrade programme that quasi-randomly upgraded roads to villages across India. Results: We find large effects of the road construction programme on the probability that a woman delivered in a health facility: moving from an unconnected village to a connected village increased the probability of an institutional delivery by 13 percentage points, with the biggest increases in institutional delivery observed in public hospitals and among women with lower levels of education and from poorer households. However, we find no evidence that increased institutional delivery rates improved rates of neonatal mortality or postpartum complications, regardless of whether the delivery occurred in a public or private facility, or if it was with a skilled birth attendant. Conclusion: Policies that encourage institutional delivery do not always translate into increased health outcomes and should thus be complemented with efforts to improve the quality of care to improve neonatal and maternal health outcomes in LMICs.

Using newly digitized U.S. city-level data on hospitals, we explore how pandemics alter preferences for healthcare. We find that cities with higher levels of mortality during the Great Influenza of 1918-1919 subsequently expanded hospital capacity by more than cities experiencing less influenza mortality: cities in the top half of the mortality distribution increased their count of hospitals by 8-10 percent in the years after the pandemic. This effect persisted to 1960 and was driven by increases in non-governmental hospitals. Growth responded most in richer cities, exacerbating existing inequalities in access to healthcare. We do not find evidence that government-run hospitals or other types of city-level spending related to healthcare responded to pandemic intensity, suggesting that large health shocks do not necessarily lead to increased public provision of health services.

In this paper, we analyze income per capita absolute and club convergence of 383 metropolitan areas of the U.S. To our knowledge, we are the first to provide club convergence analysis of income per capita at the metropolitan area-level. Using data for 1969-2017 containing 18,767 observations , we employ a data-driven convergence model that also allows heterogeneity in the panel. We demonstrate evidence of the absence of absolute convergence and the presence of club convergence. Furthermore, we use the ordered logit model to analyze the economic, demographic, and several other factors influencing club membership. Our results reveal a diversity of growth experiences across the U.S. metropolitan areas by extensively accounting for heterogeneity using a novel transition dynamic modeling framework.

This paper studies racial segmentation in the US housing market since 1960. I document large differences in housing outcomes for Black and White households. In 1960, Black households on average are 20 percentage points less likely to own a house (relative to White households with the same income); if they owned, their house values are lower by the equivalent of almost one year of annual income; and even when renting they spend less by the equivalent of one month of rental expenditures. By 2019, the rent and price gaps have declined by about half, whereas the gap in ownership rates has not changed. To interpret these facts, I use a dynamic housing assignment model with a choice to buy or rent housing. I estimate the degree of market segmentation by inferring differences in the quality of housing available to Black and White households, and the resulting differences in rents, prices, and the cost of owning a home. The model infers that Black households pay higher quality-adjusted rents and prices, especially at the high end, and thus sort into lower quality homes. In terms of lifetime consumption-equivalent welfare, relative to an integrated market, the average Black household is five percent worse off in 1960 and remains one percent worse off in 2019.

Immigrant populations typically have lower mortality rates and longer life expectancies than their nonimmigrant counterparts. This immigrant mortality advantage has been a recurrent finding in demographic and population health research focused on contemporary waves of immigration. However, historical data suggest that European immigrants to the United States in the early twentieth century had worse health and higher rates of mortality, yet it remains unclear why a mortality advantage was absent for immigrants during this period. This article combines Vital Statistics records and Lee–Carter mortality models to analyze mortality by nativity status for the U.S. White population from 1900 to 1960, examining variation by age, sex, time, and place. Contrary to contem porary expectations of a for eign-born mortality advantage, White immi grants had higher mortality rates in the early 1900s, with the largest foreign-born disadvantage among the youngest and oldest populations. Although for eign-born and U.S.-born White mortality rates trended toward convergence over time, the foreign-born mortality penalty remained into the 1950s. A decomposition analysis finds that immigrants’ concentration in cities, which had higher rates of infectious disease mortality, accounted for nearly half of the nativity difference in 1900, and this place effect declined in subsequent decades. Additional evidence, such as a spike in mortality inequalities during the 1918 influenza pandemic, suggests that common explanations for the immigrant mortality advantage may be less influential in a context of high risk from infectious disease.

What is the contribution of demographic changes to house prices? We answer this question by analyzing various channels through which changes related to demographics may affect aggregate housing demand and supply differently. Specifically, we focus on the changes in life expectancy, international immigration, urbanization and fertility. As these changes are sustainable and predictable, understanding the role they play in the real estate markets is important for predicting future house prices. We develop and estimate a general equilibrium supply-demand model to evaluate the contribution of these factors and find that they can account for 40.54% of the observed house price growth from 1970 to 2010. Adopting the projection of these four factors provided by Census and the United Nations, we use our model to predict house prices through 2050. We find that the house price will keep growing from 2010 to 2050 by 4.42%–18.85%, depending on urbanization rates and the level of immigration.