Labor income inequality among couples has increased by 33 percent in the U.S. over the past half-century. Over the same period, the correlation of labor income within couples has also increased sharply. Is this increase in sorting over labor income a cause for the rise of labor income inequality among couples? Using the March supplement of the CPS, first, I find that there has been a sharp increase in positive sorting over labor income in the U.S. in the 1970-2018 period. The top decile of men's earners married to the top decile of women's earners has doubled from 10.6 percent in 1970 to 23.3 percent in 2018. Second, I use a bounded copula framework as a reference distribution to track the relative changes in labor income inequality among couples. Using this framework, I find that positive sorting over labor income did play a role in increasing labor income inequality among couples in the 1970-1990 period; however, I find little evidence to suggest that this relationship existed in the 1990-2018 period.

The study of occupations has a long history in the social sciences. The United States Census Bureau creates occupation codes to systematically identify similar jobs and classify them into occupations; since 1980 Census occupation codes are based on the Standard Occupation Classification System. Periodic updates to occupation codes are necessary for documenting emerging and disappearing occupations and appropriately reflecting the labor market. While these updates are important for capturing the changing jobs that are performed in the United States, occupation code changes make it more complicated for researchers to make appropriate comparisons over time. This paper presents the methodology used at IPUMS to harmonize 2010 and 2002 Census occupation codes. Our work builds upon the foundation of existing harmonized occupation data available through IPUMS (variables OCC1950 and OCC1990). Combining our contribution with existing IPUMS infrastructure creates a consistent measure of occupation for all seven coding schemes over the 1950-2017 period.

Women and men work in markedly different jobs, leading to persistent occupational segregation by gender. This paper provides evidence that gender differences in how individuals value activities performed at work, termed job tasks, can help explain these sorting patterns. I conduct a hypothetical choice experiment to elicit workers’ willingness to pay for a set of tasks that are more frequently performed by one gender than the other. The experimental scenarios ask participants to choose between two hypothetical jobs that differ in terms of pay and the amount of time spent on a gender-typical task, but are otherwise the same. I find significant gender differences in willingness to pay for three of the five tasks that I examine. Willingness to pay is significantly higher among participants who report spending more time on a task in their current job, suggesting that estimates are correlated with actual sorting behavior. I show that gender differences in preferences for the tasks that I investigate can account for a substantial portion of occupational segregation in the U.S. labor market.

Ridesourcing services provided by Transportation Network Companies (TNCs) such as Uber and Lyft are spreading across the United States and are thriving. As a result of TNCs' expansion, there has been concern that ridesourcing is disrupting the traditional for-hire vehicle market, and those drivers are suffering. Based on 12-year Integrated Public Use Microdata Series (IPUMS) datasets from 2005 to 2016, we investigate how the income, worker classification, and employment status of for-hire vehicle drivers in the United States had changed after Uber entered their local markets. We find that with the entry of Uber, the hourly wage income of for-hire drivers had decreased, the percentage of self-employed drivers had increased, and the likelihood of being employed had increased in “Uber-adopted” metropolitan areas. The results confirm the disruptive effect of ridesourcing services on the for-hire vehicle industry and its labor force in the United States. The analysis for the five largest metropolitan areas provides more detailed evidence of the effect of ridesourcing.

Residential segregation in Boston has drawn considerable attention in recent years, but much less notice has been given to the issue with respect to the rest of New England. This regional brief focuses on residential segregation between all minority groups and non-Hispanic white residents in metro areas throughout the region. New England’s population is predominately non-Hispanic white; however, the region has diversified considerably since 1990, as most of the population growth has occurred among minority groups. Residential segregation by race/ethnicity declined over that same period in nearly every medium-sized and large metro area in New England, though the levels remain modestly higher than the national averages. Contrary to the broader trends, many of New England’s small metro areas saw increases in residential segregation during the 1990–2017 period. While racial and ethnic minorities in the region tend to be clustered in fewer neighborhoods relative to the rest of the United States, they are less isolated from non-Hispanic whites; that is, on average, a minority resident of a New England metro area has a considerably higher share of white neighbors compared with minorities elsewhere in the country.

This paper aims to assess the effects of industrial pollution on infant mortality between the years 1850-1940 using full count decennial censuses. In this period, US economy experienced a tremendous rise in industrial activity with significant variation among different counties in absorbing manufacturing industries, which is used as a proxy for variation in space-time industrial pollution. To compensate for lack of data, measures of gender ratio are instrumented for infant mortality since male embryos are more vulnerable to external stressors like pollution during prenatal development. Consistent significant results in different scenarios asserts that infants and more specifically male infants had paid the price of pollution during upswing in industrial growth at the dawn of the 20th century. Contemporary datasets are used to verify the validity of the proxies and the instruments.

We contend that the rise of mass incarceration in the United States can be framed through the lens of stratification economics, which views race- and class-based discrimination as a rational attempt on behalf of privileged groups to preserve their relative status and the material benefits which that status confers. Using the first (to our knowledge) local-level data set on incarceration rates by race, we explore the relationship between income inequality, poverty, and incarceration at the commuting zone level from 1950 to the present. Consistent with Michelle Alexander’s hypothesis that expansion of the penal system and the rise of “tough on crime” policy were efforts by privileged groups to drive a wedge into working-class political coalitions formed out of the Civil Rights Movement, we find that labor markets with greater inequality experienced larger increases in the overall incarceration rate. Furthermore, we find that relative rates of poverty play a key role in explaining differential effects of mass incarceration across race. Areas where White poverty rates were large relative to non-White poverty rates experienced no significant change in White incarceration, but an expansion of non-White incarceration. These findings have implications for policies related to economic and judicial systems.

This report investigates factors in metropolitan statistical areas (MSAs) associated with higher price appreciation at the low end of the market relative to appreciation at the high end. US home prices have exceeded the 2006 peak, just before the 2008 housing market collapse. Between 2000 and 2019, the house price index has almost doubled, with a larger increase occurring at the low end of the market. Home prices in the lowest 20th percentile have increased 126 percent while prices in the top 20th percentile have increased 86 percent. The relative price appreciation varies substantially across MSAs. For example, in Los Angeles, the appreciation rate of low-tier homes was 133 percentage points higher than for high-tier homes, while the difference was only 18 percentage points in Chicago. We find that in MSAs with higher employment growth, stronger zoning and land-use regulation, and less land available for development, prices for low-tier homes have increased more than for high-tier homes. The investor share and its growth in the home transaction market is not associated with the price growth rate differences between low-tier and high-tier homes. The relatively greater increase in housing costs for low-income households has caused residual income inequality (household income minus housing costs) to increase more than income inequality. Additionally, because MSAs with lower home price growth rates also experienced lower employment growth, housing cost burden and residual income inequality increased at a similar level across most MSAs. This suggests the need for policies to mitigate high housing cost burdens at the local level because the reasons for the increased burden—a weak labor market or stringent housing supply— differ by MSA.

This dissertation explores the extent to which policy alleviates or exacerbates labor market risks borne by workers looking at three different policies: Right-to-Work (RTW) laws, disability insurance, and state-level minimum wages. My first chapter examines impact of state RTW laws using Current Population Survey (CPS) data. Employing a state fixed effects model and a Synthetic Control design focusing on Wisconsin’s RTW legislation, I find that RTW laws are associated with sizeable increases in public sector free riding by union covered workers and considerable decreases in private sector union membership and coverage. RTW laws also substantially decrease private sector employer health insurance contributions and marginally reduce private sector wages. Given the role that union decline has had increasing worker precarity, it appears that RTW laws exacerbate labor market risks for workers. My second chapter examines the impacts of changes to disability insurance eligibility in the United Kingdom using an individual fixed effects model and data from the United Kingdom Household Longitudinal Study (UKHLS) and the Department of Work and Pensions (DWP). Increased reassessment using more stringent eligibility criteria at the local level is associated with a higher probability of disability receipt. However, my findings are suggestive of policy spillover effects, wherein individuals assigned to work-conditioned benefits seek out other benefits. While relatively well-targeted, this reform created temporary financial uncertainty for a vulnerable population and induced those assigned to workconditioned benefits to find other methods of support. My third chapter employs the American Community Survey (ACS) data and a Quadruple Difference (DDDD) model to examine the compositional effects of state minimum wage increases, specifically on the rates of self-employment within occupations. A state minimum wage increase is associated with a decrease in self-employment, particularly unincorporated self-employment within lowwage occupations with flexible work hours. Self-employed workers are induced into hired employment when the wage they can expect to earn from hired employment matches or exceeds expected earnings in self-employment. Therefore the minimum wage, in allowing for workers to select into work arrangements that are inherently less risky, may present another policy lever for reducing the labor market risks borne by workers.

This paper applies information-theoretic measures to consider the systemic effects on individual incomes of complex patterns of social and economic discrimination by race, ethnicity, and gender in the U.S. It estimates non-parametric indices of joint, conditional or incremental, and mutual information between income, social identity, and observable economic characteristics obtained using large-scale cross-sectional data from that economy. The paper advances new conceptual and empirical approaches to the nature and measurement of economic discrimination and inequalities of opportunity, founded on the formal informativeness of measures of social identity on economic outcomes. Estimated values for indices of informational association also cast new light on the effects of the intersections of gender and race/ethnicity on income, perverse patterns in the effects of education across different groups, and a few notable dynamic changes in patterns of income distribution in that economy over the past 40 years.

During the nineteenth century, Dutch female labor force participation (FLFP) was relatively low. Most scholars argue that social norms and rising wages were driving this development. However, their conclusions principally apply to married women. We study unmarried women’s LFP (UFLFP) and investigate a third driver: shifting sectoral employment shares. We include all three drivers in a logistic regression based on nearly 2 million marriage records from 1812 to 1929. We conclude that social norms and income levels mattered, but that shifting sectoral employment shares were driving the decline in UFLFP because sectors with low demand for female laborers expanded.

Population studies of Puerto Rico have recently gained salience as recent environmental and economic shocks have affected demographic behavior. In this review, we describe the major population databases on Puerto Rico and discuss the strengths and weaknesses of each. These databases include three from the US Census Bureau—the Decennial Population Census, the Puerto Rico Community Survey, and the Population Estimates Program—and three additional databases used to study migration (the Bureau of Transportation Statistics), health (the Behavioral Risk Factor Surveillance System), and labor (the Puerto Rico Labor Survey). These databases can be used to study emerging demographic topics, as well as inequities in population outcomes.

Purpose U.S. Asian/Pacific Islander (API) communities often reside in metropolitan areas with distinct socioenvironmental attributes. Residence in ethnic enclaves, socially distinct areas, is associated with lower gestational diabetes (GDM) risk, yet exposure to air pollution is associated with increased GDM risk. We examined joint effects of ethnic enclaves and air pollution to better understand GDM risk among API women, the group with the highest GDM prevalence. Methods We examined 9,069 API births in the Consortium on Safe Labor (2002-2008). Ethnic enclaves were defined as areas ≥66th percentile for percent API residents, dissimilarity (distribution of API and White residents), and isolation (degree that API individuals interact with API individuals). High levels of 14 volatile organic compounds (VOC) were defined as ≥75th percentile. Four categories were created for each VOC: LowVOC/Enclave (reference), LowVOC/NoEnclave, HighVOC/Enclave, HighVOC/NoEnclave. GDM was reported in medical records. Hierarchical logistic regression estimated odds ratios(OR) and 95% confidence intervals(95%CI) between joint exposures and GDM, adjusted for maternal factors and area-level poverty. Risk was estimated for 3-months preconception and first-trimester exposures. Results Enclave residence was associated with lower GDM risk regardless of VOC exposure. Preconception benzene exposure was associated with increased risk among those outside enclaves [HighVOC/NoEnclave (OR: 3.45, 95% CI: 1.77, 6.72)], and risk was mitigated within enclaves, [HighVOC/Enclave (OR: 2.07, 95%:1.09, 3.94)]. Risks were similar for 12/14 VOCs during preconception and 10/14 during the first trimester. Conclusions API residence in non-enclave areas is associated with higher GDM risk, regardless of VOC level. Enclave residence may mitigate effects of VOC exposure due to lower stress levels.

In this paper we discuss some strands of the recent literature on the evolution of gender gaps and their driving forces. We will revisit key stylized facts about gender gaps in employment and wages in a few high-income countries. We then discuss and build on one gender-neutral force behind the rise in female employment, namely the rise of the service economy, which is also closely related to the polarization of female employment and to the geographic distribution of jobs, which is expected to be especially relevant for female employment prospects. We finally turn to currently debated causes of remaining gender gaps and discuss existing evidence on labor market consequences of women's heavier caring responsibilities in the household. In particular, we highlight how women's stronger distaste for commuting time may feed into gender pay gaps by making women more willing to trade off steeper wage gains for shorter commutes.

This paper examines the effects of the economic conditions – unemployment rate and unemployment change – on the rate of returns to schooling, as well as the effects on the gender difference in the returns to schooling during business cycles. Data from American Community Survey 2000–2017 shows that economic conditions have significant effects on the rate of return to schooling for both men and women, and the gender difference exists. Unemployment has positive effects on the rate of return to schooling for both men and women. The returns to schooling are expected to increase in recessions and decrease in expansions for both men and women. In addition, the effects are greater for men than for women. The female-male difference in the rate of return to schooling is expected to be smaller as unemployment rate goes up and greater as unemployment rate goes down. We also find men’s returns to schooling are more sensitively responded to the economic conditions change than women’s.

Jim Crow segregated African Americans and whites by law and practice. The causes and implications of the associated de jure and de facto residential segregation have received substantial attention from scholars, but there has been little empirical research on racial discrimination in public accommodations during this time period. We digitize the Negro Motorist Green Books, important historical travel guides aimed at helping African Americans navigate segregation in the pre-Civil Rights Act United States. We create a novel panel dataset that contains precise geocoded locations of over 4,000 unique businesses that provided non-discriminatory service to African American patrons between 1938 and 1966. Our analysis reveals several new facts about discrimination in public accommodations that contribute to the broader literature on racial segregation. First, the largest number of Green Book establishments were found in the Northeast, while the lowest number were found in the West. The Midwest had the highest number of Green Book establishments per black resident and the South had the lowest. Second, we combine our Green Book estimates with newly digitized county-level estimates of hotels to generate the share of non-discriminatory formal accommodations. Again, the Northeast had the highest share of non-discriminatory accommodations, with the South following closely behind. Third, for Green Book establishments located in cities for which the Home Owner’s Loan Corporation (HOLC) drew residential security maps, the vast majority (nearly 70 percent) are located in the lowest-grade, redlined neighborhoods. Finally, Green Book presence tends to correlate positively with measures of material well-being and economic activity.

Coal-fired power plants release substantial air pollution, which included over 60% of US sulfur dioxide emissions in 2014. Such air pollution may exacerbate asthma, but direct studies of the health impacts linked to power plant air pollution are rare. Here we take advantage of a natural experiment in Louisville, Kentucky, where one coal-fired power plant was retired and converted to natural gas, and three others installed SO2 emission control systems between 2013 and 2016. Dispersion modelling indicated that exposure to SO2 emissions from these power plants decreased after the energy transitions. We used several analysis strategies, which include difference-in-differences, first-difference and interrupted time-series modelling to show that the emissions control installations and plant retirements are associated with a reduced asthma disease burden related to hospitalizations and emergency room visits at the ZIP-code level, and to individual-level medication use as measured by digital medication sensors.

The purpose of this white paper is to explore best practices in the world of private workforce development in Texas—with an eye toward those programs that demonstrate the most promise for successfully lifting people out of poverty, promoting individual and family success, moving individuals to self-sufficiency, and leading to general human flourishing. In broad economic terms, Texas is one of the most prosperous states in America. One recent study ranked Texas the third best state for business in 2020 (DePietro); another gave the Lone Star State the distinction of the best state in the U.S. to start a new business (McCann). Forbes has ranked Texas as the fourth best state for economic climate and the first for growth prospects—and what’s more, 100 of the 1,000 largest public and private companies in the U.S. are based in Texas. In terms of pure scope, Texas’ economy is the second largest in the U.S., only trailing California (Forbes). And if Texas were a nation, it would have the 10th largest gross domestic product in the world—ahead of Canada and behind Brazil—at around $1.7 trillion in economic output (Perry).

Recently, life expectancy in the U.S. has stagnated or declined for the poor and working classes and risen for the middle and upper classes. Declining labor union density – the percent of workers belonging to labor unions – has contributed to burgeoning income inequity. In this dissertation, we examined the relationship between unionism and health, and tested whether declining union density has exacerbated racial and educational mortality inequities. In the first study, we analyzed the state-level relationships between union density, mortality, and mortality inequities from 1986-2016. State-level all-cause mortality and overdose/suicide mortality overall and by gender, gender-race, and gender-education came from CDC, while state-level union density came from the Current Population Survey. Using inverse-probability-of-treatment-weighted Poisson models with state and year fixed effects, we estimated three-year-moving-average union-density’s effects on the following year’s mortality rates. Then, we tested for gender, gender-race, and gender-education effect-modification. Finally, we estimated how racial and educational all-cause mortality inequities would change if union density increased to baseline levels. Overall, a 10% increase in union density was associated with a 17% relative decrease in overdose/suicide mortality (95% CI: 0.70, 0.98), or 5.7 lives saved per 100,000 person-years (95% CI: -10.7, -0.7). Union density’s absolute (lives-saved) effects on overdose/suicide mortality were stronger for men than women, but its relative effects were similar across genders. However, our estimates were sensitive to the analytic approach. Moreover, union density had little effect on all-cause mortality overall or across subgroups, and modeling suggested union-density increases would not affect mortality inequities. In the second study, we analyzed the individual-level relationship between union membership, self-rated health (SRH), and moderate mental illness. The union membership and SRH analyses used data on 16,719 Panel Study of Income Dynamics (PSID) respondents followed between 1985 and 2017, while the union membership and mental-illness analyses included 5,813 PSID respondents followed between 2001 and 2017. Using the parametric g-formula, we contrasted cumulative incidence of the outcomes under two hypothetical scenarios, one in which we set all employed-person-years to union-member employed-person-years (union scenario), and one in which we set no employed-person-years to union-member employed-person-years (non-union scenario). We also examined whether the scenarios’ effects varied by gender, gender-race, and gender-education in stratified models. Overall, the union scenario did not reduce incidence of poor/fair SRH (RR: 1.01, 95% CI: 0.95, 1.09; RD: 0.01, 95% CI: -0.03, 0.04) or moderate mental illness (RR: 1.02, 95% CI: 0.92, 1.12; RD: 0.01, 95% CI: -0.04, 0.06) relative to the non-union scenario. These effects largely did not vary by subgroup. In the final study, we examined the individual-level relationship between union membership, mortality, and mortality inequities using data on 23,022 PSID respondents followed between 1979 and 2017. First, using the parametric g-formula, we contrasted cumulative incidence of mortality in the union and non-union scenarios. Next, we examined whether the scenarios’ effects varied by race or education in stratified models. Finally, we estimated how racial and educational mortality inequities would change if union density had remained at 1979 levels throughout follow-up rather than that at 2015 levels. Overall, the union scenario modestly reduced mortality (RR: 0.90, 95% CI: 0.80, 0.99; RD per 1,000: -18.7, 95% CI: -36.5, -0.9) relative to the non-union scenario. However, the scenarios’ effects largely did not vary by subgroup, and modeling did not suggest racial and educational mortality inequities would lessen if union density had remained at baseline levels. Overall, we found little evidence that declining union density (at least as operationalized in this dissertation) explained changing racial and educational mortality inequities in the U.S. over the last several decades. However, our results did suggest that increasing state-level union density might decrease overdose/suicide mortality, and that increasing individual-level union membership might decrease all-cause mortality

We collected 92,832 user-contributed and publicly available family trees from rootsweb.com, including 250 million individuals who were born in North America and Europe between 1630 and 1930. We cleaned and connected the family trees to create a population-scale and longitudinal family tree dataset using a workflow of data collection and cleaning, geocoding, fuzzy record linkage and a relation-based iterative search for connecting trees and deduplication of records. Given the largest connected component of nearly 40 million individuals, and a total of 80 million individuals, we generated, to date, the largest population-scale and longitudinal geo-social network over centuries. We evaluated the representativeness of the family tree dataset for historical population demography and mobility by comparing the data to the 1880 Census. Our results showed that the family trees were biased towards males, the elderly, farmers, and native-born white segments of the population. Individuals were highly mobile – in our 1880 sample of parent-child pairs where both were born in the U.S., 47% were born in different states. Our findings agreed with prior studies that people migrated from East to West in horizontal bands, and the trend was reflected in the dialects and regional structure of the U.S.