This paper investigates the effect of a large negative agricultural shock, the boll weevil, on racial income gaps in the first half of the twentieth century. We draw on complete count census data to generate a new large linked sample of Black and white fathers and their sons. Fathers are observed before and after the arrival of the boll weevil in their county, and their sons are observed in their father’s household as children and again in 1940 as adults. In race specific difference-in-differences specifications and in triple differences specifications, we find that the boll weevil differentially affected wages of Black sons born after its arrival. Relative to white sons born after the boll weevil, Black sons born after the boll weevil saw a 6%increase in their wages. The magnitude of the effect is similar when the sample is constrained to sons whose father stayed in the South and to sons who stayed in the South. Evidence on changes in fertility and heights suggests that the relative gains were driven by improvements in early life conditions for Black sons born after the weevil’s arrival. The magnitude of the effect of the boll weevil on the Black-white wage gap can be better understood by comparing it with changes between 1940 and 1950, a period of rapid convergence. The boll weevil caused the Black-white wage gap to fall by roughly half of the decline between 1940 and 1950 in the South or a third of the decline nationally.

Self-employment among immigrants is a key source for income and social assimilation with natives. Rate of self-employment is significantly higher for immigrants than for native-born individuals, and the causal reasons behind this differential are still not well understood. We hypothesize that a key factor is that domestic employers often cannot accurately assess the quality of higher education received by the immigrants in their home countries. This lowers immigrants’ return to human capital in the traditional job market relative to natives. Our hypothesis predicts that this factor should be reflected in higher relative rates of self-employment for immigrants that rises with the level of education. We test and confirm this hypothesis using IPUMS micro-data from the USA.

Household food insecurity is a concern in the US given the negative effects associated with food insecurity. An interesting finding is that elderly households tend to be more food secure than younger households, even though many are on fixed income. A relevant question is what might elderly households be doing that is resulting in greater food security? One potential explanation is that in retirement, elderly households can invest in more time intensive activities that provide greater food security. In this study, we combine time use diaries with food security surveys to examine whether time spent on food production is associated with lower levels of food insecurity for elderly households. The data show that time spent in meal preparation and eating is increasing with older age cohorts. At the same time, food insecurity is declining steadily with older households. Grocery shopping and non-grocery food shopping do not show any relevant trends. Comparing households above and below 70 years of age, we find that time spent in food production does not explain differences in household level food insecurity.

As Vermont grapples with the dual issues of high greenhouse gas emissions from the transportation sector and inequitable transportation burdens, greater electric vehicle (EV) adoption will undoubtedly be a part of state policy solutions. However, similar to other technologies and products in the early stage of market adoption, EV distribution remains concentrated in the state's more urban and affluent areas. This report: • Provides an overview of Vermont’s progress electrifying transportation to meet climate goals; • Explains the importance of factoring transportation equity and justice into future program design and implementation; • Examines the extent to which state-funded EV incentive programs are (or are not) currently reaching those most in need; and • Recommends new research, data collection, and analysis that would further embed equity and justice in future development and evaluation in Vermont. Initial spatial analysis and linear regression models conducted for this report indicate that the uptake of program funding for EVs is largely concentrated in geographic areas of the state that do not, generally, include those communities with the highest transportation burdens. With this in mind, the report recommends that future program implementation be enhanced to more effectively reach the highest burdened communities. It is also recommended that data collection and program evaluation be improved to effectively monitor the success in providing those most burdened with increased access to affordable and clean electric vehicles.

We estimate the effect of employers providing health insurance and contributing towards health insurance premiums on employee job tenure. Using merged 2008-2018 data from the Job Tenure Supplement and Annual and Social Economic Supplement of the Current Population Survey, we find that employer health insurance is associated with one additional year of job tenure. We find that a thousand dollar increase in employer contributions to health insurance is associated with at least 83 additional days of job tenure, compared to less than 10 additional days for a thousand dollar increase in wages.

Using a unique dataset, this paper examines the extent to which streets named after prominent Confederate generals are related to Black-White labor market differentials. Examining individual-level data shows that Blacks who reside in areas that have a relatively higher number of Confederate streets are less likely to be employed, are more likely to be employed in low-status occupations, and have lower wages compared to Whites. I find no evidence that individual characteristics, local characteristics, or geographic sorting explain these results.

We investigate levels and trends in literacy--being able to read and write--among American Indians in the United States. Using 1900-1930 decennial census data, we document levels and trends in reading and writing for the 1900 through 1930 period and for birth cohorts from 1830 through 1920. We thus provide for American Indians a large-scale picture of the history of literacy. We document the pace and extent of American Indian literacy from very low for the birth cohorts of the early 1800s to fairly universal for the cohorts of the early 1900s. We also demonstrate that the increases in Native literacy were closely related to birth cohort, with successive new birth cohorts having higher levels of literacy. We found little evidence that increases in reading and writing from 1900 to 1930 happened because adults increased their literacy after the school years and as they matured across the adult life course. We also document important gender differences in American Indian literacy, with the proportion literate being lower for women than for men, but with the gender gap decreasing in later birth cohorts. There were also substantial literacy inequalities across geographical regions of the country—ranging from 19 to 74 percent literate across regions in 1900. The trajectories of literacy attainment also varied across regions in interesting ways. We also document that American Indian literacy was higher among those living in urban areas, those more integrated into the Euro-American community, and those with Euro-American ancestry.

We link newly-digitized personnel records of the U.S. government for 1907-1921 to census data to study the segregation of the civil service by race under President Woodrow Wilson. Using a difference-indifferences design around Wilson's inauguration, we find that the introduction of employment segregation increased the black-white earnings gap by 3.4-6.9 percentage points. This increasing gap is driven by a reallocation of existing black civil servants to lower-paid positions, lowering their returns to education. Importantly, the negative effects extend beyond Wilson's presidency. Using census data for 1900-1940, we show that segregation caused a relative decline in the home ownership rate of black civil servants. Moreover, by comparing children of black and white civil servants in adulthood, we provide suggestive evidence that descendants of black civil servants who were exposed to Wilson's presidency exhibit lower levels of education, earnings, and social mobility. Our combined results thus document significant short and long-run costs borne by minorities during a unique episode of state-sanctioned discrimination.

This paper studies how birth town migration networks affected long-run location decisions during historical US migration episodes. We develop a new method to estimate the strength of migration networks for each receiving and sending location. Our estimates imply that when one randomly chosen African American moved from a Southern birth town to a destination county, then 1.9 additional Black migrants made the same move on average. For White migrants from the Great Plains, the average is only 0.4. Networks were particularly important in connecting Black migrants with attractive employment opportunities and played a larger role in less costly moves.

Part-time work is a common work arrangement in the United States that can be precarious, insecure, and lacking opportunities for advancement. In turn, part-time work, especially involuntary part-time work, tends to be associated with worse health outcomes. Although prior research documents heterogeneity in the health consequences of precarious work across countries, we do not know whether state-level institutional contexts shape the association between part-time work and self-rated health in the United States. Using data from the Current Population Survey (2009–2019; n = 813,077), the present study examined whether linkages between part-time work and self-rated health are moderated by state-level social policies and contexts. At the population level, we document differences in the prevalence of fair/poor health among part-time workers across states. For instance, 21% of involuntary part-time workers reported fair/poor health in West Virginia compared to 7% of involuntary part-time workers in Massachusetts. Findings also provide evidence that voluntary (β =.51) and involuntary (β=.57) part-time work is associated with greater odds of fair/poor health among individuals. Moreover, the association between voluntary part-time work and self-rated health is weaker for individuals living in states with higher amounts for maximum unemployment insurance, higher minimum wage, and lower income inequality. State-level policies did not moderate the association between involuntary part-time work and health. The present study points to the need to mitigate the health consequences of part-time work with social policies that enhance the health of workers.

Worldwide, an estimated 2 billion people, or 26% of the global population, lack safely managed drinking water services and 3.6 billion (46%) do not have safe access to sanitation. In 2017, nearly 460,000 U.S. households—some 1.1 million people, enough to fill a large city—lacked piped running water in their homes, a state we call “plumbing poverty.” While existing research has documented household water insecurity in the global South, this study aims to understand problems of infrastructural injustice in the places where many would least expect: cities in the United States. Our research documents emerging trends and persistent gaps in urban water access over the past 20 years, before and after the Great Recession of 2008. Focusing on 15 major metropolitan areas and drawing on analysis of nearly two decades (2000-2017) of Census data, we identify racialized disparities in household water access, compare trends between cities and over time, and point to worsening conditions for urban dwellers, especially renters. To understand the different pathways of water insecurity relative to housing and urban development, we profile three indicative metros: San Francisco, Phoenix, and Milwaukee. Together, these metros exemplify key trajectories of plumbing poverty: (a) San Francisco is a strong example of 'worsening' trends in a dense and relatively wealthy city; (b) Phoenix captures the 'stagnation' of plumbing poverty in a fast-growing Sun Belt city; and (c) Milwaukee, too, reflects the ‘stagnation’ of plumbing poverty but in a deindustrialized Rust Belt context. Our study finds that plumbing poverty is a social, not technical problem, with roots in unaffordable housing conditions, widening wealth gaps, and racialized inequality in some of the fastest-growing U.S. metros. While some cities have managed to decrease their unplumbed populations over the past two decades, others have made no progress or, worse, seen alarming jumps in the share of people living without piped water access. Many of the worst offenders in terms of plumbing poverty—such as San Francisco, Portland, Austin, Nashville, and Seattle—are also some of the wealthiest and fastest-growing American cities. This research underscores a need for coherent federal policy and significant infrastructure spending to rectify ongoing spatial and social inequalities of water access in U.S. cities.

East-West Gateway’s August 2021 Map of the Month highlights Missouri’s Bicentennial. This month, Missouri celebrates its 200th anniversary as the 24th state of the United States. Missouri entered the union on August 10, 1821, with its first capitol in St. Charles. This Map of the Month reproduces a portion of a map that was created in 1822, the year after statehood, by cartographers Henry Charles Carey and Isaac Lea. In the 1820 territorial census, which occurred in the year before statehood, the counties of Franklin, Jefferson, St. Charles and St. Louis (inclusive of the city of St. Louis) made up 27 percent of the state population. Enslaved persons accounted for 16 percent of the population of these counties and 15 percent of the state’s population. This census excluded Native Americans. By 1920, a century later, the boundaries of these counties and of the city of St. Louis had assumed their current form. At Missouri’s centennial, the five county-level jurisdictions represented 28 percent of the state’s population. Over 40 percent of people in the region were either foreign-born or the children of immigrants. According to the most recent population estimates from the U.S. Census Bureau, the four Missouri counties in the East-West Gateway region, along with the city of St. Louis, were home to 33 percent of the state’s population in 2020. Today, the St. Louis region accounts for a large portion of the state’s economic activity. The five county-level jurisdictions account for 39 percent of the state’s wage and salary employment, generate 46 percent of the state’s wages, and contribute 43 percent of the state’s gross domestic product.

We aim to answer whether expected occupational gains motivated rural-urban and rural-rural migration in nineteenth-century Norway. Human capital theory indicates that the higher expected gains, the more prone an individual will be to migrate. We use a micro-level data set of over 42,000 rural sons linked to their fathers based on 1865 and 1900 Norwegian censuses and employ a switching endogenous regression model controlling for the endogeneity of migration decisions. Our main finding is that the effect of expected occupational gain on the probability of rural-urban migration differs according to the rural sons’ destination and parental occupational status: the sons from low status families were migrating motivated by expected occupational advancement. Sons from families with higher occupational status were motivated by expected occupational gains only in the case of rural-urban migration.

Puerto Rico is a US territory and a popular destination for Latino immigrants in the Caribbean. Even with few language and cultural barriers, however, many Latino immigrants in Puerto Rico are uninsured. Using data from the 2014-19 Puerto Rico Community Survey, we examined inequities in health insurance coverage for non-Puerto Rican Latinos ages 18-64 living in Puerto Rico according to citizenship status and Latino subgroup (Dominican, Cuban, Mexican, and other Latino). After controlling for potential confounders, we found that noncitizen Dominicans had a significantly lower probability of having any health insurance (57.2 percent) and having any private insurance (31.5 percent). Regardless of similarities in culture and language, Latino immigrants on the island, particularly Dominicans, experience major health insurance coverage inequities. Considering that Puerto Rico's immigration system is regulated by US federal statute, both federal and local policy makers should acknowledge and focus on reducing these immigrant disparities in health insurance coverage.

The first chapter studies the life-cycle behavior of two cohorts of American women: those born in the 1960s and those born in the 1980s. Millennial women are more likely to work full time, work in professional, health, and education-related occupations, and be childless in their mid-thirties than women born in the 1960s. I build a life-cycle model that incorporates labor supply, occupation, and fertility choices, and estimate the model for the older cohort. I analyze the role of two forces in explaining the data patterns: (i) labor market factors, including changes in the wage structure and in the initial joint distribution of workers’ skills and occupations’ skills requirements, and (ii) family factors, including changes in marital status across cohorts. I find that both mechanisms are important and together are able to (i) explain the changes in occupational sorting across cohorts; (ii) predict 74% of the changes in the share of women in full-time work; (iii) explain 85% of the decrease in the share of women with two children and (iv) explain 81% of the increase in the share of childless women in their mid-thirties. The second chapter, which is work performed jointly with Lucas Finamor and Boryana Ilieva, studies women and men’s labor market and insurance decisions around childbirth in Chile, a country with widespread informality. We identify three sectors of employment: formal, informal and self-employment. An individual in the informal sector works in a private firm without a labor contract and a self-employed person is an independent worker. We document the following findings. First, there are no significant changes in the share of workers with no labor contract after childbirth for both men and women, but women are more likely to switch into self-employment where the effect is larger for those highly educated. Second, we show that highly educated women are more likely to work remotely after the first birth. In contrast, low educated women do not change work location. Third, women are also more likely to switch to less cognitive intensive occupations after childbirth, which may explain the fall in wages after the event. Fourth, women are less likely to keep private health insurance after their first birth. Finally, we explore the effects of the 2008 Chilean pension system reform on formal work decisions. We observe that women who had children after 2008 are less likely to leave formal employment, in comparison to women who had children before the reform was implemented. In the third chapter, which is joint work with Paula Calvo and Zhengren Zhu, we investigate the role of maternal mental health on children’s cognitive and mental health development. We propose a model that incorporates maternal mental health as a separate input in the human capital production function, different from cognitive and non-cognitive skills. We employ the National Longitudinal Survey of Youth 1979, where we link mothers and their children, to document the empirical patterns that motivate this study: First, poor maternal mental health is positively associated with poor mental health of her child and negatively associated with her child’s cognitive development (which includes math and reading recognition). Second, poor maternal mental health is associated with worse parental practices at different ages. Third, children’s mental health problems affect their cognitive outcomes in school. Fourth, children with poor mental health are more likely to have mental health problems in adult life, have lower wages and lower educational attainment. Our model incorporates these key mechanisms. We describe the estimation steps and propose counterfactual exercises.

Background: Data anonymization is an important building block for ensuring privacy and fosters the reuse of data. However, transforming the data in a way that preserves the privacy of subjects while maintaining a high degree of data quality is challenging and particularly difficult when processing complex datasets that contain a high number of attributes. In this article we present how we extended the open source software ARX to improve its support for high-dimensional, biomedical datasets. Findings: For improving ARX's capability to find optimal transformations when processing high-dimensional data, we implement 2 novel search algorithms. The first is a greedy top-down approach and is oriented on a formally implemented bottom-up search. The second is based on a genetic algorithm. We evaluated the algorithms with different datasets, transformation methods, and privacy models. The novel algorithms mostly outperformed the previously implemented bottom-up search. In addition, we extended the GUI to provide a high degree of usability and performance when working with high-dimensional datasets. Conclusion: With our additions we have significantly enhanced ARX's ability to handle high-dimensional data in terms of processing performance as well as usability and thus can further facilitate data sharing.

Entry rates into self-employment increase during recessions and decrease during economic upswings. I show that this is mostly explained by the higher unemployment rate during a recession, together with the fact that at all times, unemployed persons have a relatively high propensity to become self-employed out of necessity. I use econometric decomposition techniques to quantify these effects based on the monthly matched US Current Population Survey before, during, and after the Great Recession. I also document that the entry rate into self-employment with unincorporated businesses strongly increased during the recession, but not into self-employment with incorporated businesses. This highlights the association of unincorporated and incorporated self-employment with necessity and opportunity entrepreneurship, respectively. The results are useful for policymakers and practitioners to understand, forecast and act on the different types of self-employment that can be expected over the business cycle. There are also important implications for theories of the cyclicality of unemployment and entrepreneurship. Plain English Summary Self-employment will increase during recessions when unemployment is high, but it may not boost innovation. During recessions, increased unemployment underlies the higher entry rate into self-employment. Our evidence is from representative survey data from the USA covering the Great Recession. The upside is that self-employment enables workers who lose their jobs to continue to work, which can speed up the subsequent economic recovery. Thus, public policy should enable people to start businesses. However, as during recessions the unemployed mostly start unincorporated businesses, one cannot expect them to boost innovation as much as start-ups during better economic times. These insights also speak to the 2020 recession triggered by COVID-19. If unemployment remains high after the relaxation of the lockdowns, a rise can be expected especially in non-innovative self-employment. Thus, the principal policy implication of this study is that policymakers should ensure that their expectations for new businesses started during deep recessions are realistic for the circumstances.

The proliferation of the incarcerated population in the late 1900s and early 2000s has resulted in a large-scale loss of economic opportunities for incarcerated people, as the label “felon” follows ex-offenders throughout their employment search process after release. In 2020, over 608,000 sentenced inmates were released from state and federal correctional institutions (Carson, 2020). In the nine years following release, five in six state inmates released in 2005 were arrested at least once, and in the three years following release, two-thirds were arrested for a new crime (Alper et al., 2018). Research has shown that ex-offenders who are able to find employment after release are much less likely to recidivate, meaning that they have a decreased probability of engagement in criminal activity again (Holzer et al., 2003; Yang, 2017). As crime is costly, both in terms of psychological and financial damage, reducing recidivism is key to improving the overall wellbeing of society. Thus, increasing employment opportunities for qualified ex-offenders produces social, economic, and individual benefit by reducing future crime, decreasing costs of crime, and providing meaningful employment.

The geography of education issues has long been an under-researched area in geography literature. This article reviews studies on school segregation and white flight and emphasizes the importance of investigating these educational issues within U.S. metropolitan areas. We discuss the characteristics of school segregation and white flight as a consequence of contemporary educational policies and the increasing diversity of public school populations. In consideration of the intrinsic geographic nature of the aforementioned educational phenomena, we call for spatial social science research to explore new trends and patterns of white flight and school segregation in U.S. metropolitan contexts using advanced geospatial approaches as location-referenced educational data become increasingly available.

The purpose of this study is to describe population-level trends in intentions to remain childless (ITRC) among men and women in the United States in the first decades of the twenty-first century. We use a sample of 31,739 women and 24,524 men aged 18–44 from a cross-sectional and nationally representative survey, the National Survey of Family Growth. Our analyses utilize five of the survey's cycles: from 2002 through 2015–2017. Trends in the unadjusted proportions of men and women who report ITRC are presented, along with predicted probabilities of reporting ITRC from multivariate regressions. At the beginning of the twenty-first century, the prevalence of ITRC increased in the general population of men and women in the United States as well as among the youngest adults aged 18–24. If ITRC are realized, permanent childlessness rates could increase in the near future, contributing to the ongoing fertility decline in the United States. Further analyses uncover similarities and differences in the ITRC trends and correlates between men and women. Increases in ITRC among women are connected to changes in the socio-demographic composition of the population but ITRC increases among men are not connected to population composition changes. In addition, a positive education gradient is observed in ITRC among women but not among men. These variations in ITRC prompt a call for further research into gender and intentions for childlessness.