After the U.S. foreclosure crisis, institutional investors purchased thousands of homes and converted them into rental properties. Past research links these entities to a number of negative outcomes, including rent increases and eviction filings. This paper examines a larger variety of investors, including private equity firms and contract sellers. Using a national dataset of real estate transactions from 2010 to mid-2017, we examine the inter- and intra-metropolitan geography of institutional investors. Consistent with prior research, we find that large publicly traded entities purchased homes in growing Sunbelt metros, yet some specific firms target weaker-market metros. Large, publicly traded firms have concentrated investment in higher-value neighborhoods with larger shares of white residents. In contrast, private equity firms and contract sellers tend to invest in relatively lower-value neighborhoods with larger shares of Black residents. Results suggest different potential implications for these diverse actors’ investments, from crowding out prospective homebuyers to racial targeting.

Background: Black Americans have a higher prevalence of stroke and stroke-related deaths than any other racial group. Racial disparities in stroke outcomes are even wider among women than men. Conventional studies have cited differences in lifestyle (i.e. smoking, alcohol consumption, etc.) and vascular risk factors between races as the source of these disparities. However, these studies fail to account for the higher prevalence of minoritized populations at the lower end of the socioeconomic distribution. Objectives: This study explores differences in stroke risk factors across age and socioeconomic cohorts to determine whether comorbidities can sufficiently explain disparities at all ages and income levels. Design: Using the 2006–2018 National Health Interview Survey data, statistical analysis evaluated differences in risk factors among a full sample cohort (aged 18–85 years; n = 131,091) and a “young” subsample cohort (aged 18–59 years; n = 6183) of women. Methods: Logistics and unconditional quantile regression models assessed the relationship between stroke and comorbid, demographic, and behavioral characteristics across socioeconomic classes. Results: Results suggest that Black women had a 1.415-fold (confidence interval = 1.259, 1.591) higher likelihood of stroke compared with White women after controlling for age, behavior, and comorbidities. Racial disparities were not statistically significant at the higher income ranges for either the full (odds ratio = 1.404, p = 0.3114) or young samples (odds ratio = 1.576, p = 0.7718). However, Blacks had significantly higher odds of stroke in the lower quartiles (lower odds ratio: 1.329, p = 0.0242; lower middle odds ratio: 1.233, p = 0.0486; and upper middle odds ratio: 1.994, p = 0.0005). Disparities were larger among young women (odds ratio = 1.449, confidence interval = 1.211, 1.734). Conclusion: While comorbidities were highly associated with stroke prevalence in all socioeconomic cohorts, Blacks only had higher relative odds in the lower income classes. Lack of biological or behavioral explanations for these findings suggests that unobserved or uncontrolled factors such as systemic racism, prejudicial institutions, or differential treatment may contribute to this.

Food delivery platforms have become an established part of the urban dining culture, but their success is reliant upon gig workers. While prior research has uncovered substantial socioeconomic consequences associated with the platform economy, little is known about how delivery platforms affect local employment. Using a quasi-experimental research design, this paper explores the impact of the spatiotemporal market entry of Grubhub, Postmates, DoorDash, and UberEats on local restaurant employment. Expanding upon prior theoretical work, our analysis suggests that the entry of delivery platforms does not affect restaurants’ demand, as the number of food preparation-related workers remain unchanged. However, as those platforms fundamentally reduce the number of dine-in service workers, we find an overall negative impact on local restaurant employment, which is only partially compensated for by an increase in gig workers (i.e., delivery drivers). Our findings inform policy makers and the restaurant industry on the macroeconomic impact of such platforms.

Answering database queries while preserving privacy is an important problem that has attracted considerable research attention in recent years. A canonical approach to this problem is to use synthetic data. That is, we replace the input database R with a synthetic database R* that preserves the characteristics of R, and use R* to answer queries. Existing solutions for relational data synthesis, however, either fail to provide strong privacy protection, or assume that R contains a single relation. In addition, it is challenging to extend the existing single-relation solutions to the case of multiple relations, because they are unable to model the complex correlations induced by the foreign keys. Therefore, multi-relational data synthesis with strong privacy guarantees is an open problem. In this paper, we address the above open problem by proposing PrivLava, the first solution for synthesizing relational data with foreign keys under differential privacy, a rigorous privacy framework widely adopted in both academia and industry. The key idea of PrivLava is to model the data distribution in R using graphical models, with latent variables included to capture the inter-relational correlations caused by foreign keys. We show that PrivLava supports arbitrary foreign key references that form a directed acyclic graph, and is able to tackle the common case when R contains a mixture of public and private relations. Extensive experiments on census data sets and the TPC-H benchmark demonstrate that PrivLava significantly outperforms its competitors in terms of the accuracy of aggregate queries processed on the synthetic data.

In the United States, the current medical system has been controversial due to its non-universal design. Typically those who’re employed and middle or upper class have the most access to health care due to this design. Even among those with access to health insurance the amount of out-of-pocket medical expenses that families are exposed to after insurance can cause a burden in itself for staying financially afloat that year, especially when paired with the increasing cost of living in the U.S. from 2009-Present. For the purpose of this paper, we’ll be examining the relationship between migration & high medical expenses in the United States from 2009-2019. Our findings find a relationship between medical expenses and migration, and also a relationship between State-level medicaid expansion with decreased migration/medical costs.

This dissertation explores how financial literacy, financial capability, financial self-efficacy, and future time perspective affect the likelihood of low-moderate income (LMI) households in Los Angeles County owning a home and holding a mortgage. It draws on existing literature on financial literacy, financial capability, financial self-efficacy, future time perspective, and homeownership to develop a theoretical framework that identifies the factors that influence LMI households’ access to homeownership. Using secondary data merged from six surveys conducted by the University of Southern California (USC) Understanding America Study (UAS) from 2015 to 2022, it analyzes the relationships among financial literacy, financial behaviors, financial attitudes, and mortgage holding among 2,098 participants. The findings revealed significant positive associations between holding a mortgage and financial literacy, income, age, Hispanic ethnicity, and specific levels of educational attainment. However, financial self-efficacy, financial capability, and future time perspective did not demonstrate significant moderating effects in the relationship between financial literacy and holding a mortgage. The dissertation concludes that enhancing financial literacy among LMI households is crucial for increasing their access to home ownership and suggests possible interventions and policies for doing so.

This paper studies how physical climate risk affects corporations through the labor channel. By quantifying occupational climate exposure, I document that climate-exposed jobs have shorter working hours, lower productivity, and higher employment (especially of part-time workers) as workforce supplements. Firms with more climate-exposed workers adapt to unfavorable climate trends by retaining more employees, increasing insurance, and expanding offshore inputs. However, these firms have more workplace injuries and worse performance during climate surprises, indicating limitations of adaptation. I also explore various incentives and constraints for firms’ labor adaptation strategies and make further causal inferences by studying the implementation of the California Heat Standard.

Little is known about occupational SARS-CoV-2 exposures and COVID-19 outcomes. We established a Doctor’s First Reports of Occupational Injury or Illness (DFR)-based surveillance system to study cases of work-related COVID-19 exposures and disease. The surveillance data included demographics, occupation, industry, exposure, and illness, details including hospitalization and lost work. We classified workers into ‘healthcare’, non-healthcare ‘public-facing’, or ‘other’ worker groups, and rural–urban commuting areas (RUCAs). We describe worker exposures and outcomes overall by worker group and RUCA. We analyzed 2848 COVID-19 DFRs representing workers in 22 detailed occupation groups and 19 industry groups. Most DFRs were for workers in metropolitan RUCAs (89%) and those in healthcare (42%) and public-facing (24%) worker groups. While DFRs were from 382 unique worksites, 52% were from four hospitals and one prison. Among 1063 DFRs with a suspected exposure, 73% suspected exposure to a patient or client. Few DFRs indicated hospitalization (3.9%); however, the proportion hospitalized was higher among nonmetropolitan (7.4%) and public-facing (6.7%) workers. While 56% of DFRs indicated some lost work time, the proportion was highest among public-facing (80%) workers. Healthcare and prison workers were the majority of reported occupational COVID-19 exposures and illnesses. The risk of COVID-19 hospitalization and lost work may be highest among nonmetropolitan and public-facing workers.

The increase in rate of international migration to the United States since the late 1960s, coupled with a generally high rate among minority populations, altered the racial and ethnic composition of America’s urban neighborhoods. The changing demography and increase in shares of minority subpopulations underscore the salience of conducting multigroup studies of residential and socioeconomic segregation beyond the traditional white versus black dichotomy. Segregation based on subgroup characteristics (de facto or de jure) is problematic, particularly for racial minorities and low-income residents who are limited in moving to areas they can afford. These minority neighborhoods are associated with physical and socioeconomic disadvantage due to public and private deinvestment. The undercurrents of segregation were explored in the racial tipping point and white flight literature where non-Hispanic white majority residents exit old inner and central city neighborhoods when the share of minority populations increase beyond a critical threshold. Due to strong correlations between race and income, white flight also tends to concentrate poverty in the abandoned neighborhoods. Beyond this relationship iv between personal choice and segregation however, local and federal public policies have also been historically linked with segregating urban America. Federal highway programs, mortgage loan underwriting processes, suburban housing developments, and restrictive local zoning laws have created race and income-based segregated spaces. Also, reinvestment programs aimed revitalizing physical and socially distressed neighborhoods tend to yield minimal outcomes. This is often due to either limited funding compared to the magnitude of the problem or lack of sustained political commitment, overemphasis on market-based ideas which alienate minorities and low-income residents, and emphasis on new urbanism housing designs associated net losses in the public housing stock. In this dissertation, I utilized a new method (the Hsd. Index) as a composite measure of segregation to study the joint contributions of multiple socioeconomic factors (race and ethnicity, educational attainment, immigration, income, unemployment, and poverty) in understanding the nuanced dynamics of segregation in and around the immediate neighborhoods of HOPE VI, a major federal mixed-income housing and revitalization program, between 1990 and 2010. I used projects from Ohio’s two largest counties, Cuyahoga and Franklin Counties, as case studies. Among other concerns, I was interested in understanding whether HOPE VI projects decreased segregation in these neighborhoods and whether there were any differences in outcome between the two counties. I also provided the procedure for deriving and replicating Hsd. Index for future studies. I found mixed results. First, while overall segregation decreased in Franklin County, it generally increased in Cuyahoga County. Second, there was overall increase in the incidence of segregation in the immediate HOPE VI neighborhoods for both counties. v Finally, minimal improvements observed in Franklin County were within shorter radius (< 1 mile) of the HOPE VI project but in Cuyahoga County the minimal improvement was observed for census tracts located farther away from the projects (≥ 2 mile). Based on these results, HOPE VI projects in Cuyahoga and Franklin Counties created small changes in the levels of segregation around the projects. The differences observed in both counties are likely attributable to forces acting in each county beyond the influence of HOPE VI. This study and its findings provide another perspective for studying dynamics of multigroup segregation associated with the implementation of HOPE VI or other public interventions in areas with similar characteristics as found in Cuyahoga and Franklin Counties.

Racism named as a public health issue largely focuses on those at the care-receiving end of health care. Home health care workers (HHCWs) are predominantly Black, immigrant, and women; in New York state it is illegal for HHCWs to work 24 consecutive hours, but homecare agencies assign HHCWs to 24-hour shifts and only pay for 13 hours of work. The demanding work of HHCWs increases their risk for and experiences of injury, depression, cardiovascular disease, and cerebrovascular events. This manifestation of exploitation illustrates that racism affects health care workers as well as the public.

Exposure to singular or overlapping external shocks, such as rainfall extremes, armed conflict, and COVID-19, may catalyze a shift in gendered power dynamics within affected households as they cope with associated threats to their safety and livelihoods. Despite evidence that women are disproportionately affected by such shocks, little scientific work has assessed the separate and combined impacts of these three external shocks on women’s lives. In this study, we examined the distinct and overlapping associations between extreme events—growing season rainfall anomalies, armed conflict during the growing season, and COVID-19—and women’s daily decision-making power in Burkina Faso. We employed longitudinal survey data from IPUMS Performance Monitoring for Action (PMA), a complex and spatially referenced dataset. These data were collected from a population-representative sample of women of reproductive age (15–49&nbsp;years) in Burkina Faso across two timepoints: 2019/2020 (December 2019–February 2020) and 2020/2021 (December 2020–March 2021). PMA data from Burkina Faso contain detailed questions on women’s sociodemographic characteristics, health, and household dynamics. We spatially linked these data with (1) external rainfall data, (2) armed conflict event data, and (3) PMA coronavirus-specific follow-up survey data (containing COVID-19 knowledge and prevention behaviors) collected in June and July of 2020. Using log-binomial general estimating equation (GEE) models, we examined the relationship between extreme events—wetter-than-usual growing season, armed conflict (that resulted in at least one death), and COVID-19—and increased daily decision-making power among women. We found strong and significant associations between experiencing a wetter-than-usual growing season (i.e., greater than 1 standard deviation above 10-year mean) and women being less likely to have increased daily decision-making power in the household compared those experiencing usual rainfall during the growing season [prevalence ratio (PR): 0.70, 95% confidence interval (CI): 0.56, 0.87]. Similarly, residing in an area that was more affected by the COVID-19 pandemic (i.e., where 80% or more of respondents in the community reported staying home to avoid COVID-19) was also associated with women being less likely to have increased daily decision-making power in the household [PR: 0.75, 95% CI: 0.61, 0.91]. We did not observe any significant association between armed conflict and increased daily decision-making among women [PR: 1.15, 95% CI: 0.84, 1.57]. These trends indicate that women’s decision-making power within partnerships may be negatively impacted by certain household shocks. Centering women (and other marginalized and vulnerable communities) in the leadership, implementation, and as key beneficiaries of crisis response efforts may be an effective strategy to combat some of these constraints on women’s decision-making and even empower them within their households and communities.

Theories of endogenous economic growth propose that the exchange of information between workers with different knowledge facilitates innovation. In this study, we explore whether occupational gender segregation poses a barrier to endogenous growth by limiting the extent to which women and men workers exchange information, ideas and perspectives. Focusing on technology industries as a sector where information exchange plays a large role, we use error correction models to test for the potential long-term costs of occupational gender segregation on industry growth and productivity across US local labor markets from 2005 through 2019. Results indicate that occupational gender segregation stalls tech growth, while also hindering productivity. Conversely stated, the occupational integration of women and men provides a boon to growth and productivity in local tech sectors. As regions strive to be competitive in the new economy, those fostering collaboration between women and men will be better positioned to succeed.

This paper estimates the effect of Temporary Protected Status (TPS), a temporary legalization policy, on the incomes and asset ownership of Salvadoran recipients over 20 years. We compare likely undocumented Salvadoran immigrants eligible for TPS to a control group of likely undocumented immigrants ineligible for TPS in a flexible event study design that allows us to observe growth dynamics in the policy's effect over two decades. We find that earnings, homeownership, and the likelihood of using a car increased considerably for Salvadoran adults for at least 15 years following the granting of TPS. Our study suggests that even temporary and limited legal status can have substantial and sustained economic benefits for recipients.

As legal marijuana is emerging as an important component of cities across the United States, it is important to understand the factors that contribute to legal marijuana facility siting. Although land use and zoning are expected to determine where commercial marijuana facilities are located, if residential characteristics also enter siting considerations, some neighbourhoods may either bear the burden of undesired facilities or reap the benefits of legal marijuana, underscoring equitable considerations in marijuana facility siting. Thus, this study examines how neighbourhood change associated with locally unwanted land uses and gentrification influences the amount of retail marijuana facilities across three US cities: Denver, Colorado; Portland, Oregon; and Seattle, Washington. Using a series of Poisson-related regressions, this study finds that neighbourhood residential characteristics influence retail marijuana facilities in ways exceeding siting restrictions alone, like zoning. Notably, quantitative results suggest that there are fewer retail marijuana facilities in neighbourhoods experiencing locally unwanted land use-related change than non-locally unwanted land use neighbourhoods in Denver and Seattle; and more retail marijuana facilities in gentrified compared to non-gentrified neighbourhoods in Denver. Overall, these findings advance understanding of the connection between legal marijuana and neighbourhood changes and aim to influence guidelines for integrating legal marijuana facilities into communities.

We propose a theory-inspired measure of the accessibility of a city's center: the size of the surrounding area from which it can be reached within a specific time. Using publicly-available optimal routing software, we compute these "accessibility zones" for the 109 largest American and European cities, separately for cars and public transit commutes. Compared to Eu-ropean cities, US cities are half as accessible via public transit and twice as accessible via cars. Car accessibility zones are always larger than public transit zones, making US cities more accessible overall. However, US cities' car orientation comes at the cost of less green space, more congestion, and worse health and pollution externalities.

Using loan-level data, this analysis considers the intersection of race, subprime home loans, and the current vacancy crisis in St. Louis, Missouri. Borrowers in Black areas in the north of St. Louis City and St. Louis County received subprime home loans at higher frequencies during the subprime boom period of 2003–2007 than those in White areas, with differences in balloon loans especially stark. Specifically, borrowers in Black neighborhoods received subprime loans more frequently than those with equal FICO scores in White neighborhoods. As a result of these differential loan terms, North City and inner ring “First Suburb” areas saw more foreclosure and borrower payment delinquency, which in turn were highly associated with home vacancy, controlling for other risk factors. However, foreclosure was no longer a significant predictor of home vacancy after controlling for demographic factors and FICO score, indicating that the unequal loan terms may have driven much of the increase in home vacancy in the St. Louis area since the Great Recession.

Research on cross-sector differences has long demonstrated that workers in public and nonprofit sectors are more prosocially and intrinsically motivated, which in turn shapes their work behavior and outcomes, but little evidence exists on how public service employment is associated with workers’ life quality outside of the organizational context. Using cross-sectional data pooled from the Well-being Module of the American Time Use Survey, we investigate whether sectors of employment can predict workers’ subjective well-being (SWB) beyond the work domain. The results suggest public servants across all levels of government as well as nonprofit organizations experience higher life satisfaction than their for-profit counterparts. However, there are different patterns in daily experiential well-being across levels of government. Further analyses by work and non-work days demonstrate that work itself is likely to be a key factor affecting workers’ daily experience across sectors. Implications for theory and practice are discussed.

Objective This study aims to show the usefulness of incorporating a community-based geographical information system (GIS) in recruiting research participants for the Asian Cohort for Alzheimer's Disease (ACAD) study for using the subgroup of Korean American (KA) older adults. The ACAD study is the first large study in the USA and Canada focusing on the recruitment of Chinese, Korean and Vietnamese older adults to address the issues of under-representation of Asian Americans in clinical research. Methods To promote clinical research participation of racial/ethnic minority older adults with and without dementia, we used GIS by collaborating with community members to delineate boundaries for geographical clusters and enclaves of church and senior networks, and KA serving ethnic clinics. In addition, we used socioeconomic data identified as recruitment factors unique to KA older adults which was analysed for developing recruitment strategies. Results GIS maps show a visualisation of the heterogeneity of the sociodemographic characteristics and the resources of faith-based organisations and KA serving local clinics. We addressed these factors that disproportionately affect participation in clinical research and successfully recruited the intended participants (N=60) in the proposed period. Discussion Using GIS maps to locate KA provided innovative inroads to successful research outreach efforts for a pilot study that may be expanded to other underserved populations across the USA in the future. We will use this tool subsequently on a large-scale clinical genetic epidemiology study. Policy implication This approach responds to the call from the National Institute on Aging to develop strategies to improve the health status of older adults in diverse populations. Our study will offer a practical guidance to health researchers and policymakers in identifying understudied and hard-to-reach specific Asian American populations for clinical studies or initiatives. This would further contribute in reducing the health and research disparity gaps among older minority populations.

The present research looked at the return to schooling (Public and Private schools) in the census region of the East South-Central Division of the United States. The demographic diversity of this region is the primary reason for its selection. The functional form of return to schooling on education has the specified wage as the long-linear function of school attendance (in years), controlling for gender, race, work experience, occupation status, and educational attainment. We followed the extant literature to control a few socioeconomic variables. The article used 2020 sample data from Integrated Public Use Microdata Series (IPUMS) to estimate the parameter coefficients. The findings show private schools yield a higher return in terms of subsequent wage earnings compared to private schools. The results highlight a higher differential between private schools & public schools for Caucasian Americans over African Americans. Similarly, the differential is higher among females over males.

Service in the military has meant different things during different eras. As illustrations, veterans of earlier conflicts, such as World War II, were afforded educational access at the end of service that outmatched access among the general population during this time period. In contrast, veterans of the Vietnam War returned home to fewer benefits and fewer accolades for their service. People who served in both eras were more likely to be exposed to risks of combat, including mental and physical wounds, than were those who served during peacetime. These differences may lead to cohort-specific mortality risks and protections for veterans compared to non-veterans. We use 11 years of the National Health Interview Survey (1986–1997) linked to 30 years of prospective mortality status to examine the mortality risks for veteran and non-veteran men across cohorts stretching from before World War II to the post-Vietnam War era. We focus on how these risks are associated with education. In support of our General cohort hypothesis, we find that after controlling for confounding factors, veterans faced heightened risks of mortality compared to nonveterans, particularly those in earlier cohorts. We find conflicting evidence related to turning point narratives of military service, namely that an increased mortality risk among veterans with lower levels of education further widens the veteran/nonveteran mortality gap among cohorts who came of age in the middle of the twentieth century. The conclusions vary somewhat depending on how cohorts are defined. Contrary to expectations, the association is not shaped by educational attainment.