This paper provides a unified framework for practitioners who wish to estimate alternative indices of multidimensional poverty. These alternative indices are used to estimate multidimensional poverty in the USA over the last decade with a focus on analyzing trends by race and ethnicity. Individual level data on five different dimensions of well-being are compiled over the last decade using annual Census surveys. We find that multidimensional poverty in the USA declined over time regardless of the index used. A higher incidence of multidimensional poverty was observed among Hispanics, American Indians and Blacks. Poverty ranking among racial/ethnic groups was robust to the indices used. Estimates of alternative indices highlight different aspects of multidimensional poverty and provide complementary information on poverty in the USA in the last decade.

Fairness in Machine Learning (ML) has mostly focused on interrogating the fairness of a particular decision point with assumptions made that the people represented in the data have been fairly treated throughout history. However, fairness cannot be ultimately achieved if such assumptions are not valid. This is the case for mortgage lending discrimination in the US, which should be critically understood as the result of historically accumulated injustices that were enacted through public policies and private practices including redlining, racial covenants, exclusionary zoning, and predatory inclusion, among others. With the erroneous assumptions of historical fairness in ML, Black borrowers with low income and low wealth are considered as a given condition in a lending algorithm, thus rejecting loans to them would be considered a "fair" decision even though Black borrowers were historically excluded from homeownership and wealth creation. To emphasize such issues, we introduce case studies using contemporary mortgage lending data as well as historical census data in the US. First, we show that historical housing discrimination has differentiated each racial group's baseline wealth which is a critical input for algorithmically determining mortgage loans. The second case study estimates the cost of housing reparations in the algorithmic lending context to redress historical harms because of such discriminatory housing policies. Through these case studies, we envision what reparative algorithms would look like in the context of housing discrimination in the US. This work connects to emerging scholarship on how al-gorithmic systems can contribute to redressing past harms through engaging with reparations policies and programs.

Using a US nationally representative sample and a double list experiment designed to elicit views free from social desirability bias, we find that anti-transgender labor market attitudes are significantly underreported. After correcting for this concealment, we report that 73 percent of people would be comfortable with a transgender manager and 74 percent support employment non-discrimination protection for transgender people. We also show that respondents severely underestimate the population level of support for transgender individuals in the workplace, and we find that labor market support for transgender people is significantly lower than support for gay, lesbian, and bisexual people. Our results provide timely evidence on workplace-related views toward transgender people and help us better understand employment discrimination against them.

We estimate the long-run effects of the 1930s Home Owners Loan Corporation (HOLC) redlining maps by linking children in the full count 1940 Census to 1) the universe of IRS tax data in 1974 and 1979 and 2) the long form 2000 Census. We use two identification strategies to estimate the potential long-run effects of differential access to credit along HOLC boundaries. The first strategy compares cross-boundary differences along HOLC boundaries to a comparison group of boundaries that had statistically similar pre-existing differences as the actual boundaries. A second approach only uses boundaries that were least likely to have been chosen by the HOLC based on our statistical model. We find that children living on the lower-graded side of HOLC boundaries had significantly lower levels of educational attainment, reduced income in adulthood, and lived in neighborhoods during adulthood characterized by lower educational attainment, higher poverty rates, and higher rates of single-headed households.

Having a criminal record and employers having access to that record pose a severe barrier to employment and other beneficial labor market outcomes. Recent research has primarily focused on evaluating Ban-the-Box policies implemented after 2000 that limit employer access to this information in hopes of improving outcomes for people with records. However, the initial labor market consequences of making this information available to employers starting in the 1970s has yet to be studied. In this paper, I use variation in the timing and geography of employers’ inaugural access to criminal record information via state central repositories to estimate the e↵ect on labor force participation and employment for various race-gender-education groups. I find that employer access to criminal record information led to decreases in labor force participation (4pp) and employment (1.3pp) of non-college educated black men. However, these declines were o↵set by increases for whites, chiefly white women. These results imply that employer access to central repositories had two e↵ects: One, it pushed noncollege educated black men with criminal records out of the labor force by discouraging job seeking. Two, it potentially helped non-college educated black men without criminal records get jobs due to information flow and reduced competition. While, I am unable to explicitly test this in my study, it is consistent with the literature on racial discrimination and criminal record information.

Introduction: This report examines makes estimates about the Latino Population for 2020—in the United States, Los Angeles, New York City, Miami, and Houston—that differ dramatically from those published by the Census Bureau. Methods: This report uses population growth rates calculated from the raw data found in the American Community Survey (ACS) five-year files for each year between 2010 and 2019 and 2015 to 2019 to project ‘assumed’ population totals for 2020. It 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, 2017. Discussion: These data indicate that the Latino population of the U.S. was undercounted in 2020 by 0.8% using the growth rates between 2010 and 2019 to project the 2020 Latino population: and 0.4% using the rate between 2015 and 2019. This is significantly lower than the 4.99% estimated by the Census Bureau using a different methodology For Los Angeles, the Latino undercount was estimated at 1.4% using the population growth rates between 2010 and 2019 and 1.0% using the growth rate derived from the 2015-19 data. For New York City, the Latino undercount was estimated at 1.8% using the population growth rate between 2010 and 2019 and 1.1% using the growth rate derived from the 2015-19 data. For Miami, the Latino undercount was estimated at 1.0% using the population growth rate between 2010 and 2019 and 1.4% using the growth rate derived from the 2015-19 data. For Houston, the Latino undercount was estimated at 3.9% using the population growth rate between 2010 and 2019 and 2.7% using the growth rate derived from the 2015-19 data. There is no way to statistically determine why the Houston, Texas undercount was significantly higher than in the other three cities or the whole United States.

The problem this study addressed is that undergraduate, female, single parent students are less likely than other groups of women to complete college. The purpose of this basic qualitative study was to explore perceived challenges of persistence to degree completion of undergraduate, female, single parent students in a Southern state, at a suburban community college. The research question used in this basic qualitative study determined what perceived challenges undergraduate female single parents in higher education face during their first year as a student in a community college regarding persistence to degree completion. Using Tinto’s theory of student integration as a conceptual framework, 10 semistructured interviews were conducted with participants from the study site who met the criteria of being female, single parent students, and completing their first year of higher education with a C or better average. Two rounds of open coding were used to identify codes and themes. The data were synthesized into three major themes: academic integration, financial commitment, and goal commitment. This study found that to be successful, single parent female students depended on academic integration, in the form of teacher and school support, financial commitment to pay for college and life expenses and goal commitment providing a motivation to continue even when confronted with challenges. Recommendations include accessing other single parent female students, and childcare, expanding financial support, and cultivating outside support systems. Positive social change can result if institutions of higher education support students and guide them through the challenges associated with the first year of college.

Demographic shifts over the last half-century have resulted in dramatic changes in family structure. These changes have implications for the social safety net because public assistance programs define families differently. This article focuses on a critical poverty-alleviation policy, the Earned Income Tax Credit (EITC), to document family complexity in the United States. We find that more than 60 percent of children in lowerincome families reside in households with ambiguity in tax filing and thus in claiming valuable credits. Tax filing ambiguity driven by family complexity is especially common among households with Black children, highlighting significant racial inequities in the tax treatment of complex families. We also consider two reforms to reflect the realities of families today: the childless EITC and a noncustodial parent EITC.

Introduction: Cardiovascular disease (CVD) mortality rates are heterogenous among Asian American subgroups. To inform more precise prevention strategies, we identified patterns of hypertension-related CVD mortality in Asian American subgroups. Methods: Among deaths with CVD (ICD-10: I00-I99) as the underlying cause and hypertensive disease (ICD-10: I10-I15) as underlying or contributing cause in 2018-2021 mortality data from CDC WONDER with concurrent population estimates from the IPUMS Current Population Survey, we calculated age standardized mortality rates (ASMR) and proportional mortality for non-Hispanic Asian (Asian Indian, Chinese, Filipino, Japanese, Korean, Vietnamese) and Hispanic adults, with non-Hispanic White as the reference. Results: There were 37,746, 95,404, and 867,599 deaths in non-Hispanic Asian, Hispanic and non-Hispanic White groups, respectively. Among non-Hispanic Asian females, ASMR ranged from 41.6 (95% CI 40.0-43.3) per 100,000 population in Japanese to 52.6 (51.0-54.2) per 100,000 in Filipina women. Among non-Hispanic Asian males, ASMR ranged from 45.8 (43.3-48.2) per 100,000 in Korean to 81.0 (78.5-83.5) in Filipino men (Table). Proportional mortality was higher for all Asian American subgroups vs. non-Hispanic White individuals. Proportional mortality ratios ranged from 1.11 (Korean males) to 1.38 (Filipino males; Chinese and Filipina females), vs. non-Hispanic White individuals. Conclusions: There was substantial variation in hypertension-related cardiovascular mortality among Asian American subgroups. All Asian subgroups had higher proportional mortality compared with non-Hispanic White individuals.

This thesis studies how individuals who look for same-sex relationships sort in different states within the United States and its effect on the rate of same-sex relationships. These questions require the development of a comprehensive general equilibrium marriage model which accounts for premarital migration between states. The first chapter extends marriage matching models to consider individuals with different sexual orientations. The chapter gives sufficient assumptions to identify the match gains. Combined with the existence and uniqueness of a general equilibrium in marriage, we can analyze the effect of match gains on the marital rate through counterfactual analyses. The second chapter focuses on adding premarital geographical sorting to the general equilibrium matching model without same-sex couples. The model combines the marriage matching models with location choice models, where individuals can sort in different local marriage markets before choosing a partner. The model gives sufficient assumptions for the existence and uniqueness of an equilibrium given pre-migration population supplies, exogenous labor market earnings, and match gains. Identification follows a typical discrete choice strategy, with an additional discussion on handling the endogeneity of individual match utility on migration propensity. This model analyzes the interaction between the marital rate and migration decisions. The chapter illustrates the model's utility by examining the effect of earnings on migration and the marriage rate in the United States from 1960 to 2000. Both chapters' theoretical contributions are brought together in the final chapter to discuss the impact of sorting between states on the rate of same-sex relationships. The chapter outlines how combining both models leads to further difficulties in identifying match gains. The chapter gives additional assumptions sufficient for identifying match gains in this marriage model with premarital regional sorting. The model is then used to quantify the importance of moving between states for same-sex relationships. The model predicts that same-sex marital rates would decrease by almost 50% if individuals did not have a chance to move. Furthermore, 4% of lesbian women and 7% of gay men would change which states they live in if there were no difference in match gains between states. These rates decrease over time, indicating that social progress affects migration and same-sex marital rates across the United States.

Prior research on adoption of rooftop solar has investigated various economic and psychological factors contributing to or impeding adoption. One psychological factor that has been linked to environmental behavior in other settings, but not yet in the context of rooftop solar, is place attachment, an individual's sense of attachment to their community. Using a survey of over 3700 homeowners in Los Angeles County, we examine the impact of place attachment, relative to that of pro-environmental attitudes, on the decision whether to consider rooftop solar and on the decision whether to adopt it. We find that an otherwise average homeowner with pro-environmental attitudes one unit above the mean is 5.66 percentage points more likely to consider rooftop solar, while the effect of place attachment on consideration is not significant. However, among respondents who consider rooftop solar, an otherwise average homeowner with one unit stronger place attachment is 7.59 percentage points more likely to adopt, while the effect of pro-environmental attitudes on adoption is not significant. Policymakers seeking to encourage adoption of rooftop solar should contemplate leveraging place attachment, and should recognize that different policy mechanisms will be effective for homeowners in general than for those who are already considering.

Purpose: There has been little research on the healthcare cost-related coping mechanisms of families of patients with cancer. Therefore, we assessed the association between a cancer diagnosis and the healthcare cost-related coping mechanisms of participant family members through their decision to forego or delay seeking medical care, one of the manifestations of financial toxicity. Methods: Using data from the National Health Interview Survey (NHIS) between 2000 and 2018, sample weight-adjusted prevalence was calculated and multivariable logistic regressions defined adjusted odds ratios (aORs) for participant family members who needed but did not get medical care or who delayed seeking medical care due to cost in the past 12 months, adjusting for relevant sociodemographic covariates, including participant history of cancer (yes vs. no) and participant age (18–45 vs. 46–64 years old). The analysis of family members foregoing or delaying medical care was repeated using a cancer diagnosis * age interaction term. Results: Participants with cancer were more likely than those without a history of cancer to report family members delaying (19.63% vs. 16.31%, P < 0.001) or foregoing (14.53% vs. 12.35%, P = 0.001) medical care. Participants with cancer in the 18 to 45 years old age range were more likely to report family members delaying (pinteraction = 0.028) or foregoing (pinteraction < 0.001) medical care. Other factors associated with cost-related coping mechanisms undertaken by the participants’ family members included female sex, non-married status, poorer health status, lack of health insurance coverage, and lower household income. Conclusion: A cancer diagnosis may be associated with familial healthcare cost-related coping mechanisms, one of the manifestations of financial toxicity. This is seen through delayed/omitted medical care of family members of people with a history of cancer, an association that may be stronger among young adult cancer survivors. These findings underscore the need to further explore how financial toxicity associated with a cancer diagnosis can affect patients’ family members and to design interventions to mitigate healthcare cost-related coping mechanisms.

We examine the causes and consequences of an important cultural and psychological trait: the extent to which one views the world in zero-sum terms-i.e., that benefits to one person or group tend to come at the cost of others. We implement a survey among approximately 15,000 individuals living in the United States that measures zero-sum thinking, political and policy views, and a rich set of characteristics about their ancestry. We find that a more zero-sum view is strongly correlated with several policy views about the importance of government, the value of redistributive policies, the impact of immigration, and one's political orientation. We find that zero-sum thinking can be explained by experiences of an individual's ancestors (parents and grandparents), including the amount of intergenerational upward mobility they experienced, the degree of economic hardship they suffered, whether they immigrated to the United States or were exposed to more immigrants, and whether they had experiences with enslavement. These findings underscore the importance of psychological traits, and how they are transmitted inter-generationally, in explaining current political divides in the United States.

The first chapter of this dissertation constructs original estimates of output per worker, physical capital per worker, physical capital-output ratio, and average years of schooling for the states of the United States from 1800 to 2020. It also provides these estimates for agriculture, manufacturing, and non-manufacturing nonagriculture (NMNA) sectors. Using this data, I find multiple “reversal of fortunes” in the South before the Civil War and also provide nominal price of slave. This chapter provides the sources of data, methodology of estimations, and validity comparison of estimates whenever possible using alternative benchmarks. The second chapter finds life cycle age-earning estimates adjusting for cohort effects and helps shield light on inter-generational income mobility. It provides ageearning profiles and age-earning percentile for birth cohorts from 1902 to 2004. The 18 year olds do not have the same income as the current 50 year olds in their 50th birthday. The earnings among cohorts vary even after accounting for economic and social variables; it happens due to cohort-level changes. I find flattening life cycle earning profiles of recent cohorts.

This dissertation examines spatial heterogeneity that results from various environmental policies. In Chapter 1, I provide a comprehensive overview of each dissertation chapter. Chapter 2 (with Ed Rubin) demonstrates that most coal-fueled power plants are located on or near jurisdictional (county or state) borders. We find that coal-fired power plants are disproportionately sited on downwind borders (within county or state). Natural gas plants---much lower polluters---do not exhibit this behavior. Motivated by the inferred strategic siting, we use an atmospheric dispersion model developed by NOAA to estimate various aspects of the ``pollution transport problem.'' We find that nearly 90% of coal-based particulate matter leaves its state of origin within 48 hours of release. Chapter 3 (with Mark Colas) examines the effects of stringent land-use regulations on national carbon emissions. We develop and estimate a general equilibrium model of residential sorting and energy consumption. We find that relaxing land-use restrictions in California leads to a 0.6% drop in national carbon emissions. The mechanism behind this drop is straightforward. California cities have a temperate climate, carbon-efficient power plants, and high land-use regulations. These land-use regulations inflate housing prices, thus keeping households out of California cities. When households live outside of California, they emit more carbon on average, and therefore national carbon emissions are higher due to California's land-use regulations. In Chapter 4, I simulate the labor market effects of a carbon tax across the continental United States. To recover the welfare impacts of a carbon tax, I build and estimate a spatial equilibrium model that features heterogeneous households. I incorporate a rich level of heterogeneity into the model that allows me to answer: (1) who is most affected by a carbon tax, (2) how much the burden of a carbon tax is borne on different households, and (3) where the households are that bear the greatest burden from the tax. I find that workers without a college degree in manufacturing bear a disproportionate share of the tax incidence. Chapter 5 concludes this dissertation.

The General Educational Development (GED) degree is designed to be a credential equivalent to the high school diploma. However, growing evidence indicates that GED recipients have worse outcomes than high school graduates. Such findings raise the question: is the GED socially equivalent to the high school diploma? Although educational assortative mating patterns have long been used as a barometer of the social distance across educational groups, there has not been a study that has addressed this question by examining the marital sorting patterns of GED recipients. Using log-linear models, our study shows that the odds of intermarriage between GED recipients and high school graduates resemble those between GED recipients and those without a secondary degree. Racial/ethnic minorities had greater difficulty crossing the GED/high school graduate boundary when they married. Our findings detract from the purported view that the GED degree is equivalent to a traditional high school diploma.

The share of children living in a two-parent family has declined sharply in the past 40 years, driven by a decline in marriage among parents without a four-year college degree. This paper presents a number of facts about these trends, drawing on US Census data, the Current Population Survey, the Survey of Income and Program Participation, and US vital statistics birth data. First, there is a large gap in the share of children living with married parents (or two parents) that favors the children of college-educated mothers, both overall and within race and ethnic groups. Second, the decline in the share of children living in married parent families primarily reflects an increase in non-marital childbearing, not a rise in divorce. Third, the widening college gap in children's family structure corresponds to a widening college gap in marriage rates, both overall and within race and ethnic groups. The paper briefly discusses evidence suggesting a causal link between the eroding economic position of men without a four-year college degree and their declining marriage rates. Fourth, the rise in the share of children living with an unpartnered mother has happened despite a sizable decrease in births to teens, women in their 20s, and women with less than a high school degree. Fifth, the college gap in family structure has contributed to the widening college gap in household income, accentuating widening earnings inequality. These trends have the potential to exacerbate class gaps in children's outcomes and undermine social mobility.

The U.S. Census Bureau’s plans for release of data from the 2020 Decennial Census of Population and Housing will publicly and transparently address the unavoidable trade-off between data privacy and data accuracy. Statistical analysts can and should, therefore, take into account the planned presence of well-specified, well-justified noise in data releases based on the 2020 Decennial Census.To aid researchers’ preparations, this article highlights both what is new as well as what seems new but is actually little changed. We examine strategies, trade-offs, and rationales associated with processing and releasing the decennial results. Based on this review, we offer specific conclusions to help promote appropriate and well-informed usage of the 2020 Census. Our strongest recommendation is that, in addition to publishing official tables, the Census Bureau also make either the noisy measurements file (NMF) or unbiased estimates of released table entries available for research purposes. To create official counts, the Census Bureau applies processes to restore face validity to privacy-protected counts (that is, they eliminate disturbing features such as negative and fractional counts). These processes also introduce statistical bias and intractable distortions that researchers may wish to avoid whenever possible. By contrast, the NMF entries do not suffer from the statistical ills added by restoring face validity, and can be easily interpreted by trained analysts. Our other recommendations address critical needs for input to Census Bureau decisions from researchers, for development of suitable statistical tools that work with privacy-protected data, for expanded options with regard to microdata, and for steps to improve the accuracy of decennial census data overall.

Policy debates about whether wages and benefits from work provide enough resources to achieve economic self-sufficiency rely on data for workers, not working families. Using data from the Current Population Survey, we find that almost two-thirds of families working full time earn enough to cover a basic family budget, but that less than a quarter of low-income families do. A typical low-income full-time working family with wages below a family budget would need to earn about $11.00 more per hour to cover expenses. This wage gap is larger for black, Hispanic, and immigrant families. Receipt of employer-provided benefits varies— health insurance is more prevalent than pension plans—and both are less available to low-income families, and black, Hispanic, and immigrant working families. Findings suggest that without policies to decrease wage inequality and increase parents’ access to jobs with higher wages and benefits, child opportunity gaps by income, race-ethnicity, and nativity will likely persist.

Much research has debated whether immigrants' health advantages over natives decline with their duration at destination. Most such research has relied on (pooled) cross-sectional data and used years since immigration as a proxy for the duration of residence, leading to the challenge of distilling the duration effect from the confounding cohort-of-arrival and age-of-arrival effects. Because longitudinal studies tend to use self-rated health as the outcome, the changes they observed may reflect shifts in immigrants' awareness of health problems. We illuminate the debate by examining how immigrants' mortality risk—a relatively unambiguous measure tied to poor health—changes over time compared to natives' mortality risk. Our analysis uses the National Health Interview Survey (1992–2009) with linked mortality data through 2011 (n = 875,306). We find a survival advantage for U.S. immigrants over the native-born that persisted or amplified during the 20-year period. Moreover, this advantage persisted for all immigrants, regardless of their race/ethnicity and gender or when they began their U.S. residence. This study provides unequivocal evidence that immigrant status' health protection as reflected in mortality is stable and long-lasting.