This paper estimates the long-run effects of school accountability on educational attainment by exploiting two sources of variation: staggered implementation of accountability across states and individuals’ exposure to accountability. I find 12 years of exposure to school accountability leads to an increase in the likelihood of graduating high school by 2.3 percentage points but has no statistically significant effect on college attendance or the likelihood of receiving a Bachelor's degree. However, racial heterogeneity shows Hispanic students experience a significant increase in the likelihood of attending college. I rule out changes in school expenditures and teacher characteristics as potential mechanisms and present suggestive evidence that schools are classifying more students as learning disabled. Lastly, accountability is more effective in conjunction with promotion gates.

Purpose This cross-sectional study examined the relationship between indicators of workplace nancial stress and obesity among working adults. Methods Respondents were sampled from the 2021 National Health Interview Survey. Adults were included if they worked in the last week were between 18 to 65 years (n = 14, 976). A multivariable model predicted having obesity using the following the independent variables: working full time, dened as working 35 hours per week or more, working while sick, perceived likelihood of job loss, food insecurity and sociodemographic characteristics. The multivariable model was based on bivariate analyses comparing these variables, and serious psychological distress to having obesity and the number of chronic health conditions. Results At the bivariate level 6.5% of full-time workers had food insecurity, and 9.1% perceived they were fairly likely or somewhat likely to lose their job and 45% had serious psychological distress. In a multivariable model, there was a positive association between obesity and full-time employment (AOR = 1.5 CI = 1.3, 1.7), workers perceptions of being fairly or somewhat likely to lose their job (AOR = 1.2 CI = 1.0, 1.4), and obesity and food insecurity (AOR = 1.2 CI = 1.0, 1.4) after adjusting for sociodemographic characteristics. Conclusions The relationship between having obesity and working full time, serious psychological distress, perceptions about job loss and food insecurity suggests that employment may be stressful. Despite working full time, many adults experienced food insecurity. Suggestions for increasing the safety net for workers to protect against nancial and work instability are discussed.

The appendixes in this annex provide estimates of the number of U.S. military veterans who experience financial burden from housing costs. They also offer context for how these estimates have changed over the past 15 years by comparing trends among veterans to nonveterans. These findings should be of interest to policymakers focused on veteran welfare and housing policy and to the general public. Our research provided the following findings: •
Fewer veteran households than nonveteran ones are financially burdened by housing costs (defined as spending more than 30 percent of gross household income on housing). •
This lower level of housing cost burden (HCB) reflects both higher incomes and lower costs of homeownership among veterans, although income growth among nonveterans has outpaced veteran income growth and reduced the differences over the past 15 years. •
Veterans are more likely to be homeowners than nonveterans, and veteran homeowners have lower housing costs than nonveteran homeowners. However, for veteran households that rent, housing costs are similar or, in some regions, larger than those for comparable nonveteran households. •
The gap in HCB experienced by veterans relative to nonveterans shrinks as income lowers; veterans and nonveterans with the lowest income levels have similarly high levels of HCBs. •
In contrast to veterans overall, veterans who enlisted after September 11, 2001, (post-9/11) experience a greater HCB than nonveterans. This is due, in part, to recent increases in housing prices and rents and to the fact that more post-9/11 veterans are renters, but more research is needed on the factors driving this relationship. •
Although veterans are less likely to be female, the female veteran population is expected to grow over the next few decades. We found evidence consistent with past research showing that female veterans are more likely to face housing instability than male veterans, suggesting a need to focus on better meeting the housing needs of this population.

This dissertation consists of three essays regarding topics of immigration and education. The first essay is a case study of California’s DREAM Act, a state provision that began awarding qualified undocumented Californians access to state-funded financial aid in 2013. I explore effects of the program on various educational outcomes, using a synthetic control framework to accommodate the single-treatment scenario. Regression estimates are obtained by using foreign-born non-citizen Mexicans + Central Americans (FBNCMCA) as proxies for undocumented status. Results show a modest 4% effect on college enrollment of FBNCMCA Californians relative to that of Synthetic California. The increase is partly attributable to higher enrollment and graduation rates at the high school level, as evidenced by Post/Pre-RMSPE rankings and graphical comparisons between California and Synthetic California. Other educational outcomes (e.g., college graduation) show economically meaningful synthetic control graphs (i.e., good fit in the pre-period), but are prone to lack of power and precision which may be brought on by the synthetic control framework and narrow sample of interest. The second essay continues this line of inquiry by exploring adoption of in-state tuition (ISRT) and state-funded financial aid (FA) provisions across the United States. To exploit variation from staggered adoption of these policies, I employ a two-way fixed effects (TWFE) difference-in-differences (DiD) strategy. Furthermore, in light of the continued development of the generalized DiD framework, I use several alternative identification strategies, as a means to alleviate any concerns regarding known-heterogeneity issues (over time and across sections). Results show a clear, positive impact on college-going rates (9.6% and 15.6% for ISRT and FA provisions relative to the mean) with p = 0.04 but fail to show a significant effect for college completion or high school rates. This suggests that many undocumented individuals were already attending and finishing high school prior to the introduction of these provisions and that such policies may have pushed marginal students to enroll. Results also fail to show these students finishing their college education, which dampens the average graduation rate. Evidence of U.S. citizens being negatively impacted through lower college enrollment rates were present. Specifically, fewer Black Americans enrolled in postsecondary education following such provisions, leading to a smaller cohort group. Among these students, a larger proportion went onto graduate since individuals may have been positively selected. Finally, robustness checks showed that certain estimates under the TWFE strategy may be capturing a weighted average rather than the actual treatment effect. Use of never-adopting controls and only-adopting treatments were able to shed light on the complex relationship between ISRT and FA provisions and their impacts on educational outcomes. The third essay investigates labor market outcomes of Hispanic immigrants in the United States following the 2016 election cycle. Using the sharp change in immigration policy from the Obama to Trump administration, I model an individual’s labor market choice through a nested logit framework. Within the multi-level strategy, the first choice deals with one’s decision to participate in the labor force, which is followed by a degenerate branch if individuals choose not to participate. Conditional on being in the labor force, individuals can be employed, unemployed, or self-employed (exclusive). This strategy relaxes the independence of irrelevant alternatives assumption across my two nests: in the labor force and not in the labor force. Regression results show a statistically significant decrease of ≈29% (relative to the mean) in the number of unemployed and an increase in self-employed individuals (≈9%). Descriptively, some of the previously unemployed undocumented individuals may be giving up in the labor market, while others are transitioning to self-employment.

The estimation of racial disparities in health care, financial services, voting, and other contexts is often hampered by the lack of individual-level racial information in administrative records. In many cases, the law prohibits the collection of such information to prevent direct racial discrimination. As a result, many analysts have adopted Bayesian Improved Surname Geocoding (BISG), which combines individual names and addresses with the Census data to predict race. Although BISG tends to produce well-calibrated racial predictions, its resid-uals are often correlated with the outcomes of interest, yielding biased estimates of racial disparities. We propose an alternative identification strategy that corrects this bias. The proposed strategy is applicable whenever one's surname is conditionally independent of the outcome given their (unobserved) race, residence location, and other observed characteristics. Leveraging this identification strategy, we introduce a new class of models, Bayesian Instrumental Regression for Disparity Estimation (BIRDiE), that estimate racial disparities by using surnames as a high-dimensional instrumental variable for race. Our estimation method is scalable, making it possible to analyze large-scale administrative data. We also show how to address potential violations of the key identification assumptions. A validation study based on the North Carolina voter file shows that BIRDiE reduces error by up to 84% in comparison to the standard approaches for estimating racial differences in party registration. Open-source software is available which implements the proposed methodology.

This research examines employment discrimination against custodial single fathers in the United States. Fatherhood is associated with breadwinning, and employers expect full work commitment. Yet, caregiving constrains breadwinning because family demands are time-consuming and labor-intensive. This raises the following questions: In what ways, if at all, do employers discriminate against single fathers with primary caregiving responsibility? How do custodial single fathers experience their roles as primary breadwinners and primary caregivers? My dissertation offers answers to these questions by considering how gender, breadwinning, and caregiving roles operate in employers’ hiring preferences and single fathers’ efforts to meet work and caregiving demands.

An estimated 990,000 Hispanics with origins from Spain – that is, those who are immigrants from or who trace their family ancestry to Spain – resided in the United States in 2021, according to a Pew Research Center analysis of the U.S. Census Bureau’s American Community Survey. Spaniards in this statistical profile are people who self-identified as Hispanics with origins from Spain.

This paper explores the underlying factors contributing to the recent decline in labor share, focusing specifically on the roles of automation and the development of new tasks that are exclusive to humans. First, our paper strengthens the argument that automation has a negative impact on labor share. Second, we are the first to empirically estimate the influence of new human-exclusive tasks on labor share. Our findings suggest that the positive impact of human-exclusive tasks dominates the negative impact brought about by automation. Third, we find that the elasticity of substitution between labor and capital is less than one, offering a coherent framework for predicting how various factors-capital price, robot price, and wages-impact labor share. We identify two distinct mechanisms through which robots negatively affect labor share: automation and a reduction in the price of robots. Our general equilibrium model predicts that the latter will gain increasing importance in the future as robots become more prevalent. Lastly, we estimate the elasticity of substitution between tasks to be one, empirically validating an assumption that many existing studies have made.

Background: Diabetes and hypertension are common in Asian Americans and vary by subgroup. There may be further variation by social determinants of health (SDOHs), but few studies have examined this previously. Objective: To examine the associations of SDOHs and diabetes and hypertension within and across Asian subgroups in the USA Design: Series cross-sectional analyses Setting: National Health Interview Surveys (NHIS) from 1999 to 2018 Participants: Asian-American adults (Chinese, Filipino, Asian Indian, and Other Asian [Korean, Vietnamese, Japanese, and other]) Measurements: Self-reported diabetes and hypertension prevalence in pooled 5-year increments over 1999–2018 and multivariable regression models to assess the adjusted prevalence of diabetes or hypertension by poverty, marital status, education, and years in the USA, adjusting for age, sex, BMI, and health insurance status Results: From 1999–2003 to 2014–2018, the age- and sex-adjusted prevalence of diabetes increased for Other Asians (absolute change: 4.6%) but not for other subgroups; age- and sex-adjusted hypertension prevalence significantly increased for Asian Indians and Other Asians (absolute change: 5–7.5%). For Filipinos, high school education or less was associated with an increase in diabetes prevalence over time (difference from 1999–2003 to 2014–2018: +6.0 (95% CI: 2.0–10.0)), while for Asian Indians, college education or higher was associated with an increase in diabetes prevalence for the same period (difference: +2.7 (95% CI: 0.01–5.4). Differences over the 2 time periods (1999–2003 and 2014–2018) show that Filipino and Other Asians, who lived in the USA for ≥10 years, increased in diabetes prevalence. Similar variations in associations of SDOHs by Asian subgroup were seen for hypertension. Limitations: Self-reported primary outcomes and multi-year data were pooled due to small sample sizes. Conclusions: The influence of SDOHs on cardiometabolic risk is not uniform among Asian Americans, implying tailored strategies may be needed for different population subgroups. Primary Funding Source: NIH

High-wage regions have frequently been models for economic development policy. However, these places have also been more likely to experience a “hollowing out” of the middle of the labor market, whereby college-educated residents experience high-wage growth, while residents without a college degree do not. This paper studies regions where—in contrast to hollowing out—the share of middle-wage jobs has grown since 1980. Its aim is to understand how—if at all—the characteristics of these regions could suggest a model of economic development that prioritizes opportunities for noncollege graduates. The authors find that regions with a growing share of middle-wage jobs have been associated with higher levels of upward mobility and wage growth for workers without a college degree. These places, such as Wausau, Wisconsin, and Manhattan, Kansas, are associated with higher local school performance and have experienced comparatively high growth in production jobs.

Using American Community Survey data, we describe the labor market experiences of sexual minority artists and arts majors and explore the differences in their earnings and employment. We identify workers in cohabiting relationships as being either in same-sex or opposite-sex relationships. We find that artists are more than twice as likely to be in same-sex relationships when compared to the overall workforce, largely driven by higher shares of males in same-sex relationships. A similar pattern is observed for arts majors when compared to all college graduates. We find significant heterogeneity in the share of workers in same-sex relationships across individual artist occupations and individual arts majors. In testing for differences in earnings and employment, we use regression analysis, finding an increased likelihood of unemployment for male and female artists in same-sex relationships, earnings penalties for males in same-sex relationships, and earnings premiums for females in same-sex relationships. However, like other studies using coupled data, we hesitate in using causal language in interpreting the earnings regression results due to data limitations.

How people spend their time is an indicator of how they live their lives, with time use over the life course conditioned both by age and by participation in age-graded institutions. This chapter uses nationally representative data from the pooled 2008–2020 American Time Use Survey (ATUS) to evaluate how time use in 12 activity categories varies by age, gender, and disability status among 137,266 respondents aged 15 and older. By doing so, we quantify the “disability gap” in time use between men and women with and without disabilities, identifying at what age and by how much people with disabilities experience time differentials in activities of daily living (ADLs), instrumental activities of daily living (IADLs), and other indicators of social participation. Results indicate that – at many ages – patterns of time use for people with disabilities deviate from those of people without disabilities, with more pronounced differences in midlife. Further, the magnitude of women's disability gaps equals or exceeds men's for sleeping, and nearly all ADLs and IADLs, indicating that disability gaps are also gendered.

An analysis by the Pew Research Center of the U.S. Census Bureau's American Social Survey. The U.S. estimates that 990,000 Hispanics of Spanish descent, that is, those who immigrated from Spain or traced their family ancestry, will live in the United States. U.S. in 2021. Spaniards in this demographic profile self-identify as Hispanics from Spain.

Last year, a survey conducted on how well students were treated after graduating from colleges in the US according to their majors showed that the Department of Chemical Engineering took the top spot. Conversely, it was found that the majors with the worst treatment were theology or religious studies. This is the gist of the 2022 college graduate employment status report recently published by the Federal Reserve Bank of New York. This is the result of analyzing data from the Bureau of Labor Statistics under the US Department of Labor and data from the Population Census Microdata Service System (IPUMS) of the University of Minnesota Population Center, the world's largest data research institute.

Starting with Executive Order 13166 in 2000, the United States federal government began to address the language disparity issues in health care. Around the same time, several states have begun to pass language access (LA) legislation mandating translation and interpretation services at hospitals for limited English proficient (LEP) individuals. This study uses these multiple discontinuities to evaluate the effect of language access laws on infant mortality rates, adequacy of care, Apgar scores, and the number of prenatal visits from the years 1995 to 2004 for limited English proficient families. I find ambiguous results of language access laws positively impacting infant mortality rates or Apgar scores, but I find clear positive impacts on the adequacy of care and the number of prenatal visits. These findings suggest that language access laws have a clear effect on reducing barriers for limited English proficient mothers, and improving the care mothers receive. Furthermore, there is limited evidence that it improves infant health or outcomes, but the increase of prenatal visits and adequacy of care likely indirectly leads to improving infant mortality rates and Apgar scores. More research is needed into discovering how those mechanisms work and the costs of language services.

Students have several options for life after high school, including enrolling in college, pursing a technical training program, starting a career, or enlisting for military service. While college has been a popular choice, college enrollment for recent high school graduates has dropped from its peak of 70% in 2009 to 61% in 2021. 1 In fact, some people are challenging the notion that college is the best route for the majority of students. A March 2023 survey found that only 42% of Americans believe college is worth the cost because it leads to better job opportunities and higher income, while 56% believe that earning a college degree is not worth the cost. That has changed a lot in 10 years: A 2013 study found that 53% believed college was a good decision, while 40% believed it wasn't. 2 Of course, attending college is an individual decision, as each person must weigh the costs and benefits of their options. While some costs of college are immediate (your tuition payments), the benefits are spread over an entire career. This article looks at the costs and benefits of a college education and explains the rate of return of going to college, viewing higher education as an investment. Economists often use the word investment to refer to spending on capital, but that does not mean just physical capital (tools and equipment); it can mean investment in human capital (educa-tion and training) too. Costs and Benefits of Attending College It's true that the cost of going to college has risen significantly in recent decades. The first row of Table 1 shows the average annual tuition for colleges and universities in 1980, 2000, and 2020. The last row of the table shows how much college tuition costs in terms of 1980 dollars, showing that in real (inflation adjusted) terms, attending college cost over twice as much in 2020 as it did in 1980.

The United States (U.S.) witnessed considerable reduction in cigarette smoking prevalence in the recent past. While the correlates of smoking prevalence and related disparities among U.S. adults are well documented, there is limited information on how this success was shared among different population sub-groups. Based on data from the National Health Interview Surveys, 2008 and 2018, representative of non-institutionalized U.S. adults (18 years and above), we applied the threefold Kitawaga-Oaxaca-Blinder linear decomposition analysis. We decomposed the trends in cigarette smoking prevalence, smoking initiation, and successful cessation into changes in population characteristics holding smoking propensities constant (compositional change), changes in smoking propensities by population characteristics holding population composition constant (structural change), and the unmeasured macro-level changes affecting smoking behavior in different population sub-groups at differential rates (residual change) to quantify the shares of population sub-groups by sex, age, race/ethnicity, education, marital status, employment status, health insurance coverage, family income, and region of residence in the overall change in smoking rates. The analysis shows that decreases in smoking propensities regardless of the changes in population composition accounted for 66.4% of the reduction in smoking prevalence and 88.7% of the reduction in smoking initiation. The major reductions in smoking propensity were among Medicaid recipients and young adults (ages 18–24 years). The 25-44-year-olds experienced moderate increase in successful smoking cessation, while the overall successful smoking cessation rate remained steady. Taken together, consistent reduction in smoking among U.S. adults by all major population characteristics, accompanied by disproportionately larger reduction in smoking propensities among the population sub-groups with initially higher smoking propensity compared to the national average, characterized the decline in overall cigarette smoking. Strengthening proven tobacco control measures with targeted interventions to reduce smoking propensities among underserved populations is key to continued success in reducing smoking overall and remedying inequities in smoking and population health.

This is an original investigation of self-reported health status among Hispanic adults from 1997 to 2018 in the United States (US). Previous research has shown there is a widening gap in poor/fair self-reported health between Hispanics who answer health surveys in English and those who answer in Spanish that cannot be explained by demographic/socioeconomic characteristics, assimilation or region of residence. Using data from the National Health Interview Survey (1997–2018), this study explores the patterns underlying the recent increase in self-reported health among Hispanic adults in the United States by estimating the percent of the population reporting poor/fair health status by language of interview and place of birth. Central to this study is the use of ‘regular’ as a translation to “fair” which has been poised to be a non-equivalent translation. This investigation reveals that the increase is highly concentrated among non-US born Hispanic adults who answer health surveys in Spanish with increase in reports of “regular” health status driving this trend. The results presented in this short communication underscore the importance of language of interview when collecting key measures of health often employed to study health disparities.

Over the past century, we have witnessed significant changes in economic circumstances that are interrelated. In this dissertation, I focus on three key aspects of these changes. Chapter 1 examines the effect of computerization on educational attainment. I demonstrate that routine-intensive occupations have increased the share of college graduates more rapidly than abstract-intensive occupations over the last 40 years. Based on this fact, I construct a model that explains that half of the increase in the share of college graduates is due to computerization. Chapter 2 reveals the effect of changes in population growth on TFP (total factor productivity) growth. Using a firm-dynamics model extended to include endogenous growth, I show that a significant portion of the decline in TFP growth is due to the decrease in population growth in the US and Japan. Chapter 3 constructs a model to show the increasing trend of the share of investment in intangible assets. Using this model, I demonstrate that the decline in the relative price of investment goods can account for the increasing share of intangible investment. Overall, this dissertation contributes to our understanding of the complex interplay between technology, demographics, and investment patterns in the modern economy.

Background: By the end of 2022, nearly 20 million workers in the United States have gained paid-sick-leave coverage from mandates that require employers to provide benefits to qualified workers, including paid time off for the use of preventive services. Although the lack of paid-sick-leave coverage may hinder access to preventive care, current evidence is insufficient to draw meaningful conclusions about its relationship to cancer screening. Methods: We examined the association between paid-sick-leave mandates and screening for breast and colorectal cancers by comparing changes in 12- and 24-month rates of colorectal-cancer screening and mammography between workers residing in metropolitan statistical areas (MSAs) that have been affected by paid-sick-leave mandates (exposed MSAs) and workers residing in unexposed MSAs. The comparisons were conducted with the use of administrative medical-claims data for approximately 2 million private-sector employees from 2012 through 2019. Results: Paid-sick-leave mandates were present in 61 MSAs in our sample. Screening rates were similar in the exposed and unexposed MSAs before mandate adoption. In the adjusted analysis, cancer-screening rates were higher among workers residing in exposed MSAs than among those in unexposed MSAs by 1.31 percentage points (95% confidence interval [CI], 0.28 to 2.34) for 12-month colorectal cancer screening, 1.56 percentage points (95% CI, 0.33 to 2.79) for 24-month colorectal cancer screening, 1.22 percentage points (95% CI, −0.20 to 2.64) for 12-month mammography, and 2.07 percentage points (95% CI, 0.15 to 3.99) for 24-month mammography. Conclusions: In a sample of private-sector workers in the United States, cancer-screening rates were higher among those residing in MSAs exposed to paid-sick-leave mandates than among those residing in unexposed MSAs. Our results suggest that a lack of paid-sick-leave coverage presents a barrier to cancer screening. (Funded by the National Cancer Institute.)