Using a state panel from 1940–2010, I examine the impact of immigration on the high school completion of natives in the United States. Immigrant children could influence native children’s educational experience as well as their expected future labor market. I find evidence for both channels and a positive net effect. An increase of one percentage point in the share of immigrants in the population aged 11–64 increases the probability that natives aged 11–17 eventually complete 12 years of schooling by 0.3 percentage point. I account for the endogeneity of immigrant flows by using instruments based on 1940 settlement patterns.

Gary McDowell, Misty Heggeness and colleagues present census data showing how the biomedical workforce is fundamentally different to those of past generations – academia should study the trends, and adapt.

We study an economy with capital-skill complementarities, an endogenous human capital and occupational choice decision and firm-level financing constraints. We ask: to what extent do financial frictions reduce output per worker across countries? In our economy, firm-level frictions depress physical capital accumulation and, in equilibrium, also reduce the acquisition of human capital, thus amplifying the decline in output per worker. We estimate the model using repeated cross-sections of individual workers' educational attainment, labor earnings and occupational choice, both for U.S. time-series, as well as for a cross-section of countries. We find that financial frictions have much larger effects on output per worker in our economy than they do in economies with a fixed supply of human capital.

The first chapter of this dissertation considers a new class of robust estimators in a linear instrumental variables (IV) model with many instruments. The estimators are generalized method of moments (GMM) estimators, and the class includes the limited maximum likelihood estimator (LIML) as a special case. Each estimator in the class is consistent and asymptotically normal under many instruments asymptotics, and this chapter provides consistent variance estimators that are of the “sandwich” type and can be used to conduct asymptotically correct inference. Furthermore, this chapter characterizes an optimal robust estimator among the members of the class. Compared to LIML, the optimal robust estimator is less influenced by outliers and more efficient under thick-tailed error distributions. In an empirical example (Angrist and Krueger, 1991), the optimal robust estimator is approximately 80% more efficient than LIML. The second chapter of this dissertation provides a central limit theorem based on Stein’s method (Stein, 1972) which is an integral component in the proof of the main theorem in the first chapter. It also appears to be general enough in its scope that it can be applied to a variety of other problems.

Over the last few years, so-called “legacy of slavery” research has made great strides in helping us to understand how the Trans-Atlantic slave trade continues to affect contemporary life. New and improved data sources have allowed this work to become increasingly complex, with a combination of sub-national, cross-national, and individual level analyses. This research focuses on the former, where a number of questions remain, namely: do the long-term effects of slavery remain robust to other historical factors; does slavery exacerbate color stratification among black Americans; and does slavery have a net-positive influence on the contemporary social outcomes of white Americans? I use regression analysis and Census data to answer these questions. Ultimately, I find that the answers to these questions are “yes” to varying degrees. The effect of slavery remains robust to historical covariates, though the relationship is complex. Slavery seems to exacerbate color stratification among black Americans through its disproportionately negative effect on darker skinned black people. And on four of six contemporary measures, slavery improves the life outcomes of white Americans. I discuss the implications of these findings for the future of sociological research and the discussion of reparations for black Americans.

Although gender inequality has been recognized as a crucial factor influencing population health in the developing world, research has not yet thoroughly documented the role it may play in shaping U.S. infant mortality rates (IMRs). This study uses administrative data with fixed-effects and random-effects models to (1) investigate the relationship between political gender inequality in state legislatures and state infant mortality rates in the United States from 1990 to 2012, and (2) project the population level costs associated with women's underrepresentation in 2012. Results indicate that higher percentages of women in state legislatures are associated with reduced IMRs, both between states and within-states over time. According to model predictions, if women were at parity with men in state legislatures, the expected number of infant deaths in the U.S. in 2012 would have been lower by approximately 14.6% (3,478 infant deaths). These findings underscore the importance of women's political representation for population health.

Expanding employment opportunities for citizens has become an increasingly central goal of public policy in the United States. Prior work has considered that the inability of households to spatially access jobs may be a driver of unemployment. The provision of public transportation provides a viable policy lever to increase the number of job opportunities available to households. Previous research has yielded mixed results regarding whether household location is an important factor in determining employment status. Several papers have identified mobility as a limiting factor for obtaining a job, particularly in regards to private vehicle ownership. The location of economically developed neighbourhoods and the citing of public transportation are conceivably codetermined, presenting an endogenous relationship. It is therefore unclear if public transportation access is actually contributing to neighbourhood job market outcomes. This paper will use the incidence of Hurricane Sandy striking New York City on 29 October 2012 and the resulting exogenous reduction in public transit access to particular neighbourhoods as a natural experiment to test for the effect of public transportation on employment outcomes. This study identifies a significant causal effect linking public transportation access to neighbourhood unemployment rates, particularly amongst subgroups dependent on public transit.

Where do immigrants choose to settle and what are the consequences of their location choices for economic activity and welfare? This paper provides a new perspective on these questions by investigating the causes and effects of the spatial distribution of immigrants across US cities. We document that, over the last decades: a) immigrants have increasingly concentrated in large, high-wage, and expensive cities, and b) the earnings gap between immigrants and natives is higher in larger and more expensive cities. This relationship between city characteristics and the wage gap is stronger for immigrants from low-income countries and those who have spent fewer years in the United States and is robust in controlling for immigration networks. In order to explain these findings, we develop a simple spatial equilibrium in which immigrants consume (either directly, via remittances, or future consumption) a fraction of their income in their countries of origin. Thus, immigrants not only care about local prices, but also about price levels in their home country. Hence, if foreign goods are cheaper than local goods, immigrants prefer to live in high-wage, high-price cities, where they also accept lower wages than natives. Given that large and more expensive cities tend to be more productive, immigrant location choices move economic activity toward more productive cities, which results in total output gains. We estimate that current levels of immigration increase total aggregate output per worker by around .15 percent. We also discuss welfare implications.

METHODS Most quantitative research on fertility decline in the United States ignores the potential impact of cultural and familial factors. We rely on new complete-count data from the 1880 US census to construct couple-level measures of nativity/ethnicity, religiosity, and kin availability. We include these measures with a comprehensive set of demographic, economic, and contextual variables in Poisson regression models of net marital fertility to assess their relative importance. We construct models with and without area-fixed effects to control for unobserved heterogeneity. CONTRIBUTION All else being equal, we find a strong impact of nativity on recent net marital fertility. Fertility differentials among second-generation couples relative to the native-born white population of native parentage were in most cases less than half of the differential observed among first-generation immigrants, suggesting greater assimilation to nativeborn American childbearing norms. Our measures of parental religiosity and familial propinquity indicate a more modest impact on marital fertility. Couples who chose biblical names for their children had approximately 3% more children than couples relying on secular names, while the presence of a potential mother-in-law in a nearby household was associated with 2% more children. Overall, our results demonstrate the need for more inclusive models of fertility behavior that include cultural and familial covariates.

Transformation models are a very important tool for applied statisticians and econometricians. In many applications, the dependent variable is transformed so that homogeneity or normal distribution of the error holds. In this paper, we analyze transformation models in a high-dimensional setting, where the set of potential covariates is large. We propose an estimator for the transformation parameter and we show that it is asymptotically normally distributed using an orthogonalized moment condition where the nuisance functions depend on the target parameter. In a simulation study, we show that the proposed estimator works well in small samples. A common practice in labor economics is to transform wage with the log-function. In this study, we test if this transformation holds in CPS data from the United States.

Due to health and wellness benefits associated with city dwellers’ proximity to parks and urban greenspace (UGS), equitable access to these spaces can be considered an environmental justice concern. However, in attempting to address the greenspace inequities that often exist within cities, planners may inadvertently contribute to processes of gentrification and the displacement of current residents by encouraging real estate speculation. Using San Francisco as a model, this thesis explores UGS distributions and assesses their relationships with socioeconomic demographics and gentrification. It also considers a set of investment strategies that may help to avoid gentrification and displacement. The identification of several forms of economic, educational, and agegroup disparity in residents’ access to UGS in San Francisco were among the key findings. Racial-ethnic minority groups in San Francisco experienced greater than average greenspace access. Gentrified neighborhoods were also positively correlated with greater access, supporting the idea that greenspace investments might contribute to real estate speculation and so-called “green gentrification”. The results of this thesis’s analyses did not support the strategy of favoring smaller, more distributed UGS installations as a method of avoiding gentrification and displacement in San Francisco. This thesis expands upon the existing greenspace equity literature by incorporating into its definition of UGS several innovative forms of green infrastructure that have not been well-studied in the past.

There is limited evidence justifying the economic case for state education policy. Using new measures of workers’ cognitive skills that allow for selective internal migration and foreign immigration, we provide preliminary estimates of growth regressions that incorporate worker skills. Our descriptive models show that educational achievement predicts economic growth across US states over the past four decades. Projections from our growth models show the substantial potential scope for state economic development through improving school quality. While we consider the impact of a range of educational reforms, an improvement that moves each state to the best-state level would in the aggregate yield an estimated present value of long-run gains of 8 percent of discounted future GDP.

In the first paper I investigate whether higher returns to education for women lead to a reduction in teen birth rates. I first estimate the impact of the growth in the female college wage premium on teen birth rates in the U.S. An increase in the college wage premium is a proxy for higher expected earnings after completing college. Higher future earnings create incentives to pursue further education and they decrease the relative returns to having a birth as a teen. Using state variation, I find that an increase in the female college wage premium does lead to a reduction in teen birth rates. To address endogeneity and measurement error issues, I use a shift-share instrumental variable for women's wages to construct the female college wage premium. Results from three supplemental analyses support the main results. First, teen birth rates respond less to the growth in the female high school wage premium. Second, the estimates suggest that growth in the male college wage premium has a lower impact on teen birth rates but the results are not conclusive. Third, the elasticity of fertility with respect to the female college wage premium decreases as women grow older. This research can explain approximately 30\% of the fall in teen birth rates that has occurred from the early 1990s up to the present.

The analysis of public policies, even when performed by the best non-partisan agencies, often lacks credibility (Manski, 2013). This allows policy makers to cherry- pick between reports, or within a specific report, to select estimates that better match their beliefs. For example, in 2014 the Congressional Budget Office (CBO) produced a report on the effects of raising the minimum wage that was cited both by opponents and supporters of the policy, with each side accepting as credible only partial elements of the report. Lack of transparency and reproducibility (TR) in a policy report implies that its credibility relies on the reputation of the authors, and their organizations, instead of on a critical appraisal of the analysis. This dissertation translates to policy analysis solutions developed to address the lack of credibility in a different setting: the reproducibility crisis in science. I adapt the Transparency and Openness Promotion (TOP) guidelines (Nosek et al, 2015) to the policy analysis setting. The highest standards from the adapted guidelines involve the use of two key tools: dynamic documents that combine all elements of an analysis in one place, and open source version control (git). I then implement these high standards in a case study of the CBO report mentioned above, and present the complete analysis in the form of an open-source dynamic document. In addition to increasing the credibility of the case study analysis, this methodology brings attention to several components of the policy analysis that have been traditionally overlooked in academic research, for example the distribution of the losses used to pay for the increase in wages. Increasing our knowledge in these overlooked areas may prove most valuable to an evidence-based policy debate.

This report describes preliminary work in support of an early childhood needs assessment for American Indian and Alaska Native (AI/AN) children prenatal to age five. The report uses existing data to describe the population of AI/AN children and families and their participation in early childhood services. This work is part of a larger AI/AN Early Childhood Needs Assessment design project (AI/AN EC Needs Assessment). Mathematica Policy Research convened a Community of Learning (CoL) to inform three design topics on describing the AI/AN population, studying early childhood services organization and delivery, and assessing . . .

Recent immigration and migration patterns have altered the ethnoracial composition of Alameda County, California. Sociopolitical leaders have struggled to adjust to these changes. In an effort to facilitate limited English speakers access to critical municipal services, Oakland - the largest municipal in Alameda County - passed an Equal Access to Services Ordinance on May 8, 2001, which is a groundbreaking language access legislation for the City of Oaklands public administration. Using data from the 2000 Census and the 2005-2011 American Community Survey, this study examines the impact of bilingual employment policies on the ethnoracial segmentation of Alameda County workers. Logistic regression reveals that bilingual employment policies have reorganized both targeted (i.e., public contact) and non-targeted occupations within the local government public administration sector. Specifically, Spanish/Chinese bilingual speakers made gains in the public administration sector (the intended effects), while Black monolingual English speakers experienced losses (the unintended effects). The representation of Black monolingual English speakers in public contact jobs within the local government public administration sector declined by as much as 18 percentage points after the implementation of the nations first municipal-level bilingual employment policy. The impact of bilingual employment policies on the East Bays Black/Brown relations and African Americans hold on low-skilled jobs in service industries is examined.

The spatial isolation of workers from firms is frequently considered as a cause of localized unemployment. Increasing the capacity and efficiency of urban transportation systems is a habitually considered solution to worker isolation. This paper will quantify metropolitan level mobility with respect to home-work commuting. Commuter mobility in the US is found to vary substantially across metros in both levels and trends during the 2005-2014 study period. The impact of mobility on locational access is shown to be theoretically ambiguous due to a dispersion effect. Increased commuter mobility has proceeded contemporaneously with increased metropolitan employment dispersion. An instrumental variable method will exploit random variation in the political process governing transportation infrastructure funding. Results provide causal evidence that increased commuter mobility degrades employment outcomes, particularly for black residents and youth. Findings are consistent with increased commuter mobility exacerbating spatial mismatch through employment sprawl.

Online advertising is the main source of revenue for social media companies. Facebook allows advertisers to target users with certain characteristics, such as age, gender, country of origin, education level, or topical interest. Before an ad is launched Facebook’s advertising platform provides an estimate of how many users match the provided criteria. This functionality, akin to a ‘digital census’ over Facebook users, has so far been untapped for demographic research. We show the feasibility of estimating stocks of migrants within and across countries and discuss the potentials and limitations of the data. The lack of much needed information about migrants, together with the rapid global expansion of social media use and the online advertising industry, o↵er new opportunities to study migration as well as monitor progress towards achieving the goals of the United Nations 2030 Agenda for Sustainable Development.

This study examines intergenerational social mobility pathways within families of mixed migratory status. Focusing on the residents of Leon, Mexico, I use a mixed methods analysis of interviews and survey data on 50 return migrants to Leon to identify signs of intergenerational social mobility within their families. The study finds that parental return migration plays a role in determining a family’s potential for upward mobility. Findings show that migrants’ labor market experiences, both abroad and upon return to the origin country, influence their likelihood of intergenerational social mobility. Labor market experiences prove especially influential over these families’ mobility pathways, with these experiences directly affecting the occupational and educational opportunities of the return migrants and their children, respectively. The study ultimately demonstrates international migration’s potential to disrupt social class reproduction and highlights the implications of migration on the future economic success of migrant families with low levels of traditional human capital.

This thesis introduces computer vision algorithms that harness street-level imagery to con- duct automated surveys of the built environment and populations at an unprecedented resolution and scale. We introduce new tools for computing quantitative measures of urban appearance and urban change. First, we describe Streetscore, an algorithm that quantifies how safe a street block looks to a human observer, using computer vision and crowdsourcing. We extend this work with an efficient convolutional neural network-based method that is capable of computing several perceptual attributes of the built environment from thousands of cities from all six inhabited continents. Second, we introduce a computer vision algorithm to compute Streetchange-a metric for change in the built environment-from time-series street-level imagery. A positive Streetchange is indicative of urban growth; while negative Streetchange is indicative of decay. We use these tools to introduce new datasets. We use the Streetscore algorithm to generate the largest dataset of urban appearance to date, which covers more than 1 million street blocks from 21 American cities. We use the Streetchange algorithm to also gener- ate a dataset for urban change containing more than 1.5 million street blocks from five large American cities. These datasets have enabled research studies across fields such as economics, sociology, architecture, urban planning, and public health. We utilize these datasets to provide new insights on important research questions. With the dataset on urban appearance, we show that criminal activity has a robust positive corre- lation with the spatial variation in architecture within neighborhoods. With the dataset on urban change, we show that positive urban change occurs in geographically and physically attractive areas with dense, highly-educated populations. Taken together, the tools, datasets, and insights described in this thesis demonstrate that computer vision-driven surveys of people and places have the potential to massively scale up studies in social science, to change the way cities are built, and to improve the design, execution, and evaluation of policy and aid interventions.