Research in statistical data privacy (SDP) has traditionally self-organized into two disjoint schools of thought: statistical disclosure limitation (SDL) and formal privacy (FP). Both perspectives rely on different units of analysis, measures of disclosure risk, and adversarial assumptions. Yet in recent years, differential privacy (DP), a particular variant of FP, has emerged as the methodologically preferred perspective by analyzing release mechanisms and database schemas under the broadest possible adversarial assumptions. To do so, DP quantifies privacy loss by analyzing noise injected into output statistics. For non-trivial statistics, this noise is necessary to ensure finite privacy loss. However, data curators frequently release collections of statistics where some use DP mechanisms and others are released without additional randomized noise. This includes many cases where DP mechanisms are implemented in such a way that depends on the confidential data, such as by choosing the privacy loss parameter based on confidential data (or synthetic data highly correlated with the confidential data). Consequently, DP alone cannot characterize the privacy loss attributable to the entire joint collection of releases, nor decisions that were made in implementing the mechanism. Such problems pose an existential threat to building DP systems in practice that DP alone cannot answer. In this dissertation, we study the privacy and utility properties of “partially private data" (PPD), collections of statistics where only some are released through DP mechanisms. In particular, we define the random variable Z as “public information" not protected by DP. PPD is inherently statistical, as it relies on assumptions about the correlation structure between private and public information. We present a privacy formalism, (ϵ, {Θz}z∈Z)-Pufferfish (ϵ-TP for short when {Θz}z∈Z is implied), a collection of Pufferfish mechanisms indexed by realizations of Z. First, we prove that this definition has similar properties to DP. Next, we introduce two release mechanisms for publishing (PPD) satisfying ϵ-TP and prove their desirable properties. We additionally introduce perfect sampling algorithms to exactly implement these mechanisms, as well as approximate Bayesian computation algorithms for sampling from the posterior of a parameter given PPD. We then compare this inference approach to the alternative where noisy statistics are deterministically combined with Z. We derive mild conditions under which using our algorithms offers both theoretical and computational improvements over this more common approach. We demonstrate all the effects above on two case studies: one on COVID-19 data, and one on rural mortality data. Finally, we discuss the implications of all the above from a social and legal perspective, with the end goal of using PPD to make FP technologies more accessible to essential social science data curators.

Immigrants—many of whom are women—make up a significant portion of the U.S. population. In 2021, 23.2 million lived in the United States, outnumbering immigrant men, according to recent analysis of the 2021 American Community Survey. Immigrant women made up 14% of the country’s overall female population. While their work adds critical value to our economy and society, immigrant women in the labor force earn less than foreign-born men, and less than U.S.-born men or women.

This paper describes a dataset mined from the public archive (1999–2020) of the US National Incident Management System Incident Status Summary (ICS-209) forms (a total of 187,160 reports for 35,170 incidents, including 34,478 wildland fires). This system captures detailed daily/regular information on incident development and response, including social and economic impacts. Most (98.4%) reports are wildland fire-related, with other incident types including hurricane, hazardous materials, flood, tornado, search and rescue, civil unrest, and winter storms. The archive, although publicly available, has been difficult to use for research due to multiple record formats, inconsistent data entry, and no clean pathway from individual reports to high-level incident analysis. Here, we describe the open-source, reproducible methods used to produce a science-grade version of the data, including formal connections made to other published wildland fire data products. Among other applications, this integrated and spatially augmented dataset enables exploration of the daily progression of the most costly, damaging, and deadly environmental-hazard events in recent US history.

There are significant costs to rising unemployment. For workers, particularly those being paid low wages, higher unemployment rates nationally are associated with worse employment prospects, health outcomes, and general well-being. In 2020, these dynamics played out rapidly. The COVID-19 pandemic brought a massive shock to the labor market and the unemployment rate climbed from 3.5 percent, a low level not seen since the 1960s, to 14.7 percent.1 By July 2022, 28 months after the start of the crisis, sizable public investments helped foster a remarkably rapid turnaround and bring the unemployment rate back to its prepandemic level. The speedy recovery of the US economy, in contrast to the slow rebound after the Great Recession, minimized many of the severe problems of sustained high unemployment. By understanding the dynamics of high unemployment, policymakers, practitioners, and employers can work to alleviate its detrimental consequences. This report summarizes existing research on the effects of rising unemployment on three different areas: first, on workers, their workplaces, remuneration, and job quality; second, on the social factors that are indirectly affected by rising unemployment, such as enrollment in education, health outcomes, crime rates, and family well-being, particularly for low-income workers; and third, on the wider economic effects of unemployment on social spending, productivity, national income, and future unemployment rates. One strand of research explores the impact of long-term employment on factors such as income, labor market attachment, and health (Nichols, Mitchell, and Lindner 2013). This paper aims to frame the parameters of a different but complementary research area: the impacts of rising unemployment at large, both in the short and medium term. The contemporary labor market is unusually tight and provides a prime opportunity for researchers to better understand the benefits of low unemployment as well as the risks of rising unemployment. This report serves as a first step in that process and as a potential foundation for future research.

This paper quantifies and decomposes recent trends in US particulate matter (PM2.5) disparities from the electricity sector using a high-resolution pollution transport model. Between 2000 and 2018, PM2.5 concentrations from electricity fell by 89% for the average individual, more than double the decline rate in overall US ambient PM2.5 concentrations. Across racial/ethnic groups, we detect a dramatic convergence: since 2000, the Black-white PM2.5 disparity from electricity has narrowed by 95% and the Hispanic-white PM2.5 disparity has narrowed by 93%, though these disparities still exist in 2018. A decomposition reveals nearly all of these disparity trends can be attributed roughly equally to improvements in emissions intensities and compositional changes in electric generators, with small contributions from scale and residential location changes. This suggests both local air pollution policies and recent coal-to-natural gas fuel switching have played major roles in reducing US racial/ethnic pollution disparities from electricity. Although we detect similarly large PM2.5 improvements for the average low- and high-income individual, PM2.5 disparities by income are relatively small, with little change over time.

Chapter 1 and Chapter 2 explore the implications of geographic labor mobility in the con- text of the US labor market. These chapters share a common motivation based on statistics presented herein describing the geographic dimension of job switching behavior; primarily that a surprisingly high share of job-to-job flows within the US take place across metropoli- tan areas. Chapter 1 explores the microeconomic implications of this for workers through the lens of non-local outside options. In labor markets with frictions, outside options are a key determinant of workers’ wages and firms’ rents. When outside options improve, work- ers can benefit by switching jobs, but even those that remain in the same job can realize gains through leveraging their improved bargaining position to renegotiate wages. In this chapter, I study the geographic dimension of workers’ outside options and effects on labor mobility and wages for job stayers. I show that a large share of job-to-job flows are across metropolitan areas (MSAs), which suggests that the non-local dimension of labor market opportunities may be substantial. To obtain causal estimates of the effect of non-local out- side options on wages and geographic labor mobility, I construct measures of exposure to changes in labor market conditions in other markets, and use a shift-share instrumental variable strategy to identify exogenous variation in non-local labor demand. I find that increases in labor demand in an MSA’s network of labor markets are associated with in- creased job-to-job outflows with an elasticity of about .30, and higher wage growth for job stayers with an elasticity of .11. The effect of non-local shocks on job switching and wage growth is 30-50% of similar estimates of the effect of local labor demand shocks. Labor mobility is much more responsive to demand from MSAs with the strongest historic labor flows, which account for about 70% of the total mobility effect. I find similar mobility re- sponses across education levels, but the effect of non-local outside options on wage growth is concentrated on workers without a 4-year college degree and in industries with lower average education levels. Chapter 2 turns to the macroeconomic implications of geographic labor mobility in de- termining long-run labor market outcomes. Using U.S. data, I show that job-to-job flows across metro areas are about 40% of all metro area job-to-job flows, and that there is sub- stantial heterogeneity across metro areas in the rate of incoming and outgoing job-to-job flows. I introduce a general equilibrium model of spatial on-the-job search that provides a framework for studying the effects of labor markets’ heterogeneous geographic positions on long-run outcomes. In the model, labor demand is endogenous and wage bargaining al- lows me to explore the implications of worker’s labor mobility on employer market power. I calibrate a simple version of the model and find that relative to an economy with no mo- bility, there is a moderate increase in average wages and a fall in unemployment. Changes in firm rents due to workers’ stronger bargaining position account for about half of the increase in average wages. In Chapter 3, I present coauthored work studying methods for the measurement of inflation using large, micro-level retail sales data sources. In particular, this chapter explores alter- native methods for adjusting price indices for quality change at scale. These methods can be applied to large-scale item-level transactions data that includes information on prices, quantities, and item attributes. The hedonic methods can take into account the changing valuations of both observable and unobservable characteristics in the presence of prod- uct turnover. This chapter also considers demand-based approaches that take into account changing product quality from product turnover and changing appeal of continuing prod- ucts. This chapter provides evidence of substantial quality-adjustment in prices for a wide range of goods, including both high-tech consumer products and food products.

Background: Several studies have evaluated whether the distribution of natural environments differs between marginalized and privileged neighborhoods. However, most studies restricted their analyses to a single or handful of cities and used different natural environment measures. Objectives: We evaluated whether natural environments are inequitably distributed based on socioeconomic status (SES) and race/ethnicity in the contiguous United States. Methods: We obtained SES and race/ethnicity data (2015–2019) for all U.S. Census tracts. For each tract, we calculated the Normalized Different Vegetation Index (NDVI) for 2020, NatureScore (a proprietary measure of the quantity and quality of natural elements) for 2019, park cover for 2020, and blue space for 1984–2018. We used generalized additive models with adjustment for potential confounders and spatial autocorrelation to evaluate associations of SES and race/ethnicity with NDVI, NatureScore, park cover, and odds of containing blue space in all tracts (lowercase italic n equals 71,532n=71,532 ) and in urban tracts (lowercase italic n equals 45,338n=45,338 ). To compare effect estimates, we standardized NDVI, NatureScore, and park cover so that beta coefficients presented a percentage increase or decrease of the standard deviation (SD). Results: Tracts with higher SES had higher NDVI, NatureScore, park cover, and odds of containing blue space. For example, urban tracts in the highest median household income quintile had higher NDVI [44.8% of the SD (95% CI: 42.8, 46.8)] and park cover [16.2% of the SD (95% CI: 13.5, 19.0)] compared with urban tracts in the lowest median household income quintile. Across all tracts, a lower percentage of non-Hispanic White individuals and a higher percentage of Hispanic individuals were associated with lower NDVI and NatureScore. In urban tracts, we observed weak positive associations between percentage non-Hispanic Black and NDVI, NatureScore, and park cover; we did not find any clear associations for percentage Hispanics. Discussion: Multiple facets of the natural environment are inequitably distributed in the contiguous United States.

The COVID-19 pandemic has presented a social dilemma; “social distancing” was required to stop the spread of disease, but close social contacts were needed more than ever to collectively overcome the unprecedented challenges of the crisis. How did Americans mobilize their social ties in response to the pandemic? Drawing from a nation-wide daily online survey of 36,345 Americans from April 2020 through April 2021, we examine the characteristics of Americans’ core networks within which people discuss “important matters.” Comparing the COVID-19 networks to those previously collected in eight national core network surveys from 1985 to 2016, we observe remarkable stability in the size and relationship composition of core networks during COVID-19. In contrast to the robust nature of core networks, we discover a significant rise in racial homophily among kin ties, and political homophily among non-kin ties. Simultaneously, our study reveals a significant surge in the adoption of remote communication technology to connect with individuals who are geographically distant. We demonstrate that the changing mode of communication contributes to increases in racial and political homophily. These results suggest that the COVID-19 pandemic may bring people remotely together but only with the like-minded, deepening social divides in American society.

Rapid changes in the U.S. economy have made it increasingly important to be able to produce estimates of poverty on a timely and frequent basis. Despite the demand for current and frequent statistics, there is a lag between the reference period and annual publication of poverty statistics. This paper builds on existing studies combining the basic monthly Current Population Survey with the Annual Social and Economic Supplement (CPS ASEC) to create a subannual measure of poverty with reference periods of 1, 3, and 4 months. I present subannual estimates of the Official Poverty Measure (OPM) and the Supplemental Poverty Measure (SPM) for 2009-2022. I also examine various methodological issues around the design of a subannual poverty measure. I also present corroborating results from the Survey of Income and Program Participation (SIPP) and the Household Pulse Survey (HPS). I suggest that a monthly poverty measure, to supplement annual statistics on poverty, may be appropriate for publication as a research series by the U.S. Census Bureau.

This paper estimates experimental impacts of a supported work program on employment, earnings, benefit receipt, and other outcomes. Case managers addressed employment barriers and provided targeted financial assistance while participants were eligible for 30 weeks of subsidized employment. Program access increased employment rates by 21 percent and earnings by 30 percent while participants were receiving services. Though gains attenuated after services stopped, treatment group members experienced lasting improvements in employment stability, job quality, and well-being, and we estimate the program’s marginal value of public funds to be 0.64. Post-program impacts are entirely concentrated among participants whose subsidized job was followed by unsubsidized employment with their hostsite employer. This decomposition result suggests that encouraging employer learning about potential match quality is the key mechanism underlying the program’s impact, and additional descriptive evidence supports this interpretation. Machine learning methods reveal little treatment effect heterogeneity in a broad sample of job seekers using a rich set of baseline characteristics from a detailed application survey. We conclude that subsidized employment programs with a focus on creating permanent job matches can be beneficial to a wide variety of unemployed workers in the low-wage labor market.

Service offshoring raises the fear of job loss for high-skilled workers, unlike goods offshoring, because workers at home compete with highly educated workers in low-income countries. This paper examines whether the increase in the United States's service offshoring to India has reduced the domestic employment of the occupations with greater exposure to Indian service imports. To account for endogeneity, I instrument for the growth of the United States's service imports from India by exploiting the change in Indian exports to European countries. Service offshoring reduces total employment from 2000 to 2006; however, this effect disappears overall and becomes positive for college-educated workers in the later period from 2006 to 2016. Unlike goods offshoring, the employment effect is largely driven by college-educated workers, and the employment growth in the later period is larger.

In the early twentieth century, the mining industry was characterized by isolation, dangerous working conditions, employer power, and declining employment. However, miners also enjoyed high earnings, flexible schedules, and company housing. In this article, I explore intergenerational economic mobility for miners’ sons. Using linked full-count US Census data to explore outcomes for miners’ sons compared to other sons, I find that miners’ sons usually do worse than manufacturing workers’ sons but better than farmers’ sons. Successful sons of miners grew up in urban neighborhoods that were mining-dependent, had access to education, and moved from their childhood counties. Sons of miners in the coal industry, which was shrinking, also did worse than sons of miners in the oil industry, which was expanding. This article sheds light on the effect that industry growth and geographic isolation has on intergenerational outcomes.

Every year in AUgust and September, over 13,000 K-12 schools across the United States open heir doors to start their semesters, welcoming over 55 million children. In 2020, most schools did not physically open. They went fully online due to the conintuing health risks of COVID-19 and the lack of planning and resources devoted to ensuring the possibility for safe in-person return. The decision to move to online education was controversial and contested, reviving well-worn tropes, pitting teachers against children and the economy. The federal government, for example, argued that closed schools "could damage our children's education for years to come and hinder our nation's economic comeback." While business leaders and newspaper editorials described teachers as "incredibly selfish, putting their fears, largely unfounded, ahead of the needs of their students, their communities and their country." Others expressed well-founded concerns about how shuttered schools and online only instruction would deepen existing inequalities and put additional stress on families.

Millennials are buying homes at a lower rate than generations before, according to an analysis by Investopedia. Substantially fewer of those born between 1981 and 1996 are homeowners today than Gen X and baby boomers were at the same age. Housing affordability is taking a toll on all generations, but the lack of entry-level homes and the dearth of new builds are particularly impacting millennials. According to data from the Census Bureau’s Survey of Construction, the proportion of new homes built that cost under $200,000 dropped from 10% of new home stock in 2019 to less than 1% in 2022.

Since 2015, several U.S. states have begun granting professional or occupational licenses to recipients of the Deferred Action for Childhood Arrivals (DACA) program. In this study, we examine the differences in the issuance of such licenses at the state level to estimate their effect on the labor market outcomes of DACA recipients. Using data from the Current Population Survey spanning the period from 2012 to 2020, we find that granting access to occupational licensing has had a significant positive impact on the wages of DACA recipients, raising them by 12.5 percent. We also observe that this wage premium is greater for male recipients than for female ones. Furthermore, we find that the wage premium and increased hours of work resulting from access to licenses are most pronounced for recipients over 24 years of age who have completed their formal education and have already established their careers. These findings suggest that access to occupational licenses can improve the labor market outcomes of Dreamers. Therefore, expanding access to public benefits, such as licensing, can substantially benefit DACA recipients in states where such access has been broadened.

Despite the rapid growth of the Bangladeshi diaspora in the USA, knowledge about this new diasporic community remains very limited. This study argues and demonstrates that the Bangladeshi diaspora in the USA is a fast-growing and sizable diasporic community that requires systematic research and better understanding. It delineates the history of the Bangladeshi diaspora to the USA in four periods and documents the phenomenal growth of the Bangladeshi diasporic community in the USA since 1981, using data from the US Department of Homeland Security (DHS). By taking into account the legal Bangladeshi immigration as well as the emigration and mortality rates of immigrants and undocumented Bangladeshi immigration, it estimates the current size of the Bangladeshi diasporic community in the USA at about 500,000 instead of a range of low-to-mid 200,000s normally cited. Additionally, using the pooled samples of the 2001–2019 American Community Surveys (ACS) and other ACS data, as well as the DHS data, this paper provides a demographic and socioeconomic portrait of the Bangladeshi diasporic community in the USA. The findings are generalizable to the population and fill some important gaps in the literature.

Efforts to document long-term trends in socioeconomic mobility in the United States have been hindered by the lack of large, representative datasets that include information linking parents to their adult children. This problem has been especially acute for women, who are more difficult to link because their surnames often change between childhood and adulthood. In this paper, we use a new dataset, the Census Tree, that overcomes these issues by building on information from an online genealogy platform. Users of the platform have private information that allows them to create links among the 1850 to 1940 decennial censuses; the Census Tree combines these links with others obtained using machine learning and traditional linking methods to produce a dataset with hundreds of millions of census-to-census links, nearly half of which are for women. With these data, we produce estimates of the intergenerational transmission of socioeconomic status from fathers to their sons and daughters. We find that for married men and women, the patterns of mobility over this period are remarkably similar. Single women, however, are less mobile than their male counterparts. We also present new estimates that show that assortative mating was much stronger than previously estimated for the US.

This study examines the creation of Black communities in the context of the Exoduster movement, the first major migration of African Americans out of the southern and border states. We focus initially on Nicodemus, Kansas, a site with well-preserved archival information, and then turn to census microdata on roughly three-hundred African-American communities that emerged in Kansas, Colorado, and Oklahoma between 1880 and 1920. Analysis of these communities informs a general model of minority-majority group relations that is differentiated along two dimensions: spatial segregation and the distinctiveness of business activities. Under conditions of prejudice from the majority ethno-racial group, the model predicts that rates of minority business proprietorship and community growth will increase with segregation and distinctiveness, but that the joint occurrence of these conditions presents an existential threat. We draw conclusions for the trajectories of several well-known Black communities, including Nicodemus, Tulsa, and Langston, Oklahoma.

We argue that the Democratic Party's evolution on economic policy helps explain partisan realignment by education. We show that less-educated Americans differentially demand "predistribution" policies (e.g., a federal jobs guarantee, higher minimum wages, protectionism, and stronger unions), while more-educated Americans differentially favor redistribution (taxes and transfers). This educational gradient in policy preferences has been largely unchanged since the 1940s. We then show the Democrats' supply of predistribution has declined since the 1970s. We tie this decline to the rise of a self-described "New Democrat" party faction who court more educated voters and are explicitly skeptical of predistribution. Consistent with this faction's growing influence, we document the significant growth of donations from highly educated donors, especially from out-of-district donors, who play an increasingly important role in Democratic (especially "New Democrat") primary campaigns relative to Republican primaries. In response to these within-party changes in power, less-educated Americans began to leave the Democratic Party in the 1970s, after decades of serving as the party's base. Roughly half of the total shift can be explained by their changing views of the parties' economic policies. We also show that in the crucial transition period of the 1970s and 1980s, New Democrat-aligned candidates draw disproportionately from more-educated voters in both survey questions and actual Congressional elections.

This study aims to understand the health outcomes of parents with children with autism spectrum disorder (ASD) and the interactive effect of child health insurance status. The study utilized 2014-2018 pooled National Health Interview Survey data to construct weighted national estimates and assess main and interaction effect logistic regression models. Findings show parents of children with ASD experienced significantly poorer health compared to parents of children without autism. Insurance status was found to significantly interact with child ASD status. Compared to parents of children without ASD who used private insurance, parents who had a child with ASD who used private insurance, public insurance, or were uninsured were found to have 1.5-, 3.2-, and 2.1-times higher odds of poorer health, respectively. Future research and implications on policy and practice are discussed.