In this paper we consider the problem of minimizing composite objective functions consisting of a convex differentiable loss function plus a non-smooth regularization term, such as $L_1$ norm or nuclear norm, under R\'enyi differential privacy (RDP). To solve the problem, we propose two stochastic alternating direction method of multipliers (ADMM) algorithms: ssADMM based on gradient perturbation and mpADMM based on output perturbation. Both algorithms decompose the original problem into sub-problems that have closed-form solutions. The first algorithm, ssADMM, applies the recent privacy amplification result for RDP to reduce the amount of noise to add. The second algorithm, mpADMM, numerically computes the sensitivity of ADMM variable updates and releases the updated parameter vector at the end of each epoch. We compare the performance of our algorithms with several baseline algorithms on both real and simulated datasets. Experimental results show that, in high privacy regimes (small $\epsilon$), ssADMM and mpADMM outperform other baseline algorithms in terms of classification and feature selection performance, respectively.

What determines the joint dynamics of aggregate employment and wages over the medium run? This classic question in macroeconomics has received renewed attention since the Great Recession, when real wages did not fall despite a crash in employment. This paper proposes a microfoundation for the medium-run dynamics of aggregate labor markets which relies on worker heterogeneity. I develop a model in which workers differ in their skills for various occupations, sectors employ occupations with different weights in production, and skills are imperfectly transferable. When shocks are concentrated in particular industries, the extent to which workers can reallocate across the economy determines aggregate labor market dynamics. I apply the model to study the recessions of 2008-09 and 1990-91. I estimate the distribution of worker skills using two-period panel data prior to each of these recessions and find that skills became less transferable between the 1980s and 2000s. Shocking the estimated model with industry-level TFP series replicates the increase in aggregate wages in 2008-09, and decline in 1990-91. The model implies that if either the composition of industry shocks or the distribution of skills in the economy had been the same in the 2008-09 recession as in the 1990-91 recession, real wages would have fallen, while employment would have declined less. The declining industries during 2008-09 all employed a similar mix of skills, which induced many low-skill workers to leave the labor force and limited downward wage pressure on the rest of the economy. Finally, the model inspires a novel reduced form method to correct aggregate wages for selection in the human capital of workers, which accounts for cyclical job downgrading by focusing on the wage movements of occupation-stayers. This correction recovers pro-cyclical wages, suggesting the changing composition of the workforce was crucial for aggregate wage dynamics during the Great Recession.

This paper introduces a new innovation data source to re-examine how spatial distance affects the dif- fusion of ideas and innovations in an economy. We exploit the descriptions of products and services contained in U.S. trademark registrations during 1980-2012 to identify terms (tokens) not previously used by firms to describe products and services. From these we select tokens frequently re-used by follower firms. By linking the new tokens to the business addresses of innovator and follower firms, our data encompass all instances in which innovations captured by trademark tokens arise within and diffuse across the United States. We aggregate innovations at the year and ZIP code level and estimate Poisson models of the likelihood and intensity of diffusion between locations. After endogenising the creation of new diffusion links between ZIP codes, our results show that spatial distance no longer affects the creation of diffusion links within the US after 1996. However, contingent on previous diffusion from a sending to a receiving ZIP code, we find persistent, strong and negative effects of greater spatial distance on the intensity (extent) of diffusion for existing transfer links between locations within the US.

This research illuminates the historical development of creative activity in the United States. Census data is used to identify creative occupations (i.e., artists, musicians, authors, actors) and data on prominent creatives, as listed in a comprehensive biographical compendium. The analysis first sheds light on the socioeconomic background of creative people and how it has changed since 1850. The results indicate that the proportion of female creatives is relatively high, time constraints can be a hindrance for taking up a creative occupation, racial inequality is present and tends to change only slowly, and education plays a significant role for taking up a creative occupation. Second, the study systematically documents and quantifies the geography of creative clusters in the United States and explains how these have evolved over time and across creative domains. Third, it investigates the importance of outstanding talent in a discipline for the local growth of an artistic cluster.

The primary objective of “all-payer” rate setting—regulatory regimes through which states set hospital payment rates for all insurers—was to control costs through consistent, centrally regulated payments. These regimes were often linked, however, to an ancillary goal of financing care for the uninsured. We show that the surcharge mechanism used to achieve this secondary objective decreased the stability of these payment regimes. This instability reflected a feedback loop from surcharge rates to insurance coverage and back to the quantities of uncompensated care in need of financing. Instability was exacerbated when Health Maintenance Organizations were exempted from surcharge collections, creating a regulatory arbitrage opportunity. Legal challenges connected to the incidence of uncompensated care surcharges contributed to the abandonment of all-payer rate regulation by several states. These developments illustrate the wisdom of the Tinbergen Rule, which recommends that independent policy objectives be met with independent policy instruments.

How does the appearance of a new immigrant group affect the integration of earlier generations of migrants? We study this question in the context of the first Great Migration (1915-1930), when 1.5 million African Americans moved from the US South to northern urban centers, where 30 million Europeans had arrived since 1850. We exploit plausibly exogenous variation induced by the interaction between 1900 settlements of southern-born blacks in northern cities and state-level outmigration from the US South after 1910. Black arrivals increased both the effort exerted by immigrants to assimilate and their eventual Americanization. These average effects mask substantial heterogeneity: while initially less integrated groups (i.e. Southern and Eastern Europeans) exerted more assimilation effort, assimilation success was larger for those culturally closer to native whites (i.e. Western and Northern Europeans). Labor market outcomes do not display similar heterogeneity, suggesting that these patterns cannot be entirely explained by economic forces. Our findings are instead more consistent with a framework in which changing perceptions of outgroup distance among native whites lowered the barriers to the assimilation of white immigrants.

This chartbook illustrates the characteristics of the county’s 65-plus population in 2015*, and changes that have occurred since 2000. It also includes population characteristics, such as education, income level, and marital status, that are shown to be associated with the need for long-term services and supports. There are charts that compare the older population of the county to the state as a whole, and charts that illustrate change over time within the county. The data presented in this chartbook are intended to assist planners, decisions makers, and service providers to understand the growth in numbers and proportion of older adults, particularly those who will likely need assistance. An online interactive data center is available for you to define your own topic, county, and population of interest to see current figures and change over time. Please visit www.ohio-population.org.

Scholarship has long recognized the centrality of white racial sexual fears in the rhetoric and practice surrounding the lynching of African Americans in the U.S. South in the decades around 1900. The topic has not previously been taken up for systematic study beyond event-level analyses. This article presents theoretical and empirical evidence that whites’ intersecting racial and gender concerns converging in racial sexual fears were conducive to lynching related to interracial sex, but not to those unrelated to interracial sex, under certain conditions. The empirical findings, based on lynchings in 11 southern states from 1881– 1930, demonstrate that lynchings related to interracial sex were more likely to occur in contexts characterized by higher levels of white female dependents residing with white male householders, higher levels of white female school attendance, and higher levels of adult black male literacy. These findings suggest that interracial sex-related lynching served to recover and retain white men’s racial and gender status, which postbellum developments had undermined, by oppressing not only African American men and women but disempowering white women as well. White racial sexual fears during the lynching era should, therefore, be seen as constituting a social force in their own right with long-term consequences for race and gender relations and inequalities.

“Superstar effects” generate large compensation differentials among similarly talented individuals. Are superstar effects amplified by technological innovations that extend the scale over which talent is deployed? I test this idea in the market for entertainers, using the roll-out of television as a natural experiment which provides clean variation in a scale-related technological change. The launch of a local TV station increases top entertainers’ incomes, resulting in a twofold increase in top-percentile income share, while reducing employment and incomes of lower-level talents. These results show clear evidence of superstar effects and are inconsistent with canonical models of skill-biased technological change.

Electricity generation is a large contributor to fine particulate matter (PM2.5) air pollution. However, the demographic distribution of the resulting exposure is largely unknown. We estimate exposures to and health impacts of PM2.5 from electricity generation in the US, for each of the seven Regional Transmission Organizations (RTOs), for each US state, by income and by race. We find that average exposures are the highest for blacks, followed by non-Latino whites. Exposures for remaining groups (e.g., Asians, Native Americans, Latinos) are somewhat lower. Disparities by race/ethnicity are observed for each income category, indicating that the racial/ethnic differences hold even after accounting for differences in income. Levels of disparity differ by state and RTO. Exposures are higher for lower-income than for higher-income, but disparities are larger by race than by income. Geographically, we observe large differences between where electricity is generated and where people experience the resulting PM2.5 health consequences; some states are net exporters of health impacts, other are net importers. For 36 US states, most of the health impacts are attributable to emissions in other states. Most of the total impacts are attributable to coal rather than other fuels.

The costs and benefits of environmental policy depend crucially on the assumed micro-elasticities between market and non-market goods. In their absence, general equilibrium models have assumed environmental amenities are perfect substitutes with market goods, such as consumption and leisure. The first part of the paper compiles the most comprehensive micro-data to date between 1980 and 2014 containing individual and county outcomes from the Census Bureau and Environmental Protection Agency. Using the Consumption Expenditure Survey, I create individual-level predictions of consumption expenditures in the census micro-data. With the newly created data, I document a strong negative association between ozone pollution and both consumption and leisure at the county-level. The second part of the paper develops a partial equilibrium sorting model with non-separable preferences between each of the market goods (consumption, leisure, and housing) and pollution. The third part of the paper uses the equilibrium conditions implied by the model to estimate the elasticities between each of the market and non-market goods. The fourth part of the paper uses the estimated elasticities to revisit the welfare gains of the Clean Air Act Amendments (CAAA) in the presence of non-separabilities. Preliminary results indicate that initial estimates of the welfare gains are drastically overstated.

Rural Maine was in a bad way. Young people were leaving in droves, headed to the cities, or to other states to seek their fortune. Farms sat empty, while the farmers that remained couldn’t find enough help. It’s a story that has resonance today, when economists are warning that Maine, especially its rural areas, are seeing economic stagnation and a lack of population growth. In 1908, the Maine Bureau of Labor Statistics summed up the problem like this: The great need of the State is for young blood, people who will increase the population by rearing families that will have a love and desire for agricultural pursuits. This is the kind of people that will solve the question of abandoned farms and decrease in population

With more than 30 million people moving to North America during the Age of Mass Migration (1850–1913), governments feared that Europe was losing its most talented workers. Using new data from Ireland in the early twentieth century, I provide evidence to the contrary, showing that the sons of farmers and illiterate men were more likely to emigrate than their literate and skilled counterparts. Emigration rates were highest in poorer farming communities with stronger migrant networks. I constructed these data using new name-based techniques to follow people over time and to measure chain migration from origin communities to the United States.

According to the 2017 American Community Survey (ACS), there are nearly 1.8 million occupied rental units in Los Angeles county. Within that huge rental supply, more than one-third (36.2%) are occupied by renters with very low-income (defined as total household income that is less than half of the county median household income of $65,000 in 2017). This very low-income group of renters is the principal focus of this Brief. These low-income renters have great difficulty finding a rental housing unit they can afford without public assistance. Yet, even though the very low-income segment is the major beneficiary group targeted by public programs for rental assistance, less than half receive any public assistance. A key policy question is why is it so hard for low-income renters to find affordable housing? What hurdles impede their success in housing? Ultimately, what . . .

Over the last decades, regions in the United States have been diverging. More skill-intensive areas have experienced a higher wage and skill premium growth at the same time that they became even more skill-intensive. This process deepened inequality both between and within urban areas, concentrating educated and high-earning workers into few cities. In this paper, I show that the substantial decline of manufacturing industries following the sharp rise of Chinese exports, and how local economies adapted to the loss of employment in those sectors, contributed significantly to the divergence among metropolitan areas in the US. Nonetheless, differences in local outcomes are not only the consequence of variations in industrial composition or exposure to foreign competition. Instead, I show in this paper that the consequences of international competition on local labor markets are highly heterogeneous. Even conditional on having manufacturing sectors with similar size and characteristics, the sign and magnitude of the effects of rising import competition depend critically on the characteristics of the rest of the local economy. In particular, I focus on how the share of local workforce with college education shapes the reaction to adverse shocks. Among more skill-intensive regions, greater exposure to import competition makes cities to attract college-educated workers and to increase college wages and skill premium. On the other hand, among less educated regions, foreign competition has negative effects in terms of college-educated workforce and wages. This result highlights that the contribution of trade to regional divergence critically depends on the ability of cities to adapt to adverse shocks.

In this brief, we use American Community Survey data from 2010 to 2017 to examine trends in uninsurance among non-Hispanic white, non-Hispanic black, and Hispanic women who reported giving birth in the past 12 months. We find the following: ? Uninsurance among new mothers fell between 2010 and 2017 for all three groups: by 47 percent for white new mothers, 41 percent for black new mothers, and 39 percent for Hispanic new mothers. ? The Hispanic-white uninsurance disparity among new mothers decreased by 26 percent between 2013 and 2017, and the black-white disparity decreased by 37 percent between 2013 and 2016. ? Despite coverage gains, nearly half a million new mothers remained uninsured in the United States in 2017. ? Coverage disparities remained in 2017: uninsurance rates were 24.4 percent for Hispanic new mothers, 12.1 percent for black new mothers, and 7.0 percent for white new mothers.

This chartbook illustrates the characteristics of the county’s 65-plus population in 2015*, and changes that have occurred since 2000. It also includes population characteristics, such as education, income level, and marital status, that are shown to be associated with the need for long-term services and supports. There are charts that compare the older population of the county to the state as a whole, and charts that illustrate change over time within the county. The data presented in this chartbook are intended to assist planners, decisions makers, and service providers to understand the growth in numbers and proportion of older adults, particularly those who will likely need assistance. An online interactive data center is available for you to define your own topic, county, and population of interest to see current figures and change over time. Please visit www.ohio-population.org.

Low birthweight (LBW) is associated with myriad health and developmental problems in childhood and later in life. Less well-documented is the variation in the relationship between LBW status and subsequent child health by socioeconomic status—such as education levels and income. This paper examines whether differences exist in the relationship between LBW and subsequent child health by maternal education.

Drug abuse costs the US more than $740 billion annually, with more than 70,000 overdose deaths every year (National Institute of Drug Abuse [NIDA], 2017; NIDA, 2019). Despite the high addiction disease burden, we have little causal evidence on factors that lead to illicit drug use. In particular, while many papers have studied the impact of economic conditions on legal substances like alcohol and tobacco use (see Pacula, 2011, and Henkel, 2011, for reviews), few have studied illicit drugs, and only one study (Carpenter, McClellan, & Rees, 2017) has systematically investigated all categories of illicit drugs....

This report documents how metropolitan planning organizations (MPO), State departments of transportation (DOT), and other transportation agencies are adapting Environmental Justice (EJ) analysis techniques to understand transportation impacts in communities undergoing rapid demographic change. The report also highlights national demographic trends, showing significant change in the size and location of low-income and minority households. These changes are documented in a series of State- and county-level maps demonstrating absolute and percent change in minority populations, low-income populations and limited English proficiency populations between the 1990 – 2010 decennial census periods. Based on the state of practice review, the report also highlights strategies for addressing changing demographics in EJ analysis and provides five case studies that document best practices from State DOTs and MPOs in areas undergoing rapid demographic change. Practices highlighted in this report offer practitioners improved methods for understanding and responding to demographic change in EJ communities across the lifecycle of a transportation action.