ABSTRACT. Unemployment insurance (UI) was a key aspect of the U.S. government’s fiscal response to the COVID-19 pandemic.  The sharp rise in aggregate UI benefits was due in large part to temporary UI programs created by the CARES Act at the onset of the pandemic.  These programs (1) expanded UI to workers not covered by the regular UI program, including the self-employed, independent contractors, and part-time workers; (2) supplemented weekly benefits by $600 per week; and (3) extended the duration of benefits by 13 weeks.  In the federal-state UI system, UI policy varies considerably across states along multiple dimensions, including weekly benefit amounts and the maximum duration of benefits.  In this paper, I focus on the success rate of UI applicants—the share of UI applicants who received benefits.  The success rate captures the effect of the eligibility expansions and is a key factor in the UI recipiency rate (the share of the unemployed who received UI benefits).  The paper investigates two questions.  First, what was the impact of the eligibility expansions under the Pandemic Unemployment Assistance program (PUA) on success rates, and how did the impact vary by demographic group?  Second, what was the impact of the pandemic programs on state variation in success rates?

The paper uses individual-level data from the Household Pulse Survey (HPS), an experimental rapid-response survey that measured how the pandemic affected U.S. households from a social and economic perspective.  The HPS, an online survey that the Census Bureau conducted regularly throughout the pandemic, asked respondents whether they had applied for UI benefits and whether they had received benefits since a reference date that was several months before the survey date.  The HPS data used in the paper cover UI application and receipt from the beginning of the pandemic in March 2020 until August 2022.  This period includes the staggered exit of states from PUA in which 20 states terminated PUA in June or July 2021 and the remaining states ended PUA at the federal expiration in September 2021.  To complement the HPS, I use data from a supplement to the Current Population Survey in 2018 and 2022 that asked respondents whether they had applied for and received UI benefits since their last job.

Exploiting the staggered exit from PUA, I find that the eligibility expansions under PUA increased the success rate of UI applicants.  The estimated effects of PUA are larger for demographic groups with lower success rates before the pandemic: females, those with some college, young workers, and those who had never married.  This evidence is consistent with PUA expanding access to UI.  I also find that PUA reduced variation across states in success rates.  The variation in success rates fell with the introduction of the temporary UI programs early in the pandemic and increased when PUA ended.  Potential reasons for this result include an increased federal role in the UI system during the pandemic and more standardization across states.

ABSTRACT. The United States experienced an unprecedented increase in unemployment insurance (UI) claims starting in March 2020. State UI-benefit systems were inadequately prepared to process these claims. In states that used an antiquated programming language, COBOL, to process claims, potential claimants experienced a larger increase in administrative difficulties, which led to longer delays in benefit disbursement and may have led to an increase in discouraged filers. Using daily debit and credit card consumption data from Affinity Solutions, I employ a two-way fixed-effects estimator to measure the causal impact of having an antiquated UI benefit system on aggregate consumption. Such systems led to a 2.8-percentage-point decline in total credit and debit card consumption relative to card consumption in states with more modern systems. I identify two factors that likely contributed to this outcome: delays in benefit disbursement and an increase in discouraged filers. I estimate that the share of claims whose processing was delayed by over 70 days rose by 1.4 percentage points more in COBOL states relative to non-COBOL states. I also find suggestive evidence that the increase in administrative burden for claimants in COBOL states led to an additional 4.5 million discouraged filers relative to the number that would have been expected had they faced the same increase in administrative burden as claimants in non-COBOL states. Based on a back-of-the-envelope calculation using 2019 data, my results suggest that the decline in consumption in COBOL states in 2020 after the pandemic-emergency declaration corresponds to a real-GDP decline of at least $90 billion (in 2012 dollars).

ABSTRACT. As a response to COVID-19 shutdowns starting in 2020, the US federal government dramatically increased weekly unemployment insurance (UI) payments, expanded the eligibility of UI recipients, and extended the amount of time recipients could receive UI payments. These programs were ultimately extended to September 2021 by the American Rescue Plan Act. In June and July of 2021, 26 individual states at least partly opted out of these programs before they were set to expire, with 21 states fully opting out. This acts as a natural experiment in reducing UI compensation and restricting UI eligibility, which together we call “restricting UI” along intensive and extensive margins, respectively. We leverage this withdrawal to analyze the effect of restricting UI on general violent and property crime and, separately, intimate partner violence. We use a difference-in-differences framework comparing states which restricted UI to states which never or had not yet restricted UI. In our preliminary analysis using calls for service data in major cities, we find that restricting UI decreased calls for intimate partner violence. We will expand our analysis first by using a broad, detailed national crime data set, and then by attempting to disentangle effects of restricting UI along intensive and extensive margins.

ABSTRACT. Changes in pay and benefits alone incorrectly predict self assessed changes in overall job quality 30 percent of the time according to survey evidence from job-changers. Job-changers also place more emphasis on their interest in their work than they do on pay and benefits in evaluating whether their new job is better. Parents particularly emphasize work-life balance, and we find some indications that mothers value it more than fathers. Improvements in pay are highly correlated with improvements in other amenities for workers with less education but not for workers with a bachelor's degree or more. The higher positive correlation implies that differences in pay and benefits understate differences in total job quality to a greater degree among workers with less education.

ABSTRACT. Just cause employment policies have long been used to discourage the arbitrary firing of workers. Recent efforts at passing such laws have been motivated by deterring discrimination; such motivation was prominent in the recent passage of New York City's just cause law for fast food workers. However, the economics literature on employment protection legislation has largely ignored this rationale. This paper presents a framework to study the effects of just cause in an environment with taste-based discrimination. We allow for the possibility that JC regulation can enhance fairness by stopping managers from indulging their personal biases at the expense of shareholders and disfavored worker. This framework generates predictions that we test on New York City's legislation. Using a synthetic difference-in-differences design on publicly available data, we find results consistent with taste-based discrimination against older workers, but not for such discrimination against women or racial minorities. Further analysis suggests other mechanisms of discrimination may also be at play. In particular, our findings are consistent with screening discrimination - when managers tend to hire or keep workers from groups they can better judge - against black workers. These results suggest that managers have relatively less confidence in the productivity of black workers on average.

A draft of this paper can be found at the following link: https://drive.google.com/file/d/1R6fK2B4eeuVAYYwvH6kksLeJ2IUh5v1X/view?usp=sharing

ABSTRACT. Remote wage employment gradually increased in the United States during the four decades prior to the pandemic, then surged in 2020 due to social distancing policies implemented to stem the spread of COVID-19. Using the 2010–2021 American Community Survey, the authors examine trends in wage and hours differentials for full-time remote workers and office-based workers as well as within occupation differences in wage growth by work location. Throughout the period, remote workers earned higher wages than those working on-site, and the difference increased sharply during the pandemic. Real wages grew 4 percent faster for remote workers than on-site workers within detailed occupation groups and increases in remote work intensity were positively associated with wage growth across occupations. Before the pandemic, remote workers worked substantially longer hours per week than on-site workers, but by 2021, their hours were similar.

ABSTRACT. panel

ABSTRACT. The Standard Occupational Classification (SOC) system is a federal statistical standard used by federal agencies to classify workers into occupational categories for the purpose of collecting, calculating, or disseminating data.

To reflect changes in the economy and the nature of work, the SOC system is revised periodically, with the interagency SOC Policy Committee (SOCPC) making recommendations to the Office of Management and Budget (OMB). OMB requires the use of the SOC when publishing Federal statistics about occupations, makes the final decisions about the SOC, publishes the SOC Manual, and charters the SOCPC.

The SOCPC is in the process of drafting a federal register notice announcing an upcoming revision and OMB is likely to publish the initial Federal Register notice soliciting public comment in Spring 2024.

We think your conference would be a great venue for our outreach efforts in letting the government economist community know about the upcoming revision, its importance, and implications. We are flexible in where you think this topic fits best, whether that be on the new and revised data products panel, or a poster session.

ABSTRACT. Increasing income and wealth inequality has characterized the U.S. economy for several decades. While researchers differ on the extent of the increase and its causes, they no longer disagree on the essential phenomenon. Yet the nation's disparate federal statistics make it difficult to accurately measure income and wealth inequality and other aspects of economic well-being for the nation's households and families. The Committee on National Statistics (CNSTAT) of the National Academies of Sciences, Engineering, and Medicine appointed a panel to review the major income, consumption, and wealth statistics currently produced by U.S. statistical agencies and provide guidance for modernizing and integrating the information to better inform policy and research.

This paper will discuss the forthcoming report from the Panel on An Integrated System of U.S. Household Income, Consumption and Wealth Data and Statistics to Inform Research and Policy. This report builds on the current effort at CNSTAT to create a 21st Century Data Infrastructure, as described in the report, Toward a 21st Century National Data Infrastructure: Mobilizing Information for the Common Good. The panel’s report, as stated in the Statement of Work, will (1) review the major income, consumption, and wealth statistics currently produced by U.S. statistical agencies, and (2) provide guidance for modernizing the information to better inform policy and research (such as understanding trends in inequality and mobility). The panel’s report will evaluate the need for and value of a fully integrated system of income, consumption, and wealth statistics to provide consistent macro-level statistics (e.g., total household or family income) and micro-level statistics (e.g., income for households in each quintile of the distribution), and include recommendations regarding the relevance, accuracy, timeliness, geographic and population detail, and consistency of statistics on income, consumption and wealth, and the need for an integrated system of these statistics. As part of its review, the panel’s report will also discuss a range of issues, including the following: • appropriate definitions of household, family, and individual income, consumption, and wealth; variations in definitions that would be useful for particular purposes (e.g., market income before government taxes and transfers); inclusion of particular components (e.g., capital gains and wealth transfers); different reference periods (e.g., sub-annual versus annual); and comparability with commonly used international concepts and methods; • the appropriate breakouts of income, consumption, and wealth statistics, population groups, the level of geographic granularity, frequency and timeliness of updated estimates; • the treatment of and method(s) used to value in-kind benefits and services (e.g., health insurance or transfers of food or housing); the treatment of retirement income and plan contributions; and • needed quality improvements for relevant data collection programs (including cross-sectional and longitudinal data sets), the potential for using multiple data sources, including surveys, state and federal administrative records, and commercial data, as well modeling, to produce integrated, highest-quality estimates.

ABSTRACT. The HBCU Environmental Justice Technical Team is a collaborative group of HBCU-affiliated economists, data scientists, geoscientists, and GIS specialists that visualize and analyze impacts of environmental policy for various stakeholders, including institutions, government agencies, and community organizations.Join us for an engaging and informative session examining economic and environmental policies of the Biden Administration using a socioeconomic analytics lens. We will explore the pivotal role of maps and other Geographic Information Systems (GIS) tools in shaping environmental policy, promoting justice, and addressing environmental disparities. We will delve into several key topics, including the Justice 40 Executive Order, the Climate and Economic Justice Screening Tool (CEJST), and the ubiquitous nature of GIS as it leverages itself to analyze economic and environmental disparities, and promote equitable policies.By the end of this session, you will gain a deeper understanding of the critical role that maps and GIS play in shaping the political landscape, fostering justice, and ensuring equitable access to resources and opportunities. This session will empower you with knowledge and insights to advocate for change and promote more inclusive and environmentally sustainable policies. You can use this opportunity to engage with field experts and participate in this essential conversation.

ABSTRACT. The HBCU Environmental Justice Technical Team is a collaborative group of HBCU-affiliated economists, data scientists, geoscientists, and GIS specialists that visualize and analyze impacts of environmental policy for various stakeholders, including institutions, government agencies, and community organizations.Join us for an engaging and informative session examining economic and environmental policies of the Biden Administration using a socioeconomic analytics lens. We will explore the pivotal role of maps and other Geographic Information Systems (GIS) tools in shaping environmental policy, promoting justice, and addressing environmental disparities. We will delve into several key topics, including the Justice 40 Executive Order, the Climate and Economic Justice Screening Tool (CEJST), and the ubiquitous nature of GIS as it leverages itself to analyze economic and environmental disparities, and promote equitable policies.By the end of this session, you will gain a deeper understanding of the critical role that maps and GIS play in shaping the political landscape, fostering justice, and ensuring equitable access to resources and opportunities. This session will empower you with knowledge and insights to advocate for change and promote more inclusive and environmentally sustainable policies. You can use this opportunity to engage with field experts and participate in this essential conversation.

ABSTRACT. The HBCU Environmental Justice Technical Team is a collaborative group of HBCU-affiliated economists, data scientists, geoscientists, and GIS specialists that visualize and analyze impacts of environmental policy for various stakeholders, including institutions, government agencies, and community organizations.Join us for an engaging and informative session examining economic and environmental policies of the Biden Administration using a socioeconomic analytics lens. We will explore the pivotal role of maps and other Geographic Information Systems (GIS) tools in shaping environmental policy, promoting justice, and addressing environmental disparities. We will delve into several key topics, including the Justice 40 Executive Order, the Climate and Economic Justice Screening Tool (CEJST), and the ubiquitous nature of GIS as it leverages itself to analyze economic and environmental disparities, and promote equitable policies.By the end of this session, you will gain a deeper understanding of the critical role that maps and GIS play in shaping the political landscape, fostering justice, and ensuring equitable access to resources and opportunities. This session will empower you with knowledge and insights to advocate for change and promote more inclusive and environmentally sustainable policies. You can use this opportunity to engage with field experts and participate in this essential conversation.

ABSTRACT. The Americans with Disabilities Act (ADA) considers people who ever had a physical or mental impairment that substantially limits major life activities, including major bodily functions, as having a disability. However, the standard set of six questions used to identify people with disabilities in federal surveys, such as the American Community Survey (ACS), focus on activity limitations, not bodily function limitations. This paper uses four common impairments substantially limiting major bodily functions (i.e., cancer, diabetes, depressive disorders, and asthma) to analyze the share of people with such an impairment, who meet the ADA disability definition, but are not identified as having a disability by the six questions. The findings show that of people aged 18-64 who ever had such an impairment, nearly half answered “no” to all six questions. Nevertheless, the findings suggest that the six questions have identified people with disabilities who were more vulnerable (i.e., not employed or poor health). This paper estimates that if major depressive disorders were added to the six questions, disability prevalence in the 2021 ACS would increase from 10.8% to between 12% and 13.4%. That prevalence would increase to between 15.8% and 21.5% if depressive disorders were added and between 22.1% and 34.8% if all four impairments were added. This paper recommends that the Census Bureau continue to measure disability using a definition following the ADA and include questions about impairments substantially limiting major bodily functions when revising the six questions in the ACS.

ABSTRACT. The Earned Income Tax Credit (EITC) is the largest needs-tested cash assistance program for low- and medium-income individuals in the U.S. and is heralded as one of the most effective anti-poverty programs in the country’s social safety net. Through its impact on an individual’s health and labor force attachment, EITC exposure has the potential to affect an individual’s incidence of work disability. Over 2 million Americans apply for Social Security Disability Insurance (DI) benefits annually, with nearly four in five DI beneficiaries older than 49 years. Using data from the Panel Study of Income Dynamics (PSID) that spans over 50 years, we adopt a life-course approach to estimate the policy effects of EITC exposure from birth to mid-adulthood, i.e., ages 0-49, on the probability of individuals experiencing work disability in the years leading up to standard retirement ages, i.e., ages 50-59. Within the PSID, work disability is richly characterized by self-reports of the duration and severity, with individuals reporting both chronic and severe work limitations expected to be at the highest risk of applying for DI benefits. Furthermore, we use early Medicare receipt to proxy for DI claims in our investigations of whether social policy effects extend to DI awards. Our results show that exposure to EITC during adulthood can significantly and substantially reduce the probability of acquiring a work disability and subsequently needing DI later in life. These findings imply that the EITC could play a pivotal role in influencing DI application and award trends. To contextualize our findings, we also provide back-of-the-envelope calculations to quantify the economic impacts of augmented EITC exposure on DI awards.

ABSTRACT. Structural barriers to equality of opportunities, including many discriminatory practices produce socio-economic inequalities in health, education, and labor market outcomes of individuals of various race and ethnic groups. Social scientists and public policy makers are interested in estimating such inequities and designing and evaluating public policies and programs that can reduce such gaps. Data on race and ethnicity for individuals is much needed to that end. The Social Security Administration (SSA) and other Agencies collect such data on the US population. But under the Privacy Act, many individuals do not report such information. Generally, there are patterns in the names of individuals within racial groups; often individuals within a racial group live closer to each other in geographical clusters; and often the race of an individual is associated with the individual’s country of origin or the birth country (which was the original basis of race and ethnicity categories anyway). It is possible to use legally observable individual characteristics such as first name, last name, sex, geolocation of residence, country of origin to estimate statistical models and train Natural Language Processing (NLP) Machine Learning models to impute missing race information. Accuracies of such imputations are hampered by the statistical problem that some of the minority racial populations are very small fractions of the majority population, creating unbalanced data biases in imputation of race. Another problem is the anglicizations of names of some individuals in some of the racial groups such Black and Chinese, which also lead to poor performance of the race imputation models.

In this paper I briefly survey the existing statistical and machine learning literature, focusing on the recent state-of-the-art NLP Machine Learning models of two popular network architectures --- LSTM and very recently introduced Transformer architecture in a paper, “Attention is all you need”, 2017. A few specialized Transformer models such as the BERT and GPT models involving tens of millions of parameters are trained on large corpuses of text data for the purpose of sentiment analysis, text summarization, text translation and text generation. A few researchers applied those trained models for the race imputations (e.g., raceBERT model), without producing significant improvements in the predictive performance. I build a Transformer model of the type introduced in the 2017 “Attention is all you need” paper, consisting of small number of parameters tailored to the race imputation problem. I train LSTM and transformer models using the SSA’s administrative datasets --- RECS, NUMIDENT and Geolocation data --- having data on individual characteristics mentioned above. I compare model performances of the two models over various sets of feature variables, character, and word level tokenization methods amenable for generalization to out-of-vocabulary cases and a data balancing method of oversampling the minority races in the training data. As an application, I estimate the pattern of racial differences in some health outcomes, namely the disability and mortality risks, using the SSA ‘s one percent Continuous Work History Sample (CWHS) data set of size around 3 million.

ABSTRACT. Official Bureau of Labor Statistics (BLS) estimates of productivity growth in the retail trade sector indicate that productivity has grown at a moderate rate of 2.8 percent per year between 1987 and 2017, and that there is considerable variation in growth rates across 4-digit industries. But the official data, which can be thought of as weighted averages of establishment-level productivity, tell us nothing about what goes on within industries. Given the transformation of retail trade over the past three decades, this information could provide more insight. In this paper, we present productivity dispersion statistics for industries in the retail trade sector. These statistics are similar to the BLS-Census Bureau Dispersion Statistics on Productivity for manufacturing industries and complement the official BLS industry-level productivity statistics. We find that from 1987 through 2017, productivity dispersion increased slightly on average. Surprisingly, the tails of the retail productivity distribution have similar dispersion as we find in the middle. Firm dispersion has increased more than establishment dispersion.

ABSTRACT. This paper studies the adverse long-term impact of a decline in lender health on aggregate productivity. I develop a simple model of productivity-enhancing investment where firm exposure to fragile banks leads to losses on both the intensive and the extensive margin. The model is consistent with the surge in exits and prolonged drop in productivity growth observed in Spain in the aftermath of the 2008 financial crisis. The model also highlights the existence of a bias in the measurement of observable TFP growth during an episode of heightened exit. Using data on Spanish firm-bank relationships and bank bailouts, I implement an exit-adjusted measure of productivity growth and use it to quantify the output loss attributable to the financial friction. A decade after the crisis, output growth from the extensive margin recovers but the same is not true of the output level. The output shortfall from the intensive margin proves much more persistent, with the growth gap only beginning to narrow towards the end of the sample period. Together, these dynamics amount to a cumulative loss of 3% of pre-crisis GDP over ten years.

ABSTRACT. Employment in the informal sector of developing economies is characterized by lower pay, higher uncertainty, and lower productivity (Ulyssea 2018; Kochar 1995; Kochar 1999; Laporta and Shleifer 2014). Labor-market interventions, such as minimum wage, that induce transfers to workers in the informal sector may improve their welfare and productivity. It is unclear, however, if price floors would deliver intended outcomes on workers' earnings in informal and casual labor markets. In a neoclassical setting, a binding price floor would reduce quantities demanded and induce excess supply. However, empirical findings deviate from this intuition; in the context of formal, salaried employment, the literature on minimum wage has found mixed evidence on employment that deviate from a neoclassical intuition (e.g., Card and Krueger 1994; Cengiz et al. 2019; Jardim et al. 2018).

In this paper, we study the market-wide impact of labor regulation directly imposed on a casual and informal labor market: two-wheel taxi rides on mobile ridesharing apps in Indonesia. Digital platforms provide visibility into a type of informal labor market on which there had been limited, high-frequency data. We study the effects of a price-floor policy in the ridesharing market intended to raise workers' earnings. By exploiting the city-level variation in the timing of policy implementation, we estimate its causal effects with difference-in-differences and synthetic control methods.

We find that, on average, the policy increases the trip price by 4.6% but does not significantly affect the overall transaction volume nor increase driver earnings or wages. These effects are driven by an increased excess labor supply by 24.3%, reducing the number of transactions per driver and bringing down average earnings to the pre-policy levels. We also find that the minimum-fare policy and associated adjustments come at the cost of reduced driver productivity and higher price for consumers. We find that the policy lowers driver productivity by increasing the share of less productive drivers in the workforce and reducing individual productivity due to crowding on the supply side. When we examine the differential effects by customers' exposure to the policy, we find that customers in the top deciles pay 20% more per trip and on daily expenditures. However, we also find a homogeneous, 5% reduction in a proxy of customers' wait time regardless of their policy exposure, suggesting that the incidental benefit of the policy is suboptimally allocated relative to its cost.

To rationalize these findings, we introduce a static matching model of the ride-hailing market in which prices shift exogenously, and workers respond by choosing labor supply. We then select key outcomes and compare effect sizes with a similar study by Hall et al. (2021) that evaluates a platform policy by Uber in the United States. Both papers find increases in labor supply, but the magnitude in our analysis from Indonesia is more than five times larger than in the U.S. counterpart. Our findings are consistent with the idea that high levels of informality and large labor stocks in Indonesian cities trigger larger supply responses than in the United States.

ABSTRACT. This paper examines the economic impact of the U.S. financial sanctions against Russian companies in the aftermath of Russia’s 2014 annexation of Crimea. It shows that this sanctions program, which primarily cut off sanctioned firms’ access to international financial markets, produced the unintended consequence of strengthening the sanctions targets relative to their unsanctioned peers. Specifically, while the policy successfully halted new international borrowings by sanctioned companies, the spillover impact of the policy resulted in these targets shrinking in size by less than unsanctioned Russian firms. To explain these results, I argue that sanctions led to a reallocation of domestic resources in favor of sanctioned firms. I present a heterogeneous firm model with segmented capital markets and a borrowing constraint in which sanctions against international borrowers led to capital crowding out and credit rationing among domestic borrowers. This research highlights the limitation of targeted sanctions, identifies factors for policymakers to consider in calibrating future programs, and analyzes policy alternatives. It also offers insights for the 2022 sanctions and sheds light more broadly on the impact of international financial integration and capital flows on firm size dynamics.

ABSTRACT. This research paper presents a comparative analysis of social and economic dynamics, focusing on the Varna and Ashrama systems in ancient India and Adam Smith's theory of the division of labor. Delving into the principles of the ancient Indian systems, namely Varna and Ashrama, the study examines their influence on social structures, labor allocation, and economic roles. Simultaneously, it explores Adam Smith's influential theory on the division of labor during the Enlightenment era, emphasizing its conceptualization and implications for economic development. By comparing these distinct frameworks, the research seeks to unveil commonalities, disparities, and the enduring impact of these historical systems on societal organization, economic efficiency, and individual contributions. Employing a multidisciplinary approach, the paper contributes to a nuanced understanding of the historical roots of economic philosophies and their implications for social and economic dynamics.

ABSTRACT. This research delves into the enduring roots of corruption in Haiti and aims to unravel the factors, variables, and conditions exacerbating the nation's prolonged economic stagnation. Despite Haiti's remarkable triumph in achieving independence as the first nation founded by former slaves, its journey towards economic prosperity has been fraught with challenges. By meticulously examining the persistent factors contributing to corruption and economic difficulties, this paper seeks not only to deepen the understanding of Haiti's unique situation but also to inform and shape actionable policy recommendations. These recommendations are designed to address the critical issues at hand and pave the way for meaningful economic advancement in Haiti.

In 1804, Haiti emerged as the first country founded by former slaves after a ferocious struggle for independence from French colonial rule, led by Toussaint Louverture (PBS). Despite its valiant beginnings, Haiti has yet to realize economic prosperity for its citizens. The root of Haiti’s contemporary challenges lies in its inception as a nation. Upon independence, Western powers imposed an embargo on Haiti, only lifted upon payment of exorbitant reparations to France (Quigley 2010). This severely impaired the nation’s economic development and its ability to forge trade relations. Compounded by environmentally detrimental policies and natural disaster susceptibility, Haiti’s path to prosperity has been fraught with obstacles. Moreover, political corruption and the absence of robust institutions have exacerbated the country's plight.

In stark contrast, the Dominican Republic—sharing the same island—has charted a markedly different social, economic, and political trajectory. By benchmarking against the Dominican Republic’s successes, we hypothesize that policies and economic decisions beneficial to the Dominican Republic could potentially be adapted to aid Haiti. This study leverages data from the World Bank, focusing on various regional and income groups, to draw comparisons and inform policy recommendations.

ABSTRACT. This paper addresses dishonesty in economics: the factors that enable it, and what can be done to stop them. It also discusses various instruments that economists can use to identify and dismantle dishonesty in the profession. These instruments include: (1) the study of how to change the incentives within the economics profession, (2) exposure of the organizations that promote dishonesty in economics, (3) appeals to economics students before they become indoctrinated (or before they self-select out of the economics profession), (4) government responsibility to ensure the elimination of government funding for dishonest economic research, (5) exposure of the authors of dishonest work in economics, and (6) a changes in tone of discourse on the topic, with an emphasis on the need for scientific integrity. The paper argues, however, that scholarly discourse, itself, on the topic may do more harm than good, but drawing attention away from the factors that really matter when it comes to encouraging the profession to be more honest. The study draws upon previous literature on these topics, while also taking a fresh approach to examining some of the psychological and sociological factors that influence the profession’s incentive system.

ABSTRACT. This paper seeks to explain the movement of the first three waves of COVID-19 over time and geographical space for readers who are less familiar with how demographic factors may intersect with a cause of death characterized by large surges, or waves, of deaths. We use publicly available data from January 2020 through February 2022 from the National Center for Health Statistics for our analysis. We find that—by race and ethnicity group—the level of urbanicity associated with county of residence, the distribution of residence by U.S. geographical area, and the distribution of socioeconomic status (SES) levels by density of county within varying geographical areas may have all contributed to changing levels of differential COVID-19 mortality as the virus gradually spread throughout the country. In contrast, the age distribution of COVID-19 deaths remained roughly constant over the first three waves of COVID-19. This paper also analyzes the extent to which preliminary COVID-19 mortality outcomes reflect pre-pandemic preventable, or premature, mortality patterns by race and ethnicity and finds movement towards those pre-pandemic patterns over time. Overall, our findings suggest that, with more comprehensive data, changes in the relationship of race, ethnicity, and SES with the risk of death from COVID-19 over the course of the pandemic may help researchers learn more about contributing factors to both pandemic-based mortality differentials and pre-pandemic all-cause mortality differentials. These contributing factors may be more difficult to observe for causes of death that have more stable trends over time. The improvement of public health in the United States could directly enhance economic growth through increased labor force participation. More speculatively, to the extent that any conditions hindering better U.S. performance during the pandemic were also conditions hindering greater potential economic growth, improvement in those conditions may have the potential to enhance the standard of living of all members of society. We stress that current knowledge remains nascent and present open questions for future researchers to consider.

ABSTRACT. We estimate healthcare utilization changes in response to non-pharmaceutical interventions by using cell phone traffic data and a new dataset on social mobility restrictions on and off Native American reservations. Our findings suggest that reservation residents experienced a smaller reduction in total visits to healthcare facilities compared to off-reservation residents; we also find that reservation residents reduce their average distance traveled to healthcare facilities by between 9 to 20 miles depending on the type of facility as a result. This can be attributed to the lack of alternative healthcare options and more acute health conditions prevalent among American Indian populations. Interestingly, we observed a stronger reaction from reservation residents to casino closures than to stay-at-home orders issued by their respective counties, highlighting the importance of considering reservation-specific policies in public health policy analysis. Our study sheds light on critical health issues faced by a vulnerable population that is often overlooked due to the lack of appropriate data. By using an alternative data source that accurately represents reservations, we contribute a novel method to address the undercoverage of this population in traditional surveys.

ABSTRACT. Recent research has shown COVID-19 pandemic caused the United States healthcare industry to grow. In fact, the data from this research show that there was an increase in healthcare entrepreneurship. What this research has not shown, however, is the change in the structure of industry after the COVID-19 pandemic. Namely, there has not been much analysis on the types of healthcare start-ups during and after the pandemic. Nor has there been much analysis on composition of the healthcare industry—the number of employees and firms as a percentage of all healthcare employees and firms—before and after the pandemic. As such, this paper examines the changes in types of healthcare start-ups before and after the pandemic, and; the relative composition of the healthcare industry before and after the pandemic. Various United States Census data—including the Statistics of United States Businesses, Business Formation Statistics, and Business Dynamics Statistics—are used.

ABSTRACT. Governments frequently turn to private markets to deliver public benefits. This structure can lower the government's costs if it designs a payment system that attracts competitive firms with cost controls the government lacks. In this paper, I analyze the implications of this system for the Medicare Advantage program. I use administrative data to develop and estimate a model of firm entry and product offering decisions that captures how firms endogenously respond to government policies as well as consumer sorting and utilization of health insurance. I then use the model to simulate other payment policies in Massachusetts. I find that under the current design, the government overpays firms for their participation and the enrollment they generate. Under a policy that lowers firm payments and transfers a portion of this money to consumers, the government can reduce spending by roughly \$276 million (\$350 per enrollee). This policy incentivizes similar firm participation and enrollment, while more equitably distributing surplus across healthy and sick consumers.

ABSTRACT. The housing market is highly seasonal. The number of home purchase applications that mortgage lenders receive fluctuates depending on the season. Using the confidential version of the 2018 to 2022 mortgage application data collected through the Home Mortgage Disclosure Act (HMDA), we examine if lenders ration credits when their mortgage processing capacity is constrained due to an increase in demand for home purchase loans during a high season. We show that seasonal variations in home purchase mortgage applications exist both intertemporally and across geographic areas, coinciding with the climate patterns. Using this variation, we find that refinance applications are more likely to be denied when home purchase application volume is high compared to when it is low, even after controlling for extensive sets of loan and credit characteristics. Our finding is robust to various specifications. This study is the first to examine lenders’ response to capacity constraints driven by a demand shift using a novel identification strategy. Our identification strategy relies on the fact that (1) the demand for home purchase loans increases during high seasons, increasing the likelihood of lenders’ capacity to process applications to be constrained; and (2) unlike the credit characteristics of home purchase applications, those for refinance applications are not correlated to the seasonality. Lastly, we find that Fintech lenders are less likely to ration credits during high seasons since their origination process is more automated and relies less on human workforce. Our finding has an important welfare implication for consumers since borrowers’ access to credits vary based on the seasons.

ABSTRACT. I show that a low capital gains tax exacerbates instability in asset markets. This outcome arises naturally in various asset pricing setups due to the higher elasticity of prices to investors’ expectations when taxes are low. This mechanism is reinforced when agents learn from prices, as low taxes make self-fulfilling booms and busts more likely. Applying this theory to the United States, I establish a connection between capital gains tax cuts and the proliferation of stock market fluctuations since the 1980s. A structural estimation of the model attributes approxi- mately 30% of the increase in the Price-Dividend ratio variance to these tax cuts. Moreover, these cuts significantly contribute to explain the reduction in return predictability, the increase in the elasticity of prices to beliefs documented using surveys and the equity premium. The findings remain robust to the rise in stock buybacks and the decline in interest rates.

ABSTRACT. How do housing prices respond to global financing conditions, fiscal policy, and productivity shocks? This paper addresses these issues using a stylized model that takes into account the interactions between international capital movements, relative prices, and income inequality. The reference is to a small open economy with free international capital movements, where two types of goods (tradable and nontradable) are produced by households endowed with different skills but identical preferences. The analysis highlights that international capital flows affect real housing prices both by changing the equilibrium real exchange rate (and hence the price of non-tradable real estate) and by altering income distribution (and hence the relative income of the marginal home buyer). Fiscal policies aimed at reducing income inequality have different impacts on housing prices depending on their targeting, coverage, and on the share of homeowners in the population. Exogenous productivity shocks have a different impact depending on which sector is hit, and external financing can mitigate these effects in real terms but in some cases amplify the change in nominal prices.

ABSTRACT. We study agents' responses to changes in the slope or the intercept of a piecewise linear schedule of incentives. For example, past theory predicts that for increasing marginal income tax rates, many taxpayers will report income exactly at the threshold where the tax rate increases. Many empirical settings that would in theory predict such a mass-point, instead have a diffused mass near the threshold, or no excess mass at all. We attribute these diffuse mass points to optimizing frictions and introduce new theory and new estimation methods to allow these frictions to depend on observables. Our methods are not limited to public finance and apply to a general class of mixture models and any of the four possible piecewise linear constraints, 1) slope increase, 2) slope decrease, 3) intercept increase, or 4) intercept decrease. We demonstrate these methods in three of these four settings. We document which covariates account for a substantial share of optimizing frictions and provide elasticity estimates that explicitly control for frictions.

ABSTRACT. The Employment and Training Administration recently created a series of Workforce Services Dashboards that will be made available to the public in mid-April. These dashboards examine WIOA Titles I & III program data quality, system reach, and differences in access to services and outcomes between demographic groups by state. These analyses were produced in response to Executive Order 13985, Advancing Racial Equity and Support for Underserved Communities Through the Federal Government and they provide valuable context and insight on participants in the WIOA Adult, Youth, Dislocated Worker, and Wagner-Peyser programs by various demographic characteristics. Comparisons to national-level data or other outside data sources are included where appropriate.

ABSTRACT. The Employment and Training Administration’s Office of Apprenticeship promotes and helps employers and other organizations develop quality, accessible Registered Apprenticeship opportunities for workers seeking higher-skilled, higher-paying jobs and organizations seeking to build a qualified, diverse, and inclusive workforce. Apprenticeship utilizes the Registered Apprenticeship Partners Information Data System (RAPIDS) system for registration, oversight, and tracking of apprentice progress and apprenticeship compliance. A core component of the Office of Apprenticeship’s work is ensuring all Americans have access to Registered Apprenticeship opportunities and RAPIDS data are a key component. OA works to help employers increase diversity, equity, inclusion, and accessibility in their workforce through Registered Apprenticeship and can help them connect with targeted populations such as women, Veterans, people of color, youth, people with disabilities, justice-involved individuals, and other underserved communities.

ABSTRACT. This presentation will provide an overview of how to summarize information by industry, sectors, and occupations in various data sources; review state and federal LMI data sources and know which to use to understand a sector’s current employment and wages, worker demographics and education, and projected employment. The presentation will conclude with a demo on data visualizations to analyze and depict a sector’s workforce. These dashboards can be a resource for policymakers, states, workforce system practitioners and many other stakeholders as we look to better understand emerging and critical sectors, and how the public workforce system can leverage it for effective sector strategies.

ABSTRACT. Occupational segregation explains a meaningful portion of the contemporary gender and racial wage gaps. The public workforce development system is an important context in which to study occupational segregation because each year thousands of workers pursue occupational skills training to prepare for careers in a range of occupations. Individual training accounts (ITAs) are the most common way workforce participants pursue skills training. This presentation explores the extent of occupational segregation in the training choices of ITA recipients and the gender and racial pay disparities that could be expected to result from these selections. The presentation will touch on implications for how occupational segregation and pay disparities might be reduced.

ABSTRACT. When studying the effects of US state level policies, a common approach is a difference-in-difference (DiD) analysis of border counties comparing neighboring counties subject to the policy with control counties that are not. The common trend assumption, necessary for causal identification with DiD, seems reasonable since adjacent counties are separated only by a state line. While border counties are nearby and probably geographically similar, there can be substantial demographic differences, which can compromise identification and external validity. Using data from the American Community Survey (ACS), the Small Area Income and Poverty Estimates (SAIPE) Program, and the Local Area Unemployment Statistics (LAUS), this paper investigates demographic differences between border and non-border counties across state lines. It includes an extensive analysis of racial composition, poverty, and unemployment rates, aiming to determine if border counties are representative of national populations and consistent across state lines. The findings reveal significant differences in the racial composition of border counties compared to non-border counties, with a higher share of white population and under-representation of Hispanic and other minority groups in border counties. These disparities raise questions about the external validity of policy evaluations using border counties. The paper presents a case study of a DiD analysis by Lyu and Wehby (2020), which examined the impact of COVID-19 stay-at-home orders using border counties. Our analysis suggests that while border counties may offer some advantages in reducing locational biases, their demographic differences could impact the validity and generalizability of policy evaluations. This paper highlights the importance of considering demographic differences in border county-based policy analysis, particularly when the unit of analysis is a county rather than residents of border counties. Richer econometric specifications and additional robustness checks are needed to account for these demographic differences and ensure the reliability and applicability of border county DiD analyses.

ABSTRACT. Does increased labor market scarcity lead to more local anti-immigration enforcement? We answer this question by evaluating the impact of three national economic shocks on the likelihood that a county sheriff forms a partnership with US Immigration and Customs Enforcement (ICE) in the form of 287(g) contracts. Looking at the universe of signed county-level agreements between 2002 and 2020, we separately use a long-difference, instrumental variables approach and a difference-in-differences design to evaluate the effects of the rise of automation, import competition from China, and the Great Recession. We find that effects of economic shocks on immigration policy depends on their nature. While increased exposure to the rise in automation and import competition did not lead to differential adoption of a 287(g), commuting zones severely impacted by the Great Recession saw large increases. A possible mechanism to explain the differing results is the differential migration response of foreign-born individuals. The results of this paper highlight that general economic anxiety can lead to the enactment of anti-immigration policy; therefore, it is an important channel to consider when discussing the forces that drive sentiment towards immigrants.

ABSTRACT. We document the prevalence and amount of debt delinquency in the US among older adults, overall and by racial/ethnic group characteristics. Using administrative data on credit use for a sample of 4 million older adult borrowers, we show that about one in five consumers age 50+ with a credit bureau record had delinquent debt in August 2022, suggesting difficulties meeting financial obligations in retirement. The prevalence of debt delinquency decreased with age, with about one in four adults age 50–61 and one in six adults age 62+ holding delinquent debt. Consumers living in local areas where a majority of residents identify as American Indian or Alaska Native, Black, Hispanic, or Asian American or Pacific Islander were more likely to have delinquent debt and/or higher median amounts of delinquent debt, relative to consumers in majority-White areas, for various types of delinquent debt.

ABSTRACT. A deep body of literature suggests that young adults who experience a recession may experience long-term declines in their employment and earnings potential that undermine their ability to establish financial stability. Such challenges may especially be relevant for young adults of color and young adults living in communities of color, as prior research has found that macroeconomic downturns may worsen longstanding, structural disparities in wealth and financial well-being along racial lines. To protect the most disadvantaged, many states have implemented emergency measures in the aftermath of the pandemic recession, such as more generous public supports and consumer protection policies.

Using 18 waves of credit data on 800,000 young adults ages 20-29 from February 2020 to August 2023, I answer three research questions: (1) How have community-level racial disparities in young adults’ financial well-being changed between 2020 and 2023?, (2) Did young adults fare better or worse during the pandemic compared with less economically volatile times (e.g., 2016-2019)?, and (3) To what extent have state policies (including extended unemployment insurance programs and utility shutoff moratoria) protected young adults or exacerbated pre-COVID inequities in credit health?. I use descriptive regression analysis, propensity score matching, and two-way fixed effect (TWFE) difference-in-difference models to answer these animating questions. I test the robustness of TWFE estimates on subsamples of young adults (a) living in communities of color and (b) without student loans at baseline as well as using policy discontinuities at state borders to estimate policy impacts similar to work by Dube, Lester, and Reich (2010) and Schmidt, Shore-Sheppard and Watson (2020).

I find that young adults living in communities of color experienced relative improvements in their credit health early in the pandemic that show signs of reversing by August 2023, signaling worsening racial inequities three years after the pandemic recession ended. Despite elevated levels of economic volatility between 2020 and 2023, young adults ages 20-29 see sharper gains in credit health compared with similar young adults in 2016, with the largest relative gains occurring between 2020 and 2021. I also find evidence that state policies, such as utility shutoff moratoria and extended unemployment insurance, likely contributed to protecting the most vulnerable populations during this time, helping to marginally improve young adults’ credit scores and decrease delinquencies. Despite these effects, young adults in 2023 may be experiencing elevated levels of financial distress and greater challenges staying current on bills, which could signal the need for additional policy action to help young adults build and preserve financial stability moving forward.

ABSTRACT. The supply of competitively-effective trademarks is limited, leading different companies to often attempt to register the same or very similar marks for the same types of products. This results in trademark races, where trademark protection is generally granted to the firm that files first. We analyze the impact of such races for trademark protection on firms' product introductions. We focus on races where the trademark filing dates are sufficiently close so that applicants are unlikely to be aware of each other’s interest in the contested mark. This allows us to treat the outcome of the race among contestants as if it were random. We first confirm empirically that races are more likely to occur for competitively effective trademarks. We then show that relative to their earlier-filed, “race winner” counterparts, later-filed applications experience an average of 1.5 years of additional examination pendency, and are 50% less likely to eventually obtain trademark registration. Tracking subsequent applications from the same firm, we also find that “race losers” exhibit a 20-30% lower hazard rate for their next product introduction in the same product class as the original application. Our findings have important implications for firms and policy-makers operating in crowded product markets.

ABSTRACT. This study applies the real options framework to explain the behavior of patenting under uncertainty -- in particular the resilience of innovation investments during periods of uncertainty and business fluctuations. In so doing, it bridges research on uncertainty and investments and emerging work on secondary patent markets. The paper establishes that a firm's patenting has a negative association with sales uncertainty, but that sufficiently high levels of exposure to secondary patent markets mitigates the effects of uncertainty shocks and may even induce a positive association between uncertainty and patenting. A critical reason is that, owing to patent trading, a firm's investments in patent assets are reversible, if only partially. They are not necessarily sunk costs. Through the secondary market for patents, firms are able to use patent protection to store value and recoup investments in knowledge. Firms highly exposed to secondary patent markets tend to be in the technology hardware industries. The mitigating effects of reversibility hold for different measures of uncertainty and hold even after controlling for liquidity, credit ratings, and industrial concentration.

ABSTRACT. Private information causes markets to fail. Markets for innovative technologies are especially fascinating from an information perspective because both buyers and sellers may possess private information about the quality of an innovative technology. Additionally, the quality of innovations are highly uncertain, and uncertainty can exacerbate market failure by increasing the value of private information. In this paper, we develop a generalized Akerlof-style model that incorporates private information on the part of both sellers and buyers as well as flexible uncertainty. Importantly, key parameters of our model map to observable outcomes. This allows us to ``invert'' the generalized Akerlof model to obtain estimates of parameters that govern private information, uncertainty, and other frictions. We apply the model to U.S. patent data. We present estimates for private information and uncertainty for a baseline sample of patents as well as documenting how those estimates vary across technology classes and over time. We validate our estimates by showing that sellers' private information is negatively correlated with disclosure. We then conduct several counterfactual exercises. We find that eliminating sellers' private information, for instance through very strict disclosure rules, has only a negligible effect on welfare. Reducing buyers' private information or overall uncertainty, both of which are difficult to influence by policymakers, have much larger effects on welfare. Finally, we show how different parties obtain private information or reduce uncertainty over the lifetime of a cohort of patents.

ABSTRACT. We examine patterns of inventors entering a new field based on data from a curated collection of patent and publication records related to aeronautics around the world from 1800-1918. For about 2000 of these patentees, we have some biographical information and can see their technological focus. Somewhat over half were engineers, and the proportion rose over time. The term "aeronautical engineer" first appears in about 1909. Relatively few inventors switched from working on balloons, helicopters, or ornithopters (with flapping wings) to working on fixed-wing aircraft. We compare some fixed-wing aircraft patentees who had patented previously to some who had not, to get a sense of the technological background of the inventors in the new field. We can make some inferences about the inventors' interests and strategies. When airplane manufacturers appeared starting in 1908, relatively few of the inventors started companies. They patented in multiple countries however, which was a kind of business strategy. As best we can tell, only a few sold their patent rights, and not many went to work in the new industry. The new airplane manufacturing companies did not all have patent portfolios; we compare their business strategies based on some examples.

ABSTRACT. We document that the recovery of workers' earnings, wages, and labor supply after a job separation is affected by housing characteristics. Homeowners suffer larger and more persistent earning losses than renters, and losses increase on home equity availability and decrease on housing payments. To rationalize our findings, we propose an island search model with endogenous savings and housing decisions, where human capital dynamics depend non-trivially on workers' job status. The calibrated model recreates the larger unemployment scar for homeowners through different channels: an initial higher fall off the job ladder due to human capital decay while unemployed, an intensive use of home equity to smooth consumption, a pickier attitude towards reemployment, and lower migration to better local job markets.

ABSTRACT. Measuring wealth inequality in the United States relies on survey data, as limited administrative data sources exist. Survey respondents provide detailed answers to their net wealth's disaggregated components, including a self-assessment of the value of their home, which is typically their largest single asset. In this paper, we investigate whether self-reported home values align with market transaction values; and; to the extent they diverge, why do they, and what are the implications for racial wealth inequality and other key statistics? To answer these questions, we exploit a unique dataset linking a large national dataset of home transactions to internal American Community Survey (ACS) microdata at the address-level. Comparing transactions that occur in close proximity to the survey date from 2008-2016, our initial results suggest that homeowners modestly overestimate (5%) their home's value on average, but with substantial variation by region and over the business cycle (as much as 15% in 2008). Importantly, we find misperception in home value varies systematically across demographic, income, and economic characteristics of the household, most notably by race. Our findings imply heterogeneity in misperception can significantly alter recent trends in racial wealth gaps, as the Black-White wealth gap may be understated by as much as 40% by official statistics in some years. For context, we compare these results with other macroeconomic measurements that rely on self-assessed values, like Personal Consumption Expenditures (PCE), which are generally less distorted than wealth measurements over the period we study.

ABSTRACT. Market liquidity conveys how long a home takes to sell. This measure is often studied in conjunction with real estate prices. Previous research has documented location gradients, where homes nearer to city centers tend to have higher prices. However, little is known about localized housing market liquidity. This paper presents spatial patterns both within and across cities using real estate listings data. Preliminary findings show homes nearer to city centers take less time to sell than homes farther way. We also find evidence that these spatial patterns vary by property type.

ABSTRACT. When the Federal Open Market Committee (FOMC) makes monetary policy announcements, liquid markets tend to react immediately to both the direct change in short term rates and expectations about the future path of monetary policy. In this paper, we examine the extent to which a much less liquid market, residential housing, responds to monetary policy announcements using a novel micro dataset that covers millions of individual property transactions nationally. Rather than using monthly or quarterly aggregated data, we use the underlying microdata obtained from Zillow (“ZTRAX” data set) that includes rich information on individual transactions as well as corresponding home characteristics for each property. Methodologically, transactions-level intra-monthly data better exploits the timing of the announcements for cleaner identification, providing new insights into how monetary policy shocks affect a market that makes up a substantial portion of the economy, where interest rates are thought to play a key role. Empirically, we compare the effect of “surprise” announcements to “expected” announcements on home prices using a regression discontinuity design (RDD), finding that monetary policy surprises generally have a more potent, immediate impact on home prices. Further, we explore the effects of quantitative easing on this market, as well as geographical variation in home price responses to monetary policy more generally.

ABSTRACT. Jerome Powell in a 2018 speech noted that a tight labor market from a long economic expansion may help support wage growth among some demographic groups that traditionally had not seen much progress on wages. This paper examines the links between economic slack and wage growth. Using the unemployment gap as a measure of economic slack, I estimate the effect of economic slack on the growth rate of wages for individuals included in the Atlanta Fed’s Wage Growth Tracker data from January 1983 to December 2022. These data contain not only the wage growth for a particular individual but also basic demographic data about gender, race, age, and education. By dividing the sample using these characteristics, I compare the effect of economic slack on wage growth and test if there are different effects for individuals with differing demographic characteristics. I find mixed results. There seem to be different effects for male and female workers yet not for White and Nonwhite workers. Younger workers may see larger wage growth from slack compared with prime age and older workers. Yet, educational level does not seem to play a role. These mixed results hold when I separate the data into expansions and recessions, and job stayers and job switchers.

ABSTRACT. This paper studies the problem of a labor market regulator who knows that workers prefer to work fewer hours at their current wage, but lacks specific knowledge of production, labor disutility, and the bargaining protocol. We show that for a large class of bargaining protocols, moderate regulation (such as a small minimum wage) is counterproductive in that it results in hours that exceed the efficient quantity. We find that a combination of the minimum wage, overtime pay, and a cap on hours is optimal in a novel robust regulatory setting where the regulator has neither a prior nor exogenous bounds on model parameters.

ABSTRACT. This paper links American Community Survey data and postsecondary transcript records, leveraging a unique partnership between the U.S. Census Bureau and state higher education systems, to examine how the gender pay gap varies across the distribution of education credentials for a sample of 2003-2013 graduates. Although recent literature has emphasized gender inequality among the most-educated, we find a smaller gender pay gap at higher education levels. Field-of-degree and occupation effects explain most of the gap among top bachelor’s graduates, while labor supply and unobserved channels matter more for less-competitive bachelor’s, associate, and certificate graduates. Our results indicate that contemporary gender inequality lacks a unified explanation and requires different policy interventions depending on socioeconomic context.

ABSTRACT. Traditional marriage patterns have shifted rapidly in recent decades. In the US, the Census Bureau reports that fewer than half of all US households in 2020 include married couples, down from a peak of 78 percent of households in 1950. Many factors have contributed to this trend, including rising cohabitation rates, legalization of same-sex marriage, and changing divorce patterns. Although changes in labor market opportunities are often cited as a contributor to changing marriage patterns, researchers have not examined whether policies such as sex discrimination laws, which influence women’s labor market opportunities, are a contributing factor to these changing marriage patterns.

This paper uses decennial U.S. Census data from 1930-1980 to examine whether laws designed to reduce discrimination against women in the workplace influenced women’s marital decisions. We use difference-in-difference methods to exploit cross-state and time variation in state-level antidiscrimination legislation in place prior to the Equal Pay Act of 1963 and Civil Rights Act of 1964. We compare the probability of marriage and age of first marriage among young women who come of age in states with sex discrimination laws relative to similar women in states without these laws. Preliminary findings indicate that women who grew up in states with antidiscrimination laws are less likely to marry, exhibit later ages at first marriage, and achieve higher levels of education relative to their counterparts in states without antidiscrimination laws. The results are consistent when we examine laws passed very early (before 1930) or passed relatively late (in the 1950s). To our knowledge, this is the first paper to examine the impact of antidiscrimination laws on women’s marital choices.

ABSTRACT. The panel discussions will focus on the current and future ability of survey and administrative data to identify and study the well-being and program participation of people with disabilities. Specific issues addressed the complexities in measuring disability, survey instrument design, the capacity of available administrative records from disability-related programs.

ABSTRACT. The panel discussions will focus on the current and future ability of survey and administrative data to identify and study the well-being and program participation of people with disabilities. Specific issues addressed the complexities in measuring disability, survey instrument design, the capacity of available administrative records from disability-related programs.

ABSTRACT. The panel discussions will focus on the current and future ability of survey and administrative data to identify and study the well-being and program participation of people with disabilities. Specific issues addressed the complexities in measuring disability, survey instrument design, the capacity of available administrative records from disability-related programs.

ABSTRACT. The panel discussions will focus on the current and future ability of survey and administrative data to identify and study the well-being and program participation of people with disabilities. Specific issues addressed the complexities in measuring disability, survey instrument design, the capacity of available administrative records from disability-related programs.

ABSTRACT. Climate change calls for a dramatic transformation of the energy system toward low-carbon sources. The path to net-zero emissions, however, could witness increasing commodity price volatility due to fossil fuel demand declines and supply disruptions. Fossil fuel exporters are key to the clean energy transition, but few papers have examined their macroeconomic vulnerability to rising commodity price volatility. In this paper, we create a new panel dataset of 172 countries spanning three decades between 1990 and 2019 to provide empirical estimates for the macroeconomic impact of exogenous oil price shocks on the real, fiscal, monetary, financial, and external sectors. We show that global estimates for the macroeconomic impact of oil price shocks mask considerable heterogeneity across fossil fuel exporters and importers. Using a structural decomposition of oil price fluctuations, we find that positive oil supply shocks are much more detrimental to fossil fuel importers on impact, whereas positive aggregate demand and oil-specific demand shocks pose downside risks for both exporters and importers. In contrast, speculative and oil inventory demand shocks create a major windfall for fossil fuel exporters. Over the medium term, a positive oil supply shock depresses real output and depletes reserves in fossil fuel exporters, whereas positive aggregate demand, oil-specific demand and inventory demand shocks boost both output and reserves. To the extent that the energy transition will be accompanied by lower-than-expected future oil demand, these results imply that fossil fuel exporters will face significant economic headwinds with potentially large output loss and deteriorating external balance. Overall, results of this paper help inform structural macroeconomic reforms for a smooth clean energy transition.

ABSTRACT. How will climate change affect risks to economic activity? Research on climate impacts has tended to focus on effects on the average level of economic growth. I examine whether climate change may make severe contractions in economic activity more likely using quantile regressions linking growth to temperature. The effects of temperature on downside risks to economic growth are large and robust across specifications. These results suggest the growth at risk from climate change is large—climate change may make economic contractions more likely and severe and thereby significantly impact economic and financial stability and welfare.

ABSTRACT. Recent trends show that some wealthy nations have increased GDP while reducing Material Footprint (MF). While this fact seems consistent with the downward side of an Environmental Kuznets Curve (EKC), they might also reflect the bust phase of a cycle. To explore that possibility, we introduce Environmental Kuznets Swings (EKS) – investment-driven cycles of rematerialization and dematerialization. Our nonparametric analysis separates the data into a long-term EKC trend and a cyclical EKS component. Surprisingly, the noted dematerialization largely stems from these cycles. Using input-output data, we decompose EKS by materials and sectors and show that EKS are mostly driven by construction and construction-related sectors. This challenges the sustainable growth narrative in developed economies and calls for a reevaluation of economic-environmental links.

ABSTRACT. In this paper, we explore the impact on housing prices from the strike in Cali, Colombia, in 2021 that lasted two months and affected many areas of the city. We use data from advertisements from the website www.fincaraiz.com to explore the prices from the offer side and explore a temporary drop in prices in areas neighboring strike locations. Our analysis uses a diff-in-diff model to consider these locations before, during, and after the strike. We find a drop in prices of 9 \% for apartments and 15\% for houses on average during the strike. We separate impacts from police interventions and the handling of the protest by policemen. In addition, we found an important recovery after the strike activities stopped.

ABSTRACT. I document substantial price dispersion in the U.S. mortgage market, where each given lender charges observably similar borrowers different rates for similar loans. To explain the conditional price dispersion, I estimate a structural model that features borrowers' demand for mortgages and lenders' individualized optimal price decisions. I find evidence of adverse selection in the form of a positive correlation between the unobserved determinants of borrowers' demand and lenders' costs. Moreover, lenders' ability to learn borrowers' private information and set individualized prices alleviates the severity of adverse selection. I decompose the unexplained price dispersion into cost-based price adjustments and demand-based price discrimination and find that cost factors contribute more. The counterfactual analysis reveals that consumers benefit from the status quo individualized pricing compared to more uniform pricing as it reduces equilibrium interest rates through enhanced competition. The effect is heterogeneous across income groups with higher-income borrowers benefiting more.

ABSTRACT. Despite the housing sector’s importance to the macroeconomy, the ability of forecasters to predict developments in this sector has not been extensively studied. This paper analyzes the private sector’s ability to forecast housing starts, a key and widely followed economic indicator. We provide evidence on the accuracy, bias and efficiency of Consensus Forecasts of U.S. housing starts. We also study the behavior of forecasts around turning points—including the Great Recession and the 1990 recession—and the consistency between forecasts of housing starts and GDP forecasts. Finally, the performance of Consensus Forecasts of housing starts is compared to the predictions of estimated models of the economic drivers of housing starts.

ABSTRACT. This paper explores the connections between housing affordability, property ownership, and income inequality. Through a comprehensive analysis, this paper explains how the disparities in housing accessibility contributes to broader economic inequalities by examining the correlation between property ownership patterns and income distribution. By assessing the extent to which various levels of housing affordability intersect with the income inequality, studying the impact on different demographic groups and regions, this paper aims to produce a nuanced understanding of how disparities in housing opportunities can amplify broader economic inequalities.

Additionally, the research provides a set of targeted policies aimed to mitigate housing related disparities. The policy recommendations address challenges such as astronomical increase in property prices, limited affordable housing options, and unequal access to mortgage opportunities and proposes a more inclusive housing market. This study aims to highlight the important connection between affordable housing and income equality and provides sustainable solutions for housing opportunities that contribute to shared prosperity.

ABSTRACT. IRS audit rates have fallen for over a decade due to declining resources. In addition to the loss of direct audit revenue, decreased enforcement likely results in lower tax compliance by taxpayers who were not audited. This is referred to as lower "voluntary compliance" in the tax literature and is driven by a taxpayer's behavioral response to the perceived risk of being audited. We contribute to a small literature on how IRS enforcement affects the voluntary compliance of the general taxpayer population. Using microdata from random audits conducted for research purposes, we find that noncompliance increases on certain line items as overall audit rates decline. This compliance response varies by the visibility of the line item to the IRS through third-party reporting (e.g. W2s). We find that a one percentage point increase in the audit rate decreases misreporting of wage and salary income by 5.1 percent; this effect grows to a 21.8 percent drop in misreporting of income subject to much less information reporting. Notably, we do not find a discernable deterrence effect for income subject to little or no information reporting. Our findings suggest that the deterrent effects of IRS enforcement depend on how well noncompliance is able to be measured and validated.

ABSTRACT. A recent working paper by a team of academic and government researchers (Elzayn et al, January 2023) found that Black taxpayers in 2014 were audited at 3 to 5 times the rate of non-Black taxpayers. Because the IRS does not collect data on the race and ethnicity of taxpayers, these findings were based on estimated race probabilities derived using public data on the race/ethnicity distributions of first names, last names, and Census block groups. Since that time a team of IRS researchers has been working to better understand the mechanisms that generate this result. This presentation summarizes their preliminary findings.

ABSTRACT. This project will investigate how the possibility of scrutiny by the public or by watchdog organizations affects taxpayer behavior. Somewhat uniquely among taxpayers, tax-exempt organizations operate in a tax compliance environment in which many of their reporting decisions are visible to the public. Although called "tax-exempt," exempt organizations pay taxes on unrelated business income and file a return, the Form 990-T, to report that income. Until 2006, the 990-T was confidential, but was made public for charitable exempt organizations (but not exempts) by the Pension Protection Act. Prior research (e.g., Hoffman 2007, Omer & Yetman 2003, Cordes et al. 2002) finds that in the era before the Form 990-T was public, firms managed the joint reporting of charitable and unrelated expenses in order to minimize taxable income. Did publicizing the 990-T reduce firms’ propensity to manage taxable income? Alternately, do firms shift unrelated business activity to a controlled for-profit subsidiary, which does not file a 990-T and so escapes disclosure?

To answer these questions, we rely on a sample of confidential pre-2006 990-Ts of charities, as well as post-2006 990-Ts of non-charitable exempt organizations, merged with the universe of confidential returns filed by for-profit subsidiaries, all under a research agreement with IRS. We find that, as expected, the PPA reduced the expense-to-revenue ratio at treated firms relative to control firms, and increased tax paid. Although the overall number of for-profit subsidiaries rises in the treatment population, this pattern does not seem to be the result of PPA, as the firms with higher expense ratios before PPA are actually the least likely to create new subs. We also find no evidence that expense ratios are higher among controlled subsidiaries after reform.

We argue that these results may have implications for other reform efforts related to possible shifting of expenses across related entities, such as the recent move towards country-by-country reporting for MNEs.

ABSTRACT. Nonfilers contribute 9 percent, or $32 billion, towards the individual income tax gap. Audits of this population have decreased due to declining resources. Fewer audits not only result in loss of direct revenue, but also in lower voluntary compliance. In this paper, we estimate the effect of in-person audits of nonfilers on their future filing behavior. We compare the filing behavior of taxpayers audited during Tax Years 2009-2014 against a group of unaudited taxpayers who were eligible for audit. We find that audited taxpayers are 5.3 to 13.8 percentage points more likely to file a return in subsequent years, an indirect effect that attenuates over time. These findings are qualitatively similar to other studies of the indirect effects of audits on future compliance.

ABSTRACT. Recency bias---the tendency to give excessive weight to recent developments---can distort our perception of technological change. The last fifteen years of smartphone technology has been marked by a dramatic shift to large touch screens, computational photography, integration of machine learning, and more. Some observers of recent cell phone technology see these changes, as well as the advent of smartphones themselves, as representative of a revolution in cell phone technological innovation beginning in the mid-2000s. Yet, the early cellular era (1987-1992) was also characterized by rapid technological change, with car phones succeeded by bag (portable) phones, brick (handheld) phones, and flip phones. Exploiting a novel dataset collected from advertisements in The New York Times, we introduce a new Jevons price index for the early cellular era based on model-level prices in the New York City market from 1987 to 1992, a period with no existing price index for this product. We show that prices for cell phones have been characterized by a fairly steady pace of rapid declines since their introduction in the mid-1980s: Price declines typically fall between 15 and 20 percent from 1987 to 2022, with no apparent change in the secular trend. That the change in price levels has held relatively constant over a 36-year period is consistent with a steady march of technological change rather than a new era of accelerated cell phone innovation or a ``smartphone revolution.'' In the course of the analysis, we also explore two broader questions. First, is distinguishing products sold at different establishments materially important for price measurement? Second, how can we leverage the rich and largely untapped resource of advertisements in newspaper archives for price research?

ABSTRACT. The U.S. Bureau of Labor Statistics (BLS) Consumer Price Index (CPI) is the principal federal data source for tracking food price changes. Although the CPI is a timely source of quality price information, it has shortcomings for food and nutrition policy analyses. First, the CPI only reflects temporal price variation but several studies have found that prices vary both temporarily and spatially, which is useful for analyzing trends in food prices and in econometric models of the effects of nutrition and food policies (Zhen et al., 2019). Second, the product aggregation in the CPI is too coarse for food policy analysis. In particular, the CPI food categories do not align well with the Dietary Guidelines for Americans or more generally with foods that are currently relevant in the nutrition and food policy literature (e.g., whole grain versus refined-grain products). Scanner data can be used to address these limitations of the CPI. Circana’s OmniMarket Core Outlet retail scanner data are a nonprobability sample of approximately 60,000 retailers from various retail channels (e.g., supercenters, grocery stores, dollar stores). It contains all products sold in these stores, including UPC-level quantities and revenues from which prices, and thus contain significantly more price observations than the CPI. The granular UPC-level information across time and geography allows for construction of panel price measures for foods that are relevant to nutrition and food policy discussions. An issue with scanner data is little is known about the representativeness of stores in this nonprobability sample. Levin et al. (2018) found that sales reported in the Circana retail scanner data were roughly 50% of those reported in the Census of Retail Trade with participation being mostly from large stores located in coastal and urban areas. Although these shortcomings can be somewhat addressed using post-stratification projection factors or sampling weights, it is still unclear whether the prices reported by the stores in the retail scanner data are representative of the prices offered by stores in the United States. We examine the variation between and long-term trends in the scanner data-based price indexes and the CPI to benchmark the quality of price information in the retail scanner data. We first construct price indexes approximating the CPI approach for similar food categories and geographies using the retail scanner data. We then examine correlation and direction of price movement from month-to-month between the CPI and the scanner-based price indexes. We find the correlation coefficient between the CPI and the scanner-based price index for all food at home across all geographies to be 0.90, which is consistent with another validation study for Nielsen retail scanner data (Ehrlich et al., 2022). However, we also see quite a bit of heterogeneity in correlation across region and food category. As survey data become increasingly more expensive to collect and curate, alternative sources of data like retail scanner data have become popular, and the findings from this analysis will help elucidate some of the benefits and limitations of using this price information for food and nutrition policy analysis.

ABSTRACT. The U.S. Copyright Office is conducting a study on the geographic distribution of copyright registrations received by individuals and corporations within the United States. We hope to better understand where the copyright system is used, where it is under-used, and the demographic profiles of these areas. Understanding these patterns of copyright registrations is a helpful step for better understanding the socio-economic factors that influence creative and innovative activity.

The study first presents our data of registration counts by three geographic units: states, counties, and metropolitan/micropolitan statistical areas. The data is disaggregated by fifteen different types of copyrighted works, especially those most registered - Literary works, Artistic works, Serials, Dramatic works, Motion pictures, Computer programs, Musical works, and Sound recordings. Registration data are reported as totals, relative to population, and relative to the number of people working in industries and occupations that are thought to rely on copyright protection. The following section compares registration counts per 1,000 people to demographic covariates, finding that registrations are more prevalent in areas with higher incomes, more racial diversity, and a greater share of people working in creative fields.

ABSTRACT. panel

ABSTRACT. panel

ABSTRACT. panel

ABSTRACT. panel

ABSTRACT. Our ability to identify and research business demographic trends and performance disparities across demographic groups hinges upon the availability of reliable, frequent, and timely business demographics data. In response to declining response rates, and increasing imputation rates and costs, starting with reference year 2017, the Census Bureau began providing nonemployer demographics not through a survey, but a program that leverages administrative and census records to assign demographics to the universe of nonemployer firms: the annual Nonemployer Statistics by Demographics series (NES-D). Given NES-D’s success, Census is now evaluating the feasibility of assigning demographic characteristics to U.S. employer businesses using administrative records. In the presentation we will describe the background, methodology, ongoing challenges, current results and next steps of this ongoing effort. The use of administrative records in business demographics statistics should be viewed as a complement to surveys, and a vehicle to unburden respondents and allow the survey to measure issues that cannot appropriately be captured with administrative records or third-party data.

ABSTRACT. This paper explores the impact of two fiscal programs, the Economic Impact Payments and the Paycheck Protection Program, on vehicle purchases and relates our findings to post-pandemic price pressures. We find that receiving a stimulus check increased the probability of purchasing new vehicles across the income distribution. In addition, the disbursement of funds from the Paycheck Protection Program was associated with a rise in local new car registrations. Our estimates indicate that these two programs account for a boost of around 1 million units—or about 7 percent—to new car sales between 2020 and 2021. Furthermore, the induced boost in sales coincided with the presence of significant production constraints and exacerbated an inventory drawdown, thereby contributing to the rapid increase in new vehicle prices that prevailed in the subsequent year.

ABSTRACT. The employment rate in the United States fell dramatically in many industries during the COVID-19 pandemic—the childcare industry, in particular, was hit hard. In this article, we use data from the U.S. Bureau of Labor Statistics (BLS) Business Response Surveys and other BLS sources to examine employment, wages, telework, benefits, and the inner workings of the critical child daycare services industry before and during the pandemic and most importantly how the industry has managed since. Although the childcare industry’s wages are low and it has high labor turnover, our findings show that it is critical in supporting workers across all industries.

https://www.bls.gov/opub/mlr/2024/article/childcare-employment-before-during-and-after-the-covid-19-pandemic.htm#top

ABSTRACT. Introduction The 21st century has seen rapid growth in the commercial space sector. Two subsectors of particular interest are firms operating small satellites and firms offering small launch vehicles for these “smallsats”. While most smallsats have historically rideshared as secondary payloads on large rockets, 48 different small vehicles are currently in development to provide more timely, dedicated launch options for smallsats.

The government believes that flexible and timely U.S.-based smallsat launch options are important for building economically and strategically important space-based services, resilient to safety and security threats. Therefore, the Air Force Office of Commercial and Economic Analysis (OCEA) worked with The MITRE Corporation (MITRE) to understand which factors have the most impact on the market shares of commercial smallsat launch providers.

Data and Model Structure We estimated the parameters of a discrete choice model to explain the vehicles chosen by 857 commercial smallsats launched between 2018 and 2021 Q2. We hypothesized that each satellite operator chooses a vehicle based on its cost, perceived reliability, average delay, days between consecutive launches, and whether the launch provider is in the same country as the operator. We then estimated a conditional logit model to determine the probability that each vehicle will be chosen, based on the characteristics of both the vehicle and the launch provider. We focus on the seven most widely used vehicles for launching commercial smallsats during this sample period: the Antares 230/230+, the Electron, the Falcon 9, China’s Long March 2, Russia’s Soyuz 2.1a/b, India’s Polar Satellite Launch Vehicle (PSLV), and Europe’s Vega.

Model Results All variables were found to be statistically significant and impacted the choice of launch vehicle in the expected way. The utility of the model was also demonstrated by exploring a series of counterfactual scenarios, estimating what would have happened if the features of existing vehicles were different or if new vehicles had been available in the market.

Launch customers are sensitive to cost, but perhaps less so than one might speculate. However, customers appear to be very sensitive to perceived vehicle reliability, with a 10% increase in reliability translating to an increased likelihood of being selected of 18% to 25%, depending on the vehicle. To achieve a high perceived reliability, a vehicle must have a strong record of repeated successes, implying that launch providers have a strong incentive to focus on reliability.

Vehicles with frequent launches and minimal delays are also more likely to be chosen, all else equal. This finding is promising for smaller launch vehicles such as Rocket Lab’s Electron and Firefly’s Alpha, which can offer a dedicated alternative to ridesharing.

We also found that there is a substantial preference for a commercial smallsat operator to choose a domestic launch provider. This may be due to transportation costs, legal complications, or other trade frictions. Regardless, this suggests that the strength of the U.S. smallsat launch sector is closely related to the strength of U.S. smallsat manufacturing and the growth of U.S.-based smallsat operators.

ABSTRACT. USDA is tasked with the mission of ensuring that agricultural producers have access to credit at reasonable terms and rates to maintain a vibrant and diverse family farm population (1). To fulfill this mission, the USDA’s Farm Service Agency (FSA) provides direct and guaranteed loans to credit worthy family-sized farms who are unable to obtain credit elsewhere, at reasonable rates and terms, through loans directly to qualifying producers and guarantees on loans made by approved lenders. While loans directly to producers were less than 10 percent of all U.S. Agricultural debt in 2020 (2), these loans serve an important role in ensuring that the most vulnerable producers have access to farm credit to establish, maintain and expand their operations.

To evaluate how well FSA’s farm loan programs are serving their intended populations data from the annual Agricultural Resource Management Survey (ARMS) over the 2013-2021 time period is merged with a set of unique NASS provided ARMS-FSA identifiers to calculate farm debt market penetration rates for FSA loan programs across different categories. Market penetration is defined as the weighted number of FSA borrowers divided by the weighted number of indebted farms in the survey. Categories chosen include groups believed to be highly credit constrained.

Initial results presented at the 2024 ASSA Annual Meetings (3) indicate that FSA direct borrowers tend to have large market penetration among small and mid-sized family farms, single operator beginning farmers, and racial-ethnic minority producers. Guaranteed loans tended to have a large market penetration among mid-sized and large family farms, single and multigenerational begging farming operations, and those with more moderate financial stress. These results support the conclusion that FSA direct and guaranteed loan programs are serving their intended population- credit constrained small and mid-sized farms- with each focusing on a different subset of that population.

These initial findings should generate lively discussion among conference participants interested in the role of federal credit programs and agricultural loans, the impact of credit constraints on producers, and related topics. Future research arising from these initial findings is bound to only further illuminate the role that FSA loans play in providing credit to support small and mid-sized family farmers and ranchers. programs.

(1) Ahredsen et al. “Beginning farmer and rancher credit usage by socially disadvantaged status.” Agricultural Finance Review. October 2021. (2) Monke. "Agricultural Credit: Institutions and Issues. Congressional Report Service. July, 2022. (3) Atkinson. “USDA Farm Service Agency’s Loan Programs: Evaluating their mission of providing loans to credit-constrained agricultural producers.” 2024 ASSA Annual Meetings in San Antonio TX January 5-7, 2024.

ABSTRACT. Approved for Public Release; Distribution Unlimited. Public Release Case Number 24-0026

Title: Using Mobility Data to Analyze Healthcare Markets

Researchers commonly rely on claims datasets to derive competition and access metrics for healthcare markets, but these datasets are typically lagged and fall short for services where self-pay is common such as chiropractic and dental care. Insight into all aspects of healthcare—not just those covered by insurance—is necessary to create innovative and effective policies aimed at improving quality, access, and outcomes across the population. Mobility datasets contain up-to-date information on foot traffic to locations across the nation regardless of payment source and can be used to fill these gaps. This analysis explores the use of mobility data to create competition and access metrics for chiropractic and dental markets.

The team used de-identified and aggregated mobility datasets from SafeGraph and Advan to analyze chiropractic and dental markets in Ohio between 2019 and 2022. The team created a scalable database of market characteristics including the number of providers located within a geographic boundary (county, Census Tract, and Census Block Group), the number of providers visited by people residing in a geographic boundary, the average distance traveled to providers, and multiple formulations of the Herfindahl-Hirschman Index (HHI). Initial analysis of travel distances indicates that patients travel farther for dental care than chiropractic care on average with a mean distance traveled in 2019 of 21.7 miles for chiropractic care and 28.4 miles for dental care. These averages have grown to 23.4 miles and 30.15 miles for chiropractic and dental care, respectively. Initial competition results indicate that the chiropractic market is more concentrated than the dental market. The average patient-flow HHI in 2019 across counties was 1,534 for chiropractic markets and about 1,046 for dental markets. Both are becoming more concentrated over time—the average HHI grew to 1,723 (1,223) for chiropractic (dental) care across counties in 2022.

There are important limitations to consider. Some are inherent to the data, for example the inability to disentangle workers from visitors from patients, and some which may improve with time as vendors continue to improve their data collection techniques, for example location accuracy and how to assign visits in a multi-level building. Notwithstanding the limitations, this work illustrates the potential for mobility data in creating a complete data-informed insight into healthcare, which can inform the decision-making process, help contain costs, and improve quality and access to care. For example, mobility data can be used to analyze activities that affect health beyond participation in the formal healthcare system (e.g., gym use and fast-food restaurants) and to perform large-scale national analyses of important policy questions (do public transit stops near providers increase foot traffic from low-income areas?).

This work was funded by The MITRE Corporation’s independent research and development program.

ABSTRACT. Small businesses are a strong driver of economic growth in the United States. Between 59.6% (Bureau of Labor Statistics, Business Employment Dynamics) and 79.8% (U.S. Census, Business Dynamic Statistics) of net job growth was created by small businesses, over the past 10 years. Yet over time small business’s share of the economy has shrunk. In 1963 small businesses with less than 500 employees received 55.7% of all revenue. In 2017 this number had fallen to 35.6%. In this paper, I work to better understand this trend by looking at it from several angles using data from the Statistics of U.S. Business. First, by the size of firm where the smallest sized firms were hit hardest. Second, by industry where over the last 15 years the outcomes are widely dispersed with about 30% of industries increasing in small business share of revenue, while 10% decreased by more than 10 percentage points. And third, by state where outcomes were mostly uniform across geography with the few outliers explained by changes in industry composition. Lastly, I conclude with regression analysis associating the change in the share of small business revenue with potential causes including economic growth, firm entry, and regulatory barriers. Economic growth was associated with declining small business share of revenue, while firm entry was associated with the opposite. Changes in regulatory barriers did not appear to affect small business share of revenue in this work.

ABSTRACT. Debt level has increased in most African countries in recent years; furthermore, in some countries at a worrisome pace. However, the continent is not yet experiencing a systemic debt crisis risk. On a regional level, several countries in the Economic Community of West African States (ECOWAS) experienced major episodes of financial crisis characterized by unsustainable fiscal deficits several decades after independence. During this period, however, current account deficits were considered normal. Therefore, ECOWAS countries were encouraged to borrow from abroad to finance their deficits and to create a conducive environment that attracts foreign investment to boost economic growth. Meanwhile, little attention was paid to the individual countries’ absorptive capacities and ability to repay the borrowed funds. Suma (2007) posited that external funding has been crucial in developmental projects, financing capital, and budgetary support for developing countries. This research will build upon the framework of Suma (2007), covering the ECOWAS countries from 1970-2022 via spatial regression methods. The general objective of this research is to examine the impact of external debt on economic growth in ECOWAS countries. To achieve this general objective, the specific objectives of this research are  Investigating the link between external debt and economic growth of ECOWAS countries;  Examining the structure, type and composition of ECOWAS’ external debt;  Identifying the transmission mechanism of external debt influences on economic growth of the ECOWAS countries.

ABSTRACT. Peru has had immense success in combating child malnutrition. Between 2007 and 2014, stunting in Peru halved for children under 5 years of age (measured using Height-for-Age Z Scores). Since the 2000’s, there has been a committed effort in Peru to improve child nutrition. Various public health policy programs were aimed at improving maternal and child health. Peru’s performance has been exemplary in reducing child stunting. The malnourishment of children is a big challenge in developing nations. It causes high child-mortality rates and disease. It affects the future lives of millions of children. By exploring the channels through which one country was able to make huge strides to improve child nutrition, we can draw lessons for many other countries to benefit from. This study explores the factors that enabled Peru to have great success in improving child nutrition. It focuses on the effect of maternal socioeconomic conditions on child nutrition. By looking at Peru’s health data over different points in time, the study investigates how the socioeconomic makeup of women changed over time as a result of targeted public health policy. It examines the relationships between a mother’s nutrition, well-being, education, and employment and explores how these factors affect the health of her child.

ABSTRACT. This work is an update on the recent development of the Quarterly Personal Consumption Expenditures by State. We will talk about source data, methodology, and show the recent experimental estimates. We will show how these data will complement Bureau of Economic Analysis' other regional statistics. Moreover, they help the stakeholders with regards to timeliness and higher frequency to do economic research and policy making.

ABSTRACT. Heart disease remains the leading cause of death in the United States since 1921 despite various efforts in health promotion. Using CDC WONDER data from 1999-2020 and Robert Wood Johnson Foundation’s County Health Rankings Data from 2020, this study provides descriptive statistics and visualizations of ethnic/racial, geographical, and temporal variations in heart disease mortality across the United States at the county level. By creating multiple choropleth maps over time, we find that heart disease mortality rates decrease overall; however, different races and ethnicities experience varying changes in heart disease mortality rates with Hispanic/Latino, American Indians/Alaskan Native, and Asian/Pacific Islander populations witnessing increased mortality rates while African American and White mortality rates decrease slightly. In addition, we also explore how specific social determinants of health such as median household income, food insecurity, and obesity correlate geographically with racial/ethnic health disparities. This research indicates geographic areas where, potentially, efforts are needed the most, which populations are disproportionately affected, and possible factors to lower heart disease mortality for all populations.

ABSTRACT. This study employs advanced text mining methods to comprehensively understand the developmental landscape and historical milestones of artificial intelligence (AI) in healthcare diagnosis in the United States. Given the widespread attention and societal impact of AI, our focus centers on patents related to health data diagnosis in the U.S., aiming to visualize and predict the popular technologies and future trends of AI in American public health. The research provides predictive insights into relevant government policies in the healthcare field and policies related to AI in healthcare.

Utilizing the BERT model, this study illustrates the advancements in deep learning and the transformative influence of AI technologies, such as Convolutional Neural Network (CNN) and Artificial Neural Network (ANN). Patent analysis reveals a significant expansion of AI applications in computer hardware, software, and medical equipment. The COVID-19 pandemic has catalyzed a surge in diagnostic technologies. Notably, ethical and legal considerations, alongside concerns for data privacy and transparency, emphasize the necessity for increased regulatory oversight. Breakthroughs in diseases like diabetes, cancer, tumors, and particularly heart disease through electrocardiogram analysis underscore AI's potential in preventive medicine.

The research implications highlight AI's transformative potential in reshaping public healthcare. Policymakers are encouraged to reinforce supervision, giving due consideration to ethical and legal frameworks, especially in computer hardware, software, and medical equipment, to prevent the emergence of monopolies that could impact healthcare system efficiency. Simultaneously, the study recommends providing appropriate financial support for emerging medical treatments, including those for heart disease and diabetes. This emphasizes the need for a comprehensive approach to harnessing AI's potential while addressing ethical, legal, and regulatory considerations.

ABSTRACT. Expenditures on food and beverages in the United States reached $2.4 trillion in 2022, according to ERS’s Food Expenditure Series. Such expenditures include spending at food-at-home (FAH) establishments—grocery stores, supercenters, convenience stores, and other retailers—and food-away-from-home (FAFH) establishments—restaurants, school cafeterias, sports venues, and other eating places. While real per capita total food expenditures increased steadily through the decades, the share of expenditures at FAH establishments decreased from about 53 percent in 1997 to 44 percent in 2022. This report investigates what macroeconomic indicators are driving these changes within each State and how the extent of each driver varies across States. Previous research has highlighted the roles of rising incomes, lower saving rates, and behavioral changes in U.S. consumer spending. Additionally, the extent of the impact on food spending is expected to be heterogeneous across geographic locations. The framework used in this study separates food spending at the State level into four components: Disposable personal income (DPI) (i.e., income effect); personal consumption expenditures (PCE) as a share of DPI (i.e., propensity to spend versus save); total food spending as a share of PCE (i.e., propensity to spend on food versus non-food); and FAH as a share of total food spending (i.e., substitution between FAFH and FAFH). Using data from 1997 to 2022 at the State level, this paper assesses the relative effects of the drivers of food spending and focuses on comparing the magnitude of these drivers during non-recessionary periods, the Great Recession (December 2007 to June 2009), and the COVID-19 Recession (February to April 2020).

ABSTRACT. The aim of this study is to investigate how the lifting of travel bans as part of the bilateral trade agreement and exposure to unfamiliar visitors has affected the political landscape in Taiwan. To accomplish this, we use the commute times to the main airport as a proxy for exposure to Chinese tourists in different regions of Taiwan. The driving time to major cities was used as the cutoff point. A new dataset was built by merging geographical information into the Taiwanese electoral database, and the regression discontinuity (RD) design model was applied. The study would like to explore the treatment effects between the different tourist-exposure regions before and after the Economic Cooperation Framework Agreement (ECFA) was rectified. The study also ran a robustness check by testing various local polynomial functions in the RD model. The results show that after the ECFA went into effect, the local treatment effect became more significant. In other words, electoral districts exposed to more Chinese tourists encountered a larger scale of political-ideological realignment. The study's findings demonstrate that the ECFA is a counter-example of the progressive economic integration theorem.