Household surveys are a vital source of information for understanding economic well-being.  But traditional survey methods often fail to capture accurate data on income and government program participation, particularly for low-income and minority groups. This grant supports research by Derek Wu at the University of Virginia to improve the nation’s official income and poverty statistics by linking major household surveys with administrative records. The team will develop new coverage-adjustment methods using federal and state administrative data to fill gaps and correct inaccuracies in survey samples. They will create imputation models to replace misreported survey entries with values derived from administrative records. The project will reconcile inconsistencies between surveys and administrative data to produce more accurate estimates of household income, poverty rates, and other key economic indicators.  The project’s expected output includes revised survey weights, imputed statistical results, and openly shared methods to enhance data accuracy for research on poverty, labor force participation, and social benefit programs.

This grant supports research led by Christopher Snyder of Dartmouth College, in collaboration with Michael Kremer and the University of Chicago’s Market Shaping Accelerator, to explore how microeconomic theory can inform the design of mechanisms for funding innovation. The team will compare “push” strategies, such as research grants, with “pull” strategies, like prizes or Advanced Market Commitments, to identify optimal approaches for driving socially beneficial innovation.  Beyond cataloguing examples that seem to work under certain circumstances, the goal here  is to develop a comprehensive theory for explaining how and when to design better incentives both for the generation and uptake of new ideas.  Outputs will include academic papers and practical frameworks to fund innovation effectively, given real-world constraints faced by funding agencies and philanthropists.

Research led by Jules Polonetsky at the Future of Privacy Forum (FPF) will study the implications for eldercare of “AgeTech.”   This includes AI-powered monitoring systems, assistive robotics, and other caregiving technologies. The project will document how seniors and caregivers use AgeTech, assess privacy risks and economic barriers, and analyze regulatory approaches in the U.S. and abroad. The research team will catalog existing and emerging AgeTech products and services, conduct surveys and interviews with seniors and caregivers, and examine global policy frameworks to identify best practices. The study will also evaluate how AgeTech adoption is influenced by cost, privacy concerns, and caregiving needs, using systematic reviews and survey-based analysis. Basic findings necessary for understanding the economics of AgeTech will be shared through policy reports, academic publications, and public engagement activities. The project will produce a taxonomy, educational materials for seniors and caregivers, and recommendations for regulators and industry stakeholders.

A multi-institutional team led by Gerald Marschke from the State University of New York at Albany will link Census Bureau data on businesses and workers with the Survey of Earned Doctorates (SED) data collected by NSF.  The project will produce a new dataset tracking the employment of PhD holders and use it to analyze their impact on regional economies. The research team will integrate 20 years of SED data with longitudinal Census data to document where PhD recipients work, how they contribute to firm growth and innovation, and whether their presence influences wages and economic activity. To estimate causal effects, the study will use event studies and shift-share instrumental variable techniques, leveraging funding shocks to PhD programs as a source of variation. The resulting data will be made available to researchers through the Federal Statistical Research Data Center (FSRDC) network, and findings will be shared through academic publications and conference presentations.

Advances in Artificial Intelligence are transforming business processes and labor market dynamics.  A large research team coordinated by Christophe Combemale of Carnegie Mellon University will compile new evidence about the impact of AI on the workforce by focusing on four interrelated themes. Adoption: Researchers will analyze how AI adoption alters business processes, job roles, and skill requirements using data on job postings, occupational requirements, and firm-level technology adoption. Transition: Researchers at UCLA will use 20 years of administrative worker and firm-level data from California’s Unemployment Insurance system to study the impact of AI adoption on employment, wages, and policy interventions such as job training and unemployment insurance. Geography: Researchers will develop an “Observatory of US Job Disruption,” a national dataset and online visualization tool providing real-time, county-level unemployment risk data disaggregated by occupation and industry. Talent: By leveraging JEDx, a workforce data platform developed by the Chamber of Commerce Foundation, researchers will evaluate how employer-provided data can improve estimates of AI’s impact on labor markets, identify emerging skill shortages, and enhance workforce development strategies. The project will produce academic papers, policy reports, data tools that include the Observatory of US Job Disruption.  The research is expected to inform workforce training policies and provide empirical evidence to guide AI-related labor policy decisions.