SOS:BIO: AI-enabled measurement of place-based biomedical policies and their effects on firm dynamics and PHD careers - state and local governments increasingly use place-based biomedical…

The National Institute of General Medical Sciences awarded New Mexico Institute of Mining and Technology $148,495 on September 1, 2026, under the Biomedical Research and Research Training program (CFDA 93.859) to develop and apply an AI-enabled measurement framework evaluating how place-based biomedical policies affect biomedical PhD careers, innovation, and firm-level dynamics.

The project pursues three integrated aims. Aim 1 constructs a validated place-by-year database of U.S. state- and county-level biomedical policies, using a large language model-based agentic workflow with human-in-the-loop validation to extract policy instruments, eligibility rules, geographic scope, timing, and implementation intensity from legislative and administrative texts. Aim 2 links these policy measures to restricted-access longitudinal administrative microdata at federal statistical research data centers to estimate how biomedical policies affect biomedical PhD wages, employment stability, mobility, employer type, startup participation, and innovation outcomes. Aim 3 examines firm-level responses including firm entry, survival, employment, growth, and innovation, and assesses whether observed effects reflect net local gain or primarily reallocate firms and scientists across regions.

The award is classified as a Project Grant and supports the principal investigator's long-term goal of providing evidence that informs public investments designed to strengthen the biomedical research workforce and innovation ecosystem. Performance occurs in Socorro, New Mexico, with a period of performance ending May 31, 2029.

SOS:BIO: AI-enabled measurement of place-based biomedical policies and their effects on firm dynamics and PHD careers - state and local governments increasingly use place-based biomedical initiatives to attract biomedical firms, expand the scientific workforce, stimulate innovation, and promote regional economic development. Despite the growing use of these policies, limited evidence exists on whether they generate durable gains in biomedical workforce outcomes and firm growth or primarily reallocate firms and highly trained scientists across regions. A major barrier to answering this question is the lack of systematic, high-resolution measurement of heterogeneous biomedical policies across places and time. The long-term goal of this project is to provide evidence that informs public investments designed to strengthen the biomedical research workforce and innovation ecosystem. The objective of this project is to develop and apply an AI-enabled measurement framework to evaluate how place-based biomedical policies affect biomedical PHD careers, innovation, and firm-level dynamics. To achieve this objective, the project will pursue three integrated aims. Aim 1 will construct a validated place-by-year database of U.S. State- and county-level biomedical policies. Using a large language model-based agentic workflow with human-in-the-loop validation, the project will extract policy instruments, eligibility rules, geographic scope, timing, and implementation intensity from legislative and administrative texts. Aim 2 will link these policy measures to restricted-access longitudinal administrative microdata at the federal statistical research data centers to estimate how biomedical policies affect biomedical PHD wages, employment stability, mobility, employer type, startup participation, and innovation outcomes. Aim 3 will examine firm-level responses, including firm entry, survival, employment, growth, and innovation, and assess whether observed effects reflect net local gains or geographic reallocation. The empirical strategy will use difference-in-differences and event-study approaches that account for staggered policy adoption, heterogeneous treatment effects, and dynamic adjustment. By integrating AI-enabled policy measurement, causal inference, and longitudinal administrative microdata, this project will provide evidence on which place-based biomedical policies strengthen the biomedical workforce, support biomedical firms, and advance long-term public health goals.