Graduate Theses and Dissertations (2019 - present)

Date of Award

12-2025

Document Type

Dissertation

Degree Name

Ph.D.

Department

Business Administration

Committee Chair

Ermanno Affuso, Ph.D.

Abstract

This study examines whether who immigrates, rather than how many, matters for state economic growth in the United States. It integrates a policy-relevant proxy for skill (H-1B approvals) into an augmented Solow framework that separates immigration's quantity channel from its human capital channel and estimates dynamic effects in a balanced quarterly panel of 50 states (2010 to 2023 ). The empirical strategy estimates a two-step difference GMM Arellano-Bond model that reinforces identification using a double/debiased machine learning (DML) variant that orthogonalizes high-dimensional nuisance components via cross-fitting. This design targets the distinct roles of immigrant headcount versus skill in per capita income dynamics. Three findings emerge from this analysis. First, skill composition matters. Increases in the skilled share, proxied by H-1B approvals, are positively associated with subsequent growth in real GDP per capita. Quarterly effects are modest but accumulate to economically meaningful gains over multi-year horizons. Second, headcount alone is not sufficient. Immigration measured as labor force quantity alone is, on average, neutral to slightly negative in per capita terms after accounting for dynamics and common shocks; consistent with capital dilution in a Solow framework. Third, heterogeneity across states is economically meaningful. States with deeper innovation ecosystems translate skilled inflows into productivity growth more readily. Policy implications follow from these findings. At the federal level, expanding and smoothing high-skill pathways can raise aggregate productivity and diffuse gains across regions. At the state level, policies that attract, retain, and integrate skilled immigrants enhance the payoff to talent inflows, particularly outside traditional tech hubs. For example, states that invest in domestic human capital may realize larger gains from skilled inflows. Policies that emphasize openness to skilled immigration and robust education or training systems may complement one another. Limitations in the study point to avenues for future work. For instance, H-1B approvals are not a perfect proxy for skill and location. Here, state assignments can involve minor measurement error since workers may change location after their initial application is approved, and visa flows may respond to economic conditions despite the panel design. State-level analysis also aggregates meaningful variation that occurs at a more local level. Future studies could strengthen causal identification using shocks, such as visa lotteries or cap changes, leverage worker data, such as H-1B salaries, to trace channels from talent inflows to productivity, and consider distributional outcomes. Because the data end in 2023, the analysis largely predates the diffusion of generative AI. Future work could also test whether AI adoption amplifies the productivity return to skilled immigration.

Download

Share

COinS