Website CV Paper
Fields
Macroeconomics, Development, Labor, Innovation, Public
Job market paper
The Lasting Effects of Government R&D
Federal funding for research and development contributes to growth by creating new scientific human capital. To measure the aggregate importance of this channel, I construct a new dataset linking all U.S. patents from 1976 to 2025 to a near-universe of PhD dissertations. Individuals with research doctorates appear on roughly half of patents by U.S. inventors and are especially prevalent in critical technologies such as machine learning, biotechnology, and semiconductors. I show that the institution where individuals pursue their PhD explains surprisingly little of the variation in output, suggesting it is difficult to evaluate research skill ex ante. This implies that the marginal entrant is relatively close to the average, which mutes changes in quality as federal funding expands the pool. Moreover, the quantity of PhDs is highly sensitive to changes in federal funding. I use a new model of endogenous growth to quantify the effects of the roughly 40% reduction in federal support for R&D at universities proposed in the President's Budget Request for Fiscal Year 2026. The long-run effects of these cuts depend crucially on their duration: a 3-year cut generates a peak reduction in growth of roughly half a basis point and welfare costs of roughly 0.1% of permanent consumption, while a 23-year cut leads to a peak reduction in growth of 7 basis points and welfare losses of approximately 1%. Finally, I find that a revenue-neutral reallocation of funding from R&D tax credits to support for universities improves welfare by 0.28%.