Research

Working Papers

Leaders and the Direction of Science: Evidence from University Presidents (Solo, JMP)

Abstract: Research universities allocate substantial resources across scientific fields, yet little is known about how leadership shapes these allocations. This paper studies whether university presidents influence the direction of scientific production within their institutions. I construct a novel dataset combining hand-collected biographical information on 1,772 presidents (1,958 presidential spells) across 146 U.S. R1 universities from 1900 to 2024 with institution-by-field-by-year measures of research output from OpenAlex. I measure alignment between a president’s academic background and each field using Jaccard similarity.

The empirical strategy exploits within-institution variation across fields over time. In specifications with institution-by-field and institution-by-year fixed effects, identification comes from comparing changes across fields within the same institution-year, absorbing all institution-wide shocks. I find that fields closer to the president’s background experience increases in publications, researcher counts, and field share, consistent with a reallocation of scientific activity rather than an expansion of total output.

To address concerns about endogenous selection, I exploit plausibly exogenous leadership transitions driven by unexpected departures. Event-study estimates show no evidence of differential pre-trends and a gradual, persistent increase in aligned fields following transitions. Mechanism analyses indicate no effect on total or federal R&D, but suggest a reallocation of nonfederal resources and a significant increase in PhD production with a multi-year lag, consistent with changes in hiring and training.

The effects are present throughout the sample but are stronger in STEM fields, public universities, and in the post-2000 period, consistent with the increasing importance of coordination in modern scientific production. Finally, using a Jaffe-weighted measure of technological proximity, I show that increases in output in aligned fields at one institution are associated with increases in output in similar fields at technologically proximate universities, consistent with knowledge spillovers.

Work in Progress

  • Innovation in India (with Christian Fons-Rosen)
  • Lost Science

Publications

Coming soon