Study Finds Economists at Federally Funded Universities Avoided Flagged Keywords After 2025 Government Word List
Key Takeaways
- •The study uses the 2025 U.S. government blacklist of diversity-related terms as a natural experiment to measure academic behavioral responses to political pressure.
- •Economists at universities heavily reliant on federal funding significantly reduced their use of flagged words compared to colleagues at less-dependent institutions or in the United Kingdom.
- •The observed decline was driven primarily by research content concerning gender, race, and environmental topics rather than superficial wording changes.
- •Effects varied across department quality and author demographics, including gender and ethnicity, while individual funding status alone did not explain the pattern.
- •The paper is a CEPR discussion paper that has not yet undergone peer review.
A research paper by Dominic Rohner, Oliver Vanden Eynde, and Philine Widmer examines how academics respond to political pressure by studying what the authors describe as a natural experiment: the publication in early 2025 of a "blacklist" of words flagged by the U.S. government. The list emerged amid a broader shift in federal research policy under the second Trump administration, which moved to restrict diversity-related terminology and tied grant compliance to specific language standards.
According to the paper's abstract, the release of this list led to a sharp reduction in the use of flagged words among economists at universities that rely heavily on federal funding. This reduction was measured relative to scholars at institutions less dependent on federal funding or based in the United Kingdom. The authors report that the decline was primarily driven by content related to gender, race, and environment.
The researchers contend that the observed changes were not merely semantic but reflected actual shifts in paper content. They also found that neither individual funding status nor time-invariant author characteristics accounted for the effects. The paper documents heterogeneous effects across department quality and author gender and ethnicity, and concludes that scholars respond strongly to political pressure. The study's design—comparing institutions by federal-funding dependence—follows a standard econometric approach for isolating behavioral responses to policy changes without direct experimentation.
The authors frame their findings in terms of "Science under threat," which appears in the title of their paper. Writing on Marginal Revolution, the blogger who surfaced the study acknowledged that dynamic but offered an additional interpretation, suggesting the findings may also reflect the removal of incentives for what he termed "pandering."
The paper is available via the Centre for Economic Policy Research (CEPR) at As a CEPR discussion paper, it is a working paper that has not yet undergone peer review.