MIT enrolled a fifth fewer graduate students after losing $100 million in federal awards

The count and the percentage together imply a base. Nearly 500 fewer students at a 20% drop puts the 2024 figure at roughly 2,500 ^[7]. The thing this doesn't tell you is what that base is: the entering class or everyone enrolled. The answer decides whether the drop is one admissions cycle's decision or several cycles' worth.

Dividing the $100 million by roughly 500 students gives about $200,000 per missing student ^[14]. Treat that as a scale marker: the $100 million covers federal research awards to the institution as a whole ^[8].

On cause, the best evidence here is the decision-maker's own account. It covers one campus and one explanation, but a direct one, and a correlation across campuses would say less about why. STAT reports that graduate students are largely supported with these grants ^[11]. "If we admit someone, we don't want them to be partway through their Ph.D. and run out of funding, so we tend to be conservative on the front end," Kornbluth said ^[9]. She added: "I wish I had a crystal ball and knew when things were going to turn around, but for now we're just being cautious" ^[10].

Her case for why the intake matters rests on what universities do. Kornbluth said the very basic end of the pipeline, the unpressured curiosity-driven research behind CRISPR gene editing, microRNA and this year's Nobel-winning optogenetics work, is what has uniquely flourished at universities ^[12].

We think STAT's forecast about the midterms holds for the NIH piece. The plan for political appointees to gain more control over NIH grantmaking concerns who decides which grants are funded ^[3], and an admissions office has to price that in before it makes offers. It is still a forecast about a vote that has not been held.