American research universities have long treated the federal reimbursement of facilities and administrative costs as one of the quieter supports for scientific work. These costs, called indirect costs or F&A costs, cover expenses that cannot be assigned to a single grant but are necessary for the grant to proceed at all. The National Institutes of Health changed that arithmetic on February 7, 2025, when it announced it would cap indirect cost reimbursement at 15 percent on new and existing grants, effective February 10, 2025. Some of the affected institutions had negotiated rates above 60 percent.

That announcement followed a broader episode in January 2025. On January 27, the Office of Management and Budget directed federal agencies to pause the disbursement of federal grants and loans. A federal judge in Washington, D.C., temporarily blocked the directive the next day, and the Office of Management and Budget rescinded it on January 29. What remained was the indirect cost cap, and it did not disappear when the broader memo did.

What a 15 percent cap actually touches

Indirect costs pay for the unglamorous infrastructure that makes grant-funded science possible. Safety officers who inspect biosafety cabinets, staff who maintain animal welfare records, cybersecurity teams who protect patient data, and accountants who certify that each dollar is spent according to federal rules all draw support from the same pool. A 60 percent rate is not a bonus; it is the product of a negotiated calculation that reflects square footage, regulatory burden, safety requirements, and the particular risks of a research portfolio.

The administration's position was that many private foundations reimburse overhead at far lower rates and that universities should absorb more of their own operating costs. That argument has force when restricted to the question of whether the federal government should pay as much as it does. It does not follow, however, that a defensible policy goal can be applied retroactively to grants already awarded without notice, comment, or respect for signed award terms. Universities made multiyear commitments to maintain staff, animal colonies, long-running data systems, and specialised equipment that cannot be wound down in a week.

The first injunction: from 22 states to nationwide scope

Within three days, the cap met a lawsuit from a coalition of 22 state attorneys general. On February 10, 2025, Judge Angel Kelley of the United States District Court for the District of Massachusetts issued a temporary restraining order that protected those states. A companion lawsuit brought by the Association of American Medical Colleges, the Association of American Universities, and other research organisations sought wider relief. On March 5, 2025, Judge Kelley converted that emergency protection into a nationwide preliminary injunction.

The ruling rested on the Administrative Procedure Act. The court concluded that the National Institutes of Health had likely changed a substantive reimbursement standard without the notice-and-comment period that federal law requires. The judge also found that plaintiffs had shown a real probability of layoffs in research administration, interruption of clinical trials, cancellation of long-running studies, and a slowdown in new award spending. The NIH Guide Notice NOT-OD-25-068 remains the agency's formal statement of the cap, even as courts have blocked its enforcement. The underlying cost principles sit in Title 45, Part 75 of the Code of Federal Regulations, a framework the National Institutes of Health did not amend before issuing its notice.

Photo by Point & Shoot on Unsplash

Grant terminations as a separate legal problem

The indirect cost cap was not the only change arriving in university mailboxes in spring 2025. The National Institutes of Health also sent termination letters for existing awards, citing agency priorities and recent executive orders. The areas named in those letters included research on transgender health, LGBTQ populations, vaccine hesitancy, and programmes that use terms tied to diversity or equity. Unlike the indirect cost cap, which applied uniformly to an institution's budget sheet, each termination was addressed to a specific principal investigator and a specific laboratory.

The legal claims differed accordingly. A university harmed by the cap can show a financial loss to its entire research operation. A principal investigator harmed by a termination can show loss of data collection, loss of staff, and damage to work that may have taken years to design. In a separate action, a federal judge in Maryland has issued orders requiring the government to return certain terminated grants to active status while litigation proceeds, a step that reflected the court's view that the terminations were likely not justified under the statutes that govern National Institutes of Health awards.

The January freeze still shapes the interpretation

Courts have looked back to the January 27 memorandum because it revealed how broadly the executive branch was prepared to treat grant conditions. The memorandum from the Office of Management and Budget was rescinded before most agencies could implement it, but its logic continued in the terminations and in the rate cap. Legal scholars have read the sequence as a test of whether an administration may use existing grant agreements to impose new policy preferences after money has already been obligated.

The answer matters beyond biomedical research. A grant is a legal instrument, and the research community is learning how much of that instrument rests on statutory text, how much on administrative practice, how much on congressional appropriations language, and how much on the willingness of universities to sue. The pattern in 2025 has been that universities and states did sue, and courts responded quickly, though the substantive questions remain open on appeal.

Other systems handle overhead differently

The United Kingdom separates core research funding from project-specific grants more explicitly than the United States. UK Research and Innovation pays a separately negotiated share of full economic costs on project grants, while block funding through Research England supports the underlying research environment. Germany applies a programme allowance to many grants, and the European Commission's Horizon Europe programme uses a flat rate of 25 percent for indirect costs in most non-profit research actions. None of those systems is without argument, but none has produced the equivalent of the February 10, 2025, shock, in part because their rates are written into programme design rather than layered onto existing award terms.

The United States, by contrast, has negotiated institutional rates that vary widely. That variation is one reason the 15 percent cap was both easy to explain and difficult to defend: it treated a university with a high-security biosafety facility and a small undergraduate college as if their overhead burdens were identical.

Photo by Navy Medicine on Unsplash

What the rulings mean for researchers looking ahead

Universities are not waiting for final appellate rulings before making decisions about which positions to advertise. A biomedical department that expected to search for an assistant professor may postpone the search if its indirect recovery is frozen at a level far below its negotiated agreement. A postdoctoral researcher whose salary comes from a grant in a targeted area may receive a non-renewal notice even if the grant was not itself terminated, because the principal investigator is uncertain about the next year.

The productive response, according to research administrators, is to distinguish between the two legal tracks. An award that remains active and unobligated has different protections than an award that has been terminated, and the obligations in the notice of award do not disappear simply because a court has blocked part of the agency's policy. Researchers should keep their institutional sponsored projects office informed and should archive the original notice of award, the approved budget, the negotiated indirect cost agreement, and any dated correspondence in one place. That practice does not resolve the litigation, but it positions the university to act when a ruling lands.

The precedent being set in 2025 extends beyond indirect costs and particular grant topics. It is about whether a change in administration may lawfully alter the terms of existing research awards by notice, or whether it must go through the slower administrative path that has governed federal research funding for decades. The courts have leaned, so far, toward the slower path. Whether the executive branch accepts that answer will determine how confidently American research universities can plan their next grant, their next hire, and their next building.