The 15% Indirect Cost Cap Has Been Struck Down, Blocked by Congress, and Proposed Again. Why It Won't Die.

When a university receives a $1 million NIH grant, the researcher does not get $1 million. A portion of that award — typically between 30 and 65 percent of direct costs, depending on the institution — goes to the university itself as reimbursement for the facilities and administrative infrastructure that makes the research possible. Harvard's negotiated rate is 69 percent. MIT's is similar. A large state university might negotiate 55 percent. These are not arbitrary markups. They are rates negotiated individually with the federal government, audited by the Office of Naval Research or the Department of Health and Human Services, and tied to documented costs for buildings, laboratories, utilities, compliance offices, institutional review boards, and the administrative machinery required to manage thousands of simultaneous research projects.

In May 2025, the National Science Foundation announced it was capping these reimbursements at 15 percent for all new awards — a unilateral reduction that, for most research universities, represented a 70 to 80 percent cut in overhead recovery on NSF grants. The next day, thirteen universities and three higher education associations filed suit. Seven weeks later, a federal judge struck the policy down. Congress then blocked it in the FY2026 appropriations bill. And on April 3, 2026, the FY2027 budget proposal landed with the same 15 percent cap — this time targeting NIH, the nation's largest research funder.

The indirect cost cap refuses to die because it is not really about indirect costs. It is about a fundamental disagreement over what the federal government owes the institutions that conduct its research — and the answer to that question will determine whether American universities can continue to function as the world's dominant research enterprise.

What Overhead Actually Pays For

The term "overhead" invites skepticism. It sounds like waste — administrative bloat skimming money from the scientists who do the real work. This framing is politically effective and substantively wrong.

Facilities and Administrative (F&A) costs fall into two categories. Facilities costs cover building depreciation, operation and maintenance of research space (including specialized ventilation for chemistry labs, biosafety containment for infectious disease work, and clean rooms for semiconductor research), utilities, and library resources. Administrative costs cover departmental management, regulatory compliance, sponsored program accounting, and the institutional infrastructure that satisfies the federal government's own requirements for grant oversight.

A researcher studying protein structures needs a laboratory with temperature-controlled equipment, a building with appropriate electrical capacity, a compliance office that manages their institutional review board protocols, a grants office that tracks their expenditures against federal regulations, and a library that maintains access to the journals where they publish and read results. None of these costs appear in the researcher's grant budget. All of them are essential to conducting the research the grant funds.

The rates sound high — 55 or 65 percent — but they apply only to a subset of direct costs (not to equipment, tuition, or subcontracts above $25,000), so the actual overhead payment on a typical grant runs closer to 25 to 33 percent of the total award. And these rates are not self-reported. They are calculated according to Office of Management and Budget Circular A-21 guidelines, negotiated with federal cognizant agencies, and subject to audit. Universities cannot simply declare a rate and collect.

The Legal and Legislative Battlefield

NSF's May 2025 announcement that it would cap these rates at 15 percent — without notice-and-comment rulemaking, without congressional authorization, and without explanation for why the cap applied only to universities and not to other grant recipients — triggered an immediate legal challenge.

The plaintiffs included Arizona State, Brown, Caltech, the University of California system, Carnegie Mellon, the University of Chicago, Cornell, the University of Illinois, MIT, the University of Michigan, the University of Minnesota, the University of Pennsylvania, and Princeton — joined by the American Council on Education, the Association of American Universities, and the Association of Public and Land-grant Universities. They argued the cap violated the Administrative Procedure Act by being "arbitrary and capricious" and by bypassing required rulemaking procedures.

In June 2025, Judge Indira Talwani of the U.S. District Court for the District of Massachusetts agreed, finding "severe deficiencies in the agency's reasoning" and "errors of law." The ruling — the first final judgment against the 15 percent cap — vacated the policy entirely. Three other federal agencies that had attempted similar caps faced parallel legal challenges, and the NSF ruling set the precedent.

Congress reinforced the judicial outcome in the FY2026 appropriations package. The Commerce, Justice, Science; Energy and Water Development; and Interior and Environment Appropriations Act (P.L. 119-74) explicitly required NSF, DOE, NASA, and the Department of Commerce to continue using the negotiated indirect cost rates in effect during FY2024. The legislative language left no ambiguity: Congress considered the cap unlawful and unworkable.

And yet the FY2027 budget request, released April 3, 2026, proposes the same 15 percent cap — this time extended to NIH. The persistence reveals the administration's view that indirect costs represent discretionary spending that can be redirected to direct research, and that the political argument against "university overhead" is potent enough to survive repeated legal and legislative defeats.

The $432 Million Question

The government's own estimates put the cost of the NSF cap at $432.3 million in reduced FY2024 funding for universities. Extrapolated to NIH — which awards roughly five times as much in grants as NSF — a 15 percent cap applied to the nation's largest research funder would strip billions annually from the institutions that conduct the majority of federally funded research.

The University of Illinois calculated it would lose $23 million per year under the cap. Institutions with negotiated rates above 60 percent — Harvard, MIT, Stanford, Caltech, Johns Hopkins — would face proportionally larger losses. But the impact would not be limited to elite research universities. Regional state universities with negotiated rates of 45 to 50 percent would lose the overhead revenue that funds the laboratory maintenance, building operations, and compliance infrastructure they need to remain competitive for federal grants in the first place.

The downstream effects compound rapidly. A university that loses 70 percent of its overhead recovery on new NSF grants has three options: absorb the loss from institutional funds, reduce the infrastructure that supports research, or decline to accept the grants at all. The first option strains already-tight university budgets. The second degrades the research environment that attracts faculty and graduate students. The third — declining federal grants — sounds implausible until you consider that a grant with insufficient overhead recovery is, by definition, a money-losing proposition for the institution hosting it.

Several university administrators have described the cap's implications in precisely these terms: accepting a grant at a 15 percent rate when your actual costs run at 55 percent means the university subsidizes 40 percentage points of infrastructure costs from other revenue. For a single grant, that subsidy is manageable. For a research portfolio worth hundreds of millions of dollars, it is financially unsustainable.

Why the Cap Keeps Coming Back

The political logic is straightforward. "Cutting university overhead" sounds like eliminating waste. Voters and legislators who do not understand the audited cost-recovery system hear "universities charge 65 percent overhead on federal grants" and interpret it as profiteering. The administration's framing — redirecting overhead dollars to "direct research" — suggests that cutting indirect costs would put more money in researchers' hands.

This framing misrepresents how the system works. Indirect cost reimbursements do not come from the researcher's grant budget. They are additional funds calculated on top of direct costs. Reducing the rate does not increase the money available for experiments — it reduces the money available for the building where the experiments happen, the compliance office that ensures the experiments are legal, and the grants office that reports the experiments' outcomes to the federal government.

But the political appeal of attacking overhead is durable precisely because the system is complex and the beneficiaries — university facilities departments, compliance offices, library systems — lack the emotional resonance of a scientist losing their lab or a patient losing access to a clinical trial.

The FY2027 proposal's extension of the cap to NIH suggests the administration views each appropriations cycle as a new opportunity to advance the policy. Even if Congress blocks it again — and the bipartisan coalition that rejected it for FY2026 remains intact — the repeated proposals create planning uncertainty that erodes universities' willingness to invest in the research infrastructure the rates are designed to sustain.

What Research Institutions Should Do Now

Document actual costs with precision. The strongest defense against the 15 percent cap is transparent, auditable documentation showing exactly what indirect cost recovery funds. Institutions that can demonstrate that their negotiated rate reflects real, necessary costs — not discretionary spending — position themselves and their congressional delegations to resist future proposals.

Build the economic impact case at the district level. The same logic that makes NIH's $2.57 economic multiplier a powerful argument against direct funding cuts applies to indirect cost recovery. Every overhead dollar that funds a building custodian, a lab equipment supplier, or a compliance analyst generates economic activity in the surrounding community. Making that case district by district gives appropriators concrete reasons to preserve the current system.

Watch the appropriations language carefully. The FY2026 law requiring agencies to honor FY2024 negotiated rates expires at the end of the fiscal year. If the FY2027 appropriations process stalls — as it often does — and agencies operate under a continuing resolution, the legal status of the protection becomes ambiguous. Institutions should track this timeline and prepare for scenarios where the congressional block lapses before new appropriations language is enacted.

Diversify the funding portfolio. The indirect cost fight is one front in a broader assault on the federal research funding model. Institutions that depend heavily on a single agency — particularly NSF, which has been the primary target — face concentrated risk. Expanding into DOD, DOE, foundation, and industry-sponsored research reduces exposure to any single policy change.

The 15 percent cap has been proposed, implemented, litigated, vacated, legislatively blocked, and proposed again within 11 months. The cycle will repeat because the underlying political incentive — appearing to cut wasteful spending — persists regardless of the legal or economic merits. Universities that treat each victory as final will be caught off guard when the next proposal arrives. The institutions that survive this fight will be the ones that invest in making their cost structures transparent, their economic impact visible, and their congressional relationships strong enough to withstand the next round. Granted helps research teams find the right funding opportunities across agencies, so you can build the diversified portfolio that insulates your work from whichever version of this policy debate lands next.