The National Institutes of Health (NIH) is by far the largest provider of federal R&D funding to American universities. This month the agency issued new rules concerning how those funds are to be accounted for.
It sent academia into an existential panic.
- “The nation’s universities and academic medical centers were reeling from a directive by the Trump administration to slash funding for medical research, a decision that doctors and scientists said would have a devastating effect on studies aimed at finding treatments for diseases such as cancer, diabetes and heart disease.” – The New York Times (February 7, 2025)
Elite R&D-heavy universities claim they may lose billions of dollars in federal support for medical research. Smaller schools fear they may be shut out of the game entirely. Criticism of the measure has been severe.
- “[It] will cause chaos and harm to biomedical research and researchers in hospitals, schools and institutes nationwide. A sane government would never do this.” – Jeffrey Flier, Former Dean of the Harvard Medical School
- “Illegal and arbitrary… nothing short of catastrophic” –U.S. Senator Patty Murray (D-WA), Vice Chair of the Senate Appropriations Committee
- “A disaster for science… an affront to the separation of powers” -– the Universities’ Lawsuit (February 10, 2025)
- “‘It would be absolutely devastating. In fact, it would close down some research institutions,’ said Dawn Bonnell, vice provost for research at University of Pennsylvania. ‘It’s just unthinkable…’” – reported by PBS
- “Poorly conceived… devastating” – U.S Senator Susan Collins (R-ME), Chair of the Senate Appropriations Committee
The broader societal consequences are portrayed in dire terms: “decreased innovation” … “loss of jobs”… “shuttering of labs”… “a decline in the nation’s capacity to compete globally.”
Twenty-two states filed a lawsuit to stop the NIH from proceeding with the measure. A group of leading universities filed separate lawsuit calling the move “flagrantly unlawful.”
At risk – some say – is “Americans’ quality of life.”
All this, triggered by… an accounting maneuver?
The Cap on University ‘Overhead’
The NIH announcement imposes a cap of 15% on the amount of that universities can charge for indirect costs –- often called “overhead” – on federal grants received for medical research.
The term refers to the percentage of government R&D funding that the universities can use to pay for things other than the actual research itself – such as administrative expenses, depreciation of buildings, interest on debt, legal bills, facilities maintenance, library operations, liability insurance… heat, light, water, lawn care, snow removal, roof repair… i.e., “overhead.”
Accounting for indirect costs is a tricky business. But why is there such alarm over what sounds like a bookkeeping matter?
A Bad Look?
Because it is not about bookkeeping. The outrage is not over what “indirect cost” means, or even how it is accounted for. It is not really about the 15% cap (a version of which has actually been on the books – the Federal Code of Regulations – for decades).
What has ignited this furor is something much more basic. It is about a few seemingly simple facts that are easy to interpret in a negative light – it is a “bad look.” Whether justifiably or not, universities’ treatment of “indirect costs” for federally funded R&D has begun to exhibit an appearance of impropriety, even scandal.
[And if “scandal” sounds excessive, it is not. The issue has troubled universities for decades, and has led in the past to forced resignations of university officers, congressional investigations, prosecutions for fraud, and major financial penalties. We will review the historical context of this issue in a future column.]
The Controversy, In Four Charts
Concern about excessive overhead rates on federally funded university research is not new. It has been the focus over 30 critical reports issued by the Government Accounting Office (GAO) since 1980, along with dozens of Congressional hearings, and numerous studies by the Congressional Research Service (CRS) and other groups in and out of government.
The controversy emerges from four simple charts.
Chart 1: Overhead Charges by Leading Universities For Government-Sponsored Research
Leading American universities apply overhead rates far higher than 15%. The average rate today for the so-called “Ivy+ schools” is 63.2% – four times higher than the new cap. It is also four times higher than the guidance offered for many years by the Office of Management and Budget (OMB) as the default overhead rate for schools that do not negotiate individualized rates. (The OMB guidance rate was recently raised from 10% to 15%, effective in October 2024.)
[Sources: Harvard – Stanford – Caltech– MIT – Duke – Northwestern – Other Ivys]
Since the 1960s, universities have been allowed to negotiate with the federal government to set their overhead rates, a budgeting game – if we can call it that – which some institutions have played very skillfully. That is what changes under the new NIH directive. This is the sacred cow that is being fitted out for the slaughterhouse.
(It is important to say here that the fiscal pain that these schools anticipate may not actually entail a reduction in direct research funding. It involves a new limit on their ability to recover overhead costs not directly related to the research. There is an argument that this could even mean more funding for “real research” — more “bang for the buck.” We’ll examine that contention in the next column.)
Charts 2 & 3: Overhead Charges by Leading Private Foundations
Private foundations also fund much scientific research. Their overhead rates are capped at much lower levels than the university rates.
[Sources: Gates Foundation, Carnegie Foundation, Robert Wood Johnson Foundation, Moore Foundation, John Templeton Foundation, Russell Sage Foundation, Packard Foundation, Ford Foundation, MacArthur Foundation]
A broader picture is available from George Mason University, which maintains an up-to-date list of the indirect cost reimbursement rates for 59 major private foundations.
50 of these 59 foundations have indirect cost rates of 15% or less. 18 of them do not pay for overhead at all.
Corporate Funding for Academic Research
Another benchmark of sorts is the rate charged by profit-making corporations that sponsor university research. In the last ten years, private industry has accounted for 5-6% of total R&D funding at universities (about $6 Billion in 2023 — which is similar to the amount of funding from nonprofit foundations).
Unfortunately, data on overhead rates for corporate-sponsored research is hard to come by. However, in my experience, private sector funding sources are very reluctant accept to traditional (high) overhead rates applied to federally funded R&D. Private companies always pay close attention to the “bang for buck” – and they tend to see overhead as not belonging to the “bang” category.
One indicator of corporate willingness to fund overhead may come from the rules governing the National Science Foundation’s grants for Industry/University Cooperative Research Centers (I/UCRC’s). This type of research center has a long-established tradition and a standard fiscal model. The NSF explicitly limits overhead to 10% or less.
- At least 90% of the IUCRC program income must be used to support direct costs of the research, and [only] up to 10% may be used to support indirect costs.
Chart 4: Comparing Benchmarks
Thus, a review of current and proposed indirect cost allocation percentages makes clear that the 60-70% rates typically assessed at leading universities are much higher than other benchmarks, actual and proposed.
It is also clear that the 15% rate for the NIH cap – regardless of its merits – is in line with several precedents.
But is it the right rate?
The Right Rate
- “Indirect cost rates of 70% are likely a thing of the past.” – a Harvard professor, writing in The Wall Street Journal (February 24, 2025)
Indirect costs are real. Universities do have to pay to “keep the lights on” in order for research to take place. The question is: what is a reasonable and accountable charge for keeping the lights on? How much money can or should be diverted from the research itself to pay for overhead? If a 15% cap is too stringent (as critics of the new rule contend), and 60-70% is excessive, where is the right balance?
This question has arisen repeatedly in the past, spurring many attempts to define a fair rate. In 1994, following several episodes of university misconduct related to improper accounting for “indirect costs” (see Part 3 of this column), Congress imposed a cap of 26% on the administrative component of overhead (which still holds today). But a second component of overhead – “facilities” – was not capped. It is through this loophole that the high rates seen today at the elite schools has been able to grow.
Several private foundations have studied the question.
- Based in part on a 2009 article published by Stanford University highlighting what the authors called the “nonprofit starvation cycle,” both the Ford Foundation and the MacArthur Foundation have recently decided to raise their indirect cost caps to 25% and 29% respectively.
- The Howard Hughes Medical Institute – “the United States’s largest private funder of medical research” – does not pay for indirect costs as such, but they do provide a 25% carve out for “non-research” activities, and a negotiated “occupancy payment” to the host institution.
Thus there is support for a middle ground solution, perhaps in the 25-30% range. But the current gap between the bid (15%) and the ask (60-70%) is very large. It implies a difficult negotiation ahead.
Three Uncomfortable Questions
The charts above create prima facie case for a reconsideration of university overhead rates. But the universities don’t seem to come off well. Three uncomfortable questions arise.
Bloat?
First – there’s the aforementioned bang-for-buck problem. If 70% of the money that is supposed to go to R&D is “siphoned off” (Elon Musk’s contentious phrase) for non-research purposes, it just doesn’t sound like good business practice to most people. It starts to sound more like a tax – a rather exorbitant tax – and no one likes high taxes.
On top of that, many people have read about what has been called the administrative “bloat” in higher education. At Harvard, for example, there are now more administrators than students (!). Across the country, staffing for university administration has grown faster than faculty headcounts and faster than student enrollments. Is this linked to high overhead charges on research? It is easy to make that assumption.
Elite universities also have huge endowments, which takes some of the edge off their complaints. Harvard might “lose” $150 Million or so in R&D support for its overhead, but Harvard’s endowment stands at $53 Billion, having added more than $5 billion last year (which is about 4 times larger than the total tuition revenue of the school, by the way). In other words, Harvard could make itself whole in this crisis by applying just 3% of last year’s portfolio gains to cover the indirect costs that it can longer bill the federal government for. (Of course many smaller schools with smaller endowments don’t have that latitude.)
In any case, whether these facts are relevant or not, they certainly amplify the “bad look” problem.
Lack of Transparency
Second – “the indirect funding system is byzantine, opaque and worthy of re-examination” (in the words of a Berkeley biological research scientist quoted by NBC). No surprise here. The word “indirect” implies a degree of opacity – these are costs that cannot be itemized and attributed to individual projects, and as such, the category is often vague and open-ended. (Within universities, insiders often refer to this as the budgetary “black hole” – where overhead monies scooped from the top of the reserah award disappear, without a clear and public trace from source to use. More to say on this point in Part 2.)
Inconsistent Accounting Standards
The third and most uncomfortable question is: Why do universities employ two different standards of accounting for the same type of activity?
Universities regularly accept grants from private foundations (and private companies) to perform research that is often quite similar in nature to federally funded research, and they accept these much lower rates of indirect cost reimbursement.
Why is a dollar of research funding docked 60 or 70 cents if it comes from the government, but only 10 cents if it comes from Bill Gates?
It’s not an easy question to dodge. There are several possible answers, but none are fully satisfactory (in my view) to explain an overhead rate for federally-funded research that is about 5000 basis points higher than the rate charged private foundations. (We’ll examine the various proposed explanations in Part 2 of this column.)
Summary
This gross discrepancy – this “bad look” – looms over the political debate. There is at least a superficial inconsistency here – and maybe worse? It dampens public sympathy and encourages extreme and unfavorable interpretations. As even some who criticize the NIH policy admit, there is at least a good argument for review and possible adjustment of university overhead policies for federally funded R&D. The 15% rate may be harsh and unwise, but 70% can seem indefensible.
In Part 2, we’ll get into the weeds on the cost accounting questions, to understand better how and why these rates are so divergent. In Part 3, we will contextualize the current controversy in the long history of attempts to reform the indirect cost problem affecting federally funded R&D programs, that has bedeviled several administrations over the last few decades.