On February 7th, 2025, the National Institutes of Health
(NIH) announced a
sweeping policy change that sent shockwaves through the academic research
community: overhead reimbursements on extramural grants will now be capped at
15% for all domestic institutions.
This is a seismic shift for elite universities that have
grown accustomed to overhead rates often exceeding 60%, and their reaction has
been exactly what you would expect from institutions suddenly cut off from a
major revenue stream. Universities are in an uproar, fearing massive budget
shortfalls and disruptions to longstanding financial models.
For decades, concerns have been raised with NIH’s overhead
reimbursement system and how it shapes the culture of academic science. The
intent behind overhead payments is to cover the hidden institutional costs of
conducting research—expenses such as administrative support, facilities, and
maintenance. However, in practice, the system has metastasized into something
far beyond that.
Universities have come to depend on overhead as a critical
and flexible source of fungible income, and this has influenced institutional
priorities in ways that are not always aligned with the best interests of research.
The current funding structure has inadvertently shifted priorities, placing a
premium on securing large grants, often at the expense of fostering
intellectual curiosity and risk-taking in research.
With the recent decision to cap NIH overhead rates at 15%,
there is understandable concern and uncertainty. This shift will be painful for
many institutions that have planned their budgets around expectations of
significant overhead reimbursements that may not materialize, and adjusting to
this new reality will be difficult.
While the transition may be abrupt, it provides an
opportunity to reassess the way research funding is structured. By moving
towards a model more similar to that of the National Science Foundation (NSF),
where direct costs are emphasized and grant sizes are more modest, we have the
chance to realign incentives in medical science, away from massive,
bureaucratic studies toward more innovative and creative research.
This moment, though painful, offers a chance to correct a
long-standing imbalance. While universities adapt, this shift could ultimately
benefit the broader scientific community by prioritizing research quality over
institutional financial strategies. The challenge now is to rebuild a funding
culture that values discovery and insight, rather than continuing to reward the
pursuit of ever-larger grants because of the overhead they bring.
NIH Overhead: The Silent Engine of Medical School
Profits
The way overhead works is that if someone is awarded
$100,000 in direct costs, the university gets an additional percentage of that amount
to cover “indirect costs” of research.
Before Friday that would have been more than $60,000 extra at elite
institutions but now has been cut to $15,000. This is intended to cover the indirect
costs of doing research, including administrative support, office space, utilities,
IT support, library facilities, building maintenance, safety and research
protection services, etc.
Disparities in negotiated overhead rates have exacerbated
inequality among institutions. Elite universities, such as Harvard, Yale, and
Johns Hopkins, have secured overhead rates exceeding 60%, while
the average overhead rate has reportedly hovered
around 27%, such that many institutes must operate with significantly less
financial cushion.
Negotiating individual overhead rates separately with each
and every institution is a complex
problem, leading even President
Obama to consider imposing a flat rate for all sites, which would promote greater
equity between institutions. Currently, well-endowed universities only get
richer, not just because they get more grants, as that is compounded by their
higher overhead rates, supposedly justified because they have fancier equipment
and more expensive facilities to maintain.
While 15% seems low, it should be pointed out that foreign
entities are only eligible for 8%
overhead on NIH grants, and yet they still eagerly accept such funding,
implying that those universities recognize that NIH research funding clearly
benefits them, even without enormous overhead.
Unlike most academic departments, medical schools do not
have to pay the full salaries of their own research professors, as NIH grants
are expected to cover them. By contrast, NSF grants, which fund non-medical
scientific research do not pay faculty
salaries during the academic year, as they consider research to be part of
one’s normal institutional duties. This is yet another reason there are much
higher professor to student ratios in medical schools than elsewhere in the
university.
This fact also leads to a system where salaries are
generally higher for medical researchers than those of other scientists.
Why? Because the salaries, while
determined by the university, are paid largely by NIH, and as the salary goes
up, so does the magnitude of the university’s overhead payment. By contrast, in Mathematics or Anthropology,
salary costs are typically paid directly by the university.
The NIH Overhead Game: Bigger Grants, Bigger Profits,
Worse Science
Indirect costs are real, but the current system incentivizes
researchers to seek ever-larger grants to sustain institutional budgets. Every
new grant brings a fresh injection of overhead funds, which universities use to
prop up budgets across the board. The
more money a professor brings in, the more overhead the university gets, and
this relationship is key to career advancement for medical faculty.
This led to an institutional preference for bigger, more
expensive grants, rather than better science, leading to the devaluation of
small, hypothesis-driven research or theoretical modeling in favor of behemoth
multi-center data collection studies that guarantee massive long-term overhead
payouts, and ultimately to a system where faculty spend more time writing grant
proposals than doing research.
Worst of all, the large multi-center studies prioritized
because of their big budgets are not only the least creative research
scientifically but are also the worst for education and science culture. Because of their scale, those studies emphasize
homogenizing research protocols over many sites, to make sure everyone does
everything in exactly the same way, so their data can be easily combined for
joint analysis, guaranteeing small, incremental increases in knowledge but
rendering revolutionary innovative discoveries and creative original thinking virtually
impossible.
This leaves little room for the people doing the work to
engage in real scientific thinking.
Instead, scientists become little more than technicians, processing
samples, running prepackaged statistical analyses, executing pre-defined
protocols akin to recipe books. This is
very bad for the very culture of academic science as it concentrates thinking
in a small number of powerful voices, with everyone else following orders. While
large-scale projects provide stability, they can sometimes prioritize institutional
security over fostering groundbreaking innovation.
The system rewards universities not for good science, but
for maximizing grant revenue streams.
The more faculty applied for and secured NIH grants, the more overhead
flowed into institutional budgets. Medical
school faculty are often discouraged for applying for NSF or private
institutional grants that do not pay equally high overhead rates, with some
universities like California Institute of Technology refusing
to accept grants that pay less than 20%, for example.
Many years ago, a senior administrator came around to the
basic science departments in our medical campus and commented that although our
papers, grants and awards were as good as or better than the other top tier
medical schools, the size of our grants was smaller, and hence so was the
resulting overhead.
His response was to ask what he could do to encourage us to write
more grants - not better grants, not grants for more impactful research, but
just more grants, because he needed more overhead to keep the campus in the
black. If it were only funding the indirect
costs of the funded research, this complaint would make no sense, would it?
This past week, the Stand Columbia Society, a group of
Columbia University affiliates dedicated to “advocating for Columbia University’s
core mission of excellence in teaching, learning, research and patient care…”
sent out an alarmed message claiming that the 15% overhead cap could cost Columbia
anywhere between $114 and $202 million dollars.
They did not express concern that cutting overhead would hurt scientific
research, but rather about its devastating effects on non-scientific
programs.
In
their own words, the $348.9 million of indirect costs Columbia received in
2024 were fungible, and they “effectively cross-subsidize diverse academic
programs” that could not sustain themselves, graduate tuition subsidies,
faculty start-up and retention bonuses, and they “indirectly subsidize the arts
and humanities which receive no comparable grant support. This cross-subsidization is essential.” If
Columbia could afford to use NIH overhead to pay for all these things outside
of science, why should taxpayers keep footing the bill as research costs?
If the government wants to subsidize such things, let them
do it explicitly, rather than through the backdoor. This highlights
long-standing concerns that high overhead rates may have provided financial
flexibility beyond the direct costs of research.
And it’s certainly nothing unique to Columbia. Some institutions even give investigators
some of their indirect costs back to use as “unrestricted funds” to be spent on
whatever unexpected expenses come up in their labs. This suggests that their universities did not
need the full amount of overhead paid to cover research costs either.
To be completely fair, universities are simply playing the
game according to rules set up by the government. It is completely reasonable that they take
this approach when the system supports backdoor subsidies. By negotiating these
high overhead rates, universities benefited for decades, but now the system is
undergoing a necessary correction.
Big Genetics, Big Money, No Breakthroughs: How NIH
Funding Went Wrong
For decades, I was a
vocal critic of former NIH Director Francis Collins, long before
it became fashionable to do so, arguing that genome-wide
association studies (GWAS) and the Human Genome Project (HGP) would fail to
deliver meaningful clinical breakthroughs. I was right. The promises of
personalized medicine and revolutionary treatments never materialized. Instead,
NIH doubled down on bigger and bigger genetic studies, not because they were
finding anything, but because they had no choice but to keep going.
Collins had staked his
reputation (and NIH’s budget) on the idea that sequencing the human genome
would transform medicine. But when smaller GWAS studies failed to uncover
actionable genetic variants, they couldn’t admit defeat. Instead, they scaled
up, arguing that bigger sample sizes were needed. When those studies failed,
they scaled up again. At this point, the Human Genome Project and GWAS were too big to fail—NIH had
to keep funneling money into them to justify the massive initial investment.
Universities were all too happy to play along, because these
massive projects were ideal for securing huge indirect cost returns. Unlike
small-scale, investigator-driven research, GWAS and other massive multi-omics
projects required vast infrastructure, computing power, data coordination
centers, and armies of administrators - all of which translated into higher
indirect cost payouts for universities.
But the scientific returns never matched the investment. If
a study needs half a million subjects to detect an effect, that effect is too
small to matter. The NIH’s focus on big science crowded out hypothesis-driven,
high-risk, high-reward research, replacing it with data hoarding and
statistical fishing
expeditions.
The new NIH overhead cap may finally force a reckoning.
Without the perverse incentives of massive indirect cost payouts, universities
will no longer have a financial motive to prioritize scale over substance. NIH
itself will have to rethink its addiction to big science, redistributing funds
toward smaller, more intellectually rigorous projects that actually test ideas,
rather than just generating more data.
Maybe, just maybe, this reset will bring us back to the kind
of science that actually moves medicine forward.
Why the Trump Administration Had to Rip Off the
Band-Aid on NIH Overhead
It is fair to argue that this policy change was too radical
and too fast, and that universities should be given time to adjust. Their budgets have been developed on the
assumption they would receive a predetermined amount of overhead that suddenly
seems unlikely to materialize. While I strongly agree with the decision to reform
the overhead system, to improve the culture of science, and to set overhead
rates to a flat universal rate, changing it suddenly on a Friday afternoon
without warning could have serious short-term consequences, especially for
universities that lack a large endowment to fall back on.
Even for rich elite universities like Columbia, unexpectedly
losing $200,000,000 is no walk in the park.
Personally, I might have suggested the rate cut should apply only to
newly awarded grants, for example, to give some time for universities to gradually
adjust to their new reality. A 4-fold rate
cut from 60% to 15% might better have been phased-in gradually, to allow the
system to adjust, giving universities time to prepare by pursuing other sources
of funding and modifying their long-term budget plan accordingly.
Normally, when dealing with addiction, you don’t just rip
away the supply overnight, you wean people off.
But the Trump Administration did not give universities methadone, they
threw them straight into detox.
The backlash from elite universities was immediate, and as
expected, they’ve turned to the courts to protect their financial windfall. Earlier
this week, a federal
judge temporarily blocked this rate cut following a lawsuit filed by 22
states. If history is any guide, the
higher ed lobby will continue to push for delay or even reversal of this reform
before it can take full effect. Whether the Trump administration sticks the
landing will depend on whether they anticipate this resistance and hold firm
against the usual DC pressure.
Given the resistance to change of any kind in Washington, it
is likely that the Trump administration felt that the only way to make sure
these overhead rates really do get cut is to simply do it suddenly and
unexpectedly before the system could rally to stop it. After all, when they pursued
a similar option in 2017, it was quickly quashed.
The Overhead Gravy Train Just Stopped—And Universities
Are Scrambling
The professoriate and university administrators are out in
full force challenging the cuts. Several
recent articles have amplified the outrage from medical schools, claiming that
these cuts will cripple research and cost lives, but when you break down their
arguments, it’s clear they are more worried about losing money than losing scientific
progress.
For example, the Association
of American Medical Colleges declared that the cuts would “diminish the
nation’s research capacity, slow scientific progress, and deprive patients,
families, and communities across the country of new treatments, diagnostics,
and preventative interventions.”
NIH is still fully funding direct research costs: salaries,
lab supplies, equipment, and experiments. By reducing overhead payouts, more
funds will be available for new research projects, potentially increasing
the number of grants awarded and diversifying the range of funded science.
What they really mean is that universities will have less
“fungible” money to spend on non-research priorities, and that the institutes
that got the highest overhead rates (like Harvard and Stanford) will no longer
have such an enormous financial advantage over smaller schools.
More Reforms Are Coming—Here’s What NIH Should Do Next
Overhead reform is a good start, but NIH’s problems run
deeper. For decades, it has funneled billions into bloated megaprojects,
entrenched an old boys’ network of grant recipients, and allowed universities
to profit off taxpayer-funded research while bearing none of the financial
risk. If Trump wants real reform, he’ll need Congress—because one of the
biggest problems can’t be fixed by executive action alone.
The Bayh-Dole
Act of 1980 is why universities can patent federally funded discoveries.
NIH pays for the research—salaries, equipment, and overhead—but when those
projects yield lucrative patents, institutions keep the profits. Columbia, for
example, has made nearly $790 million from
Richard Axel’s co-transformation
patents. This isn’t a scandal - it’s how the law works. But there’s
something wrong with a system where taxpayers take all the financial risk, yet
universities get all the reward.
Bayh-Dole reform would require new legislation, but
solutions exist. Universities could be required to reinvest a portion of patent
profits into federally funded research or let NIH recover its investment before
institutions take full control of revenues. Another option is partial
government ownership of patents, perhaps through President Trump’s proposed sovereign
wealth fund, ensuring taxpayers see a return. If universities want full
control of their discoveries, they should fund the research themselves.
Beyond patents, NIH’s funding model needs a reset. A small
group of entrenched researchers control a disproportionate share of grants,
locking out early-career scientists and smaller institutions. Some
investigators are investigators on ten or more grants simultaneously, creating
a monopoly that stifles competition. Just as market monopolies limit
innovation, capping the number of simultaneous NIH grants per investigator
would open funding to a broader range of researchers, with a wider range of
ideas. This would be consistent with
global norms – in Finland, a given investigator can only be named on one active
grant at a time from the Academy
of Finland, their largest domestic science funder.
But the biggest problem is Big Science bloat. NIH has spent
decades dumping money into massive multi-center studies that generate endless
data but few breakthroughs. When early GWAS failed to deliver, NIH didn’t pivot
- it doubled down, demanding bigger sample sizes and larger budgets.
Universities adapted to this model because it guaranteed long-term funding. The
result? A system that rewards predictability over risk-taking, institutional
security over transformative science, minor incremental progress over potential
revolutionary breakthroughs.
NIH needs to break this cycle. Instead of $40 million for
one bureaucratic megaproject, fund forty $1 million grants supporting
high-risk, high-reward ideas. The safest science is rarely the most important -
but under NIH’s current structure, it’s most likely to get funded.
Part of the problem is that NIH bureaucrats - not scientists
– too often set the agenda. Funding priorities are increasingly driven by
Requests for Applications (RFAs) - pre-determined categories dictated by NIH
program officers who aren’t actively conducting research. This governmental
central-planning approach pushes politically safe or bureaucratically favored scientific
priorities over independent, investigator-initiated proposals. The best way to
fix this? Reduce RFA-driven funding and restore competition letting the
marketplace of ideas rather than government administrators set the agenda shaping
the future of medical discovery.
Finally, there’s the issue of transparency. Universities
rely on NIH overhead payments to sustain research infrastructure, but those
funds have also been used strategically beyond science. If overhead is covering
non-research expenses, the public has a right to know. The answer isn’t
eliminating overhead—it’s full disclosure of how taxpayer dollars are spent.
The backlash to Trump’s overhead cap proves how deeply
embedded NIH money is in university budgets. But universities aren’t the
problem - the system is. With its built-in cycle of government dependency, NIH
has spent decades rewarding institutional size over scientific creativity.
If Congress wants real reform, it should start with
Bayh-Dole—ensuring taxpayers see a return on investment. NIH should end
monopolized funding, prioritize scientific competition over bureaucratic
stability, and demand transparency in spending.
This isn’t an attack on universities, it’s a fight to save
science from bureaucratic inertia. The best institutions will thrive. The
bloated ones will have to adapt. Either way, the era of unchecked, runaway NIH
spending is coming to an end.
No, NIH Cuts Won’t Kill Science—They’ll Save It
The NIH overhead cap is not a crisis, it’s a long-overdue
course correction. We academics face a painful period of adjustment, but this
reform is an opportunity to prioritize real scientific discovery over
bureaucratic excess.
By shifting funds away from administrative overhead and into
direct research costs, NIH is restoring a healthier, more competitive funding
model - one that will reward creativity, risk-taking, and scientific excellence
instead of conservative megaprojects that fuel institutional bloat. This shift
will allow NIH to fund more projects, support more investigators, and drive
more genuine breakthroughs with the same budget, rather than further
concentrating resources in a handful of elite institutions.
The best universities - those that truly value the
disinterested search for truth - will adapt. Some will streamline operations,
seek private partnerships, or restructure their funding models to be less
reliant on NIH subsidies. Others may embrace industry collaborations, fueling
greater private-sector innovation and breaking free from the culture of
dependence on government-controlled research dollars.
For too long, NIH spending has sustained a culture of
administrative excess instead of fueling true scientific progress. That era is
ending. The universities that embrace innovation will lead the future. The ones
that cling to the old system will fall behind. Science isn’t driven by
government payouts - it’s driven by daring ideas, visionary research, and
fearless intellectual pursuit. It’s time to fund it that way.
