Mike Lauer is the former Deputy Director for Extramural Research at the National Institutes of Health. A cardiologist and researcher, he joined the NIH’s National Heart, Lung, and Blood Institute in 2007 as the Director of the Division of Prevention and Population Science. From 2015, he oversaw the NIH’s $32 billion funding program for external research. Since leaving NIH in 2025, he has become an outspoken advocate for fundamental reform in how the federal government supports biomedical research.
We discuss
Why the NIH used to fund 60% of grant applications, and now funds just 10%
How “soft money” forces researchers to fund their own salaries
How distributing lots of small grants wastes everyone’s time
How block grants could fund more breakthrough science
Why researchers don’t get their first independent award until their mid-40s
Thanks to Harry Fletcher-Wood, Shadrach Strehle, and Jasper Placio for their support in producing this episode.
For a printable PDF of this interview, click here:
My colleague Caleb Watney told me you’re one of the deepest thinkers on public science funding.
That’s very sweet of him.
Today we’re going to find out if he’s telling the truth.
You were the Deputy Director for Extramural Research at the NIH. Extramural is the fancy word for grants that go out the door: 80–90% of NIH’s total budget. If one thinks of the NIH as a big investment fund for biomedical R&D, extramural is almost the entire portfolio.
As deputy director, did you get to decide who got funding?
No. It’s obviously a very high-volume system. When you’re dealing with tens of billions of dollars and tens of thousands of grants, the system by definition has to be decentralized, so there’s no one person making the decisions about all those grants. We had a very elaborate system that involved tens of thousands of reviewers, as well as thousands of program staff who together would make the decisions about what grants would get funded.
What was your role?
We had an office within the Director’s office that served as the main coordinating body for the NIH. We were responsible for what you could call the corporate framework of extramural funding. It included some very down-to-earth practical things, such as our IT system, electronic Research Administration (eRA), which handled more than 50% of all the grants in the federal government. Anybody who interacts with the NIH in any way — sending in applications or serving as a reviewer — would interact with eRA.
We were also responsible for grant rules and policy. One thing that took up an awful lot of my time was dealing with non-compliance and integrity cases. We also played a major role in outreach and communications about NIH grant rules and policies.
I’m going to read you some things you said in our first call, and I want you to explain them. You said:
“The system of funding science is fundamentally broken. In some respects, it’s been an unmitigated disaster. It was a house of cards, and it’s not surprising that it’s now falling apart.”
What do you mean?
The system suffers from systemic flaws that go back way before the second and even the first Trump administration. The fundamental problem is that the system, which was originally designed to be competitive, has become hyper-competitive — so much so that it is dangerous and corrosive.
Back in the late 1940s and 1950s, when the NIH grant system started, it was relatively small, and about 60% of all applications got funded. Getting funding would not be a source of a great deal of stress. Scientists essentially had to fill out the forms — they were four pages long. Today, a typical grant application is 100-150 pages long. I’ve seen grant applications that are over 1,000 pages. It’s a different world.
Over time, more and more money was fed into the NIH system. This was a two-edged sword. On the one hand, it was great getting more money. On the other, the scientific community responded by expanding dramatically — more buildings, scientists, graduate students, PhDs — many more than the monies could accommodate.
As time went on, the likelihood of getting funding went down, and the system became excessively competitive. We went from a 60% success rate in the 1950s down to 10% in 2025. The many problems that the grant system has — incremental research, the loss of innovation — are symptoms of the fact that too many scientists are chasing after too few grant dollars. This has rendered the system quite vulnerable.
What do you mean there’s too many scientists? Your job was to encourage and support science happening out in the world. How can somebody in your role say there’s a bubble in science?
I would imagine that most readers have a job: you gain an income from your employer. Your assumption would be that as long as you did high-quality work, your employer would continue to pay you.
That is not what it’s like in science. Let’s say you’re a scientist and you work at a university. There’s a high likelihood that your university doesn’t have the money to pay your salary. Instead, it says to you, “It is up to you to get your salary. You’re going to have to apply for grant money from the federal government.” That means you have to participate in a competition in which the likelihood of success every time you ask for funding is on the order of 10–20%.
As you might imagine, this is a highly stressful state. Instead of thinking about great ideas and experiments you could be designing, you’re worried about whether or not you’re going to have enough money to pay your own salary, as well as the salaries of your staff, six months or two years from now.
The term that we use for this system is “soft money.” When a university hires you, they don’t have the money to pay you, or maybe they only have the money to pay you for a few years. They make it very clear to you that after that you’re on your own. It’ll be your responsibility to bring in your salary, or a big chunk of it. If you fail to do that — because you’re not lucky in this extremely hyper-competitive system — your lab will be closed down and you may very well lose your job.
This is obviously a very high-stress situation. It’s not one that is conducive to great science. What’s conducive to great science is where people can engage in long-term thinking and don’t have to worry about their funding.
This is a succinct and punchy critique of the current system. Did your colleagues at the NIH agree with it?
I think for the most part, yes. Certainly some of my colleagues in the top echelon did recognize that the system was fundamentally broken. There were papers that came out, say 10-15 years ago, with titles like, “Rescuing biomedical funding from its systemic flaws” and “Saving biomedical research.” I don’t think that very many of my colleagues were questioning the premise that the system was in serious trouble.
What counts as extramural? My understanding is that most of these are grants to specific researchers for specific projects.
That’s exactly right. The word “extramural” literally means outside of the walls [Latin: extra muros]. There are a couple of ways a government could support science. One is that the government has its own scientists as salaried employees. That certainly does happen. In fact, the NIH has a rich intramural program — inside the walls — that has produced great scientific discoveries and Nobel Prize winners.
The extramural approach is to fund grants that go mostly to universities and medical schools — research institutions that hire scientists to conduct specific projects. Even there, there are different approaches that can be taken. One approach might be to give universities large sums of money and rely on them to figure out what projects to conduct and which scientists to support. For the most part, that is not the system that we have. We fund lots of relatively small individual projects — something that came out of various events around World War II.
When you say “very small,” will you give me a sense of the scale of the R grants — these classic investigator-driven grants?
The most common type of grant that the NIH funds is what’s called an RPG: a Research Project Grant, not a rocket-propelled grenade. In 2024, the agency issued a little more than 40,000 research project grants, each for about $600,000. To put that in perspective, the overall NIH budget — everything extramural, intramural, and various other expenses — adds up to $48 billion. Each individual grant is a very small part of the overall whole.
I’m trying to do the math in my head — these are something like half of the NIH’s grant-making?
The R grants make up most of the grant-making. The other major component is training. Roughly $1 billion got spent in 2024 to pay for fellowships and training grants; more like $30 billion for research grants. [Here is a long list of all the different NIH activity codes.]
Some folks will not be familiar with how the federal science funding apparatus developed in this particular way. Will you give us the two-minute version?
It’s fascinating. Before World War II, government did not fund very much science. Scientists wanted government to be as far away as possible. They did not want it to be controlling their lives. World War II changes things, because suddenly the government has a need for science and needs it very fast. The government puts a huge amount of money into supporting science. Most famously the atomic bomb — the Manhattan Project. But another big project was the development of methods to mass-produce penicillin.
When the war came to an end, science had been so successful that the government decided that there was a need to continue and perhaps increase its support. The question was how to do that. The NIH started almost by accident. The agency existed, but it was a small laboratory. They found themselves with some extra money because the price of penicillin had suddenly gone down. The person who had my job way back then sent out a letter to university deans saying, “If you have any need for some extra money, please let me know.”
He got an overwhelming response. That was the beginning of the NIH grant program. It did not start with an act of Congress. It started because of a series of accidents. By 1946-47, the program was de facto running. Congress liked it so much that they did make it legal.
You can imagine a lot of systems where the federal government gives money to researchers. But the model we have is one where researchers say, “Here’s the project I’d like to do over the next few years, and I need this much money for it.” Why is it that way?
The model is based on what happened to the Rockefeller Foundation. [For much more on the Rockefeller Foundation’s role in building the modern scientific ecosystem, I recommend Eric Gilliam’s writing.] In the pre-war years, it was the main nonprofit supporter of biomedical science in the United States. Initially, they gave large institutional grants to universities. Then in the late 1920s and early ‘30s, two things happened. One is that they underwent a reorganization, and the second was the Great Depression. Their investments shrank and their financial situation became more problematic.
Lots of readers won’t be aware that the Rockefeller Foundation was the biggest funder of science in the US at this point. It’s funny to think of a private foundation being not just a plurality, but I believe the outright majority of science funding.
Today, if people think about foundations that support science, you might think about the Gates Foundation, which supports a lot of science. But compared to the NIH, even Gates is relatively small.
Facing this reorganization and financial pressures, the Rockefeller Foundation was concerned that if they gave out large institutional grants, they could put themselves at risk. They figured it would be a lot safer to give out small project grants. If a project fails, it’s no big deal. The other thing is that these would be short-term grants — for a few years. They were not obligating themselves to a lot of money down the line.
The person who headed the medical science division at Rockefeller was Alan Gregg — he later won the Lasker Prize. He was vehemently opposed to this. He said that it was going to gum up the works: there was going to be a huge amount of administrative work, and essentially the Rockefeller Foundation would turn into “a dispensary of chicken feed.”
He vehemently pushed — once the NIH became the main funder — not to follow the Rockefeller model, and to go back to the way things were before. But the reason the NIH did follow that model is because universities were now used to it — this was the accepted approach for nonprofit sponsors to support science. Nobody ever sat down and had a thoughtful policy discussion about what might be best. It evolved this way, and this is the system we have now.
In your read, ruthless competition for these grants drives a lot of the paperwork requirements and unpleasantness associated with being a scientist today. Will you say a little more about that?
Again, there’s no way to understand this without some history. During the first 15 years that the NIH ran, the paperwork was kept to an absolute minimum. Essentially scientists were given money — theoretically for a project, but the reality was they could use it however they felt fit. There was remarkably little oversight from government staff.
When this came to the attention of a particular congressman, Lawrence Fountain — a Democrat from North Carolina in the late ‘50s and early ‘60s — he was appalled. He said, “This is no way to run a government program.” To trust people, give them money, and not carefully oversee what they’re doing, made absolutely no sense. He called on the NIH to change their system.
The NIH actually resisted this. Initially, they said, “Thank you very much for your wonderful advice,” and moved on. But Fountain was insistent. There were a series of contentious hearings held in 1962 when then-NIH Director James Shannon said, point-blank, that Congress had it all wrong. “The whole idea, the reason why our system works well, is that we give scientists money and let them loose. For you to ask us to run this like we would run a weapons program makes absolutely no sense.”
Congress would not have it. They said it was totally unacceptable for the agency to run things this way. That was the beginning of the extreme bureaucratization that we have, with hundreds, if not thousands, of laws and regulations, and the insistence that scientists must propose a project and must do the project they said they were going to do.
That was a major series of events in the 1960s. There were three reports put out by the [House] Committee [on Government Operations]. Each was more caustic than the one before it. We often talk now about the poor relationship between politicians and the NIH. Unfortunately I lived through a good part of this and it was very sad to see, but we’ve been through this before. The consequences were quite serious.
I’d love you to explain that explosion in paperwork — what’s in these 150 to 200-page grant applications?
The research plan — the science itself — is relatively short. That’s about 12 pages. The rest consists of a lot of information about:
The administrative structure of the institution;
The facilities;
Performance sites;
Budget, and
Detailed biographical sketches.
Then, depending upon the specifics:
Information about the protection of human subjects;
Vertebrate animals and how they’re going to be used;
Data management and data sharing, how that’s going to happen, and
Biosafety; select agents.
I could go on and on. There’s a long list of requirements, and some people have questioned why all of this has to appear in applications when there’s only a 10–20% chance that they’re going to get funded anyway. “Wouldn’t it make much more sense to not worry about that until later?”
Another way of thinking about this is, how much time do scientists spend on administrative issues as opposed to doing science? This is something that has been carefully studied by the Federal Demonstration Partnership. They have conducted well-designed surveys, starting in 2005, then every six or seven years. They showed the same thing: scientists are spending about 45% of their time related to federally-funded research addressing administrative requirements, not doing science.
There’s no way that you’re going to be able to completely eliminate administrative requirements. The administration is important and it has to be done properly. But when people are spending 45% of their time handling paperwork and only 55% doing science, you’ve got a big problem.
How much control does the NIH have over the content of that paperwork? Some of it is required in statute, like the parts about biological agents and national security. How much leverage did you have internally to make other parts of the grant application shorter?
There is some leverage, but not much. We can think of this in two ways. On one level, there are explicit statutes regarding the protection of animals and people, export controls, agents, biosafety and so forth.
But it’s also an understanding of congressional expectations. In many respects, these have not changed since the 1960s. The expectations are that the agency will run a very tight ship. That may sound like a good thing, but it means that you err on the side of more administration, more oversight, and more data to be collected. There is some discretion, and a number of my colleagues — I give them great credit for it — have attempted to make the system simpler and remove some of the administrative steps.
Another problem is that we have dozens of federal agencies that support research, and every single one has their own culture, laws, and regulations. Sometimes they contradict each other. One agency may say, “Thou must do A.” Another agency says, “Thou must never do A.” Then if you’re the poor grants administrator at the university and you’re sending out grants to both agencies, somehow you have to manage to keep it straight. The agencies also use different computer systems. That creates all kinds of problems. It’s something where the agencies do have some control if they want. But I would say for the most part, the degree of leverage is relatively small because it’s congressionally-directed.
What are the most onerous paperwork requirements for scientists? If you could go to Congress and say, “Get rid of these two or three requirements,” what will give scientists the most time back?
One would be to figure out ways to streamline the requirements related to human subjects protections and animal welfare. These are very important, but they’re way too complicated.
Another is filling out all the information on the initial grant applications. One suggestion has been that the initial grant application should be maybe 15 pages — 12 pages of the research plan, 1 page that has a high-level look at the budget, and 1 page with additional stuff. That would take a lot less time to put together. Then — only for those 10–20% who are lucky enough to get funded — you ask for all the additional information. That could potentially save a lot of time, and it would not decrease the amount of information that the government gets on the grants that it funds.
One of the classic streams of economic evidence that we think about a lot at IFP is a set of papers that say, “You could double or triple the tax dollars that go to R&D, and the payoff for society in cancer drugs, geothermal energy, and whatever else you want to put in there would still be positive.”
How would you go about doubling the amount of money the NIH spends and limit this horrible rat-race dynamic?
One big question is whether we should be focusing on individual projects — as opposed to scientists, programs, departments, or even universities. You asked me earlier, “What’s wrong with asking scientists to stick to the projects that they proposed?” One of the problems with this project-based approach is that it’s based on a false premise. Science is fundamentally different than, say, remodeling a kitchen. I hire a contractor. I know exactly what I want. I work it out with the contractor that he’s going to remodel the kitchen in this particular way. I hold him to it. If he doesn’t do it, then there’s a potential breach of contract.
With scientific grants, it’s incompletely specified by definition. We don’t know what exactly is going to happen:
The hypothesis may turn out to be wrong.
The experiments may yield surprising results.
New technologies may come on the scene which completely change the way that one thinks.
There may be new public-health threats — nobody predicted COVID-19 until it happened.
All of a sudden the needs change.
The very nature of science is that you don’t know what you’re going to be doing over the next year — for sure not five years. I saw this interesting commentary by Peter Lawrence, a zoologist at Cambridge University, who said that we’re asking scientists to become bureaucrats and astrologers. They’re supposed to write an astrology about what they’re going to be doing over the next five years, and then stick to this particular fiction.
One of the other reasons we have a surplus of scientists is that the incentives are all wrong. The grants that we fund support salaries. If you’re a university, you like getting grants, because then you don’t have to pay your faculty’s salaries. In fact, as I mentioned before, you go so far as to say, “We’re not going to pay your salaries. You’re responsible for getting your salaries.” But it’s even worse than that, because the way grants are structured, we not only pay for the cost of the research, we also pay for the overhead. It’s called indirect costs.
We did a whole episode on this topic, going far deeper than any of our readers wanted us to.
I listened to that and I thought it was very well done. The problem here — I’m not going to talk about the wonkiness of overhead costs — I’m talking about the incentives. Not only will the grant pay for the professor’s salary, it will also pay a bonus on top of that, in the form of indirects. That means universities are incentivized to hire people that they do not have the resources to support.
It gets worse than that, because the same thing applies to graduate students and postdocs. Grants will pay for their stipends, their tuition, as well as overhead on top of their compensation. That means universities are incentivized to bring in more graduate students and postdocs — irrespective of whether they have any hope of getting an academic job when they’re finished. We have this whole layer of perverse incentives, which is creating this bloated system.
This all makes perfect sense in terms of the incentives for the institutions. But let’s say I was considering grad school in a hard science, and I’d seen this insane rat race for grants. There are way fewer opportunities at the postdoc level, or to have one’s own lab, than there are people chasing those opportunities.
If I were a talented young scientist, couldn’t I go into industry, or to some big pharmaceutical company, and avoid this whole rigmarole entirely?
That’s absolutely correct. From an economic point of view, it would be absolutely irrational to go down the academic route. We do see that fewer American students are going to graduate school and getting postdocs in the biomedical sciences. We’ve become increasingly reliant on foreign students and postdocs to fill those spots. They are willing to come in and accept a situation that an American person would not accept, because they think they’ll get outstanding training — which they could potentially take back to their home country — or that they might be able to stay and be successful here in the United States.
You mentioned a lot of your colleagues shared this general critique of the way that the NIH doles out money. So why have we been in this regime for 75 years?
There are some fundamental questions that have not been answered. One is, why should the government fund science in the first place? Is it because it’s the right thing to do — if the government doesn’t fund basic science, nobody else will? Or should it produce a direct benefit to the public? If so, what exactly does that mean? Does that mean a direct benefit next year or five years from now? Or would we accept a direct benefit that may be ill-defined, and happen 20-25 years later?
I’m going to put some meat on this, or some fat on this. Back in the ‘70s and ‘80s, NIH and the Veterans Administration (VA) funded some scientists — both on the intramural and extramural side — to conduct work on sugar metabolism. That included work on a hormone called glucagon. We can think of it as the anti-insulin: insulin reduces blood sugar; glucagon increases blood sugar. There was a group of scientists in various parts of the country who were doing very basic research on how sugar metabolism works, on how glucagon is made. It’s something that at the time seemed extremely esoteric.
“Basic research” in this context means there’s no immediate commercial application.
Yes, this is curiosity-based research. It is, “I want to better understand how nature works. I’m not thinking about a drug that I’m going to be able to produce within the next few years and get approval from the FDA, or a device I’m going to be able to sell. I’m just interested in how the world functions — pure science.” Some people might call it science for science’s sake.
Should the government be supporting this? Back in the ‘60s, ‘70s and ‘80s, when this work was being done, nobody could say where this was going, including whether it was going to go anywhere. It eventually led to the blockbuster GLP-1 weight-loss drugs, like Wegovy, Ozempic and Zepbound. They’re now the top-selling drugs and have dramatically changed the way we think about obesity — that’s why I brought up fat. These drugs were not developed to help people lose weight. They were developed as a potential treatment for diabetes. Then it turned out that they also induced a great deal of weight loss. They also prevent heart disease. It’s an amazing story.
This is a success story where the government supported basic science that a company would not support, because there was no clear benefit in the short term. Also, companies could not be guaranteed to keep all the benefits. There’s nothing to stop other companies from taking advantage of that knowledge and making money themselves.
So this is a fundamental question and it has not been answered. Back in the ‘60s — and we’re talking now about the very top of the executive branch of government — President Johnson said, point blank, “We’re spending too much money on basic research.” [For more on this episode, see here.] As you can imagine, this sent the scientific community into panic. Eventually things settled down and stayed the same. But one reason why we are where we are is that fundamental questions like this — “Should the government be spending a big chunk of its money on basic research? Should we be paying for salaries for university faculty? Should we be paying for more graduate students and postdocs?” — were never answered.
I don’t know if that answers my question.
These debates about the role of government in funding science — whether to fund basic or applied research — they’ll always be with us. That’s a question for a democratic society. Politicians and citizens can have all kinds of perspectives about what the goal of federal science funding should be. That’s totally normal and expected.
But that’s a separate question from “Which science funding mechanisms should we use?” The grants we use to fund science at the NIH are almost exclusively these time-bound, project-bound individual grants. But it’s not obvious to me that politicians firmly believe that this is the most efficient way to fund science, or to get the most high-value science.
I feel like, whether you want more basic research or more applied research, everyone should want to experiment more aggressively with the way that we fund science.
I completely agree with that. Part of the issue here is, who should be making these decisions about how the system runs? Politicians and their proxies, or technocrats — scientists, either out in the field or in the government? This question also has never been satisfactorily answered. There’s been a continuing tension, going back to the very beginning.
When the post-World War II science apparatus was being set up, there was a big debate going on. On the one hand were the Democrats, primarily led by Harley Kilgore from West Virginia, who said, “The politicians should control this. Yes, the government should be supporting science in universities. But it should be the politicians who decide how the system works — they should be all over it.”
On the other hand, you have Vannevar Bush and Cassius Van Slyke, the initial head of the NIH grants program. They were taking the exact opposite stance, saying, “No, the only role of the government is to write a check. The scientists should have complete control — not only over the science that they do, but also the policies that govern how the science should be conducted.” This tension’s been going on for a long time.
Part of the reason we’ve got the system that we have is that nobody’s ever done a careful assessment of, “What’s the way we can get the best of both worlds?” The best of the world in which the politicians are calling the shots is that there’s accountability, transparency, and responsiveness to public needs. The best of having the scientists run the show is that they have the expertise, and arguably a higher degree of competence, to make sure that a program runs well.
The other point you’ve emphasized is you don’t know in advance what you’re going to find. So pre-specifying the outcomes or gating the funding until you produce the product you promise is harder. You want to leave some room for scientists to find things out and change course as they learn.
You can almost think of this as an investment. You put together a diverse portfolio with a range of assets. Some of them are going to be extremely successful, but since you don’t know what they’re going to be, you invest in a wide range. That’s what venture capitalists do. They try to get their money into many different places. Most of their investments will not do very well, but a few are going to do so well that it more than justifies everything else. That’s very similar to what we have in science.
It’s an interesting metaphor, and one that we’ve been talking about a lot here at IFP. In terms of the number of investments, the system does very well at “diversifying” — 40,000 RPGs in a given year. But each of those grants is relatively small. You can’t fund a lot of the breakthrough science with huge upside on just $600,000 over three years.
My colleague Caleb Watney likes to say, it’s like our science portfolio is 90% bonds with a 3% expected return. We’re massively underdiversified in terms of the expected returns of the scientific portfolio.
That’s a very interesting way of thinking about it. Also what that does is fix your commitments. Yes, each individual project is very small, and so the potential loss from each project is quite small. But you’re giving up a huge amount of flexibility, which is inherent in the value of science. Overall there’s going to be a net loss. The other thing is, if 45% of your most important people’s time is being spent on administrative work, and much of that is because they’re attending to a large number of relatively small projects, that is going to be inherently wasteful.
One way I sometimes think about this is, “bureaucratic units.” I don’t know whether that’s a formal term, but say a grant application or an annual award is a bureaucratic unit. Each of those units come with a whole set of administrative tasks. NIH will issue 60–70,000 awards every year — competing awards, new awards, as well as renewals. We also get up to 80,000 applications a year. When you put all that together, that’s well over 100,000 bureaucratic units. Every single one entails work, and opportunities for mistakes. They drain the system.
An alternative way we’ve argued you could go about federal funding for science is something we’ve called “X-Labs.” The idea is a new kind of grant — a big block grant to an outside institution. Maybe it’s a university — maybe it’s a different research entity that has lower overhead or is more incentivized to pursue these breakthrough innovations.
The National Science Foundation has just announced a very similar program, Tech Labs. Generally, and for the NIH in particular, it sounds like you’d be supportive of something like this. What form do you think it should take?
I would be extremely supportive. The key to having something be successful is that it’s got to be big enough.
Let’s do a thought experiment: say you have a university that’s getting $100 million a year, and that comes in a very large number of small grants. To make it easy, assume that all of that money is going into basic science, in laboratories working on a wide variety of things.
We’ll tell that university, “We’re going to give you $90 million — one grant.” Now we only have one bureaucratic unit, not thousands. “With this one grant, you’re going to run a basic science program.” We can define — and Congress could help do this — what work would be allowed — what can be paid for and not. For example, “We can pay for salaries for faculty and staff, but not graduate students and postdocs. That should be handled someplace else.”
Then we leave it up to the university to figure out how to spend that money most wisely. They know who their scientists and team players are. They know who’s collaborative. They know what their strengths are. They would be in a position to make an appropriate judgment.
The system has to be accountable. I could imagine various ways that could be done. One is regular financial audits to make sure that the monies are being handled properly. The other would be retrospective scientific audits: we would tell the university, “We expect you to produce some great science. We are not expecting that every single scientist in your staff will produce great science, because we know that’s not possible. But we expect you to set up an environment that will ensure that somebody somewhere — actually somebodies somewhere — within your system will produce great science. We’re going to use that retrospective review, conducted by experts, to determine whether you will continue to get funding.”
In order for this to work, it has to be big enough that scientists are not going to be incentivized, implicitly or explicitly, to spend their time writing grant applications. Their incentive should be to do great science — that’s what they should be spending their time on. The administrative work should be solely that which is necessary to get the science done, not to deal with government administrative issues. If the system is too small, then the scientists are going to still be writing grant applications on the side.
If I’m filling out another application for $150,000 and I’m doing it on company time.
That completely defeats the purpose.
The other argument I’ve heard for big block grants is that they let you do a specific kind of science you can’t do on smaller budgets. You can’t pay for team-based, leading-edge science unless you’ve got capital to spend. Nickel-and-dime grants here and there will never add up to enough to buy the piece of equipment that will let you do the breakthrough experiment.
That’s absolutely true. A block grant system enables a wide diversity of work to be funded. For example, it would be very reasonable for universities to devote some of the money to what you might call “skunkworks” — relatively small exploratory projects. You give people six months to see whether there’s anything there. If it works, great. If it doesn’t work, no worries, we move on to something else. But it also gives them the resources to buy expensive equipment and put together large teams of scientists to do more ambitious projects.
I think, ironically, something where there’s only one grant would allow for a greater degree of diversity than what we have now, where the university is saddled with a whole bunch of individual small projects.
The system incentivizes research institutions not to go for big swings. There’s no upside to trying to chase down a huge grant for something that has a high likelihood of failure.
One of the upsides of what you’re recommending is you get some political guardrails, but you also say, “You’re going to be judged on your portfolio. If you can get some big hits — the next crazy weight-loss drug — we’re going to forgive shots on goal that were creative and didn’t pan out.”
Not only forgive; I would be stronger than that. We fully expect that failures will happen. In fact, we would be very worried if you come back to us and say, “I have 500 scientists in my institution and all 500 produced great work.” Then I’m worried that you’re being too conservative. I certainly expect all 500 of them to have managed their money responsibly. I expect them to run their labs in a civil and safe way. But I expect a large proportion of them won’t have much to show scientifically, because that’s the nature of the beast.
While we’re on the topic of institutional reorganizations, I want to ask you about the different Institutes and Centers in the NIH, the ICs. There’s a few handfuls. That’s why it’s the National Institutes — plural — of Health. Some are more commonly known — big disease-focused ones like the National Cancer Institute (NCI), the National Institute of Allergy and Infectious Diseases (NIAID), and the National Institute on Aging. They get the vast majority of NIH dollars. But then there’s a whole bunch of them that don’t often get public attention.
There have been arguments that the institutes should be consolidated. Project 2025, for instance, said there should be fewer. [NB: I was thinking of OMB here, not Project 2025.] It’s not just a right-wing view — there have been all kinds of perspectives on this over the years. What’s your take?
I used to say, “I work for the National Institutes of Health. The key word is the second word — institutes — and the key letter is the ‘s’ that comes at the end.” There are 27 institutes and centres. That means that at any given time, there are 54 different opinions. So it was a difficult place to work.
Why double the number of opinions?
I’m being a little sarcastic, but it’s a remarkably fragmented agency with each unit doing their own thing. They have their own culture and set of rules. There are certain rules that are common to the entire agency, but the degree of variation is remarkable.
On the one hand, that was a good thing because it enabled some innovation. But the downsides were much worse. It’s also problematic for scientists because if you submit a grant to the National Cancer Institute, your likelihood of getting funding — even before 2025 — was maybe 10-12%. If you submitted a grant to the Basic Science Institute [the National Institute of General Medical Sciences], it was 30%.
Why should there be such a big difference? It has to do with the fact that the agency is split up into these administrative units. If it were entirely up to me, there should be one institute, the National Institute of Health. That’s actually what it was. The discussion that we had about giving one block grant to a university — the same could be applied to the NIH. What Vannevar Bush proposed back in 1945, with his Endless Frontier report, was one agency to cover all civilian research for the government: physics, chemistry, computer science, and biomedicine. The more bureaucratic units you have, the messier things are, the more difficult they are to manage.
How different are the individual institutes? Does each director have their own fiefdom?
There’s an enormous amount of variation. For example, the National Cancer Institute and NIAID devote a larger proportion of their funds to intramural research. The NCI devotes a fair amount of money to research centres. That means there’s less money available for the relatively small grants.
Some institutes decide what grants they’re going to fund by scores from peer review committees — the numbers alone. There’s very little thinking that goes into it. I’m being a little glib, but not by much. Then there are other institutes, like the Basic Science Institute, where the peer review scores play a major role, but are only one component of a more holistic decision. Those cultures are quite different.
When you look at the size of the different institutes, you see a pretty consistent pattern: institutes that have an associated patient advocacy group are much bigger. The National Cancer Institute gets 20% of the budget of NIH — there are a lot of people with a direct stake in that. Whereas the Basic Biomedical Science Institute does not have the same massive advocacy group.
How do advocacy groups shape how the institutes work?
They play a major role. The reason why the National Cancer Institute is as big and powerful as it is, is because Mary Lasker was a strong proponent for research — essentially a lobbyist; an extremely effective one. She, along with a number of her colleagues, successfully lobbied the government, including President Nixon, to pass the National Cancer Act of 1971, which led to the National Cancer Institute becoming much bigger and more powerful. In 1948, the National Heart Institute was formed. That was also a result of a lobbying effort. The advocacy groups have played a major role in how the agency is organized.
The institute directors pay very close attention to the advocacy groups, because they want their support. It translates into lobbying on Capitol Hill. I was reminded a gazillion times that, as a federal employee, I was not allowed to lobby. But the advocacy groups could lobby — and they did, very effectively. It’s not a stretch to say that I’m going to pay a great deal of attention to their wants and desires, because I want them on my side.
If I’m head of the National Cancer Institute, I can’t lobby Congress, but I can take meetings with the cancer advocacy groups and say, “Your voice is so important and it needs to be heard on Capitol Hill”?
The National Cancer Institute may be a bit of a tenuous example, because the director is a political appointee. But the other 26 institutes — let’s say the head of Heart, Lung, and Blood — what you said is exactly right: “We greatly value your support.”
There’s been a lot of turnover among the heads of Institutes and Centers (ICs). Let’s say I’m suddenly placed as head of one of them. What advice would you give to a new institute head?
I did have conversations with new institute directors. One very important piece of advice I gave was, “You should have an extremely low threshold to get in touch with me and ask even the stupidest question, because I could potentially get you out of trouble.”
Let me get to something very practical. When you run a grants competition where only 10-20% of applicants are going to be successful, you’re going to have many unhappy people. Some of them are going to be high-powered academics who know you from your previous life. They will try to lobby you for help on their individual grants. Or they may ask why their grant was not funded.
The piece of advice I would give is, “You need to stay as far away from this as you possibly can. Your role is to advocate for the agency and the public, not for individual scientists. Stay out of these small fights, because otherwise they will eat you alive. They’ll take up all your time and make you miserable.” You punt it off to your staff and let them handle it. But we tell people, “We’re very sorry that we can’t fund your grant because we don’t have the money.” That’s not a bureaucratic excuse. That is the truth. You don’t want to get into an argument about details on specific projects.
The main message that I gave IC directors is, “You’re in the government now. It’s a different environment than academia. There’s a different set of norms and rules. If you’re going to be successful, you don’t have to like them, but at least you have to respect them. I don’t mean this in a pejorative way, but if you pretend that you’re still in academia, you’re going to get yourself into a lot of hot water, and you will not be able to accomplish what you want to do.”
Did that happen — IC heads come in, act like they’re still academics, and get burned? Imagine I’m that guy: what am I doing wrong?
One thing you may be doing wrong — let’s put Trump aside here, before Trump — is that the administration may have a certain view of the world, and want to impose a particular policy, and you think that it’s stupid. You have every right to think that. But then you pretend you’re still in the university, where you can essentially say what you want, because academic freedom rules above all else. You’re open about your objections — so much so that you get in the way of the administration’s desire to accomplish a particular goal. That is a great way to not do things well.
In a university — let’s say I’m a department chair and I disagree with the university policy. I can be open about it. I’m tenured. My dean may not love me, but that’s the culture. In the government, it’s not quite like that. I have been in countless situations where I did not agree with what the higher-ups were doing. The way to handle it is behind closed doors. You present your arguments. You explain why you think there may be a better way of accomplishing whatever it is they want. In this discreet way, things get negotiated.
Ultimately if you’re not happy with the way things are turning out, you have two choices. You go along with it and help them out, even though you disagree with it — or you leave. To act like you’re in academia and scream out protests is not a wise strategy.
Mike, what did I forget to ask you today?
We did talk about the questions of faculty salaries. This was a big problem that came up in the early 1960s, when President Eisenhower put together a commission to look at the government-university partnership. Eisenhower strongly supported that partnership. But just as he was concerned about the military-industrial complex, he was also concerned about a bio-technical complex, or elite.
His advisory group expressed concern about government grants being used to support faculty salaries. They were afraid that it was going to create a soft money system. Nonetheless, they said the government should go ahead and support faculty salaries. They did not come up with a way of dealing with the problem. This set off the bubble that has only gotten worse, and has led us to our hyper-competitive state.
Another reason why that block grant system would work so well is that you say to the university, “You can use this money to support salaries and pay for research. You can also use it to support your facilities and administrative costs. But you only have a limited amount. You figure out how best to use it.” They would want to spend enough on indirect costs to provide appropriate support and oversight. But they wouldn’t want to spend too much, because that would take away from the science.
How does your typical institute head think about the difference between the intramural research they support and the extramural — the grants going out the door? Is one of them more valuable to society than the other? Is one of them more exciting to work on?
Some of our institute directors care very deeply about their intramural program. They have much more control over it — these are their scientists. They’re right there in Bethesda. Intramural is fundamentally different from extramural. You’re not bound by short-term grants, so you can think much more long term and develop programs accordingly. Scientists have their laboratories, research groups, and a budget assigned to them.
After four or five years, they undergo a formal, retrospective review. It’s very different from extramural, where peer reviewers are asked to predict what’s going to happen in the future. Here, the reviewers are asked to look at what a scientist has been doing over five years and render a judgment accordingly.
If the scientist is doing outstanding work, they will maintain their budget.
If their work is not so good, their budget may be shrunk.
If it’s entirely unsatisfactory, the laboratories may be closed.
I’ve had an opportunity to see this in action. It’s very rigorous. The results of the program speak for themselves. Nobel Prize winners have come out of the intramural program.
I went to a fascinating lecture by Barney Graham. He was one of the key people who helped develop the COVID and Respiratory Syncytial Virus (RSV) vaccines. He worked in Vanderbilt, and then came to the NIH intramural program, where he developed this vaccine laboratory. He figured out a way to modify viral proteins so that they could be used for effective vaccines like the RSV and COVID vaccines.
I asked him afterwards, “Could you have done this work in the extramural world?” He said, “Absolutely not.” Within the intramural environment, he could take risks. He did. Some of what he did at the time seemed crazy. In retrospect, it was brilliant. But he didn’t have to worry about whether his lab was going to be funded next year or the year after. He could think long-term. It is a fantastic model for how to do research.
You’re saying we should create the conditions for extramural research to look more like that — to take these big swings for home runs, instead of for singles.
That’s exactly right.
One last thing I want to ask you about is the talent pool getting these extramural grants. The average age of grant recipients has crept up over the past few decades. There are arguments that, as science deepens, you’ve got to spend more time learning to get to the frontier of a field. There are counter-arguments: “If we want more breakthrough science, we should reverse these calcified structures that mean that a young gun is likely not getting a grant — it’s going to somebody who could be their father or mother.”
What do you think?
The fact that somebody doesn’t get their first independent research award until they’re in their mid-40s…
45 on average, right?
Yeah, it’s nuts. To put this in perspective, I went to medical school, went through internal medicine and cardiology training, and started taking care of patients when I was in my late 20s — as a fully-trained doctor. Somehow society thinks that’s okay.
But you need 15 more years before you should expect to start getting federal support for research.
Before you can become an independent scientist. I could be an independent doctor in my late 20s, but I could not be an independent scientist until my mid-40s. Absolutely nuts.
It’s a symptom of two things. One is what we’ve talked about: the inn is simply too crowded. It’s more difficult for people to get in. Then something very specific: back in the early ‘90s, mandatory retirement went away. That meant that successful scientists — I’ll define that as somebody bringing grant money into their university — didn’t have to retire when they turned 65. Of course they stayed on board. They were doing good work and bringing in money to the university. The universities were very happy.
As a result, the overall age of the workforce increased, because we had far fewer people retiring at the time they previously would. If they’re staying in the system and bringing in grant money, less is available for younger investigators who have not yet had an opportunity to put together a proven record. This is an unintended consequence of eliminating mandatory retirement.
