How to double science research

Scientists spend 40% of their time chasing grants according to some estimates. Suppose they spend 20% of their time doing something else, such as teaching. That means they spend no more than 40% of their time doing research.

If universities simply paid their faculty a salary rather than giving them a hunting license for grants, the faculty could spend 80% of their time on research rather than 40%. Of course the numbers wouldn’t actually work out so simply. But it is safe to say that if you remove something that takes 40% of their time, researchers could spend more time doing research. (Researchers working in the private sector are often paid by grants too, so to some extent this applies to them as well.)

Universities depend on grant money to pay faculty. But if the money allocated for research were given to universities instead of individuals, universities could afford to pay their faculty.

Not only that, universities could reduce the enormous bureaucracies created to manage grants. This isn’t purely hypothetical. When Hillsdale College decided to refuse all federal grant money, they found that the loss wasn’t nearly as large as it seemed because so much of the grant money had been going to administering grants.

20 thoughts on “How to double science research

  1. Hi John,

    From my perspective as a postdoc, this seems like a huge problem. Many PIs I’ve known spent a considerable amount of time (up to 50%) chasing funds and “administrating”.

    However, I feel a grant system is necessary because people should justify and plan out their proposed research. I wouldn’t want to see an alternative where a university or research institute gets a tranche of money and then doles out funds to researchers.

    That alternative might be more efficient on paper but realistically there’s the risk of wastage on pet projects, favoritism, etc. So you’d still have to justify asking for funds with a grant application!

    Perhaps a compromise to decrease overhead would be to make grants larger and less numerous? It doesn’t take 10x the effort to review a $2 million grant versus a $200 thousand grant.


  2. In Canada, faculty still apply for grants, but the number of grants (and granting agencies) is much, much smaller. Additionally, the “soft money” position so common in the US is almost non-existent up here. Salaries are salaries. Despite this, research doesn’t seem to top 40-50%. The “three pillars” (for tenure review, etc.) are still broken down as R (40%), T (40%) and Service (20%). Administration still eats 50% of the grant money, although no-one is really sure why — it’s some poorly worded “overhead”, which most people assume is code for “using your grant money to pay for something random that has nothing to do with you or your department”.

    At some level, there may be a limit to how much of your life you can spend doing true research anyway — 4 hours per day (your previous post re: Hamming, etc.).

  3. Grants also pay for research assistant (i.e. PhD students’) salaries in addition to faculty salaries. I would guess that research assistants don’t spend a significant amount of time chasing grants, because their faculty does the chasing mostly. I would say the contribution of research assistants would make up for the 40% of faculty’s time in most cases.

  4. Paul: I don’t think universities should doll out internal grants. I think they should pay salaries. It’s how private companies work and how universities used to work.

    I don’t think you need the grant process to evaluate performance, not when it comes as such an enormous price. There’s already the tenure process. And I believe researcher should work on pet projects! That’s what academic freedom is for. The grant system enforces groupthink.

    Wesley: Sounds like the Canadian system is better. But do they really put 40% emphasis on teaching? US universities say they do, but it’s a farce. Imagine two candidates coming up for tenure. One is an abysmal teaching with lots of publications and grants. The other is an award-winning teacher with a few papers and grants. I know who I’d bet on.

    I agree that people don’t have that many hours of concentration a day. But suppose they spent their best hours on grant proposals, as many do. Then they have even fewer mental resources for research.

  5. Universities do in fact pay professors a salary- several tenured members of my department are unfunded. At least in sciences requiring material things, the grant keeps the lab open above a skeletal level, paying for personnel and consumables well above a professor’s pay, typically by about an order of magnitude. Of course, indirects keep the lights on for everyone…

    Hillsdale is a very special case. As was clear from several summers of marching band camp in its bucolic setting, it is clearly a college, with a primarily educational mission. Its refusal of grant money is offset by its ultra conservative politics, which were part of the reason of its refusal- as such the private donations roll in in a manner incommensurate with its size from people who believe that no colleges should get any federal money. But it has no sciences research program of greater note than any other small, regional college in its class.

    What’s more disturbing than take-home pay is the influence grant money has on tenure decisions- historically, this wasn’t even discussed, but it now ranks alongside productivity in consideration at most institutions.

  6. A quote from the wild: “Here professors don’t write papers, they write grant proposals” with the expectation that they spend nearly no time on research but they achieve their research goals and publications by getting funding for their team of students/postdocs.

  7. Grants are not sought out to pay professors salaries. They already draw a salary from the university. Grant money is used primarily to pay stipends and tuitions for the army of grad students it takes to actually do any research. And equipment and travel to conferences etc.

  8. Vivek: At some universities there is an explicit requirement for faculty to cover a certain percentage of their salary in grants. Or it could be implicit: No grants, no promotion.

    Part of hiring someone is providing the equipment and travel they need to do their job. I don’t imagine Google offered you a salary but said you’d have to raise money for your computer.

  9. John: Sorry, I should have been more specific: Here in Canada, most investigators *do* get paid a salary from their university or institute, yet they still spend a lot of time chasing grants to keep their labs populated with people, equipment, reagents, etc. In principle, a single researcher on their university salary could be very productive teaching and writing papers, but the latter might get hard without a lab burning a few $100k per year (unless you’re a mathematician or historian).

    And yes, while the grant process isn’t there to evaluate personal performance (there’s tenure, as you mention), I do think it’s necessary to direct funds to projects that are worthwhile, even in an academic sense. The grant system acts like peer-review, in a way.

    Yet even under a grant system, researchers will always be able spend a part of the lab’s research capacity on pet projects (or “high fliers”) that are too premature for grant funding (I’ve seen this), as long as they fulfill the intentions of their primary grant.

  10. John, in many experimental sciences, such as biology, chemistry, physics, and experimental areas of engineering, the grant money pays for equipment, supplies, and allows the faculty to claim space for those activities. The salary of the professor, even if it is coming from the grant by say 50% is a small fraction of the grant. For example a typical RO1 biology grant might be $200k/yr for 4 years. So 150-175k/yr is going to grad students and equipment supplies etc (depending on how many grants the lab has and how much of the salary must be from grants, and the overhead is calculated separately)

    The biggest problem in my opinion is that grants rely too much on review by grant review boards. This produces a feedback loop, well known researchers who are impressive due to publications and are working on the hot topics get better reviews, and so get more money, so their topic becomes hotter, and they get more money and ….

    In my opinion grant reviews should give some score to the proposal. This score should be divided by the maximal score, add 1, and make this number proportional to the probability of being funded (ie. take all the modified scores, add them up and come up with the normalizing factor to divide each score, then draw one grant at a time at random until all the money is allocated (assuming grants are uniform size, you’d have to adjust for grants asking for much larger quantities of money)).

    In other words, only half of the information used to decide who gets funding should come from the review, the other half should come from pure uniform randomness. This would make it more likely to get funding for things that are probably a good idea, but not so hot, not performed by the trendiest people, not at the trendiest universities, etc. The justification for this is that we really aren’t that good at judging the quality of grants, because if we knew what was going to happen, we wouldn’t need to do research anyway. Picking partially at random helps avoid biases that can keep us wasting money in a small region of the search space when we could have gotten huge rewards somewhere else we didn’t know about.

  11. Daniel: I like the idea of a random component to funding. As in simulated annealing, it can get you away from a locally optimal point so you can explore more effectively.

    There’s a substantial “black market” component to grant-funded research that makes the system work better than it should, sorta like in a communist economy. If people strictly followed the rules, which they don’t, very little would get done.

  12. Another funny thing in Canada where I live the universities (at least any worth having a degree from) are all public. I’m assuming the state schools are the same in the US.

    So, lets get this straight: the education ministry or whatever of your province/state wants a good university and has partially funded it but they need more money. So they get the academics to go begging for grants. Where do the grants come from? In Canada a good 90% comes from government just maybe different branches, National Research Council, NSERC, etc.

    So in short the government spends 100% of the money that goes to the academic but gives away 40% of their time to effectively do clerical work. Then on top of it there is the whole bureaucracy that sure is needed to make sure that the projects are relevant before handing over money but are also doing a whole lot of more clerical or political non-sense (eg. making sure that the professor is from the right region speaks the right language to meet a quota) rather than helping allocate the money based on public good.

  13. Mike: There are only two completely private colleges in America. The rest take public money to some degree. And public schools take private money too, though I imagine not nearly so much as the other way around.

    It’s not just that the grant process uses up 40% of the recipient’s time. It also uses the time of countless administrators who do nothing but keep track of grants.

  14. @Paul Krzyzanowski first post

    Monolithic grants are a great idea. You kind of get that by large institutes (say NRC lab devoted to optical tech, or Perimeter Institute in Waterloo). I like that model reminds me of the stuff that the Max Planck Institutes in Gemany, EMBL etc do.

    The beauty of it is that academics will move to were the funding is. They also will have a large number of potential collaborators and access to equipment that a single lab exploring in that space at a local university might not get funding for (say spinning disk or electron microscopes).

    You also can mandate that the larger body tackle bigger problems say “explorations into the fundamentals of commercializing fusion” rather than give money in drips and drabs to places scattered all over the place and hope that it is enough and that the relevant people find each other rather than continue to churn out incremental papers in their little sub-section of the problem space.

    This also makes the process more accountable to the public: it is easier for the public to question “Do we need more or less money than $200M that went to FusionU this year” than it is for them to question the $200k that went to something like “exploring weak symmetry breaking in SO3 negative meson-baryon interactions at high temperature” + 40 other applications scattered across various levels of government hierarchies for funding. Plus it also allows for negotiations where it makes sense: hey sure you can have an extra post doc as long as I can use your ion-trap etc. Versus now where you pretty much ask for as much money as you think you can get away with and cross your fingers.

  15. universities could reduce the enormous bureaucracies created to manage grants. This isn’t purely hypothetical. When Hillsdale College decided to refuse all federal grant money, they found that the loss wasn’t nearly as large as it seemed because so much of the grant money had been going to administering grants.

    This issue isn’t limited to universities, either, and it’s important to me because I work for a consulting firm that does grant writing for nonprofit and public agencies. One thing I find striking is how little funders understand about how fundees respond to incentives and to funding requirements.

    If funders understood more about the sheer amount of administrative hassle they impose, and the costs of it (directly or indirectly: the money comes from somewhere), I think we’d see something more like a “common application” for many grant-funded projects.

    I’ve written much more about this issue in “Why Fund Nonprofits, Public Agencies, and Other Organizations Through Grant Applications At All?” and elsewhere.

    (Note that I originally wrote a version of this comment for a Hacker News submission.)

  16. In these days of shrunken state and federal funding, any money a university touches, they KEEP! Much of the struggle of a PI is minimizing the pound of flesh a university must be paid for lab space, “services” and “utilities”. I’ve seen some PIs start in-house with a small investigatory grant, then when success is immanent, create outside companies through which to funnel larger grants.

    Some enlightened universities have created “incubators” to reduce the overhead costs and keep projects on-campus. But they still get their cut.

    Other institutions have deep intellectual property hooks that complicate how research is applied and deployed. Many top schools have successfully replaced some budget losses with licensing revenue.

    In such an environment, many other things would have to change before universities could be trusted to do effective grant management, recruit and encourage researchers, and do far more than just pay salaries.

    Most important would be to somehow restore non-grant funding, so that chasing grants becomes less critical to survival.

  17. True. Usually univs will pay professors for 9 months, and then use grant money for the summer months. Also true that grants figure heavily into tenure decisions. The overhead of the grant application rat race is what drove some faculty to join industry. E.g. Matt Welsh:

    Your proposal should go further: to break the grant rat race, universities should give professors not just salaries but also funds to hire grad students. A salary is not enough because you really do need a team of good grad students. The problem with this approach is: how do you decide how to allocate funds to professors, and will that process be any less bureaucratic than grant applications?

  18. My supervisor would tell me that she would not bother with the low-success-rate kind of applications if there was not a substantial part of necessary project planning involved, that will be useful even if the application is rejected. Although this is hardly enough to compensate all the effort of a rejected application, I think it needs to be factored into the cost/benefit calculation of applying for a grant. Moreover I think it is useful to keep in mind when choosing projects. That is, choose or design the kind of projects where you can make good use of the prep work for an application even if it is rejected.

  19. I have friends at a university where they are told when hired “within seven years you will receive a major grant of over $1 million and you will publish an average of one article a year in a ‘top’ journal or you will not get tenure”. At least it is laid out honestly for them.

    As for me, I wrote a lot of grants my first several years as an assistant/ associate professor. Then I decided if I was working this hard bringing in grant money to fund overhead, I may as well start my own company.

Comments are closed.