# Entitled to a Grant: What is fair?

May 02 2016 Published by under Fixing the NIH, NIH, NIH Careerism, NIH funding

I am genuinely curious as to how you people see this. Is there any particular difference between people arguing that that acquisition of the first major grant award should be protected versus multiple award and the people arguing that acquisition of the first and third concurrent awards should be on an equal footing?

If we agree that NIH (or NSF or CIHR or whatever) grants are competitively awarded, it follows that nobody is actually entitled to a grant. And as far as I am aware, all major funding agencies operate in a way that states and demonstrates the truth of this statement.

Specifically in the NIH system, it is possible for the NIH officials to choose not to fund a grant proposal that gets the best possible score and glowing reviews during peer review. Heck, this could happen repeatedly for approximately the same project and the NIH could still choose not to fund it.

Nobody is entitled to a grant from the NIH. Nobody.

It is also the case that the NIH works very hard to ensure a certain amount of equal representation in their awarded grants. By geography (State and Congressional district), by PI characteristics of sex and prior NIH PIness, by topic domain (see the 28 ICs) or subdomain (see Division, Branches of the ICs. also RFAs), etc.

Does a lean to prioritize the award of a grant to those with no other major NIH support (and we're not just talking the newcomers- plenty of well-experienced folks are getting special treatment because they have run out of other NIH grant support) have a justification?

Does the following graph, posted by Sally Rockey, the previous head of Extramural Research at the NIH make a difference?

This shows the percentage of all PIs in the NIH system for Fiscal Years 1986, 1998, 2004 (end of doubling) and 2009 who serve as PI on 1-8 Research Project Grants. In the latest data, 72.3% had only one R01 and 93% had 1 or 2 concurrent RPGs. There were 5.4% of the PIs that held 3 grants and 1.2% that held 4 grants. I just don't see where shifting the 7% of 3+ concurrent awards into the 1-2 grant population is going to budge the needle on the perceived grant chances of those without any major NIH award. Yes, obviously there will be some folks funded who would otherwise not have been. Obviously. But if this is put through in a systematic way*, the first thing the current 3+ grant holders are going to do is stop putting in modular grants and max out their allowable 2 at $499,999 direct costs. Maybe some will even get Program permission to breach the$500,000 DC / y threshold. So there won't be a direct shift of 7% of grants back into the 1-2 grant PI population.

There has been a small trend for PIs holding more grants concurrently from 1986 to the late naughties but this is undoubtedly down to the decreasing purchasing power of the modular-budget grant.

BRDPI.
I"ve taken their table of yearly adjustments and used those to calculate the increase necessary to keep pace with inflation (black bars) and the decrement in purchasing power (red bars). The starting point was the 2001 fiscal year (and the BRDPI spreadsheet is older so the 2011 BRDPI adjustment is predicted, rather than actual). As you can see, a full modular 250,000 year in 2011 has 69% of the purchasing power of that same award in 2001. Without that factor, I'd say the relative proportions of PIs holding 1, 2, 3 etc grants would be even more similar across time than it already is. So I come back to my original question. What is fair? What policies should the NIH or any broad governmental funding body adopt when it comes to distributing the grant wealth across laboratories? On what basis should they do this? Fairness? Diversity of grant effort? PR/optics? __ *and let us face it, it is hugely unlikely that the entire NIH will put through a 2-grant cap without any exceptions. Even with considerable force and authority behind it, any such initiative is likely to be only partially successful in preventing 3+ grant PIs. DISCLAIMER: As always, I am an interested party in these discussions. My lab's grant fortunes are affected by broad sweeping policies that the NIH might choose to adopt or fail to adopt. You should always read my comments about the NIH grant game with this in mind. • mH says: As you point out, fairness is not really part of the equation here. Funders have lots of priorities and mandates that don't necessarily align with the scientific or career concerns of their applicants or their peer reviewers. Many funders, either through separate funding mechanisms or (like NIH) payline adjustments try to lower entry barriers for newcomers, limit awards per PI, etc. Funders can also decide to promote diversity, over the howls of those who think that our bench is so shallow that this is equivalent to the end of meritocracy. Whatever. We are all free to argue for how we think our respective funders should do things, but at the end of the day are responsible for getting grants in the system that exists. • qaz says: The difference is that losing the third grant is not a "death wall". The lab continues. But losing that last grant is the end. Lab dies, that science path ends. Given that science is generally run at the "lab" level - departments are really collections of labs, it seems that the organismal level is the lab. Changing techs within a lab keeps continuity (thus the "tech" level is below the organism). Changing labs within a department changes continuity (thus the "department" level is above the organism). So the question is does NIH care if the lab survives or not. • Ewan says: Is not the mandate of the NIH to improve public health? The public are their customers, not us. So: funds should be allocated to maximise this. My guess is that this means few or no labs with >1 basic research grant, but a priority on continuity of successful programs - which is going to hurt new folks trying to break in. As one of those new folks a short time ago, that would suck; but I think that's what the NIH should be aiming at. • drugmonkey says: But losing that last grant is the end. Lab dies, that science path ends. Or never starts. Don't forget that part, qaz. But so what? This sounds as though you are referring to some special entitlement to "at least one" that trumps entitlement to a person landing their seventh concurrent NIH grant. True? Is this your stance? If so, .....why? And if you meant to differentiate those who have been NIH funded and have lost it from those who never have managed to get a major award in the first place...again, why? Why are the previously funded to be protected over the never-funded? So the question is does NIH care if the lab survives or not. No, the question is whether my Readers care if NIH cares if labs survive. Do you? What constitutes "a lab"? The once-funded? Anyone who is allowed to serve as PI by a qualified applicant institution? • drugmonkey says: My guess is that this means few or no labs with >1 basic research grant And my guess is that it means more labs with 3 grants and fewer with only one grant. I'd say public health is maximized when 33% of PIs have 3 concurrent, 31% have 2 concurrent and 30% have 1 grant. That leaves 6% of the PI slots for the RealMoneyBags PIs. wheee!!!!! let's pull stuff out of our vapor clouds!!!! • Curiosity says: This isn't a policy proposal by a long shot, but in evolution, diversity is critical. We could think of the basic scientific enterprise as a vaguely random walk, benefiting from diverse viewpoints and approaches. To that end, limiting accumulation of wealth in select sectors is good for science -- so, fight monoculture by raising the bar for the third grant. But to answer specific q's on who, what, when, where? dunno dunno dunno dunno. • Ewan says: "And my guess is that it means more labs with 3 grants and fewer with only one grant." Huh. Why? Or was that pure sarcasm? • drugmonkey says: You started a discussion of what we guess would be a way to maximize public health, Ewan. You stated your guess and I countered with mine. • drugmonkey says: Curiosity- me thinks you do not understand how evolution has gone down on this planet. • Grumble says: "No, the question is whether my Readers care if NIH cares if labs survive." Of course I care. There needs to be diversity in science. That is paramount. Without a lot of different approaches and ways of thinking, there is a much reduced chance of some flash of insight producing the sort of important advance that really pushes knowledge forward. I take that as a given, and I think anyone who understands how science works would agree. So, it's great that NIH POs realize this, and are trying to make sure that labs don't collapse. I'd like to point out, though, that this is done in competition with the goals of scientific grant review, which can easily yield terrible scores enough times in a row to eject any given PI - in other words, scientific review is saying this guy's grants suck and his science is terrible, but the POs realize it's just not true. Which suggests to me, as I have said over and over again, that the whole system is bollocks. Why does everyone have to go through the whole scientific review charade if program staff are going to step in and muck about with the results to make sure the research programs they like are going to continue no matter what? Would it not make sense to, ya know, fund people instead of projects, at least some of the time? Because that's effectively what POs are trying to do, right? So why make everyone waste so much goddamn time on writing grants? [Waiting for the inevitable pearls clutched to mouth and exclamations of opprobrium that traditionally follow this suggestion, as if I walked in here wearing a Trump T-shirt.] • Curiosity says: Well, DM, I suppose it depends on whether we view the NIH as a mechanism for diversity generation or a resource to be exploited. If the latter, the successful few strategies will prevail and why are we having this conversation? If the former, then NIH ought to design policies to limit consolidation of funds. Of course, it's a bit of both. Reveals the limitations of analogies more than provides insight into how many "grants" "one" "lab" "should" "have". • drugmonkey says: As you suggest, Grumble, we already do have a hybrid system in NIH-land. There are substantial elements that do fund the people/programs over the project. And, you will note, that I do not object to this and indeed have suggested some middle ground approaches to do it in a fairer, more systematic way. I.e., expanding the use of R37 MERIT across all of the ICs and down into each and every Branch. I have also suggested a way of dealing with grant churn (basically by passing grants forward for consideration in subsequent rounds) and advocate for NIH to focus on people / career in their success evaluations rather than merely look at per-application rates of success. I see no need for MIRA type approaches that further narrow the program-based approaches to rare-winners and place additional pain on the vast majority. There needs to be diversity in science. That is paramount. Without a lot of different approaches and ways of thinking, there is a much reduced chance of some flash of insight producing the sort of important advance that really pushes knowledge forward. I take that as a given, and I think anyone who understands how science works would agree. Unfortunately, not all who understand how science works can do the math that tells them that with an essentially fixed NIH budget from Congress and the relentless advance of inflation things have to change. Nor can they follow the logic that with further and further divisions of the pie so that all comers can say they have a NIH grant will lead to serious inefficiencies. • mH says: How is current standard vanilla R01 assessment not already a hybrid of "person" and "project."? I don't get the demand for special "person" grants other than some persons think having their ideas evaluated by CSR riff raff is beneath them. "inefficiencies" heh • Grumble says: "see no need for MIRA type approaches that further narrow the program-based approaches to rare-winners" Not sure what MIRA is but I fully agree that person-not-project mechanisms should not just fund rare winners, but a broad cross section. "an essentially fixed NIH budget from Congress" I'm in the middle of a novel by Neal Stephenson, Cryptonomicon, that (very entertainingly) makes the point that nations win over other nations, both in war and economically, by liberally promoting the free development of ideas, especially in relation to technology. He makes the point that the Allies won WWII by winning on several technological frontiers (cryptography being one, but also, obviously, weapons development) - and that because of the very structure of German and Japanese societies, in which ideas were repressed, it was essentially impossible for them to win a technological race with the allies. I am worried that this fixed budget thing is leading us straight to Axis-like stagnation, and Donnie doesn't cheer me up at all on that front. At some point, if the budget stagnates enough, it will be *impossible* to maintain enough scientific diversity, so all the hair-pulling about how the NIH should apportion its pennies are just fiddling while Rome burns. • serialmentor says: I wonder how the option for multiple-PI grants factors into all of this. I'm PI on two R01s, both with three PIs total. So do I hold 2 grants or the equivalent of 0.66 grants? As more grants become multiple-PI grants, we'll certainly see the number of grants per person increase. • Ola says: Agree @serialmentor, I'm on 3 grants, but only one of them is truly "mine", the other 2 are multiple PI. Averaging them all out, the 3 of us PIs who share grants collectively have 5 between us, or about 1.67 grants each. My big worry is that going in for competing renewal on my "own" R01, reviewers will look at the biosketch, see 3 awards and decide I'm well funded or less needy than someone with fewer awards. Need to find a way to express that it's only 1.67 grants in real terms, in the biosketch. • MorganPhD says: My dream scenario is the NIH preferentially funds R01's where the PI doesn't ask for salary support. Kill the soft money PI's and you can do significantly more work for the same current price. Get universities to put some more skin in the game, and single Ro1 labs will be more sustainable. • Ewan says: "You started a discussion of what we guess would be a way to maximize public health, Ewan. You stated your guess and I countered with mine. Yes, I get that; I was curious, as I noted, whether you had any rationale for your guess. Having watched many folks with zero or one grant struggle mightily to get maximum return on their were in plentiful supply and hence watched much less carefully, my guess was based on the expectation that careful management of$across a larger number of labs is likely to produce greater scientific advance per$ (and hence assumed greater increase in public health). You apparently have a different expectation: why?

• MorganPhD says:

On the other hand, from a public health perspective, the NIH would be justified in awarding R01's preferentially to those with no university teaching or service requirements.

Give them to 100% soft money faculty who have 100% of their time to devote to basic and translational science. Require that faculty directly work at the bench/computer, instead of serving on committees, writing more grants, going to planning meetings, etc.

• drugmonkey says:

, then NIH ought to design policies to limit consolidation of funds.

From the Rockey graph...where would you draw policies to limit such consolidation? as of FY2009 only 7% of PIs had more than two grants. Or are you suggesting even the two grant level is an unneeded consolidation?

How is current standard vanilla R01 assessment not already a hybrid of "person" and "project."?

It absolutely is and has always (i.e., in my time in the system :-)) been. And when times started to get tight post-doubling I heard a lot of PO talk about needing to "save" their long-funded older and established scientists and next to zero talk about a robust launch for my generation who were in the thick of trying to get started at that time.

I fully agree that person-not-project mechanisms should not just fund rare winners, but a broad cross section.

Do you have any suggestions for how we might accomplish this?

I wonder how the option for multiple-PI grants factors into all of this.
If there is any will at the OER of NIH to keep up with periodic assessments along these lines, I do hope they figure out how to deal with this. Applicants are supposed to approximately divide the budget proposal but I doubt everyone even bothers to make up an anticipated division at this point.

• drugmonkey says:

Need to find a way to express that it's only 1.67 grants in real terms, in the biosketch.

I'd say the first thing you do is detail this in the little blurb you write after each grant. "My lab has a third of the budget of this to...". That sort of thing. This should be boiler-plate, imo. If reviewers still miss it, then you'd have to consider putting it in the personal narrative, I guess.

You apparently have a different expectation: why?

Much as with you, from personal experience. Watching labs around me in my department over time, other labs and departments I am familiar with for various reasons, my unhealthy interest in reviewing career arcs of scientists ahead of me (and now behind me) on the path, etc. Most especially including the labs that have failed in some manner, the ones that have thrived and the ones that seem chronically on life-support.

This undoubtedly has much to do with the relative expense and ballistics of the kinds of research that I am most familiar with. I am aware that starting and stopping and re-starting if grant funding gets gapped for a year or two leads to huge inefficiencies in the output. I am swayed by considering the sunk cost of investing startup grant funds in a program that is unable to continue for the second five years, the second decade or the duration of a natural career. It is also the case that I think the unique or highly impactful findings come from labs in my field when they have an extended interval of ease- that extra grant that came in, a couple of easy renewals right on schedule, PECASE, MERIT, a BigMech, three NRSAs for trainees or whatever- and not from when they were struggling to keep one R01s worth of funding in place. I am dismayed by several really good scientists I know who appeared to have made it by landing their first major funding (maybe even round 2 or three of funding) and then decided to bail on the career / institution /job type because they just couldn't take 20 or 30 more years of this crap.

I have felt the truthiness of the high efficiency of desperately running for your dinner on one award- it seems like productivity is highest. But these introspections do not take the career arc into account and ignore the longer term productivity, imo.

• jmz4 says:

" I am dismayed by several really good scientists I know who appeared to have made it by landing their first major funding (maybe even round 2 or three of funding) and then decided to bail on the career / institution /job type because they just couldn't take 20 or 30 more years of this crap. "
-I'm curious, where did they go?

As much as it pains me to think about it, since I'll be applying for a first R01 relatively soon, I think the ESI system/incentive is a bit warped. Yes, it is nice to help people establish their labs and get the funding they need. But that is absolutely meaningless if it only gets harder and harder to maintain that funding. They should take those grants and expand the K99 program.

So I think a huge issue, as Morgan brought up, is salary support. Most people would not be nearly as fearful of losing/not getting their grant if it didn't mean that salaries would be in jeopardy, particularly their own. I think having to produce (the right kind of) results to pay your own salary has huge negatives for science in that it encourages corruption and bad sportsmanship. It also, as Morgan points out, means less science for more money, and destabilizes the workforce pipeline by encouraging bloat. The NIH should phase out 100% salary payments off of grants entirely (including postdocs and grad students).

• jmz4 says:

"They should take those grants and expand the K99 program."
-sorry, that sentence should have continued.... "if they want to spend money on new PIs". I don't think they should. I think they should ramp funding relatively smoothly, not have sudden peaks and valleys in your relative funding rate.

• DO IT TO JULIA!!!!!!!

• drugmonkey says:

jmz4- ftr, most of these ppl predate the ESI program. These challenges are not brand new.

• qaz says:

We know that monopolies and oligopolies are bad for business because they cause rent-seeking (gaining $from controlling the market) rather than innovation. We know that reduced diversity is bad for ecological systems because they reduce the number of available niches (having more species in an area actually increases the number of available species who can find places to live). Although I have not worked through the math, I would be willing to bet that having a diverse scientific base is going to be better for discovery than a limited oligopoly of four-R01 empires. Moreover, for someone who complains bitterly about graduate students discovering at age 26 that they "wasted" five years of training, you are remarkably sanguine about throwing away 45 year old scientists who have done five years of grad school, five years of postdoc, and twenty years of running a lab. (And as we both know very well, running a lab is a skill that takes time to develop.) Yes, labs are hard to start when we are protecting our investment in successful labs, but they will be just as hard to start when they are competing with four-R01 empires. That's not going to change. My point was (and is) that losing the last grant is a death wall. It means the lab shuts down. That there are a large number of investments that are lost permanently (integrated lab knowledge, working protocols, PI skill sets). This means that there is a very very big difference between losing the third R01 and losing that last R01. Very importantly, losing that last R01 is NOT the sign of an unsuccessful lab. While we know that grant scores are accurate, they are not precise. There is a huge amount of noise in the system. Thus, a grant that deserves a 5% is going to get something like a 5% +/- 25%, which means it is likely that it will end up below the funding line. What we need desperately in the scientific system is the stability that if someone is doing a good job they get to keep doing that good job. I'm not arguing for a sinecure. I'm arguing for a 100% renewal rate *if* the program has been going well. You should know that if you've published X papers in X quality journals, then your grant will be renewed. This was effectively the system we had in the GoldenAge(tm). Such a system would allow labs to survive with one R01. Extinction is forever. • MorganPhD says: My scientific idol runs a single R01 lab. She's under 50. She published 3 papers in Cell, Science, and/or Nature this year. She has 7 postdocs. 6 have independent fellowships. From the NIH prospective, this person gives an incredible ROI. Some look at this lab and say "that's a lucky, one-off scientist" I look at it and say " this is how science should be" • Grumble says: "Do you have any suggestions for how we might accomplish this?" First, maintain the current system for new investigators and anyone else who want to compete in it. Second, for investigators who have had >X years of funding (X=5? 10?), set up a new system that allows them to submit renewals that consist of a single sheet of plans plus a list of accomplishments. Cap the amount any single PI can obtain with this mechanism at about a single R01 level. Allow indefinite renewals using this mechanism every 5 years - with the understanding that the review is based on productivity, and poor productivity means no renewal. Finally, make it more difficult, but not impossible, for PIs who choose to be funded this way to obtain competitive R01s in addition to past productivity-based (let's call them "lifeline") R01s. The number of competitive R01s, or perhaps the total amount of funding, for each PI who has a lifeline R01 would probably have to be capped. That is because lifeline R01s would be relatively easy to get, so the risk is that people grab their lifeline and STILL write tons of grants, so their total funding ends up being what they'd get under the current system + the lifeline amount. That is not the goal. The goal is to give PIs an alternative to the current system: to give up some amount of funding (or chance at funding) in exchange for the stability of the lifeline. The end result, I think, would be more stable funding for more mid- and late-career people. Under such a system, instead of constantly submitting grants, I would opt for my "lifeline" funding and submit competitive grants only occasionally. • Dave says: My scientific idol runs a single R01 lab. She's under 50. She published 3 papers in Cell, Science, and/or Nature this year. She has 7 postdocs. 6 have independent fellowships. That's not funded from a single R01. That's for sure. From the NIH prospective, this person gives an incredible ROI. See above I look at it and say " this is how science should be" If you mean that a PIs career should not hinge on a single NIH grant, I agree. But the example you cite is awful for many reasons. • Dave says: ^Sorry, didn't read that properly. Just saw that 6/7 postdocs have independent fellowships. If true, that's impressive. • Grumble says: "Just saw that 6/7 postdocs have independent fellowships. If true, that's impressive." And unrealistic, for most post-docs and PIs. (Although more than half of the post-docs I've had have managed to get some sort of fellowship, it has never been for the entire time they've spent in my lab.) • MorganPhD says: Oh, it's absolutely unrealistic in the current environment. Totally. Coincidentally, her R01 ends April 30th, 2016, so we'll see how long that actually lasts. I also used that anecdote of a system that might work well for maintaining/allowing more single R01 labs. I strongly believe that switching to a fellowship system for trainees will remove part of the pressure to maintain multiple R01-level grants. (I'd get rid of T32 training awards too, to be honest. Convert them to individual F awards. At NIGMS, individual F awards are funded at 5% of T awards). Also, of course, my previous comment about hard money positions being part of the equation as well. • Anon says: I gotta ask. Does she take only US Citizens, and require NRSA submission to be in her lab? I get a lot of (mixed bag of quality) international applicants that are limited in funding possibility. • Newbie(ish) says: Grumble: what you describe is similar, in essence, to the intramural NIH system. I would offer that it encourages good, strong, stable science over innovation, which is not a bad thing. In intramural setups, there is a degree of stability encouraged by appointments that renew in lengthy (what is it, 3 year at NIH?) terms. You might think that kind of appointment would encourage risk taking, but, in practice, I think it encourages incremental, high quality science. Sure, you're going to try new things, but with three years to make the new thing work, there's less pressure to throw it all at the wall and see what sticks. There are some similar setups at certain European universities, where appointments come with guaranteed staff or student positions. I think that's one of the best ways to encourage continuity and quality in a scientific program. For those of us in soft money positions (me!), with not just our salaries but our actual necks on the line every freaking year, there is tremendous, significant pressure to get external funding. We have to try anything and everything to get a buck. BIG, COLLABORATIVE, DIFFERENT, NEW. All of this encourages innovation. I'm with y'all above arguing for diversity. I think we need a little of each - risk takers and steady lines. I like the lifeline grant idea, even though it excludes me, as an ESI. My own preference would be lottery system. Score in the upper 25th %ile, your name gets thrown into the lottery. At least then it would feel *fair*. Although, still these ideas about measuring productivity and funding the person not the grant... that works in an intramural setup because an appropriate person exists to judge whether the faculty member is productive. ALL of this boils down to a problem of how to implement grant review at a large scale. In theory, it'd be great to look at professor XYZ and say, hey, they're consistent, they do good stuff, let's make sure they keep going. In practice, it's just another way to let unacknowledged bias slip into everything we do. • Newbie(ish) says: It's like tenure with NIH funding! Pre-tenure: lottery system for upper 25th %ile Tenure: an actual review panel Post-tenure: R01 lifeline so long as publications remain steady. The trick is implementing this in a way that it doesn't become the new old boys club and is actually based on real merit. I still think all of this boils down to scale up in grant (or PI) review. Heck, we just need better ways to judge scientific merit. • mH says: what do people think K99s solve? they don't change the job market or competition for grants, they just give a big advantage to a small subset of the competitors, often those who are pedigreed or otherwise already recipients of other cumulative advantages. • mH says: "And unrealistic, for most post-docs and PIs. (Although more than half of the post-docs I've had have managed to get some sort of fellowship, it has never been for the entire time they've spent in my lab.)" I have seen glamorous PIs at a glamorous institutions who just make it a requirement: you have 1 year to get a fellowship or you're out. • drugmonkey says: We know that monopolies and oligopolies are bad for business because they cause rent-seeking (gaining$ from controlling the market) rather than innovation. We know that reduced diversity is bad for ecological systems because they reduce the number of available niches (having more species in an area actually increases the number of available species who can find places to live).

The Rockey data I posted referred to some 26,000 PIs of RPGs in FY2009. Almost 19,000 1-grant labs. Many more if you count all of the mechanisms. Cutting this by what percentage would be catastrophic for diversity (remember, scientists tend to fill vacant niches because they are always seeking new stuff).

You have to get up into the 6-grant range to find 13 PIs...this is a number that might look like "monopolies and oligopolies" to me.

you are remarkably sanguine about throwing away 45 year old scientists

I disagree. I am not sanguine about ANY of this shit. I am on record saying I think we should absolutely pull another $30B out of the DoD and double the NIH next FY. With further inflationary adjustments as a default. But it ain't going to happen qaz. And my current soapbox, as it happens, is about the plight of the "45 year old scientist". I.e., Gen X, i.e., my generation, i.e., those who got absolutely screwed by the Boomer hegemony followed by the ESI "sudden discovery". I am concerned about sustaining the middle career investigator in what should be the meat of her productivity. That requires, in my opinion, a strong and increased cohort of middle career investigators with 2 to 4 grant level support. In short, I am arguing in favor of the "45 year old scientist" who has another 30+ years (going by current oldster arguments about how they are just as good if not better until they die) of productivity ahead of her. My point was (and is) that losing the last grant is a death wall. It means the lab shuts down. Yes. It does. As you very well know, we currently have too many mouths at the currently available linear footage of trough. Your solution keeps shrinking the slop per mouth until we have many more pigs "alive" in the pen but none of them can actually thrive. This is not healthy either. It doesn't make the product that the NIH desires. You don't like reducing the birth rate. You don't like culling the herd. Other people, not sure if it is you, scream bloody murder about retiring the oldest pigs. All you have done is say "we need to squeal louder for more slop! that's the ticket". This is fantasy. Very importantly, losing that last R01 is NOT the sign of an unsuccessful lab. Of course it isn't. Nor is failing to get that first one. Nor is failing to ever be in position to fight for that first R01 a sure sign of an unworthy scientist. So what? What is your point? That we should strive to award as many R01s as possible, even if they are for one$25K module in annual direct costs, so we can all wear the participation medal?

What we need desperately in the scientific system is the stability that if someone is doing a good job they get to keep doing that good job.

I agree. We need to increase stability for *some* so that their work is reasonably efficient. But.....reality. How are we going to do this when the pen is full of too many pigs?

Such a system would allow labs to survive with one R01.

No, it wouldn't. Because the system would implode due to too many mouths and not enough slop.

• Grumble says:

"implementing this in a way that it doesn't become the new old boys club"

It's already an old boys' club and nothing is going to change that, ever. That's just a sociological truth. So, new, old, what's the difference? Only that the old boys have to spin their wheels with constant grants in the current system, less so in the proposed lifeline system.

"I have seen glamorous PIs at a glamorous institutions who just make it a requirement: you have 1 year to get a fellowship or you're out."

I am not an asshole and would rather close up shop and do something else than become one.

• drugmonkey says:

My scientific idol runs a single R01 lab. ... 6 have independent fellowships.

That's not a 1 R01 lab. That's a 3 R01 lab (at the least). I hold no truck whatsoever with NIH policy proposals based on "get somebody else to pay large fractions of the charge for the NIH business".

Cap the amount any single PI can obtain with this mechanism at about a single R01 level...(let's call them "lifeline") R01s.

Devil is in the detail of what represents a "lifeline". See above about qaz's apparent argument that we need to award a dinky R01 to all comers. If you made my lifeline equivalent to the target I think I've needed for the duration of my career to date (and it's about the same number/purchasing power that it has always been) I think, right now, I would never want to get any bigger. Make it substantially less than that and I am going to fight for the rest. What I hear in your voice on this (and that of qaz) is that you assume that the lifeline number for you will be set at what you see as your target lab size. I am not sure that your improvement works if we make sure that everyone is within the same percentage of their target. So that just makes it another do it to Julia plan- you get comfort but screw those other guys. Unless I mistake you and you really mean that the "lifeline" is intended to hit most people's individually and circumstantially determined satisfaction level. In which case I submit that we come right back to where we are at present- too many mouths at the trough.

• Grumble says:

" "45 year old scientist" who has another 30+ years "

I'm around that age and am beginning to seriously consider the idea of retiring at 62. A. This grant game is bullshit and I don't need the stress. B. After observing many elders retire late and then die of some horrible disease not too long after that, it seems to me that it makes much more sense to quit early and enjoy a long retirement doing fun things. C. The ol' 403b isn't doing so badly. Could even retire now if I wanted to move to East Bumfuck and buy a 45k house on the prairie. Have a couple cats and dogs, grow a little weed, do some target practice from my back porch... • Grumble says: "I am not sure that your improvement works if we make sure that everyone is within the same percentage of their target" No, the idea is not to make sure that everyone is happy with their lifeline. The idea is that the NIH simply says, "Here is the lifeline mechanism, here is the amount that you *AND EVERYONE ELSE* gets if they opt in. Opt in if you want, opt out if you want. Your choice." Sure, this will exclude people who feel that whatever the lifeline amount is grossly inadequate, and therefore not worth the cost involved (giving up chances at other potential sources of funding). Of course, those costs are mitigated by the fact that you don't have to put in any grant writing effort at all to get the lifeline amount. I have no idea whether the NIH would choose a lifeline amount that would be acceptable to me (so no, I am not saying Do It To Julia). I am simply saying that introducing such a system is one way to both maintain diversity and reduce the need for scientists to waste so much time writing grants. • Newbie(ish) says: The lifeline idea does not have to be scaled to support a full lab. The point is to let the skeleton crew keep knowledge and expertise alive. And to take away the sheer panic (although maybe it's only me as an ESI) of having to dump everyone all at once when you lose your last straw. Support for 2 personnel and a small research program (i.e., an R01) is plenty to keep things alive. And, Grumble, I agree: it is a club mentality right now. My meandering thought was to point out that all of this discussion about how to implement structural changes to how funds are allocated doesn't address the biggest problem: that we simply don't know how to fairly assess the scientific merit of such a large number of possible grants (or PIs). We can change how funds are allocated but that still doesn't change that demographic bias (including investigator status/reputation as a demographic). The problem is review. • Newbie(ish) says: Here's a question. At my career stage, grant review is useful. (Up to a certain point). My ideas get better, particularly from that first round. Does this benefit go away when you are more senior? Is some of the purpose of NIH review to actually improve the quality/significance of the science? Talking first revision here - not round after round. • bacillus says: This question might get me laughed off this forum for good, but it is a serious one. How much severance do University PIs get if they are forced out due to a lack of grant success? Is this considered "with cause"? • Newbie(ish) says: I don't know if any severance is ever given because I would assume that is considered with cause. My contract stipulates external grant support and stipulates the terms by which my position will terminate if I fail to achieve it. Pushing someone out who doesn't have tenure takes time (going through tenure review) and pushing out someone with tenure also takes time (by making life so unpleasant that they eventually leave). At my institution, it seems to take about a year of people being truly and completely out of money before they get ousted... although it can also get stretched by some rather non-transparent administrative mechanisms, since there are various pots of money available to dip into. • drugmonkey says: I have never heard of severance, meaning cash and you leave. Tolerance and salary past the point the grant runs out? Sure. " you have 6 mo to get a job elsewhere". That sort of thing. • drugmonkey says: A skeleton crew just amps up the pressure for the non-lifeline fund. Increases the number seeking funding. • Newbie(ish) says: "A skeleton crew just amps up the pressure for the non-lifeline fund." Not if the funding *actually* supports the skeleton crew. If I knew I could guarantee two research techs, I'd have a much more stable research program. "Increases the number seeking funding." But doesn't any practice seeking to improve science funding provide an incentive to do more science? I don't think making things more difficult/less attractive is the right approach. (Better understanding of *reality*, yes...) • Grumble says: "A skeleton crew just amps up the pressure for the non-lifeline fund. Increases the number seeking funding." It doesn't necessarily increase it *beyond the number seeking funding now*. I already made the argument that this won't happen because if a feature I built into it to prevent people from trying to get (what they have now) + (lifeline): Remember that by accepting lifeline funds, you are giving up some ability to have access to the non-lifeline funds. Either non-lifeline grants become harder to get, and only worth pursuing if you have a really top-notch idea + top notch preliminary data, or the amount a lifeline recipient can get from all sources is simply capped. Or both. • MorganPhD says: @Anon, Actually only 1 NRSA out of the awards in the lab. She has a great reputation in her field and attracts great PDs with competitive awards (think Helen Hay Whitney, etc). @drugmonkey, The entire academic research system is based on "get somebody else to pay large fractions of the charge for the NIH business". I'm confused as to your stance on who should pay for research salaries? Should the universities put more skin in the game or should the NIH shoulder the entire burden? I argue universities get a really f#cking good deal, even if they always cry poverty. The NIH should use the bully pulpit to get the best value for their dollar. But they (legally) can't and won't. • All this "diverse ecosystem" business is a load of self-serving claptrap. Why does "diverse ecosystem" mean "all labs are one-R01 small-town grocers" and not "there should be different sized labs with different levels of funding to support different kinds of research"? • MorganPhD says: "I am not an asshole and would rather close up shop and do something else than become one." Here here. • Newbie(ish) says: Why does "diverse ecosystem" mean "all labs are one-R01 small-town grocers" and not "there should be different sized labs with different levels of funding to support different kinds of research"? Well, n=1 here, but that's what I mean when I am using the word diverse. "I'm confused as to your stance on who should pay for research salaries? Should the universities put more skin in the game or should the NIH shoulder the entire burden? I argue universities get a really f#cking good deal, even if they always cry poverty." YES. Here's a way to increase NIH's budget by 50% overnight: stop Universities from sucking grants dry through overhead. Frankly, it's double dipping: I pay 25k per grad student for tuition (that ostensibly goes toward salary support) AND I pay overhead. That's insane. I am so sick and tired of seeing fancy buildings and high salaries being built off of federal grants. There needs to be better transparency about overhead rates and how they are allowed to be used. • Newbie(ish) says: Why does "diverse ecosystem" mean "all labs are one-R01 small-town grocers" and not "there should be different sized labs with different levels of funding to support different kinds of research"? Well, n=1 here, but that's what I mean when I am using the word diverse. "I'm confused as to your stance on who should pay for research salaries? Should the universities put more skin in the game or should the NIH shoulder the entire burden? I argue universities get a really f#cking good deal, even if they always cry poverty." YES. Here's a way to increase NIH's budget by 50% overnight: stop Universities from sucking grants dry through overhead. Frankly, it's double dipping: I pay 25k per grad student for tuition (that ostensibly goes toward salary support) AND I pay overhead. That's insane. I am so sick and tired of seeing fancy buildings and high salaries being built off of federal grants. There needs to be better transparency about overhead rates and how they are allowed to be used. • drugmonkey says: Not if the funding *actually* supports the skeleton crew. If I knew I could guarantee two research techs, I'd have a much more stable research program. first, TWO techs? Some people call that "a lab" not just a "skeleton crew". man, you people. second, my point is that this keeps more and more people scientifically viable and submitting grants. aka, more mouths at the trough. It further amps up the stress on the fixed-budget of NIH. How is this not obvious? But doesn't any practice seeking to improve science funding provide an incentive to do more science? A practice to improve the stability of funding and decrease the grant churning for a fixed population of PIs may result in more science but it doesn't necessarily increase the grant stress. Remember that by accepting lifeline funds, you are giving up some ability to have access to the non-lifeline funds. It's zero sum. You are securing a fixed population of PIs into place. There is no way to get rid of them, save attrition by age. Within a trice that "above lifeline" pool gets shrunken to the imperceptible. Again, this is fine if the lifeline is set to what you need but it isn't so fine if it is less. Take whatever you think you'd be happy with and divide by 3. Or 4. or 5. Are you so sanguine now? I bet not. Unless, of course, your only goal for being in this business is to have the participation medal of being able to say that you are NIH funded. Sorry, but that isn't me. I am in this business to secure enough grant funding to actually do some science. I'm confused as to your stance on who should pay for research salaries? Those who are requesting that the work be done. All else causes distortions of the system. Of various types. Almost all of these come with pluses and minuses. What I tend to find is that scientists' views on this are almost invariably aligned with their own personal situations and experiences and what it is perceived would be to the benefit of themselves. Luckily, we have a broad and diverse set of Readers who can hash out much of the available discussion space. I pay 25k per grad student for tuition (that ostensibly goes toward salary support) AND I pay overhead. That's insane. Please take the time to work out the real budget for you lab. From space rental on your local market to support services like biohazard disposal, custodial and security. Animal vivaria and veterinarian work, if relevant. Figure out how to purchase IRB, IACUC or biohazards protocol review (hint, it can be done, many small biotechs in many R1 U cities need these services). Benefits for your staff. Etc. Cost out the whole damn thing. For real. Then come back and tell us how outrageous your overhead rate is. Because I had occasion to do just this a few years ago. It opened my eyes WAY the fuck up about overhead, let me tell you. • drugmonkey says: Oh and one more thing. Even if your research is cheap, your U is the recipient of grants not you and it amortizes the overhead across cheap and expensive lab types. • drugmonkey says: CPP- well sure....but some are more diverse than others, y'know? • Dave says: Those who are requesting the work be done??? Interesting way to put it. Seems like a softening from your usual stance on this. • drugmonkey says: Why is that a softening? It is a generalization of the principle. • Dave says: You're usually a hardliner when it comes to institutional salary support. I think they are at least in part requesting work be done, no? • Comradde PhysioProffe says: Jeezus motherfucke, when are dumasses ever gonna get that there is no free money, and that "tell those universities they aren't getting indirect costs" means that tons of scientists are getting fired? • drugmonkey says: CPP- but those IDCs are only used on walnut paneling, heirloom carpets and aged single malt, right? • Newbie(ish) says: Hah. You will never convince me that a university hiring world class architects to design the exterior of their buildings and paying their administrative leaders exorbitant salaries can't manage to cut a few corners to better support their overhead and reduce the burden on the federal government. The idea that they're giving us a good deal is absurd. Yes, labs are expensive. But you know what, universities make money these days, and professors fuel that economy - but the money goes straight to the top and no effort is made to invest in stability of the research endeavor. Also, you are aware how private universities - and their foundations - factor into this, right? The insane, unbelievable endowments? Yet overhead on science grants is how they keep the lights on, all while growing their little stockpile or money? Not cool. Really not cool. This is not an uninformed opinion. I was in the muds with that crap for years... a quick google search confirms a pretty serious fact: grants are a major if not the top source of operating revenue for many R1s. The economics of how science dollars support private "nonprofit" universities is really messed up. • physioprof says: In every other endeavor that the US Govt pays non-govt entities to provide products, services, etc, there is an expectation that such entities are--at a minimum--breaking even on providing such items. And in the case of for-profit entities, the expectation--obviously--is that they'll make money. When Home Depot sells a new stove and refrigerator to the White House, no one says, "Hey, wait a minute! Home Depot is a giant corporation with a171 billion market capitalization! They should be forced to sell the govt appliances at a loss!" But for some reason, when it comes to universities being paid by the US Govt to perform research, all of a sudden everyone loses their minds. And the fact that grant income is one of the largest sources of revenue for a university doesn't mean jacke shittio on its own. The cost of research is, correspondingly, one of the largest expenses. It's pathetic and embarrassing how infantile many scientists get about this.

• Grumble says:

" You are securing a fixed population of PIs into place. There is no way to get rid of them, save attrition by age. "

No, there is a way to get rid of them. They must be reviewed for productivity. A study section scores and ranks them, and program staff decides what the cut-off it for lifeline funding, just like with the current system. The number who actually get (or continue to get) lifeline funding depends on the amount allocated to the program - again, just like the current system. Some deserving people will lose out - that is going to be the case under ANY system because there is not enough money for all who deserve it.

You are confused on the whole point of this proposal. The point is not to guarantee free money to any PI who wants it. The point is to change the criteria of review from "how good is what she proposes" to "how well has she done". It reduces PI anxiety a little, but the real benefit is that it reduces the amount of time wasted on writing grants so that PIs can spend more time doing actual science.

• drugmonkey says:

Then how in practice is this any different from now, Grumble? Or is your lifeline significantly below the current modular (with reductions) limit?

• drugmonkey says:

You will never convince me

Why are you afraid to interrogate your assumptions? Do the math. Cost out your research program. All of it.

• Newbie(ish) says:

DM I'm arguing that some R1s should be feeding money back INTO the system.

Answer me this: are you really OK with overhead on government grants being the number one or number two source of revenue for private institutions with billion dollar endowments?

• Grumble says:

"Then how in practice is this any different from now, Grumble?"

As I said: the real benefit is that it reduces the amount of time wasted on writing grants so that PIs can spend more time doing actual science.

"Or is your lifeline significantly below the current modular (with reductions) limit?"

Why would it have to be?

• Dave says:

I dont care about IDCs. Over time, it has become obvious to me that they are neccesry and, for the most part, are reasonable. The issue of salary still is less clear to me. I still believe strongly that faculty should have a base salary from their institutions. If that means way less faculty then so be it.

My entire institute is now officially 75% soft-money. The remaining 25% who are tenured or on TT include clinical faculty, so the proportion of faculty doing research that are tenured/on TT is probably around 5-10%. Is that sustainable? It's a bubble that the NIH alone cannot support.

• drugmonkey says:

Newbieish- sure, why not. They do research, that's the current gig. Our national biomedical research workforce is housed in Universities. Of course they should receive and spend money to accomplish these ends.

What does their endowment have to do with it?

• Dave says:

Clinical revenue is far more important for my place than IDCs. In fact, clinical revenue is likely supporting many research faculty indirectly, or directly in some cases.

• drugmonkey says:

The health care system is supporting research????

Where are all the outraged scientist voices on this? (Also see, Pell grants)

• Newbie(ish) says:

Let's go by order of endowment size, shall we?

(My favorite, by the way, is Stanfard, who makes no bones about it to their shareholders:
"Given federal budget constraints, we expect the direct costs of university sponsored research (excluding SLAC) to grow by only 1.3%, while indirect cost recovery will increase by 2.1%")

Grant contribution to operating revenue rank __the following__ on total source of university operating revenue:

Harvard, tied with student tuition for second:

Johns Hopkins, first:
http://www.socialimpactexchange.org/sites/www.socialimpactexchange.org/files/AnnualReport%202013(2).pdf

I can also assure you, having followed this data for years, that there were many points in time where grant income became *the* number one source of revenue for some of these institutions. Stock market tanks, think they reach for their endowments? Heck no: they reduce operating expenses, pull back from the endowment, and keep government contribution the same. I've been through this. It's bullshit.

Ah, you say, but research is an expensive endeavor - this is the only model by which a university can operate. Well, a quick search at public institutions demonstrates that some manage to do high quality research and educate their undergrads without relying so heavily on grant money (UCSD, U Wash, U Mass, etc.; where grants are well in the middle or at the bottom of the revenue list). It is the private "non-profits" that are dragging every dollar they can out of the government. The bigger their own endowment, the larger share of public money they take.

You asked what is fair and my argument is that the science funding structure is flawed. You yourself have pointed out on multiple occasions that R01s *do not pay for their own research*. You get the R01 to get the overhead to make your institution happy. You get other sources of funding to enable expected productivity on the grant. That means that American taxpayers are, without their knowledge, subsidizing growth of private universities (and at a rate that seems to be starkly different from public institutions). Even the most superficial look at the economics of this should enrage anyone.

*Note: I am well aware that these reports are not all from the same fiscal year, but they are all recent; I grabbed whatever came up on google first.

• MorganPhD says:

@Drugmonkey,
The "Requesting the Work" comment is interesting to think about.

The NIH isn't requesting the work (except for contracts), yet they shoulder the majority of the burden of salaries. Universities want the work done, and by extension, so does an individual PI, who needs to do the work to keep the job.

I recognize the need for overhead in grants. I just strenuously object to universities crying poverty over funding the research enterprise.

There is a surefire way to reduce the costs of research. Don't do it. Shut down your Department of Important New Science or go back to being an undergraduate teaching university, and get "revenue" through students.

But universities won't do it. So I don't like hearing them complain about how much it costs to fund research when there is another option.

• drugmonkey says:

Of course the NIH is requesting the work. If you need evidence, every grant application has to list the FOA it is responding to.

• drugmonkey says:

I just strenuously object to universities crying poverty over funding the research enterprise.

I have little patience for such whinging as well. According to all collected whinges of University officialdom everything they do is a loss leader that doesn't pay for itself. Except, maybe, the Parking Office.

• According to all collected whinges of University officialdom everything they do is a loss leader that doesn't pay for itself.

Our medical school administration is extremely explict and open about the fact that clinical income is far from a "loss leader", and is what subsidizes the actual loss leaders: education and research. Every year, there is a presentation of detailed accounting of all of this stuff at the institutional level. Endowment income is a very small fraction of our budget, and the idea that richy rich universities should blow through their endowments instead of fairly negotiating for their indirect costs as required by federal law is fucken delusional.

• drugmonkey says:

Ok. Parking Office and clinical income.

• Dave says:

The problem with using clinical revenue is that the MDs are under enormous pressure to see more patients. These guys have much better options in the private sector, so it's a difficult balancing act. I doubt clinical revenue can grow endlessly either. Institutions have to find a balance.

• Dave says:

PhDs will of course quietly accept this.....

• Newbie(ish) says:

They don't need to "blow through their endowments". They need to stop effing growing them when the economy tanks. That's why the endowment exists, as a buffer, and yet it is the last purpose that private universities use their endowments for.

• Dave says:

Endowments have nothing to do with this convo. Regardless of what their initial purpose was (as a buffer in time of exigency) these days they essentially are dick swinging contests. The goal is to grow them, not spend them.

• Newbie(ish) says:

"The goal is to grow them, not spend them."

How about *maintain* them and use them for the purpose of supporting university endeavors - which includes science research

• physioprof says:

The problem with using clinical revenue is that the MDs are under enormous pressure to see more patients. These guys have much better options in the private sector, so it's a difficult balancing act.

Could be regional/local differences in this, but our medical school in partnership with our medical center (not owned by the university) has been swallowing up large numbers of private practices in the region. My understanding is that docs are exhausted with all the administrative form-filling they have to do, and want to get back to actually practicing medicine. So they are trading a somewhat smaller expected annual income for being relieved of administrative stuff. And the medical center/school benefit from a huge economy of scale for implementing all this administrative stuff that can't be enjoyed by sole practitioners or small group practices.

They need to stop effing growing them when the economy tanks. That's why the endowment exists, as a buffer, and yet it is the last purpose that private universities use their endowments for.

Did you look at the sizes of university endowments as the economy crashed in 2008, and then began to recover? Endowment values absolutely plummeted before bottoming and then starting to grow again, and many endowments are only recently getting back to even in nominal dollars. And as far as buffering, every university endowment spending is governed by a time-averaged buffering principle, frequently with a three-year sliding window. Spending as a percentage of value goes up when value is declining, and goes down when value is increasing, so as to smooth out the bumps in the road. Your ignorance of all of this renders your opinion valueless.

• The Other Dave says:

This recent analysis
https://nexus.od.nih.gov/all/2016/04/28/citations-per-dollar/

is helping convince NIH that they need to spread the wealth.

• Grumble says:

CPP says: "Our medical school administration is extremely explict and open about the fact that clinical income is far from a "loss leader", and is what subsidizes the actual loss leaders: education and research."

This is more or less what my med school admin has been saying as well. They've pointed out on several occasions that every R01 that we faculty get actually costs the school money. At the same time, I've seen several concrete examples of clinical departments supporting research with the profits they make.

So, why doesn't the school do what Morgan suggests:

"Shut down your Department of Important New Science or go back to being an undergraduate teaching university, and get "revenue" through students."

Because the way we attract students - med students, who pay the enormous tuition we charge - is by being well known for producing cutting edge research. They would not come here, and we couldn't charge as much to those who did, if we just gave up the research enterprise. Because the way the hospital system we're attached to attracts *patients* is by referring to our scientific research reputation. Because the people who set up the college wanted to do something for humanity, and that was to educate doctors and scientists, and to advance science. Because a lot of people have their egos and livelihoods tied up in how good the college is.

• MorganPhD says:

@drugmonkey,
I think we're digging into semantics a bit. Even if the NIH is requesting the work, they certainly aren't asking for proof that you did the work (unless you want ANOTHER grant or to renew your current one). Whereas, you can be sure that your university will request you do "more work". The NIH isn't going to knock on my door requesting I do some science.

@grumble,
I didn't want to type it out, but of course your rationale for why universities don't shut down the research enterprise, even if it costs money or doesn't pay for itself, is correct.

It's about prestige and attracting more undergrads and medical students (who become donors) or patients (who pay lots now). The majority of PhD scientists, grad students, postdocs, etc are there to promote and enhance the prestige.

• physioprof says:

Well put, Grumble. The leaders of our medical school are not evil greedy administrators, but rather people who genuinely believe in the tripartite mission of teaching, research, and clinical practice, and who genuinely want us to excel in all three areas to the best of our ability.

• Dave says:

I'm less convinced that our leaders (meaning, at my place) see the 'prestige' benefit of research as clearly as they may have in the past. I think it depends on where you are and the reputation of the school overall. At a place like mine, the sort of med students we get are bottom of the barrel and just looking to get in somewhere. Is research a big factor in that? I dunno.

Could be regional/local differences in this, but our medical school in partnership with our medical center (not owned by the university) has been swallowing up large numbers of private practices in the region.

Interesting. We definitely have been expanding into primary care services in the community. That still seems to be under the umbrella of the school, but I think offers MDs more autonomy and compensation in return. The MDs at these places are "professors" in name only and really have little else to do with the school as far as I can tell. Sounds quite similar to your situation.

• jmz4 says:

"Figure out how to purchase IRB, IACUC or biohazards protocol review (hint, it can be done, many small biotechs in many R1 U cities need these services). Benefits for your staff. Etc."
If you ever have the time/inclination to turn that into a blog post, I for one would be very interested in reading it. It would also be a convenient reference post, given how often indirects are whinged about in the comments.
I do think that the things you mentioned are one of the reasons NIH needs to pay higher indirect (vs, say NSF), but a lot of people not working with mouse models don't understand the regulatory burden.

@Dave:

At my former institution (private Med school), the leadership has definitely grown lukewarm on supporting research, at least as it has traditionally been done. They still want the prestige that comes from research, but their new model for that is a few super BSDs with an army of non-tenure track assistant professors (read: super postdocs) under them. It's one of the reasons I jumped ship to a different institution when the opportunity arose.

• drugmonkey says:

TOD- "justify", not "convince". This is a critical distinction.

• drugmonkey says:

jmz4- the lesson really works best when you work the problem set *with your own variables*. Otherwise ppl just claim that whatever example is put up surely doesn't apply to them. Really folks, it's not hard to call up a commercial real estate broker and ask about price per sq ft for bench lab space. Then start adding your specific requirements and find out what those cost.

Or maybe you can just call a CRO or incubator space which will have more of the package already put together for you.

• jmz4 says:

@DM: I understand, but don't forget a large part of your readership doesn't have their own variables yet (e.g. I have no clue how many mice our lab has, what our admins salaries are, etc).

Also, commercial space is not tax exempt, as most university space is, nor is it likely to be unmortgaged (as opposed to longstanding university space) so I'm not sure how I would correct for that. Then there's economies of scale versus independent outfits, leveraged over multiple departments. It seems like it could be quite tricky, which is why it might be nice to see it done as an example for how to do it for yourself.

@Newbi-ish
Large portions of the payout are used to fund the University budget, so I don't see how they're huge sources of untapped money. They are, however, an efficient means for the university to grow their operating budget, and you probably don't want to mess with that too much.

That's why I like the idea of taking NIH money to establish research endowments as a way to promote funding and workforce stability. Though the government is prohibited from engaging in investment and speculation, I wouldn't mind seeing them start taking some of their annual budget to seed a private/public endowment (with matching funds from universities/institutions/philanthropists), that would fund a set number of labs in perpetuity. Like establishing a mini-HHMI (HHMI endowment is 18 billion) every year. E.g. a 2 billion dollar investment could fund 25 labs every year at 1 million dollars in total costs per year, while theoretically being able to replicate itself every 14 years or so (using the real world average endowment growth rate of 7.5 percent over the last 10 years, which includes a large negative during 2008), which is coincidentally quite close to the average training period for a researcher.

Seed 2 of these (at 5% of the NIH's budget) per year for 20 years, and at the end of that two decades you'd have a funding source that would be comparable to the NIH's expenditures now (20 billion or so in dividends), but would be vastly more predictable and stable, since the rate of growth would be built in and fixed*.
You do it gradually enough, and you can wean the system off of Congress and into one where the rate of growth of the biomedical research funding is fixed and somewhat exponential.

*As this becomes a dominant source of funding, the vagaries of the market can be balanced out by bailout funds from the regular budget.

I'm just not sure if this solution would be legal or practical once things like corruption and graft come into play.

• DrugMonkey says:

I have no clue how many mice our lab has, what our admins salaries are, etc
No time like the present to find out. I'm not sure how long you've been around this blog but... Let's talk about the money.

Also, commercial space is not tax exempt, as most university space is, nor is it likely to be unmortgaged (as opposed to longstanding university space) so I'm not sure how I would correct for that.

There is no "correction". There is "you naive academics need to realize this is part of the skin in the game that you enjoy". Whether from the State which chooses not to tax that land, or chooses to build those buildings for you or from a private Uni which raises funds for a building or whatever. (the mortgage, btw, for new buildings can be rolled into the indirect calculation in some cases).

Then there's economies of scale versus independent outfits, leveraged over multiple departments.
Another benefit that accrues to you (the investigator pissed off about "your" indirects being "stolen by the dean") by virtue of your association with this larger entity. The point is for you to come to a number for your own lab (or anticipated lab) and then start asking yourself about just these modulating factors.

This is also, btw, where you might stand to start asking yourself how *you* benefit from the IDC skimmed from the next person over. Mouse colonies, for example, cost the University more than does some neckbearded Mt Dew drinker doing bioinformatics on his computer.

Also where you stand to examine the charge that each successive grant award to Dr Big shouldn't get the same IDC because *within her lab* it looks like the costs don't scale. Right? yeah well all that means is that the IDCs on Dr Big's fourth award are going to partially cover the real costs of *your* first award.

• drugmonkey says:

This probably a good time to remind you of Datahound's post Indirect Cost Rate Survey.

The indirect cost rate is calculated based on 9 "cost pools." These include 5 facilities pools (building depreciation and use allowances, interest on debt associated with selected equipment and buildings, equipment depreciation, operations and maintenance expenses, and library expenses) and 4 administrative pools (general administration, departmental administration, sponsored projects administration, and student services and administration). The administration component has been capped at 26% for a number of years. For those with severe insomnia, the University of Cincinnati website has a relatively clear description of their indirect cost calculation and allocation.

One thing that baffles me is why NIH only considers direct costs when evaluating the budget. Yes, indirect costs are real costs.

But if you can do the same experiments in an Iowa corn field or a Manhattan high rise, why pick the expensive option?

If institutions competed on total costs, there would be pressure on institutions to reduce indirect costs, or cover their indirect costs with other funds (endowments?)

• drugmonkey says:

Oh, the NIH considers IDC alright. In some cases when it is critical.

• Grumble says:

"Oh, the NIH considers IDC alright. In some cases when it is critical."

If that were the case in any significant way, then institutions with higher IDC rates would have lower overall grant funding rates. As a consequence, we would hear howls of protest from deans of private schools everywhere about how they are being penalized. I haven't heard that either of these is happening.

• qaz says:

DM - are you sure that NIH considers IDC when making funding decisions? (Do you know of specific cases?) I know that NSF does, because NSF budgets include IDC, so when NSF makes the decision about whether project X is worth spending Y dollars on, that decision is based on the total budget. But in all of my conversations with NIH program officers, it has always been talked of in terms of direct costs, not total. (And, yes, I've heard NIH program officers say "that's a cheap project, we can find a way to fund it", or "for that size budget, it will be hard to fund at that percentile score", but the definition of "cheap" and "size budget" have always been stated in terms of direct costs, without IDC.) I truly don't know whether NIH considers IDC in funding decisions or not.

PS. You can see the difference between NSF and NIH because the round numbers for NSF are round in terms of total budget, while the round numbers for NIH are in terms of direct costs.

As Grumble suggests, this is something that could be looked at - is there any correlation (negative, one would predict) between funding success and IDC rates. (Datahound, can you help us out here?)

The link from Datahound for "For those with severe insomnia, the University of Cincinnati website has a relatively clear description of their indirect cost calculation and allocation." changed. I think this is the document

https://www.uc.edu/content/dam/uc/af/budgetfinsvcs/gcc/docs/UC_F%26A_cost_primer_FY2016.pdf

Also, if you still want more to read about the finances of research, check out the Scripps Research Institute annual report:
http://emma.msrb.org/EP922147-EP715589-EP1117446.pdf
note that they lost money each of the past two years, and their indirect cost rate is 89%!

• drugmonkey says:

Qaz- yes. And yes.

• SFGiants says:

I agree with a previous poster that soft-money positions are a problem, at least at academic institutions. They only empower administrators. My institution is a relatively small medical school (faculty-wise) so it is more like traditional departments of biology where faculty salaries are on hard lines. However, for new hires this is changing to become more like most other medical schools. So, if you are a tenured Associate or Full Professor, and lose your grants, you teach more hours and still make a base salary into 6 figures. However, most of us teach quite a lot, as I am PI on 3 grants (2 large and 1 small, teach 100 didactic hours a year, and direct a course. For me, I enjoy the blend of research and teaching. I have also been in a non-medical school science department with 9-month contracts. Many medical school faculty seem surprised to learn that those 9-month positions can be quite attractive if you have a single NIH grant. This is because they have to pay you summer salary, and that can add up. If you don’t have a grant, you can take 3 months off and write a book, consult and hang out at the beach. You also have about a month off for the winter holidays as opposed to a week or so for med school faculty. The truth of the matter is that ‘tenure’ is really only meaningful if you tenure guarantees you a comfortable salary. Otherwise, it is not tenure and administrators do as they want because faculty essentially have no power.

• Disillusion Monster says:

Personally, I think having lots of grants actually encourages PI laziness and definitely dilutes the quality of scientific products.

My former boss is a soft money physician scientist with a full fledged private practice clinical career. He can only be on payroll for 50% of the time. However, his salary is the highest on the clinical compensation scale allowable (neurosurgeon) by the university. We have to obtain at least 2 R01 level or equivalent grants just to cover his salary and benefits, with enough left over run the lab comfortably.

Up until we stabilize the lab with the first 2 R01's (1 self, 1 Co-PI), he and I would work regularly and diligently on the products of the lab (grants, publications, etc). After recieving the grants, 50% of his time at lab was spent just going over the activities on the two grants, and doing his normal professorial duties.

By the third grant (joint U01 grant) a year later, he was delegating more and more work to me. For some strange reason, no additional help was hired in our lab (I wanted a post doc to help me, or at least a reassessment of my position to cover the extra duties). The quality of our product was also decreasing (publications in lower impact journals, more reviews and book chapters than original research). I think he was just trying to get to renewal of our first two R01's by demonstrating productivity in quantity over quality.

I spoke up about this, and I was "reassigned by the department" because I was needed elsewhere. I resigned.

Interesting after I resigned, I found out they reused some of the U01 prelim data to get a joint DOD grant, and some of the previous material I helped produce yielded a joint R21 and a state BRAIN grant. No idea how a 50% PI can handle that much without diluting quality control further? At some point giving a particular PI more support does not yield more returns?

• Scientopia Blogs

• DrugMonkey is an NIH-funded researcher who blogs about careerism in science. And occasionally about the science of drug use.

• Your donation helps to support the operation of Scientopia - thanks for your consideration.