Vinay Prasad's claim that some of the overhead funds go toward lavish parties is probably not completely off base. Both the author, and the universities themselves, admit that some of the collected overhead is used for startup packages to hire new faculty. Hopefully the majority of this funding does go toward building up new lab equipment and hiring initial lab staff, - but this "internal" funding is always labeled as 'unrestricted' and can easily be used to host lab parties, exotic retreats, loosely research-related trips to fancy destinations, purchases of expensive coffee machines, and the like.
Thank you for this, it is very helpful to an academia outsider.
This system still seems strange to me, that the government is being charged for the product (research), but also is charged to helped the universities build up their facilities and retain their professors (and sports stadiums?!).
To me, this would be like the Federal Government purchasing some jets from Boeing but also being charged so Boeing can retool their machinery, pay admin, and invest in facilities they can permanently keep and then use to sell jets to an airline or another country, or whoever.
It seems absurd that we are paying for research but end up gifting facilities to Universities (which also then have admin bloat and charge obscene tuition fees).
Isn't there a better way to do all this? Are running national non-university-based research institutes a better option?
There is a valid argument for recovery of indirect costs for a research. That argument, however, would never result in indirect cost rates that vary from 5 to >60% depending on the university in question. It is especially troubling that the most wealthy universities tend to charge the highest rates of indirect costs. This suggests the government if overpaying for research at some institutions that could be done for less at others. Setting a uniform overhead raid at the correct level (not too high or too low) puts expensive universities at a competitive disadvantage...and it should.
Part of the disparity in costs is the vast difference in faculty and other salaries paid in some markets compared to others. Why should the government pay 3 times as much for a faculty member at a Boston area research university for the faculty and staff time for the same work done at a university in Kansas. (I have reviewed for grants and see faculty making $70,000/year get 1 month of their salary from a grant while a peer in Boston paid $250,000/year and gets the same 1 month of their much larger salary charged to the grant.) The same triple charge then applies to the overhead. It gets worse when you consider that the month of salary for the highly paid PI is charged the grant, the university often hires a poorly paid adjunct to actually do the work. The university then pockets perhaps 3/4 of the salary "replacement" for the regular faculty member that the government is NOT paying to the adjunct. This is beyond unethical and should be considered outright fraud by the university. Such a situation offers room for massive savings for taxpayers.
If wealthy research universities on the coasts cannot compete on the salaries and overhead produced with smaller institutions in the heartland, then they must either find other revenue sources (such as their endowments) or lose out on grant funds to lower cost institutions. Cutting the salary support for faculty in over priced institutions to replace overhead for the institution is a reasonable response. Research oriented faculty will then have to choose whether they want to remain at such institutions where no salary support from grants will be forthcoming or move to a lower cost institution.
So...the change will cause disruption...but may actually for good for academia and science in the end.
Can't fire tenured faculty except under extreme circumstances, and that is the vast majority of the salary cost, which is a huge proportion of RU's budget.
I did some quick research and found estimates of 30% of total labor costs as tenured faculty at tier 1 research unis. Another 15% for tenure track. Does that sound plausible?
Even if we call tenured fixed, that's a lot of flexibility
First off, I'm always in favor of blaming the cats. I currently have my two draped on either side of my lap. The troublemaker tried to incite a fight by playing with her sister's tail, but I intercepted it, sparing me bloodshed, and now they're both snoozing.
Thanks for this thoughtful analysis. I hope you're right - but I suspect you're too hopeful when it comes to several of your assumptions/proposals.
The DOGE regime will not reallocate the money saved to actual science. If anything, we might see the establishment of for-profit laboratories - call them LabX? - that would benefit Musk. But I think the savings are gone from science for the foreseeable future.
Tenured faculty are mostly safe (unless financial exigency is declared, which will happen in some places). Non-tenured teaching faculty (me!), postdocs, and Ph.D. candidates are highly vulnerable. This will cripple labs.
I'm all in favor of trimming the DEI stuff, but most of that has been covered by tuition, as far as I understand. Plus, we have no reason to think teaching will be left untouched. I fully expect Trump will go full Orban and ban gender studies, federally, on pain of universities losing all federal funding. I'm an old-school women's studies person who's skeptical of a lot of the stuff that sets you off. I'm not financially able to retire, was never tenure-track, and I'm 60 years old. I could even imagine disciplines like sociology being banned under similar pretexts.
Defunding athletics is dear to my heart. But in 2008/09, athletic budgets remained untouched everyone. I gave up my office phone and never got one back. Athletics is a sacred cow.
I also love the idea of streamlining IRB (gawd yes), banning bullshit strategic planning, reducing bullshit admin tasks - but that crap is not visible at the federal level, and university deans, provosts, and presidents focus laser-like on anything that gives them a fat line on their CV.
I hope I'm wrong about all of this!
Perhaps we should loose the cats on upper administration and not rein them in until the Dark Pirate Jussim's proposals are adopted?
ETA: I work at a medium-sized public in Ohio that recently bumbled into R1 status. We have an osteopathic medical school and I think they're less research-heavy than most med schools.
I am a dept. chair, working with a modest mof-level dean, as an associate dean (some of might say ass. dean with justification). I am a PI on federal, state, and foundation projects. The whole economy of health science divisions of universities are fully dependent upon the acquisition and management of these facilities and administrative (F&A) payments I think to do this with existing federal awards, NIH is going to have to go the full Jackson, because the federal government, HHS, has negotiated all these rates in formal agreements into the future with each institution and have indicated which awards they apply to at the time of funding. Other parts of the federal government already indirect costs to 8%,15%, or other rates. I had colleagues in oncolcogy say their indirect costs are actually 85% or so. Mine, in mental health and rehabilitation are significantly lower.
NIH indirect money is greener than all others (as in greenbacks, not climate), because it is additional money. For other grants, indirects come out of the total cost of the grants. If the rate is 57%, when you are awarded $100,000 by NIH, that gets your institution another $57,000 for a total of $157K. But if you get money from the lowly National Institute of Disability Independent Living Research (NIDILRR), to pay the full indirect costs of a $100,000 projects you actually only wind up with $63,700, because 57% of that, $36,300 goes to direct costs, expending the entire $100,000. The very high rate makes for perverse incentives, such as making it unappealing to submit proposals, and odd policies at some institutions of charging indirect costs back to units that don;t get the full rate on successful applications. Selfishly I am rooting for lower F&A rates, to stop disincentivizing our faculty from making applications where high rates are directly eating into the direct costs of our non- NIH projects.
I agree if the overall funding and priorities are retained (which seems like a big if) it is a better solution that indirect cost rates are lowered (but not by this abrupt method) and have the projects document their actual costs as they vary by project, as we have to do with some state-funded projects now. This is pretty much anathema in the academic land I reside, so check back to see if I retain those dubious titles in the future.
Would a system like what you describe with the NIDILRR money, but where researchers have to negotiate with universities on how much indirects out of the grant make sense? I am wondering if it would encourage flow of high quality researchers to less expensive overhead universities (not sure if that would be good or bad); I am also wondering if it would encourage cost-cutting in order to attract researchers who do not want the burden of having a university take so much of their funding.
We would of course need a way to make sure funds are not misappropriated. Thoughts?
Prof. C. Northcote Parkinson, author of the eponymous law, pointed out that work will expand to fill all the time available for its completion. The law includes other available assets, including but not limited to money, space, and other resources. There's zero incentive to be efficient if you have barrels of money. Being forced to be efficient doesn't deprive Americans of life-saving discoveries.
It's unclear how equipment falls under indirect costs. If your research requires that you have a cyclotron, that's part of the grant application. That's not an indirect cost, but some parts of running it might be. Hardly 60% of the price tag on a cyclotron, though.
I agree that 60% is nuts. But here is how some equipment deserves to come under indirects. Consider an MRI machine that is used for neuroscience studies. It is used by many researchers at an institution. It costs $2m (ish) to purchase new. Let's further assume that depreciation of the machine occurs at a fixed pace of 5%/year for 20 years, i.e., 100k/year. It requires about $750k/yr in staff to maintain, service, operate, and administer. I made up those numbers but their exact values are not important. Dr. X obtains a federal grant that requires using the machine about 1% of the time over the next year. That comes to $1k in depreciation plus 7.5k in personnel costs that justifiably could come from indirects.
And that is just one concrete example to show why it can be appropriate.
Most of my experience in regular government service contracting, not grants. In those contracts, indirects include the overheads (costs not specific to one contract, such as rent, furniture, computer, utilities), bid&proposal costs (the costs to bid on contracts), general & administrative expenses (the cost of firm management, recruiting, etc.), fringe benefits (401k, healthcare, vacation, FICA, FUTA, SUTA), and profits (7-8%) are baked into the cost, but you have to reveal to the government what those rates are as part of your proposal, and they factor the total into the acquisition decision process. In those cases, direct costs are often pretty close to 40% of the total price. (As in, we pay the employee $40/hr, we charge the government $100/hr). The Professional Services Council typically shows these kinds of wrap rates. So, 60% doesn't seem off to me at, or a problem really, but it seems the way NIH does costs with F&A rates on top of everything else is pretty confusing. A lot of things that I would consider indirect costs are included in direct according to this: https://grants.nih.gov/grants/policy/nihgps/html5/section_7/7.9_allowability_of_costs_activities.htm
Very good post. I have been a Chair/Director and had substantial grants. While the dollar amounts lost from cutting indirect costs from over 50% to 15% will be substantial, my back of the envelope calculations of the loss to my state university, which is in the AAU, is this will account to about 5% in reduction of revenue. If the reductions in indirect costs are moved into funding for direct costs of research, that would be a tremendous boon, and allow NIH to fund a higher percentage of proposals. Dr. Prasad recently posted figures on his substack suggesting that proposals scored in the top 20% at NIH show little variation in productivity when they are funded https://www.sensible-med.com/p/randomize-nih-grant-giving
However I believe that will not be done, and we will see a decrease in NIH funding, and dramatic changes in funding priorities that will disrupt important research that aims to improve public health, particularly in terms of addressing health disparities.
You wanted comments from other academics that have served as department chairs in major research universities. I served as chair of physics at UCSD, the UC campus that gets the largest amount of grants, mostly government, and yes more than Berkeley and UCLA. The lafrgest share is from NIH, although we get large chunks from DOE (Energy, not education), DoD, NASA and NSF. Unfortunately, I cannot say that I can contribute very differing views, which is what you were asking for. One place where I would correct you is the level of startup funds. In physics, a theorist gets a startup up from 750K, and experimental scientists who have to set up a lab from scratch can easily get startup funds upwards from $2M. And I do agree that there is administrative bloat that should be eliminated. But the near term looks bleak: because of the fixed costs that you explained, the obvious way to address the cuts, which I am sure will be implemented, is to freeze hiring of staff and academics, and possibly eliminate cost of living increases for a long while. And reduce subsidies to pay for graduate students, both tuition and salaries.
Thanks! I agree with pretty much everything in there. Given that can be > $1m for a neuropsychologist, $2m for a physicist sounds completely reasonable. I'd also guess that at most U's physics brings in more federal grant money than does psychology.
I'm surprised NIH is so prominent in your funding stream. Curious if physics grants more typically come from DoD or DoE? As an undergrad, I worked in Research Administration at Stanford, and those were huge grants with about 60% indirect costs. This department covered just engineering, minus electrical engineering and comp sci, but including materials science.
I'm impressed at how UCSD became the studliest UC when it comes to grants!
Very good breakdown of the situation. I would point out though that salaries are not fixed costs but variable. You allude to that later in the essay when you point out that universities can just fire people and cut departments, but you list salaries as a fixed cost above (along with benefits, which are also variable spend.) It's a small point, but worth remembering when people who run organizations scream that a small cut to a budget has huge impacts to their mission. Very often such screams are not accurate to the reality of the situation. When looking at universities with massively bloated administrations, salaries are definitely not to be taken as fixed.
Good point. Not fixed, exactly, but still very hard to move muc because so much is tenured faculty. But yes, they can cut admins, some staff, and even some nontenured positions.
it wasn't Churchill, it was Voltaire (who was also notoriously ugly).
Vinay Prasad's claim that some of the overhead funds go toward lavish parties is probably not completely off base. Both the author, and the universities themselves, admit that some of the collected overhead is used for startup packages to hire new faculty. Hopefully the majority of this funding does go toward building up new lab equipment and hiring initial lab staff, - but this "internal" funding is always labeled as 'unrestricted' and can easily be used to host lab parties, exotic retreats, loosely research-related trips to fancy destinations, purchases of expensive coffee machines, and the like.
Thank you for this, it is very helpful to an academia outsider.
This system still seems strange to me, that the government is being charged for the product (research), but also is charged to helped the universities build up their facilities and retain their professors (and sports stadiums?!).
To me, this would be like the Federal Government purchasing some jets from Boeing but also being charged so Boeing can retool their machinery, pay admin, and invest in facilities they can permanently keep and then use to sell jets to an airline or another country, or whoever.
It seems absurd that we are paying for research but end up gifting facilities to Universities (which also then have admin bloat and charge obscene tuition fees).
Isn't there a better way to do all this? Are running national non-university-based research institutes a better option?
Second 2/10 update. I just added another short new section:
A Strange Ray of Hope
There is a valid argument for recovery of indirect costs for a research. That argument, however, would never result in indirect cost rates that vary from 5 to >60% depending on the university in question. It is especially troubling that the most wealthy universities tend to charge the highest rates of indirect costs. This suggests the government if overpaying for research at some institutions that could be done for less at others. Setting a uniform overhead raid at the correct level (not too high or too low) puts expensive universities at a competitive disadvantage...and it should.
Part of the disparity in costs is the vast difference in faculty and other salaries paid in some markets compared to others. Why should the government pay 3 times as much for a faculty member at a Boston area research university for the faculty and staff time for the same work done at a university in Kansas. (I have reviewed for grants and see faculty making $70,000/year get 1 month of their salary from a grant while a peer in Boston paid $250,000/year and gets the same 1 month of their much larger salary charged to the grant.) The same triple charge then applies to the overhead. It gets worse when you consider that the month of salary for the highly paid PI is charged the grant, the university often hires a poorly paid adjunct to actually do the work. The university then pockets perhaps 3/4 of the salary "replacement" for the regular faculty member that the government is NOT paying to the adjunct. This is beyond unethical and should be considered outright fraud by the university. Such a situation offers room for massive savings for taxpayers.
If wealthy research universities on the coasts cannot compete on the salaries and overhead produced with smaller institutions in the heartland, then they must either find other revenue sources (such as their endowments) or lose out on grant funds to lower cost institutions. Cutting the salary support for faculty in over priced institutions to replace overhead for the institution is a reasonable response. Research oriented faculty will then have to choose whether they want to remain at such institutions where no salary support from grants will be forthcoming or move to a lower cost institution.
So...the change will cause disruption...but may actually for good for academia and science in the end.
2/10/25 update. Based on conversations on X with Jim Coan and Richard Ebright, I have added two new sections to this post. They are titled:
The Main Source of Administrative Expansion
and
Addendum: Maybe Indirects HAVE Contributed More to Academic Bloat than the Above Suggests
If a CEO declared labor costs to be fixed costs, they'd rightly be mocked.
Can't fire tenured faculty except under extreme circumstances, and that is the vast majority of the salary cost, which is a huge proportion of RU's budget.
I did some quick research and found estimates of 30% of total labor costs as tenured faculty at tier 1 research unis. Another 15% for tenure track. Does that sound plausible?
Even if we call tenured fixed, that's a lot of flexibility
Since most faculty are contingent...they clearly can be fired. This is a strong argument for ending tenure immediately.
Most faculty at Rutgers are tenured/tenure track. I don't have the figures for other research heavy U's.
15% is fair, suck it up.
First off, I'm always in favor of blaming the cats. I currently have my two draped on either side of my lap. The troublemaker tried to incite a fight by playing with her sister's tail, but I intercepted it, sparing me bloodshed, and now they're both snoozing.
Thanks for this thoughtful analysis. I hope you're right - but I suspect you're too hopeful when it comes to several of your assumptions/proposals.
The DOGE regime will not reallocate the money saved to actual science. If anything, we might see the establishment of for-profit laboratories - call them LabX? - that would benefit Musk. But I think the savings are gone from science for the foreseeable future.
Tenured faculty are mostly safe (unless financial exigency is declared, which will happen in some places). Non-tenured teaching faculty (me!), postdocs, and Ph.D. candidates are highly vulnerable. This will cripple labs.
I'm all in favor of trimming the DEI stuff, but most of that has been covered by tuition, as far as I understand. Plus, we have no reason to think teaching will be left untouched. I fully expect Trump will go full Orban and ban gender studies, federally, on pain of universities losing all federal funding. I'm an old-school women's studies person who's skeptical of a lot of the stuff that sets you off. I'm not financially able to retire, was never tenure-track, and I'm 60 years old. I could even imagine disciplines like sociology being banned under similar pretexts.
Defunding athletics is dear to my heart. But in 2008/09, athletic budgets remained untouched everyone. I gave up my office phone and never got one back. Athletics is a sacred cow.
I also love the idea of streamlining IRB (gawd yes), banning bullshit strategic planning, reducing bullshit admin tasks - but that crap is not visible at the federal level, and university deans, provosts, and presidents focus laser-like on anything that gives them a fat line on their CV.
I hope I'm wrong about all of this!
Perhaps we should loose the cats on upper administration and not rein them in until the Dark Pirate Jussim's proposals are adopted?
ETA: I work at a medium-sized public in Ohio that recently bumbled into R1 status. We have an osteopathic medical school and I think they're less research-heavy than most med schools.
I am a dept. chair, working with a modest mof-level dean, as an associate dean (some of might say ass. dean with justification). I am a PI on federal, state, and foundation projects. The whole economy of health science divisions of universities are fully dependent upon the acquisition and management of these facilities and administrative (F&A) payments I think to do this with existing federal awards, NIH is going to have to go the full Jackson, because the federal government, HHS, has negotiated all these rates in formal agreements into the future with each institution and have indicated which awards they apply to at the time of funding. Other parts of the federal government already indirect costs to 8%,15%, or other rates. I had colleagues in oncolcogy say their indirect costs are actually 85% or so. Mine, in mental health and rehabilitation are significantly lower.
NIH indirect money is greener than all others (as in greenbacks, not climate), because it is additional money. For other grants, indirects come out of the total cost of the grants. If the rate is 57%, when you are awarded $100,000 by NIH, that gets your institution another $57,000 for a total of $157K. But if you get money from the lowly National Institute of Disability Independent Living Research (NIDILRR), to pay the full indirect costs of a $100,000 projects you actually only wind up with $63,700, because 57% of that, $36,300 goes to direct costs, expending the entire $100,000. The very high rate makes for perverse incentives, such as making it unappealing to submit proposals, and odd policies at some institutions of charging indirect costs back to units that don;t get the full rate on successful applications. Selfishly I am rooting for lower F&A rates, to stop disincentivizing our faculty from making applications where high rates are directly eating into the direct costs of our non- NIH projects.
I agree if the overall funding and priorities are retained (which seems like a big if) it is a better solution that indirect cost rates are lowered (but not by this abrupt method) and have the projects document their actual costs as they vary by project, as we have to do with some state-funded projects now. This is pretty much anathema in the academic land I reside, so check back to see if I retain those dubious titles in the future.
Such an arrangement would be a fantasy realized and could attract talent. Not sure what it does to the accounting.
Would a system like what you describe with the NIDILRR money, but where researchers have to negotiate with universities on how much indirects out of the grant make sense? I am wondering if it would encourage flow of high quality researchers to less expensive overhead universities (not sure if that would be good or bad); I am also wondering if it would encourage cost-cutting in order to attract researchers who do not want the burden of having a university take so much of their funding.
We would of course need a way to make sure funds are not misappropriated. Thoughts?
Prof. C. Northcote Parkinson, author of the eponymous law, pointed out that work will expand to fill all the time available for its completion. The law includes other available assets, including but not limited to money, space, and other resources. There's zero incentive to be efficient if you have barrels of money. Being forced to be efficient doesn't deprive Americans of life-saving discoveries.
It's unclear how equipment falls under indirect costs. If your research requires that you have a cyclotron, that's part of the grant application. That's not an indirect cost, but some parts of running it might be. Hardly 60% of the price tag on a cyclotron, though.
The comment is generalizable.
I agree that 60% is nuts. But here is how some equipment deserves to come under indirects. Consider an MRI machine that is used for neuroscience studies. It is used by many researchers at an institution. It costs $2m (ish) to purchase new. Let's further assume that depreciation of the machine occurs at a fixed pace of 5%/year for 20 years, i.e., 100k/year. It requires about $750k/yr in staff to maintain, service, operate, and administer. I made up those numbers but their exact values are not important. Dr. X obtains a federal grant that requires using the machine about 1% of the time over the next year. That comes to $1k in depreciation plus 7.5k in personnel costs that justifiably could come from indirects.
And that is just one concrete example to show why it can be appropriate.
Most of my experience in regular government service contracting, not grants. In those contracts, indirects include the overheads (costs not specific to one contract, such as rent, furniture, computer, utilities), bid&proposal costs (the costs to bid on contracts), general & administrative expenses (the cost of firm management, recruiting, etc.), fringe benefits (401k, healthcare, vacation, FICA, FUTA, SUTA), and profits (7-8%) are baked into the cost, but you have to reveal to the government what those rates are as part of your proposal, and they factor the total into the acquisition decision process. In those cases, direct costs are often pretty close to 40% of the total price. (As in, we pay the employee $40/hr, we charge the government $100/hr). The Professional Services Council typically shows these kinds of wrap rates. So, 60% doesn't seem off to me at, or a problem really, but it seems the way NIH does costs with F&A rates on top of everything else is pretty confusing. A lot of things that I would consider indirect costs are included in direct according to this: https://grants.nih.gov/grants/policy/nihgps/html5/section_7/7.9_allowability_of_costs_activities.htm
Very good post. I have been a Chair/Director and had substantial grants. While the dollar amounts lost from cutting indirect costs from over 50% to 15% will be substantial, my back of the envelope calculations of the loss to my state university, which is in the AAU, is this will account to about 5% in reduction of revenue. If the reductions in indirect costs are moved into funding for direct costs of research, that would be a tremendous boon, and allow NIH to fund a higher percentage of proposals. Dr. Prasad recently posted figures on his substack suggesting that proposals scored in the top 20% at NIH show little variation in productivity when they are funded https://www.sensible-med.com/p/randomize-nih-grant-giving
However I believe that will not be done, and we will see a decrease in NIH funding, and dramatic changes in funding priorities that will disrupt important research that aims to improve public health, particularly in terms of addressing health disparities.
You wanted comments from other academics that have served as department chairs in major research universities. I served as chair of physics at UCSD, the UC campus that gets the largest amount of grants, mostly government, and yes more than Berkeley and UCLA. The lafrgest share is from NIH, although we get large chunks from DOE (Energy, not education), DoD, NASA and NSF. Unfortunately, I cannot say that I can contribute very differing views, which is what you were asking for. One place where I would correct you is the level of startup funds. In physics, a theorist gets a startup up from 750K, and experimental scientists who have to set up a lab from scratch can easily get startup funds upwards from $2M. And I do agree that there is administrative bloat that should be eliminated. But the near term looks bleak: because of the fixed costs that you explained, the obvious way to address the cuts, which I am sure will be implemented, is to freeze hiring of staff and academics, and possibly eliminate cost of living increases for a long while. And reduce subsidies to pay for graduate students, both tuition and salaries.
Thanks! I agree with pretty much everything in there. Given that can be > $1m for a neuropsychologist, $2m for a physicist sounds completely reasonable. I'd also guess that at most U's physics brings in more federal grant money than does psychology.
I'm surprised NIH is so prominent in your funding stream. Curious if physics grants more typically come from DoD or DoE? As an undergrad, I worked in Research Administration at Stanford, and those were huge grants with about 60% indirect costs. This department covered just engineering, minus electrical engineering and comp sci, but including materials science.
I'm impressed at how UCSD became the studliest UC when it comes to grants!
Very interesting! Thank you for the context!
Very good breakdown of the situation. I would point out though that salaries are not fixed costs but variable. You allude to that later in the essay when you point out that universities can just fire people and cut departments, but you list salaries as a fixed cost above (along with benefits, which are also variable spend.) It's a small point, but worth remembering when people who run organizations scream that a small cut to a budget has huge impacts to their mission. Very often such screams are not accurate to the reality of the situation. When looking at universities with massively bloated administrations, salaries are definitely not to be taken as fixed.
Good point. Not fixed, exactly, but still very hard to move muc because so much is tenured faculty. But yes, they can cut admins, some staff, and even some nontenured positions.