How Medical Research Cuts Would Hit Colleges and Hospitals in Every State
nytimes.com
nytimes.com
Indirect costs support researchers’ needs that the institution provides. For example, who ensures the researcher’s laptop is secure? For basic research, how is lab equipment obtained? Who provides infrastructure to access observational data from an electronic health record system? Categorize these services as you wish, but these services are needed, they have costs, and the expense will need to be covered for medical research to happen. If they can no longer be considered indirect, then they will become itemized and the researcher’s direct budget will have to cover them. For example, laptops permitted to access the institutional network may have a monthly fee that covers system admin time.
In the short run, it’ll mean some research simply won’t happen as the researcher doesn’t have budget for new fees the institution charges for equipment/services that were previously “free”. For new grants, it means a greater part of the research budget will have to be itemized with these costs, eg less funding for researcher or graduate assistant effort. Either the overall grant direct costs will have to go up, or less research happens.
I think many people don’t realize this point.
The universities made it easy with large indirect costs that also covered a lot.
The biggest research universities will just be able to take a lower indirect cost and start charging for things that used to be included in indirect costs.
This may sound more “transparent” or “efficient”, and it is in some ways, but adding admin work to track stuff like this is its own form of inefficiency.
It will be like moving from an AYCE model to an a la carte model, but the grant writer will need to know the orders and special requests needed by everyone many months (or more) before the grant is scheduled to start. This is not the strong suit of researchers and/or grant writers.
This will end up being a shit show.
That's not how that works. You can't change what's direct vs indirect. Direct means the cost is directly (and solely) for what the grant was written for. Indirect means all the stuff everyone uses.
Violating those rules will get you in a lot of trouble if you're caught, and since a lot of this stuff is audited, you will likely get caught.
And yes, the audit process costs money and adds to indirect costs. Just like all the paperwork and audits and so on done for programs like food stamps often dwarfs the
I may have incorrectly used a term of art there. That said…
I am certain that universities will not come out on the short end of this deal for very long, if at all.
There are myriad ways they can do this that will pass an audit. Some games I’ve seen played before the current nonsense:
- put requests from unfavored grants at the bottom of the priority list
- move unfavored grants to crappy and/or inconvenient locations (my favorite was when they split one “group” into four different buildings, basically giving them everyone else’s space scraps)
- cut rotations for things like cleaning staff to the absolute contractual minimum
- stop providing free QoL perks
- the worst case scenario is where they just provide a lower level of service to everyone
University administrations are pros at being petty tyrants — it’s their superpower. The question is in what ways will they use these powers when their cut from certain grants is reduced.
The University of Washington took 55% off the top in "indirect costs." Then of the direct costs, a bunch more taxes went to the department, the lab, etc. I ended up getting about $40,000 total from that grant to fund my actual research.
I was told that this is just how the game works.
The Lesson: There is a huge amount of waste in academic grants.
The university does not need 55% indirect costs to fund my research. They gave me a laptop and an office to work in. But the laptop actually came out of direct costs. And as for the office... well, I tended not to use it. Coffeeshops were nicer places to work.
So that 55% — $357,500 — was not necessary for research. It disappeared in an administrative hierarchy that kept growing. I talked to professors from 20 years before, and they said that the administration had grown 3x in their time being there. They built their reputation as scientists when everything was smaller and cheaper. Now all that reputation is generating more money, which is getting gobbled up by an administration that wasn't needed when they produced their original successes.
The bloat isn't really surprising: Largely inoculated from market forces, the way modern research is conducted has enormous amounts of inefficiencies, which in turn creates plenty of ways to game the system. Without university accreditation it is very hard to publish research, get grants or invites to conferences, so there is little reason for universities to take less of your grant. Why should journals not be super profitable? It is one of the very few "objective" metrics of success in academia, Nature could probably introduce expensive publishing fees tomorrow, and people would still fall in line. And universities don't really have to adapt, because going to one is basically required to get a good paying job.
I don't know how to fix that, it seems like a very hard problem. Just shutting everything down and leaving it to the private market as the Trump administration is aiming to do seems overkill, but sometimes it might also be necessary to break a stagnating system before rebuilding it better.
I love it when people misrepresent things to people who don't know better. This happens every time research funding gets mentioned on HN, by people looking to rile up HNers who absolutely love hearing that their prejudices against the government and scientists are true and academia is robbing the dear taxpayer blind, like it's all some big shady scam.
Direct costs are things that are directly and solely related to or used for the grant. Labor. Equipment. Supplies. For the research funded by the grant, within reason (nobody accounts for or cares about, say, a lab mate asking to use some of a chemical, or one of your DVD-R disks. If you have plenty of room in that incubator that was paid for with your grant and a labmate or someone in the department could really use a corner of it to try something on their experiment, or their incubator dies and you've got room? You say yes. Captain NIH does not descend and go HALT SCIENTIST! THAT EQUIPMENT WAS PAID FOR OUT OF DIRECT COSTS FOR YOUR GRANT, NOT THEIRS!"
Indirect costs are all the costs necessary for "keeping the lights on." Indirect costs are like taxes because everyone's grants are "taxed" and that pays for all the things everyone uses. Maintaining the buildings, paying the electric bill, paying the salaries of the IT staff, paying for network equipment, servers, storage, UPSs. Indirect grant costs pay for the trash (and hazardous waste) to get collected, for the parking lot to get plowed in the winter, for the buildings to be cooled in the summer and heated in the winter. In a lab it pays for stuff like centralized purified water, central gas supplies, and so on.
You don't get to cry about how your work doesn't need all that lab stuff those bio researchers need, or that big storage cluster so you couldn't have to pay for it. Just like you don't get to cry about how the road you live off of is very short so your taxes should be lower because the town has to pay less to pave and plow it, and since you don't have any kids your taxes should be lower still because you're not using the schools. And really, your house didn't catch on fire (and you're very careful about fire safety) so why do you have to pay for the fire department? etc.
Indirect costs are negotiated between schools and the organizations that award grants. A school doesn't get to make up whatever they want. It's known by everyone involved, well ahead of time. Nobody walked into Toomim's office and said "ooo, nice shiny $650k grant you got there, Michael Toomim. We're gonna just...take...about half of that. YOINK! SUCKER! Where does everyone want to go next weekend? Bermuda, or Italy? WHY NOT BOTH!"
Further, indirect costs are accounted for when grants are awarded. The awarding organization didn't think his grant proposal was worth $650k and then oooo, the Big Mean School Administration came and took half his stuff and poor Michael Toomim was left with half the money he thought he needed...frowny face :(
The organization thought the work proposed in the grant application would require $300k in direct costs. They looked up what his school uses for indirect costs, and "wrote the check" for $650k. Other grant organizations award an amount and then an additional amount is added equal to whatever the school charges for indirect costs.
Everyone knows this, grad students included. Nobody counts indirect costs as part of their grant. Ever. When they say "I got a million dollar grant!" they mean "a million dollars plus indirect costs charged by the organization."
Third, there is no fucking way whatsoever that the school took any additional money from the grant for anything except direct costs - stuff that was used solely and exclusively for that grant. Nothing in the grant was yoinked "by the lab" or "by the department." That's called "double dipping." Every org that gives grants prohibits it, and if you're caught doing it, it is a big fucking deal.
Overheads are expressed as a fraction of direct costs, not of the total grant. If the total grant was $650k, 55% overhead would be $230k. If the direct costs were $650k, the total grant was $1M.
Universities rarely have significant amounts of money they can use freely. Most of the time, if they want to do something (such as training a PhD), someone must pay specifically for that. And if the PhD student needs a supervisor, someone must pay for the time the supervisor uses supervising the student.
Your grant was not supposed to give you a huge pile of loose money to use as you see fit. It was supposed to pay your stipend and tuition (and benefits, if there were any). And give you a small amount for research expenses on top of that. You gained independence (by not being dependent on your supervisor's funding) and credibility (by winning a competitive grant). You may have also gained the ability to continue your PhD (if your supervisor was about to run out of money). But often the biggest winner with grants like that is the supervisor, who can now use their grant money to hire another PhD student.
The entire grant was $650k. 55%, or $357k were used for indirect costs. I don't know how you multiplied 650 * .55 and got 230, or why you're thinking about a $1M grant.
Assuming the entire award was subject to the 55% indirect rate (not always true), and the total award was $650K, then the direct award amount was about $420K and the indirect amount was ~$230K.
$420K direct
$420K * 0.55 = ~$230K indirect
Total: $650K
1. Universities do not typically allow graduate students to be PIs on grants. This is for liability reasons. Even for postdocs it would be highly unusual to have full, solo PI status on a grant.
2. Federal funding agencies, and the panels that review grants, would be highly dubious about a graduate student as the sole PI on a grant.
For example, NSF's Proposal & Award Policies & Procedures Guide (PAPPG) (NSF 24-1; https://www.nsf.gov/policies/pappg/24-1/ch-1-pre-submission#...) explicitly states:
"Graduate students are not encouraged to submit research proposals but should arrange to serve as research assistants to faculty members."
I can't speak to DOE or DARPA policies off the top of my head on this point.
3. Could you provide a grant/award number to lookup on NSF Award Search (https://www.nsf.gov/awardsearch), NIH Reporter (https://reporter.nih.gov/) or the award search page of whatever federal agency awarded this grant?
I wasn't the PI, and never said I was. My PhD advisor was the PI. I wrote the grant as a graduate student, and did the research.
It was an NSF grant, with a title on something about Social Computing Systems, in like 2010 or 2011 or so. You can search for it if you want.
James Landay (Principal Investigator)
Claus Portner (Co-Principal Investigator)
Interaction Economics: Instruments that Measure Social-Computational Systems
https://www.nsf.gov/awardsearch/showAward?AWD_ID=1110965
There's a single publication listed as resulting from this award (NB: I'd guess there's probably additional papers stemming from this grant; NSF award search only lists what has been published at the time of the annual report for the FY referenced).
That publication includes 4 co-authors:
Michael Toomim, Travis Kriplean, Claus Portner, and James A. Landay "Utility of Human-Computer Interactions: Toward a Science of Preference Measurement" ACM Conference on Human Factors in Computing Systems (CHI) , v.May , 2011
This doesn't sound like a "research without depending on other people" type situation, unless you're suggesting your co-authors didn't contribute meaningfully to the research.
We can't pull up the detailed budget, but let's do some quick math for what such a grant would typically fund for a CS type project.
Assuming a 3-year grant ($140K/year) the directs would have looked like this:
* 2 graduate students - 12 months of stipend, fringe, and tuition/fees. This will be the largest portion of the grant; Total: ~$95K/year (Note: 2010 grad student support costs; would be lot more now):
* 2 PIs -- probably 0.5 or 1 month summer salary ([1] see below) for each PI each plus fringe; Total: $40K/year
* A small travel budget to pay for partipants to attend a conference or meeting (like the ACM conference this paper was presented at); Total: $5K/year
[1]: Faculty in most schools of arts and sciences schools at major universities are only guaranteed a 9-month salary (paid out over 12 months). Note this can differ even within a University. For example, A&S faculty might have one standard; the School of Medicine another; etc.
But it's all so very shortsighted. The people directing all of this clearly have no idea what they're doing or why things are the way they are.
Take USAID as a good example. The reality is that USAID is a wealth transfer from the government to American companies (where most of the money was spent) AND it was an expresion of American soft power on foreign regimes. Dismantling USAID just creates a power vacuum for some other power to fill, likely China.
You see this with talk of a massive ethnic displacement from Gaza and the West Bank into Egypt and Jordan, an action that would likely topple what are essentially puppet regimes for US Middle Eastern interests.
NIH funding is actually a massive wealth transfer from the government to Big Pharma. Why? Because Big Pharma doesn't really discover novel compounds. Federally-funded research does. A lame duck Congress passed the Bayh-Dole Act [1] in a bipartisan way that was signed into law by the outgoing Jimmy Carter in 1980. This Act basically allowed Big Pharma to profit fully from federal research.
Big Pharma spends money on marketing, lobbying and any research they do is pretty much limited to patent extension.
We are witnessing the wholesale destruction of American power here. Certain accelerationists who would otherwise be in complete opposition to the current administration are actually celebrating what's going on for that reason. The government is setting up China to be the new big bad of the 21st century but ironically is creating a massive power vacuum for China to fill and extend its influence.
This is pure short-term profit thinking and we will see over the next 4 years looting of the public purse on a scale we've never seen before.
https://www.preposterousuniverse.com/podcast/2025/02/12/bonu...
If it's true that these indirect costs only contain expenses closely related to the actual research then they can just move to the actual grant with somewhat more accounting overhead. I suspect the universities are doing that accounting already for their internal purposes so it won't be that big a change.
But if it's true that a significant part of them are not related there would need to be significant changes in budgets, and whoever benefits will have a problem.
I suspect the truth is somewhere inbetween. In any case it's a good opportunity for these organizations to figure out how to become more cost effective.
They absolutely are. The salaries of faculty and postdocs, student stipends, and administrative support staff are apportioned to each grant they work on.
I told the finops people this seemed weird especially because people kept working on the original grant, so it sounded... fraudulent (I didn't put this in email, just verbally) and they said "that's how it's done".
After that I asked the finops people if I could manage my own grants directly and they didn't allow me. It was around this time I started planning my move to industry.
This is not true due to the retroactive nature of this change. The problem is that the grant budgets are already fixed and will not increase. You can't move those costs to the grant budget since the grant money is already fully allocated to things like salary, etc.
In order for this to work, DOGE would have to automatically increase the grant budgets for already awarded grants, which they are unlikely to do.
Today there are usually only a handful of overhead rates for the entire university. Accounting only needs to be precise enough to attribute the spending to the right categories. Nobody needs to know which departments or labs benefit from that particular spending.
But if everything needs to be attributed not just to a specific department and a lab, but to a specific grant, things get complicated. Maybe one lab in a department only uses offices and meeting rooms. Another runs a bunch of servers somewhere on the premises. And a third has a wetlab and works with hazardous materials. Sometimes things go wrong and the entire building needs to be evacuated. Which means there is a need for all kinds of protocols and trainings and responsible personnel. And maybe also for more security guards at night.
Expect fun department politics when the PIs argue who needs to pay what for the infrastructure and who gets to spend more on actual research.