Academics can subsidize some research costs through tuition and splitting work researchers do to teaching. It's a business model that has been working well because of ever increasing demand for people to get at least a bachealors degree. It's not really about tenure per se, it's more that the academic business model can provide a stable revenue stream for some research work.
Labs that don't have this often require either a business to subsidize their work or they suddenly become completely reliant in grants. These environments are highly unstable in terms of job security. Positions are tied to a grant or a mixture of grants and difficult to maintain. If one or enough sources fail and the role is completely paid through grant money, it suddenly becomes untenable and people leave the role. The advantage academic environments have is that revenue stream to cushion the instability and provide stability.
It also lures in cheaper labor from grad students and post docs which helps.
I had the impression that even at very prestigious schools lab funding is mostly from grants.
I'd be interested to see how much of, say, the UNC Chapel Hill chemistry department's research expenditures come out of grants. Do you know how I'd find the information?
Two other points:
> Labs that don't have this often require either a business to subsidize their work or they suddenly become completely reliant in grants.
In what sense is the first case, a business sponsoring the lab, not "grants"?
> It's not really about tenure per se, it's more that the academic business model can provide a stable revenue stream for some research work.
But the original claim was that the cost of employing researchers is lower for an academic lab than it would be for a new lab. ("other labs could hire many more grad students for the same cost.") That has nothing to do with the availability of funding or cross-subsidies! Is it true or not?
Grants are definitely also important in academia, but at least the base salary of professors and some of the grad students ends up being covered by the university out of separate revenue sources (mostly tuition, but sometimes state funding or endowment), which makes it a bit easier. It also means that if you go a few years between grants you still get paid, and you can still get your grad students paid by having them be TAs for classes (instead of RAs on your grants). I'm not sure how to get hard data on what percentage of expenditures in a given department come from which sources, which would definitely be interesting.
There are so-called "soft money" positions that really are 100% grant-funded. In academia, people with those usually have a title like Research Professor and they're expected to pay their own salary out of their grants! They have no teaching responsibilities, since the university isn't directly paying them anything. But imo this is a much more stressful arrangement since you don't have the regular university salary to fall back on. Some researchers at small companies are essentially like that, having to bring in a stream of SBIRs to keep themselves paid.
I don't think researchers' base employment necessarily has to go through universities, it's just ended up as the most common mechanism for researchers to get a stable salary. The Institute for Advanced Study [1] is an alternate model, funding researchers' salaries out of its nearly $1 billion endowment and a stream of donations and institution-level grants, without being attached to a university. Kind of like a "think tank", but for science. But institutions along that model currently (at least in the U.S.) employ a tiny number of researchers compared to either universities or industrial research labs.
> I'm not sure how to get hard data on what percentage of expenditures in a given department come from which sources, which would definitely be interesting.
Well, I picked UNC Chapel Hill as the example mostly because it's a public school and therefore (as far as I know) its budget is public information. Is that not true?
I'm perfectly happy to do the calculation in terms of "the chemistry department spend $X this year, and received $Y in grants, so their funding comes out of grants in a Y/X proportion".
> There are so-called "soft money" positions that really are 100% grant-funded. In academia, people with those usually have a title like Research Professor and they're expected to pay their own salary out of their grants!
I was aware of this (well, not that there was a title difference), but there is something I've been wondering about.
When salaries are part of the purpose of the grant, are they itemized in the grant? It obviously doesn't constitute corruption or embezzlement to help yourself to some of the grant money -- that's what a salary is. How much of it can you take?
> it's a public school and therefore (as far as I know) its budget is public information. Is that not true?
Public university budgets should generally be public information, but tbh I don't have any idea where I'd start looking for granular data, especially in an easy to process format.
> When salaries are part of the purpose of the grant, are they itemized in the grant? ... How much of it can you take?
Yeah, pay is itemized on grants. Despite being expected to bring in your own salary, there is still some kind of official pay grade you're at. How that's set when you're hired I don't have great insight into (I imagine it varies a lot). But after the initial hire generally pay raises work the same as with any other faculty (annual cost of living or merit increases, occasional promotions between ranks, etc.). The grant agencies all work in terms of "percent effort", where you say what percentage of your time you're going to spend on this grant, and the amount requested is that percentage times your annual salary, plus a percentage for benefits. People on soft money positions sometimes piece together support from multiple grants. So you might write a grant as PI asking for 75% support over 3 years, then on a different grant your colleague is writing, they ask for money to cover 25% of your time as co-PI over 3 years. Larger institutions might also have some internal money available to smooth over shortfalls. Also, some agencies have a salary cap (e.g. NIH's is $200k).
> The grant agencies all work in terms of "percent effort", where you say what percentage of your time you're going to spend on this grant, and the amount requested is that percentage times your annual salary, plus a percentage for benefits.
This seems like it would tend to promote collusion between researchers and universities, where whenever somebody is hired into a soft money position, they would prefer to have a high official salary, and the university doesn't care because the actual salary is $0. The granting agencies aren't a party to salary negotiations -- what's keeping salaries down? Say someone's job is to get grants from NIH. Why would they ever be salaried below $200k?
Communication is a two-way street, if your questions are not being understood maybe you should put some effort into making them clearer.
Non-academic labs can't offer tenure because there's no tradition of it outside academia and by its very nature it only has value if an organisation can make a credible commitment to maintain for decades or centuries.
Labs that can dangle the carrot of a tenure-track position can pay much less than those that can't; workers overvalue it both because people generally overvalue potential prizes and because it's been gradually made rarer and pushed back further and further.
> Non-academic labs can't offer tenure because there's no tradition of it outside academia and by its very nature it only has value if an organisation can make a credible commitment to maintain for decades or centuries.
This would imply that it's impossible for non-academic institutions to offer pensions, too. Now, they are certainly moving in that direction, but it seems strange to argue that they can't.
It would imply that prospective workers don't put a lot of value on private companies' pension plans when weighing employment offers, which in my experience is true (employees do perhaps value governmental workers' pensions, because the government is the kind of stable organisation that can make multi-decade commitments). There's also a level of governmental support (high-profile bailouts and the like) of private pensions that lends credibility to pensions, whereas governments seem if anything anti-tenure.
Labs that don't have this often require either a business to subsidize their work or they suddenly become completely reliant in grants. These environments are highly unstable in terms of job security. Positions are tied to a grant or a mixture of grants and difficult to maintain. If one or enough sources fail and the role is completely paid through grant money, it suddenly becomes untenable and people leave the role. The advantage academic environments have is that revenue stream to cushion the instability and provide stability.
It also lures in cheaper labor from grad students and post docs which helps.