Hacker News Reader

Comment by seibelj | Hacker News Reader

My wife is graduating from a PhD program and all her friends spam social media with "Republicans making my taxes go up!" posts.

News Flash - universities claim they pay way more than a grad student's stipend, then get some sort of tax break by "gifting" the tuition. So if tuition is $50k, and pay a $30k stipend, total compensation is $80k - even though a student only gets $30k salary!

Why doesn't the university just say the tuition awarded to grad students is $100k? $500k? $1million? Who cares? It doesn't really exist anyway!

In reality, they should "award" grad students $0 in tuition and pay them like what they are - employees (and deeply exploited employees at that!).

EDIT - Sure, PhD students might cost more than they are paid (although I seriously doubt it given the 80 hour weeks, classes taught, and metric ton of bullshit my wife put up with). Regardless, pretending that somehow my wife is being paid $80k and the university is doing her this big favor is a ridiculous. Universities should figure out how to fund the supposed costs some other way than through a giant, made-up tuition bill that's waived. Universities are full of the smartest people on earth, surely they can figure out how to give grants properly without this insane accounting maneuver.

View on HN

We've beaten this to death in other threads, but that's not how it (often) works.

Grad students usually don't pay the tuition themselves, but someone often does. In the sciences, this often comes in the form of training grants from the NIH or NSF at the beginning of grad school, followed by individual grants to the student (e.g., NRSA) or advisor later on. Some people also get funded by private foundations.

These funders aren't going to pay $1M per student, but they might very well pay $40k. So...the university sets tuition at $40k, takes that money when it can, and waives it when it no one else will cover it. Clearly, some kind of funding is required to put on courses, organize seminar series, and things like that.

None of this is to deny that academia is broken in many ways, but it's not broken in the way you are claiming.

>Clearly, some kind of funding is required to put on courses, organize seminar series, and things like that.

Other countries manage to do all of those with a fraction of the cost of US universities. Courses should be paid for by undergraduate students tuition, grad students attend only a handful of courses throughout their studies. Universities already take “overhead” money from the NIH or NSF grants, the “tuition waiver” is just another way to get even more “overhead” than they would otherwise be able to justify.

The real problem here is the cost disease, the fact that US schools spend much, much more than schools in other countries, for no apparent reason.

The US actually does reasonably well in certain sectors -- many non-flagship state schools provide a decent undergraduate education at approximately the same cost as German universities.

The in-state tuition at Missouri State University is $7,306 (yearly!), which is actually cheaper than the typical German education. Portland State is $9,030. I choose those because Missouri and Oregon subsidize less than other states, so tuition is closer to true spend, and I'm too lazy to calculate true spend.

Germany's excellent mid-range public transit and relatively dense population makes a big difference in terms of total cost (including room and board), since many more Germans can live at home during university. But those things are more attributable to infrastructure and especially geography than cost disease.

(The case of elite private universities is pathological. Call it whatever you want, they're just charging what they know they're worth. But they're also extremely atypical -- the US has A LOT of universities, and they have different sets of problems. Central Western Flyover State University does not suffer from the same problems as the Ivy League)

>>> The in-state tuition at Missouri State University is $7,306 (yearly!), which is actually cheaper than the typical German education

No it's not. $7k+ is not a "deal", it's outrageous.

If you take a typical course load (5 courses per semester, each of which meets for 3 hrs/week for 15 weeks), that works out to about $15.55 per hour of actual class time. This isn't including ancillary benefits like access to experts in a variety of fields (e.g., through office hours), access to the library, and student activities.

That seems like a pretty decent bargin!

$7k per student per year is definitely not outrageous cost -- when adjusted by purchasing power parity, it seems on par with the costs of European schools.

What is outrageous though is that unlike in most of the Europe, students actually have to pay for it themselves. There are around 20 million university students in the US. If every school charged less than $10k tuition per year (which is totally doable in principle, which can be seen by comparing European schools cost), it would require $200B to cover that. US government is already providing $80B to support students pursuing higher education, and remaining $120B is only 20% of military budget.

> No it's not

"the average cost of an undergraduate degree in Germany is $32,000, PAID FOR BY THE STATE" [1] (emphasis mine)

7306 x 4 = $29,224 < $32,000.

Math!

Notice I put some caveats in my original post -- US states do spend some money per pupil in high ed, but those numbers are relatively small and rapidly shrinking. Tracking state cash to specific schools is pretty hard, I wasn't able to do it, so instead I just chose states known for relatively low per pupil higher ed spending.

It is possible that the schools I mentioned are more expensive than universities in Germany. But not by much.

Also. Parent claimed that education in the US is expensive in absolute terms, insinuating waste. No one is disputing that students in the US have to pay a lot for their education. But I am disputing the common myth that all colleges in the US are absurdly expensive -- in absolute terms -- compared to universities in other countries.

[1] https://www.marketplace.org/2015/03/31/education/learning-cu...

I was required to take seventeen (10 week, 3/hour week lecture) classes during my PhD. A quick check of a comparable program requires eleven 15 week courses, just about the same. Anecdote yes, but not a handful of courses.

The first college I attended was New Mexico Tech, and that school knows how to spend money properly. As an out of state student I payed less than half of what I paid at CU for an equal or better education. Sadly, schools that do this tend to be viewed as worse than average simply because they cannot create a good enough image for people to want to come and visit from farther parts of the country.

>These funders aren't going to pay $1M per student, but they might very well pay $40k. So...the university sets tuition at $40k, takes that money when it can, and waives it when it no one else will cover it. Clearly, some kind of funding is required to put on courses, organize seminar series, and things like that.

In the article, the author claims that Yale charges 67% fees for overhead from the grant, and that this does not include the money they take for tuition. Wouldn't that pay for seminar series and courses?

In some sense, a dollar is a dollar, so sure.

However, overhead rates are supposed to cover expenses that are related to the research and difficult to parcel out. For example, ordinary janitorial work is funded by overhead because it's not obvious how much (e.g.) bathroom use is attributable to each individual research project. Similarly, the library, admin staff, etc all facilitate specific research projects, but in a fairly diffuse way.

Although you could try to make the same argument for seminar series (it improves the overall intellectual tenor of the place), I doubt it would fly because it's not essential to any specific project. There are insanely specific rules--negotiated between the government and each university--about what and how much can be charged to specific accounts.

Student goes to debt for X in order to take the classes. Someone else pays for that debt X on behalf of the student. That means the student received income X, which he or she happened to spend to extinguish his or her debt obligation. Student should pay taxes on it as income.

This is especially true since the same student is likely to be running around campus and calling his or her representatives asking them to tax the rich so his or her pet causes are funded.

The more people have skin in taxes the less stupid programs would we have.

I do think we should simplify, and correct, the way grad students are paid. But I think the tax bill, with few public markups and no informed witnesses, is an extremely scorched earth way of approaching this problem. Can America afford three years for the markets to adjust and for current grad students to drop out? Does anyone know what the fallout of this tax bill becoming law will be, and will we recover?

Yes, I recently graduated from a PhD program and this is all absurd. Your cost as an employee is always more than your take home pay, this is overhead and is basically what the "tuition" of a graduate student is akin to. It pays for the operational expenses of the university, land, maintenance, taxes, etc. No company/government lab would claim that your overhead expenses are somehow something you've earned/been awarded/gifted for the opportunity to work there. These overhead expenses can often approach the cost of your salary, especially when you make less money.

You raise an interesting point - in fact, one hypothesis for burgeoning benefits offered to employees is that they hide your compensation in various tax-shielded pots of money. From this perspective, the overhead costs of an employee and the tuition/facilities fees of PhD students are not so different from what happens in the working world.

One large exception is the teaching function of PhD programs, which is often underwritten by outside funders (or by scholarships that students, themselves pursue). I'm not aware of a situation in which an employee would herself need to secure money to pay her overhead outside of academia.

Thus, the comparison is a poor one - companies 'pay' you more by covering your overhead, but universities assess a tuition that is either paid by the student (e.g., in humanities) or by a funding organization (e.g., in the sciences).

Yep. It gets to be pretty distortive. A highly paid employee might have an effective marginal income tax rate of 40-50%, so any untaxed benefit that the employer can expense is effectively being purchased at a ~45% discount (when you compare to what the employee would have to earn in order to buy it).

That's the hole in the Amazon/Netflix HR philosophy of "lol why give you free stuff? we can just pay you more and you buy it yourself".

When I eat the free food at an employer-paid conference I think, "wow, I am cheating the tax man" (because, under a proper accounting, that would be an after-tax expense but I'd being expensed by the employer and not treated as income).

Sure, but other employers won’t put their costs on employees W2. Universities obviously spend some money in the overhead, most likely way too much, but this is a topic for another discussion. The question here is why this overhead is in the form of tuition waiver, and the answer is so that they can take this money from federal grants, on top of the overhead they already take from these grants explicitly.

If my employer pays for me to get a PhD I am required to pay taxes on the value of that assitance. Why should you get a free pass just because your employer is also the university you are getting your PhD at?

I completely agree. I'm really not understanding the counter argument here.

Actually those benefits are often exempt or deductible (at least partially):

- From the feds: https://www.irs.gov/publications/p970

Furthermore, employers tend to pay for actual classes (with a professor and meetings and all that). A lot of the outrage around this is that the research portion of a PhD doesn't actually involve any of that.

>Why doesn't the university just say the tuition awarded to grad students is $100k? $500k? $1million? Who cares? It doesn't really exist anyway!

Because it does; not all PHD programs offer full stipends all the way through; it's not uncommon at all for grad students to have to get outside grants to cover their tuition as they go (or take semesters off until they do), somehow if the University is paying you not an outside source it shouldn't count as money - despite the university paying professors/facilities/etc for you? That's just off.

Agreed that private universities should pay better, but like in any capitalist society they will only pay as much as they need to. In this case, and especially in the absence of graduate student unions (which the GOP also seem primed to do away with), not very much.

The GOP plan does away with the student tax deduction, not to simplify matters, but to punish intellectualism since they're not voting GOP anyway. You can tell, because if they actually wanted to help people, there would have to have been long discussions with heads of universities to come up with a plan to not harm grad students. Obviously, that didn't happen.

I'm not sure universities are to blame for fudging things to, say, get better tax treatment. That's just how the laws are set up. The new plan will do nothing but harm universities, their teachers, and their students. Everyone is still completely correct to blame the GOP for screwing them in this instance.

More generally, there hasn't been time for "long discussions" of any kind, much less with heads of universities (tax legislation of this scope has typically entailed months of hearings, discussions with stakeholders and affected groups, negotiations, and overall vetting of the ideas and their implementation). There will be both intended and unintended consequences, and in the zeal to implement the very specific intended ones, there has been little incentive (and in fact a disincentive) to explore and reduce the space of the unintended ones.

> The GOP plan does away with the student tax deduction, not to simplify matters, but to punish intellectualism since they're not voting GOP anyway. You can tell, because if they actually wanted to help people, there would have to have been long discussions with heads of universities to come up with a plan to not harm grad students. Obviously, that didn't happen.

Alternatively, they didn't have much confidence that the folks who are exploiting grad students would be good consultants about grad student welfare.

And failing that, there are other hypotheses--the current GOP is incompetent when it comes to executing its own agenda. The GOP could generally want to lower/simplify taxes (and I think it does), and it is merely executing poorly.

"Alternatively, they didn't have much confidence that the folks who are exploiting grad students would be good consultants about grad student welfare."

Ok, but then who did they talk to about grad student welfare?

My (limited) understanding is the claimed tuition compensation plays a big role in grant funding; universities couldn't alter the accounting to get around this new tax without messing up the same grants that pay these students.

I also suspect there's rules about charging students evenly for tuition; if they say it's now $100k/yr, that means non-funded students would have to actually be charged that amount.

"universities couldn't alter the accounting to get around this new tax without messing up the same grants that pay these students."

Sure they could. For example, universities could raise the lab space fees they charge for grad student desks to match the loss of tuition. They can raise other fees professors pay out of their grants, add new fees, etc.

Research funders set rules that determine what costs are allowable. For U.S. federal funders, tuition up to a certain maximum amount is an allowable direct cost. An arbitrary "lab space fee" is not allowable for space on a university campus.

There is a facilities indirect cost that partially pays for the cost of facilities but the university cannot increase it whenever they want. They have to justify it based on their auditable actual costs, and you can rest assured most are already claiming the maximum that they can here.

> In reality, they should "award" grad students $0 in tuition and pay them like what they are - employees (and deeply exploited employees at that!).

Here's the rub.

If universities do this, then the current tax bill is basically a cut to federal funding for science and undergraduate education.

If universities do not do this, then the current tax bill places even more of the social cost of science onto graduate students.

IMO the way out of the quagmire is two-fold.

(1) Place a strict limit of 20% on overhead -- the sum of all expenses that aren't either researcher/educator salary or research/teaching equipment should not exceed 20% within certain revenue streams. This should be required for revenues from undergraduate tuition, non-taxed endowment earnings that aren't reinvested into the endowment, and revenues from research grants.

(Why not $0? Because some amount of administration is necessary, and buildings need to be cleaned/maintained/secured. "Overhead" can mean waste, but it can also mean "literally keeping the lights on". 20% is negotiable, and perhaps this is something that varies from campus to campus, but "a lowish non-zero number" seems right.)

(2) Only after doing (1), increase funding for science and public higher education.

>If universities do this, then the current tax bill is basically a cut to federal funding for science and undergraduate education.

It isn't, though. If there is 100 million allocated by the federal government for science and undergraduate education, that number gets spent whether or not universities charge tuition. It just gets distributed differently, it isn't cut.

5 - 2 is smaller than 5 - 0...

So you think nobody will apply for the grants, period?

No, how did you infer that from my post?

They'll get a $5 grant and then pay $2 back in taxes.

The universities aren't being taxed on their Federal grants, are they? I thought this whole thing was about how Graduate students were reimbursed - if that's the case, there are solutions that don't involve a 40% tax rate, which nobody would pay anyway.

Plus, if they simply paid the graduate students more, then sure - the students would pay income tax, but again, not 40%, and the student would have more choice about how to use their income. I don't have a huge amount of sympathy for a system that is built off of bleeding funding from grants to support other university programs. Sure, more might go back to the government in the form of a tax, but that's more reasonable to me than the university taxing the government for the privilege of paying them to do research.

> Sure, more might go back to the government in the form of a tax

Right, so we agree that this effectively reduces the amount of money the USFG spends on research; 5-2 < 5-0.

> but that's more reasonable to me than the university taxing the government for the privilege of paying them to do research.

We can debate about the extent to which the federal government should support education and research, but let's stop bullshitting for one damn second. Removing tax incentives for government-funded research is exactly equivalent to reducing NSF/NIH funding. This is literally a basic arithmetic fact.

>Right, so we agree that this effectively reduces the amount of money the USFG spends on research; 5-2 < 5-0.

No, we don't. If the funding actually went to research, maybe, but it doesn't, and even then it's pretty convoluted. In any other program, if I get a grant from the government and use it to pay staff, they pay income tax on what I pay them, even though the money comes from a grant. That doesn't really mean the government spends less to fund grant programs.

>Removing tax incentives for government-funded research is exactly equivalent to reducing NSF/NIH funding.

No, it isn't. It's related, but not exactly equivalent. The money still goes to fund NSF/NIH projects, even if there is tax on that income the people were paid to do the research.

Grad students have to be listed as paying tuition because that allows their tuition to be included in grant requests which is what pays their stipend. I suspect that the tuition is higher than the compensation because you have to pay overhead on grants.

I know there's insane complexity here, but why exactly would it be bad to remove a fictitious cost from a grant? The grant can still pay their stipend, for a lower total amount because $50K of it isn't set aside for the university to move from one pocket to another.

I know other people in the university need to get paid, but isn't that why there is a huge, nebulous cost called "overhead" anyway? Fictional tuition is definitely not the only thing with "overhead" attached to it.

Overhead isn't nebulous, it pays for the operating cost of the lab. At least in the case of the lab my parents worked at, that money was crucial.

I said "nebulous" because it's such a vague line item. It's probably fine that a grant doesn't have to itemize how much is paying the utility bills, how much is paying the janitorial staff, how much is paying middle management... and grant funders are fine with this too. I am not protesting the concept of overhead.

I'm just saying, as long as there already is this line item called "overhead" that pays for the university's operating expenses out of the grant, why does there also need to be a line item called "tuition" for the same purpose?

My SO is also a PhD, but had a very nice fellowship that not only paid for everything, but upped the salary ~10k. Nonetheless, ~1/3 of it went to 'tuition', despite being in the field and away from all communication for ~2 years (off and on). Yes, the 'tuition' is meaningless BS and is a shell game the Universities are playing. Unfortunately, it will not stop from the simple reason that 22 years olds are stupid. My SO sat on some admissions committees and there are literally hundreds of super-star applications for ~10 spots. Just killer people with amazing backgrounds and drive, and they get rejected because there just aren't the spot. Grad programs have NO incentive to change as the people coming in are not going to be less numerous nor less amazing. The supply is so high and the demand is so low that the price is effectively 0 anyways. Really, if STEMy programs wanted to start charging the kids, the applicant pool would not change a bit in terms of quality of student/researcher. Yes, it would be much more WASPy, but the research wouldn't diminish in the slightest as the pool is just that large and good.

Spot on, but whoever is to blame the grad students shouldn't have to bear the burden unless we want a society of undereducated people.

Not having a graduate degree hardly means you're undereducated.

True, but MANY Middle and High school teaching positions require a Masters. Despite a severe shortage of 'STEM' teachers, districts have not relaxed this requirement and would be unlikely to do so even under the proposed tax bill.

Ah, I was not aware of that. That's especially problematic given how low teacher salaries tend to be.

They make up for it with good healthcare, pensions, and unions. You may not make much, but you'l make it to retirement. Hence the push-back against charter schools which have none, essentially.

I don't think this is that straightforward. One of the comments on the article said it nicely: graduate students are still being trained and are not very effective researchers, and they are getting a significant education in doing research even if they are not taking classes (whether it is "valuable" or not today is a different discussion).

Already the cost for a postdoc is comparable to that of a student (including tuition) and probably about twice that of a student excluding tuition, and their research output can easily be more than twice that of a student (on average). There could be adjustments to the model, but charging tuition to the students is not that outrageous when you consider what the students are learning. Tuition is also typically charged on the standard credit rate for classes (pretty low for public universities).

> In reality, they should "award" grad students $0 in tuition and pay them like what they are - employees (and deeply exploited employees at that!).

It's not always the university paying the tution fee. If the PhD student is a teaching assistant, it might be the department. If the student is research assistant, it might be the NIH or whoever is funding the project. This could be public or private (e.g., Google Faculty research awards).

I seriously doubt that in any of these cases, university would be able to claim $50k as gift since the university actually received tuition fee from the student although paid by someone else.

I fully side with the idea of taxing the benefit as income for a few reasons:

1. Tuition is overly inflated

2. The "tuition benefit" is really a waver. Having a number there means that they can attempt to hardball you on compensation. (Really what you're working with is the stipend... but they claim you're getting this money + "$50000 tuition") It's pretty much the modern day equivilent of sharecropping.

I really have no idea what the system should look like. I am surprised that the tuition waivers aren’t already taxable. If you owe somebody money, and they forgive the debt, the amount forgiven is considered a taxable benefit. That’s especially true if the “somebody” you owed the money to was the IRS.

I’m not saying all current grad students are tax cheats. I’m saying that I’m surprised the law is set up how it is, since in other cases, whenever the IRS sees a dollar value, it tries to collect a tax.

Because that's the way the law is written. The law was written that way because at one point, we, as a society, decided that we value accessible education and thought that this was a good way to provide that access.

Neither of those justifies taxing a person making 30k as if they're making 80k.

They should be treated as a 30k worker and happen to be in the agreement that it's a valueless benefit. (PhD degrees are free)

That was my thought too, but Scott Aaronson had a good reply (ctrl-f for "isn’t there some workaround").

Basically, it's not entirely an arbitrary number, because other key decisions and cash flows are based on it, and you'd have to get vastly different systems to all change a bunch of things at the same time in order to "refactor" it this way, which may be an impossible burden.

https://www.scottaaronson.com/blog/?p=3542

Where does the money to pay students come from?

Often, schools receive grants designated to pay for teaching as well as a student’s stipend. It’s not clear that any granting organization will offer $1 million, though perhaps your wife’s institution is an outlier.

In my experience it comes from grants that the adviser gets for other research.

It varies a great deal from department to department, even at the same school.

Many biomedical programs start with "rotations", where students do a few brief internships (a few months each) in different labs before choosing an advisor. This is covered by an institutional training grant like the NIH's T32 (https://researchtraining.nih.gov/programs/training-grants/T3...). Once they choose an advisor, funding comes from a variety of sources: the advisor's research grants (e.g., R01s), fellowships directly to the grad student, and even money from the university in exchange for "service" like TAing.

My impression is that training grants are rarer in the physical sciences, but there are more TAships.

If the universities pay the tuition to grad students then it will skyrocket the student's taxes, which is literally their complaint about the whole thing.

Reply on news.ycombinator.com