what nonacademics should understand about taxing graduate school

There are many bad provisions in the proposed tax legislation. This isn’t even the worst of them. But it’s the one that most directly affects my corner of the world. And, unlike the tax deduction for private jets, it’s one that can be hard for people outside of that world to understand.

That proposal is to tax tuition waivers for graduate students working as teaching or research assistants. Unlike graduate students in law or medical or business schools, graduate students in PhD programs generally do not pay tuition. Instead, a small number of PhD students are admitted each year. In exchange for working half-time as a TA or RA, they receive a tuition waiver and are also paid a stipend—a modest salary to cover their living expenses.

Right now, graduate students are taxed on the money they actually see—the $20,000 or so they get to live on. The proposal is to also tax them on the tuition the university is not charging them. At most private schools, or at out-of-state rates at most big public schools, this is in the range of $30,000 to $50,000.

I think a lot of people look at this and say hey, that’s a huge benefit. Why shouldn’t they be taxed on it? They’re getting paid to go to school, for goodness sakes! And a lot of news articles are saying they get paid $30,000 a year, which is already more than many people make. So, pretty sweet deal, right?

Here’s another way to think about it.

Imagine you are part of a pretty typical family in the United States, with a household income of $60,000. You have a kid who is smart, and works really hard, and applies to a bunch of colleges. Kid gets into Dream College. But wait! Dream College is expensive. Dream College costs $45,000 a year in tuition, plus another $20,000 for room and board. There is no way your family can pay for a college that costs more than your annual income.

But you are in luck. Dream College has looked at your smart, hardworking kid and said, We will give you a scholarship. We are going to cover $45,000 of the cost. If you can come up with the $20,000 for room and board, you can attend.

This is great, right? All those weekends of extracurriculars and SAT prep have paid off. Your kid has an amazing opportunity. And you scrimp and save and take out some loans and your family comes up with $20,000 a year so your kid can attend Dream College.

But wait. Now the government steps in. Oh, it says. Look. Dream College is giving you something worth $45,000 a year. That’s income. It should be taxed like income. You say your family makes $60,000 a year, and pays $8,000 in federal taxes? Now you make $105,000. Here’s a bill for the extra $12,000.

Geez, you say. That can’t be right. We still only make $60,000 a year. We need to somehow come up with $20,000 so our kid can live at Dream College. And now we have to pay $20,000 a year in federal taxes? Plus the $7000 in state and payroll taxes we were already paying? That only leaves us with $33,000 to live on. That’s a 45% tax rate! Plus we have to come up with another $20,000 to send to Dream College! And we’ve still got a mortgage. No Dream College for you.

This is the right analogy for thinking about how graduate tuition remission works. The large majority of students who are admitted into PhD programs receive full scholarships for tuition. The programs are very selective, and students admitted are independent young adults, who generally can’t pay $45,000 a year. Unlike students entering medical, law, or business school, many are on a path to five-figure careers, so they’re not in a position to borrow heavily. Most of them already have undergraduate loans, anyway.

The university needs them to do the work of teaching and research—the institution couldn’t run without them—so it pays them a modest amount to work half-time while they study. $30,000 is unusually high; only students in the most selective fields and wealthiest universities receive that. At the SUNY campus where I work, TAs make about $20,000 if they are in STEM and $16-18,000 if they are not. At many schools, they make even less. (Here are some examples of TA/RA salaries.)

Right now, those students are taxed on the money they actually see—the $12,000 to $32,000 they’re paid by the university. Accordingly, their tax bills are pretty low—say, $1,000 to $6,000, including state and payroll taxes, if they file as individuals.

What this change would mean is that those students’ incomes would go up dramatically, even though they wouldn’t be seeing any more money. So their tax bills would go up too—to something like $5,000 to $18,000, depending on their university. Some students would literally see their modest incomes cut in half. The worst case scenario is that you go a school with high tuition ($45,000) and moderate stipends ($20,000), in which case your tax bill as an individual would go up about $13,000. Your take-home pay has just dropped from $17,500 a year to $4,500.

What would the effects of such a change be? The very richest universities might be able to make up the difference. If it wanted to, Harvard could increase stipends by $15,000. But most schools can’t do that. Some schools might try to reclassify tuition waivers to avoid the tax hit. But there’s no straightforward way to do that.

Some students would take on more loans, and simply add another $60,000 of graduate school debt to their $40,000 of undergraduate debt before starting their modest-paying careers. But many students would make other choices. They would go into other careers, or pursue jobs that don’t require as much education. International students would be more likely to go to the UK or Europe, where similar penalties would not exist. We would lose many of the world’s brightest students, and we would disproportionately lose students of modest means, who simply couldn’t justify the additional debt to take a relatively high-risk path. The change really would be ugly.

All this would be to extract a modest amount of money—only about 150,000 graduate students receive such waivers each year—as part of a tax bill that is theoretically, though clearly not in reality, aimed at helping the middle class.

It is important for the U.S. to educate PhD students. Historically, we have had the best university system in the world. Very smart people come from all over the globe to train in U.S. graduate programs. Most of them stay, and continue to contribute to this country long after their time in graduate school.

PhD programs are the source of most fundamental scientific breakthroughs, and they educate future researchers, scholars, and teachers. And the majority of PhD students are in STEM fields. There may be specific fields producing too many PhDs, but they are not the norm, and charging all PhD students another $6,000-$11,000 (my estimate of the typical increase) would be an extremely blunt instrument for changing that.

Academia is a strange and relatively small world, and the effects of an arcane tax change are not obvious if you’re not part of it. But I hope that if you don’t think we should charge families tens of thousands of dollars in taxes if their kids are fortunate enough to get a scholarship to college, you don’t think we should charge graduate students tens of thousands of dollars to get what is basically the same thing. Doing so would basically be shooting ourselves, as a country, in the foot.

[Edited to adjust rough estimates of tax increases based on the House version of the bill, which would increase standard deductions. I am assuming payroll taxes would apply to the full amount of the tuition waiver, which is how other taxable tuition waivers are currently treated. Numbers are based on California residence and assume states would continue not to tax tuition waivers. If anyone more tax-wonky than me would like to improve these estimates, feel free.]