Written quickly. Close to stream-of-consciousness.

If you haven’t kept up with recent news in the intersection between academia and politics, here is the short version: the currently debated GOP tax bill significantly increases the tax burden on graduate students, and it just passed the House.

This has been covered by Scott Aaronson, Luca Trevisan, and others. The below is a summary of information from there. Skip to the next section if you’d just like my opinion.

Currently, university’s handle PhD student tuition like this.

  • The graduate student pays \(\$X\) as tuition.
  • The university waives \(\$X\) of the tuition.
  • The university then pays a graduate student stipend of \(\$S\).
  • In the current system, stipend \(\$S\) is taxed and the waived tuition \(\$X\) is not. The student only ever receives \(\$S\) - the \(\$X\) is essentially invisible.

Under the new GOP tax bill, the waived tuition \(\$X\) will be taxed. This is a double whammy, since not only does it increase the total amount of taxable income, the increase is enough to push several students into the next federal tax bracket. A more detailed breakdown can be found here. The linked analysis shows that if nothing else changes, a typical in-state Berkeley PhD student would pay about $1400 more tax, and a typical MIT PhD student would pay about $9500 more tax. These are rough orders of magnitude for how it affects public universities vs private universities, with more damage to private universities because they have a higher tuition. Importantly, students would not get taxed because of a larger stipend - they would be taxed on money that has never entered their pockets in the first place.

As for why universities can’t simply declare that grad student tuitions are \(\$0\) - there’s some accounting trick that lets the university get more money if they give tuition waivers to grad students. I haven’t looked into the details of this.

* * *

I’m currently not in academia, for several reasons, but the big one is that I got a job offer from an industry research lab with interests close to mine. I’m certainly giving up some things, but the trade-offs fall in favor of me staying out of academia.

If I had gone to academia, I would have been okay financially, thanks to several lucky breaks. I was born in an upper-middle class family, the kind that doesn’t spend a lot of money but has money to burn. I had natural interest in math and computer science, and turns out the world’s willing to pay those people quite a bit if they enter finance or software. I liked algorithms, which happened to be the weird test of software engineering prowess in Silicon Valley - the only reason I got my first internship was because I knew the pseudocode for Dijkstra’s algorithm. And although part of my heart will always belong to the beauty of proofs, I tolerated systems enough to pick up the skills that let me handle industry.

Overall, I’ve lived a privileged life. That likely wouldn’t change in academia because CS PhDs have it easier than other departments. With careful spending, I think I could intern at a tech company some months of the year, use the money from that to fund research for the rest of the year, and still end net positive.

The thing is, these policies aren’t crippling to people like me. They’re crippling to the less fortunate.

I can’t speak for other fields, but academia for CS is increasingly a rich person’s game. Any strong PhD candidate could be at least an average software engineer, and that’s a lot of money to leave off the table. I’ve read an anecdotal story of a promising research, first to go to college in her family, and she laughed at the thought of going for a PhD. Her parents had done so much to support her. It would have been too selfish to turn down a well-paying job that could let her start paying them back.

Across all fields, the tax bill would essentially do the same: make academia more of a rich person’s game. The reason the news bothers me so much is that if it goes through, there’s going to be so much unfulfilled passion, so many students who can’t let their research interests override financial realities. It’s a duller, less colorful world.

* * *

To play devil’s advocate, the analysis above assumes nothing else about the world will change. This is very unrealistic. If the tax bill goes through as is, universities will certainly adjust - ask for more donations, decrease tuition, and make up the accounting shortfall elsewhere with even more creative costcutting. The actual tax increase would likely be lower than the current numbers.

However, I have a hard time believing that universities will be able to make up all the difference. Universities certainly have bloat, and a reduced budget provides a very strong motivation to identify that bloat - but based on what I’ve heard about university financials, I’m not convinced there’s a lot that can be cut without a fight. There are some damning numbers showing that administrators are taking up an increasingly large share of university budgets, but I’d guess that you can’t just layoff a ton of admins and expect the university to put itself back together in a reasonable timeframe.

The top-tier universities can weather this better. The lower-tier universities, less so. It’s the same rich person’s game - universities that already have trouble with recruiting grad students will have even more trouble recruiting grad students. The conclusion is similarly disappointing.

* * *

Throughout this post, I’ve been assuming academia is intrinsically valuable. That’s certainly up for debate. One argument I’ve seen is that outside of the top-tier universities, academia is a net-negative pursuit, and it would be better for society if lower-tier schools were priced out of relevance. Given the latent misery and stress of academia, and the constant self-doubt researchers have about the relevance of their own work, I think it’s worth considering this argument seriously. However, debating the merits of academia is out of the scope of this post.

To funnel everything back into RL terms (since I’m “that RL guy”) - I see academia as the ultimate extreme of the exploration-exploitation tradeoff. Industry is content to do what works, industry research labs can be more exploratory, and academia gets to consider crazy ideas that may not be relevant for decades. In my ideal society, there are always people taking crazy ideas seriously. And I mean that in a good way! Some nuts talking about water memory, some other people trying to quantify the odds we’re living in a simulation, a third group advocating that we spend the next 50 years building a model of all of ethics. The strength of academia (and the argument for tenure) is that it lets you do these things if you care about them enough.

Somewhere out in the world is a cohort of Medieval Studies PhDs, and I feel very safe saying that little of note will come from there in the next 25 years. But that doesn’t mean I want them to disappear. Do you know how insane you have to be to want to do Medieval Studies? Like, holy shit, you really really really really have to like the subject to want to spend your life doing that. How is that not crazy awesome?

The world should have room for people like that. I’m worried it won’t.