If Baumol was solely the reason for increased prices, I expect the increase to be more in line with the increase in prices for hiring tradespeople or other professional services.

FWIW Alex Tabarrok (economist) wrote a small book analyzing this and came away with the conclusion that it's mostly Baumol; if you're interested in a deeper dive:

https://www.lesswrong.com/posts/FCMfToFnRXhsfpfQB/book-revie...

Thanks for pointing this out -- I've heard some people back this claim, but I still find it confusing.

In a graph in that wikipedia article, we see that from 1998-2018, average hourly earnings increased by less than 100%, but tuition increased by more than 150%. If wages for university staff (who aren't becoming more productive) are rising to match wages for those who _are_ becoming more productive, why should tuition costs rise significantly faster than the rise in other wages?

https://en.wikipedia.org/wiki/Baumol_effect#/media/File:Pric...

Further, one statement in the education section sounds right to me:

> The economists Gary Rhoades and Joanna Frye went further in a 2015 study and argued that the Baumol effect could not explain rising tuition costs at all, as "relative academic labor costs have gone down as tuition has gone up".

And we know there's a trend towards shifting teaching responsibilities from tenure-track faculty to cheaper adjuncts. And from anecdata, I get the impression that in the humanities especially, salaries _are_ quite low and raises are currently below inflation.

And my understanding is also that the trend at many schools is towards larger classes e.g. when UC Berkeley increased the freshman class size by a third, there was not an across-the-board increase for departments to hire a proportional increase in faculty and staff, nor did existing faculty and staff get a raise ... so the same people are asked to just teach larger classes (does that count as a productivity increase?), and students just get less interaction with their professors.

Much of the money is going to construction or capital improvements, i.e. shiny new buildings.

Now look at the board of trustees of your local University, and play spot the construction firms...

Cost disease can't be the primary cause here. Cost disease happens when wages need to rise for people who have jobs that cannot become more efficient. But the wages of people teaching college courses have not gone up 710%. In fact, more and more of the teaching work is being done for lower pay and with more ruthless efficiency through the use of adjuncts or teaching faculty who teach 5/5s or worse.

Is that lack of scalability in hospital services and education completely inevitable due to the constraints of reality or is it mostly artificial? For example, would completely uncapping medical residencies increase the supply of doctors enough to significantly reduce the rate at which medical costs increase?

Hospital costs would benefit from some transparency/breakdown, but for medicine in general it's the increased complexity, more scans, more things to test for. But these tests are not cheap (nor fast enough). Manual labor is still required (to draw blood, to ship the vials, to put it into the test machine, to collect the results, upload it, review it, consult with the patient.

Then there's a lot more old people with a lot of complicated cases, which require a lot of person-hours.

And we still don't really have turnkey solutions. Sure, we have nice vaccines, but they are not 99.99%, there still are complications, there still are ridiculous inefficiencies (why do we still don't have combined yearly multivalent boosters for influenza+covid+others)?

Education also suffers from this overcomplication, but it's mostly because of credentialism. It's perfectly possible to watch lectures at home, complete coursework at home, and then go to some place fill out a test, and then do oral and practical exams. Grading can standardized (high-school exam is already done this way here in Hungary, there's a PDF booklet issued to teachers conducting the exam; and profs and TAs already do the same basically verbally).