As history has shown overwhelmingly for decades, it's perfectly possible to run a good four year college with quite reasonable costs for the student. E.g., it

As history has shown overwhelmingly for decades, it's perfectly possible to run a good four year college with quite reasonable costs for the student. E.g., it was long common for students to work off the costs just by helping in the dining hall. But, a lot of students want to go to prestige colleges. Nearly all the prestige colleges are the four year colleges of famous, high end research universities. In such universities, the teaching loads per professor are quite low, but the expenses for the research are quite high. If a student is content to make good use of an okay public four year college, hopefully live at home, keep down the cost of living, concentrate on learning and grades, then they can get a good four year education and be ready for graduate school where they need pay no tuition and maybe get a stipend. Actually, when I was a STEM field grad student, I paid no tuition, and a prof told me that they had tuition scholarship going begging from having too few well qualified applicants. A way to get an Ivy League degree is to have good grades from a four year college, good SAT and GRE scores, maybe 1-3 years of relevant work experience, and then apply to graduate school in a STEM field department of an Ivy League university. In essentially this way, I got accepted to Cornell, Brown, Princeton, and more. Much of the emphasis on research was from the US Federal government for US national security and, then, bio-medical research (all those test tubes are expensive!). The national security interest was from The Bomb, the Cold War, the Space Race, and now whatever, e.g., information security. So the DoD, NSF, and NIH made the research universities an offer they couldn't refuse: Take the money or cease to be a leading research university. Well, that Federal money is no longer so easy to get, but the research universities still want to concentrate on research, and students still think that is where the prestige and/or good education is. Then with good reputations for research, the universities get highly ranked, and students want to go there for their four year ugrad education -- which then is darned expensive. E.g., on YouTube, look at the lectures of Prof Allan Adams on introductory quantum mechanics at MIT: It appears that this is the only course he was teaching that year; it was the last year he was teaching it; and otherwise he was doing research on string theory. So, add it up anyway you want, and a seat in that course is EXPENSIVE. But, don't really need a seat in that course. Instead can buy a stack of highly regarded introductory texts on quantum mechanics (used, save money), download lectures, from Adams and possibly elsewhere, and just STUDY. If take the course at MIT and do at all well in it, still have to STUDY. Sitting in the class is not much better than watching the videos; still what's important to learn the material is to study. But if want to rub shoulders with other good students and some big time profs, then, sure, would rather be at MIT. Still, if take a course in quantum mechanics at a four year college, maybe a course not quite as well informed as at MIT, then just work to get good grades, also learn quantum mechanics from the Adams videos and the best books, do well on the physics GRE, and apply to a high end grad program in physics. Then will still have a good shot at doing as well as the MIT grads in the grad program. Similarly for other STEM fields, at least pure and applied math and, from all I can see, also computer science. Net, it really is possible to get a really good education, with really good GRE scores, in a cheap four year ugrad program, and, then, be fully competitive in a high end grad program at one of the best research universities. So, if have to sit in your parent's basement for a year getting the best ugrad STEM field education before taking your GREs and going for an Ivy League, etc. grad education, then DO that, and save some really big bucks, maybe $40,000 a year for 4 years. NBA basketball has spectators who pay big bucks and makes a good spectator sport. A good ugrad STEM field education is not a spectator sport and is essentially the same as learning to play basketball well yourself. Can't get good at a STEM field just by watching; instead have to DO it. The best video lectures, the best books, and STUDY. You CAN do that. E.g., pay all you want to MIT and take the Adams course. He still makes a total mess out of the Fourier transform. But the Fourier transform is presented with overwhelming clarity, beauty, and precision in W. Rudin, Real and Complex Analysis. At some point quantum mechanics will want to use the spectral theorem of linear operators on a Hilbert space. Well, the spectral theorem is in a finite dimensional version, intended to be a gentle introduction to the full version, in Halmos, Finite Dimensional Vector Space. There is the full version in another book, short, by Halmos, in Introduction to Hilbert Space and the Theory of Spectral Multiplicity: Second Edition (Dover Books on Mathematics) for about $10. For more, there is W. Rudin, Functional Analysis, as usual for Rudin, with astounding precision. Will have a tough time at any university finding a physics prof who will do that well with either the Fourier transform or spectral theory. So, can get the Halmos book for about $10. For Rudin, Functional Analysis, a fast Google search showed a used copy for $12.10. Why spend $40,000 a year???? Indeed, soon in high end STEM field academics it's accepted that with no more than a little guidance and some good texts, any of the good students can learn whatever they need. Basically the students are expected to be able to learn this way for their research. Once are a research prof, definitely are expected to learn this way just to keep up in the field. Net, all those $40,000+ a year ugrad expenses are for what the heck????