I fully agree that the cost of higher education is a little out of control, but it seems to me that you have to look at its historical value as well as its current value in order to determine whether there is a bubble. It seems to me that university educations were highly under-valued in the past, and so part of the price pressure is just coming from the realization that a university education is much more valuable than what it used to cost (and K-12 is even more valuable than that, but thankfully it's publicly supported).
There is also little reason to think that higher education should be bound to CPI in any linear way - there are lots of other quasi-economic influences that are perfectly valid, although greed does seem to play its part.
There's two ways to look at the issue. One is the question of value generated by the degree. If the degree adds a million dollars to your total income, it is natural for the education industry to try to capture more of that value. By that metric it all looks peachy.
In a free market, the relevant metric is something more like at how low a price you can bring the service vs. your competitors. It beggars the imagination to believe that the current price charged is the actual minimum price for the services that people are actually looking for. The question of the intangibles of the college experience is incidental, first because even those intangibles can almost certainly be had for much cheaper, and secondly because it isn't necessarily the case that people want them, regardless of how much you want people to want the same intangibles you do. In addition to the fact the intangibles could probably be had for cheaper, we have this Internet thing slowly but very, very surely eroding away the physical constraints of the old system.
They've been really shielded from the free market by easy subsidies and loans and the fact that the unaccredited competition was neither socially accepted nor able to be quite cheap enough to break through, but all of these things are suddenly changing. The costs to a startup or something of providing a college-quality education continues to drop, and the business scales fairly well. We're getting over the idea that college = good with the increasing evidence that it's not true, which will continue to open the door to non-traditional-college alternatives, some of which will begin to exceed their college breathren. The subsidies are likely to become harder to get and increasingly scary, as word of things like the non-dischargability of student debt in bankruptcy are increasingly well known. Every pillar propping up the current system is collapsing over the next few years at once.
As the college education market gets forcibly introduced to the free market, I expect the blood of the incumbents to flow freely. Like all bubble pops I can't predict exactly when it'll happen ("the market can stay irrational longer than you can stay solvent"), but I'm pretty confident about the outcome, based on the fundamentals.
(How confident? I am not planning my finances for my kids on the assumption that college is going to cost them each $50,000/yr in 2010 dollars. I computed it once and at current rates of tuition growth that's a reasonable projection for a 2-year-old and a not-yet-born. It's just physically impossible for a college-equivalent education to cost that much. So you might say I'm betting quite a lot on this analysis.)
I remember reading a number of years ago about how much it costs CalTech to educate one of its graduates, and it was something on the order of $300k/year. Granted, engineering and particularly of the intensity and small scale of CalTech makes for expensive students, but it's not true that tuition fees at many of these schools are wildly disproportionate to the amount they spend offering what they offer. Sure, many/most universities are heavy on luxury and have spending priorities that are not well-aligned to pure education, but the fact is they do offer something that cannot be had for less money - they rely on many many funding sources besides tuition in order to break even.
Here's an article about Florida universities - the costs are much more reasonable, at about $40k per graduate (all four years), but it's using 2005-ish figures, and that's an average, with degrees ranging from ~$21k to $250k+.
Of course, we're talking about a lot of 'intangibles' of the experience that are not directly related to the 'education', but for many/most applicants, the experience is a huge part of the decision about which school to go to, and so I feel it's reasonable to factor that in as a cost/expense that is not going away any time soon.
I fully agree that the free market and particularly alternatives to university will be a great influence over the next few decades - it will be a great advance once more people realize that a university education is not the best life decision.
"I remember reading a number of years ago about how much it costs CalTech to educate one of its graduates, and it was something on the order of $300k/year."
But that's not the question. With all due respect, it's not even all that interesting. If they had a million dollars to spend per student, by golly, they'd spend a million dollars.
The question is, how much would it cost in a competitive market? And given that engineers are the minority, how much should an English degree cost in a competitive market?
How much can you save by not even being on campus? By not using professors, which rhetoric aside you simply do not need for most courses? (I'd much rather see more professors doing more research and teaching grad-level courses.) How many teachers do you even need? Why are we paying people to do the moral equivalent of going to the front of the class and reading their notes? (With varying degrees of literalness.) And then paying them again to do it next year? And then paying hordes of such people to deliver the same classes to hundreds of institutions? To put it in HN terms, just how disruptive could you get while delivering the same basic product, or possibly even a better one?
"and so I feel it's reasonable to factor that in as a cost/expense that is not going away any time soon."
People say this a lot. I think people are culturally conditioned to say it. God forbid we radically rethink anything about education. But I think that you gave someone a choice of spending 20K/semester and living on campus, or spending 2K/semester and living where they chose and getting a degree just as good in the vast majority of ways, that not only would quite a lot of people made that choice ten years ago, but in the coming years, when money is going to be a lot tighter for the middle class, the 20K option is likely to just collapse. Just about the only thing holding it up right now is a whole lot of people like you who aren't willing to concede the possibility that quite a lot of the accouterments of college as we now know it are incidental to the the actual product.
Don't think with your standard cultural conditioning; think like a disruptive startup founder. The fundamentals of this industry are all wrong and it's going to burn. For better and for worse; disruptions are rarely all for the better. I'm not saying I celebrate every aspect of the change, but the fact that I think it's a mixed bag isn't going to change the absolute economic inevitability of the changes ahead.