A search for “college” and “bubble” will yield many articles about rising tuition rates (up by 440 percent since 1985), frantic efforts to attend the most selective colleges, mountains of student debt, and the possibility that this whole system will soon collapse like the market for tulips in baroque Amsterdam. See, for instance, this relatively sober piece in the Chronicle of Higher Education.

I won’t quite join in the denunciations of tuition inflation. Some students receive truly inspiring and enriching college educations whose value is hard to estimate but probably exceeds the price.

Also, at institutions like mine, middle class students qualify for financial aid. It’s smart to charge a sticker price of $50,000 if substantial numbers of families can pay that much, and then offer discounts for the many families who cannot. Setting a lower sticker price would be like leaving money on the table (money that can be used for educational and scholarly purposes). The only drawback–and it’s a serious one–is the misimpression that everyone really pays $50,000. That could (and probably does) dissuade some poor and middle-class students from applying.

In any case, I’m convinced that tuition is rising faster than inflation because of the Baumol Effect, not just a bubble. Sectors like manufacturing, transportation, and retail achieve constant productivity gains because of technology. But colleges basically sell hours of attention by professors. To be more specific, they offer this implicit formula:

Hours with a professor times reputation or qualifications of the professor divided by the number of other students in the room.

If that’s the product, there is no way to cut its cost, which constantly rises relative to other goods. The cost of maintaining a faculty with a distinguished reputation rises even faster whenever the world of science and scholarship expands and research becomes more complex and expensive.

Colleges basically try to get away with alternative products: huge lecture classes, teaching by adjuncts. But savvy students who have strong positions in the admissions market choose institutions that offer as close as possible to the real deal.

Even given these partial justifications, I would be quick to admit that the sector has a problem. Many students do not have deeply enriching experiences in college, if only because they are not prepared or motivated for the best aspects of higher education (seminars on Plato, real scientific work, service-learning). Even for students who do get a good education, the sticker price is awfully high, and neither the Baumol Effect nor financial aid discounts excuses it fully.

Part of the problem is that college is a positional good: you look better in the job market if you graduate from an institution that has a higher rank than others. Colleges that are selective tend to rise on most people’s lists (whether implicit lists or literal rankings like that of US News & World Report.) If the nation’s best students all decided to flood NoName State, its mean SAT scores would rise, its acceptance rate would plummet, and the value of its diploma would soar. But that’s not how the market actually works. Distinguished faculties, ancient campuses, and beautiful facilities are what draw competitive students. These are signals that other strong applicants will apply. Markers of prestige cost money, and once you have them, you can get away with charging high tuition because of the competitive advantage that your diploma will offer.

Because people believe that expensive diplomas offer competitive advantages, the system sustains itself. I think the bubble would burst if:

1. Prospective students decided en masse that prestigious degrees did not offer tangible benefits. There is some evidence that the benefits have been exaggerated, and if that news gets out, it could burst the bubble. On the other hand, families obtain non-economic advantages from prestige, like the opportunity to brag that their kids go to Stanford.
2. All the really smart kids decided that they didn’t care about prestige but only about learning. They would all have to decide this at once, because we learn from smart peers in college. As long as the most competitive applicants chose to attend the most prestigious and expensive schools, that is where the smartest peers –and best discussions, and hardest classes–are.
3. Lots of competitive applicants decided all at once to go to low-cost schools, in which case the prestige of those institutions would rise but the price of prestige would fall, at least temporarily. It’s hard to see this happening, because applicants can’t coordinate their decisions. Hence the most competitive ones apply to expensive and famous institutions.

Because each of these scenarios seems unlikely, I doubt the bubble will burst. But the costs of college are too high. That creates barriers to attendance and encourages harmful economies, such as exploiting adjuncts. The only way I can think of to cut the actual cost of higher education is to find valid alternatives to the formula stated in italics above. An example would be an elaborate computer-based simulation that was as valuable as time with a professor. That would address the Baumol Effect by bringing productivity gains to higher education.