A peculiar thing about American colleges and universities, whether public or private, is that they typically have to buy labor in competitive markets – but sell it at a fixed price in their internal markets. When a college wants to hire a new professor, they have to bid against other schools as well as the non-academic sector. The better are the opportunities available elsewhere, the higher the salary offer must be. Thus, for example, finance professors are typically paid very well relative to, say, English professors. Why? Because someone with financial expertise usually has very lucrative prospects in the outside world, prospects not available to someone whose expertise is late-18th century English lit.
But despite their differing costs of production, the educational services produced by these two professors are not priced separately. In most universities, as far as I can tell, tuition per credit hour is the same regardless of the discipline. If you take a finance class, you have to pay only as much as you would for an English class. Students are overpaying for the low-cost English classes and underpaying for the high-cost finance classes, which means the former are subsidizing the latter. Of course, this is only a supply-side analysis; the full analysis would take into account the differing demands for finance and English. But for many disciplinary comparisons, adding the demand side would only strengthen my point, because demand is often greatest for those disciplines most costly to provide. The same fact that makes finance professors expensive to hire – lucrative opportunities in the world of finance – makes the finance major an enticing one.
It’s difficult to avoid the conclusion that the fixed pricing scheme misallocates students among majors. If prices rose for finance and fell for English, more students would choose English relative to the status quo. That might seem like a bad outcome, given the greater job opportunities of the finance major, until you take the cost side into account: the new English majors would be compensated by having fatter pocketbooks as a result of lower tuition. The larger point is that you have to do a rate-of-return analysis: the gross return on an English major might be low, but the return per dollar invested in one's education would be higher if English tuition were lower.
The question, then, is why the system is the way it is. Is it truly economically irrational, or does some other factor – such as high transaction costs or the logistics of menu pricing – justify keeping the status quo? The best explanation I can think of (at the moment, anyway) is that since most of these institutions are non-profit, they lack the necessary incentives to get the pricing scheme right. As a test of this hypothesis, we could look at for-profit institutions like DeVry to see if they have differential pricing by subject.