Ask 100 randomly selected Americans, “What is the biggest problem with American universities?” I would bet the most frequent answer would be, “they cost too much.” The published tuition fees of American colleges have roughly tripled over the past 40 years, rising faster than income levels. Almost everything you can name today is less burdensome to buy today than in, say, the 1970s, with the major exception of higher education and, arguably, medical care.

The two big questions are: How did this happen? And what can be done about it? I will address the second of these questions in numerous future posts. But today, let’s consider why fees have risen so much. The demand for higher education has risen a lot relative to supply. But why? Here are five frequently cited explanations. First, the Baumol hypothesis (named for the late economist William Baumol): Higher education is an inherently costly labor-intensive-type activity where productivity advance is nearly impossible, leading to big price increases over time. This is also sometimes called Baumol’s Cost Disease. Second, there is Howard Bowen’s Revenue Theory (named for the late economist and college president): Colleges spend whatever revenue they raise. Absent a profit motive, there are almost zero incentives to reduce costs and conserve resources. Public university presidents, however, emphasize a third theory: public financial support for higher education in some sense is declining. Fourth, conservative theorist William Bennett’s hypothesis is that vastly expanded federal financial assistance programs have led universities to raise fees to capture increased financial aid; colleges gain more than students. Fifth, the ability to expand fees depends largely on rising demand, which in turn reflects mainly the growth in the earnings premium associated with a college diploma over time. College is expensive, but it is worth it.

In chapter six of my forthcoming book, Restoring the Promise, I discuss all this in detail, but I put a lot of emphasis on the Bennett Hypothesis. Before the massive growth of federal student assistance programs, inflation-adjusted college fees typically rose about 1% a year, reflecting some of the other factors cited above, especially Baumol’s Cost Disease. But 1% annual fee growth was bearable since incomes were rising faster. Since federal student aid started rising massively in the late 1970s, inflation-adjusted fee growth roughly tripled, going to about 3% a year, far more than incomes have risen, making college a far bigger financial burden, leading to the student loan crisis. Only in the past two to three years have there been signs that the big upward surge in tuition fees is abating, as market forces have started wreaking creative destruction on colleges previously adequately protected by large public subsidies.

In an ideal world, we would get rid of dysfunctional federal student aid programs. They are Byzantine in their complexity, have failed in the initial goal of increasing the proportion of low-income students receiving degrees, have raised costs, led to lower academic standards (e.g., grade inflation), subsidized poor academic performance, and helped promote “underemployment” among graduates. But that is probably politically infeasible, and there would be some serious transition issues if schools abandoned the current system. A second-best solution would be to curtail federal student loan programs considerably, doing away with PLUS loans, putting stricter limits on eligibility, restricting borrowing for some graduate programs (like M.B.A.s), and even implementing some modest minimum academic standards currently lacking. We could also create an environment allowing for more private funding, probably by introducing Income Share Agreements, where students sell equity in themselves (a share of postgraduate earnings for an agreed-upon number of years). Finally, to discourage excessive loan commitments to students almost certainly unlikely to graduate, we should make colleges have some skin in the game. They should pay some portion of unusually large loan defaults incurred by their students. Incentivize students to work harder, colleges to be more hesitant about luring unprepared students into school, and allow the risks associated with financing college to accrue more to seasoned investors than to relatively uneducated students.

Other things can be done to contain college costs: for example, have professors teach more, reduce the number of campus administrators, and constrain intercollegiate athletics. I will talk more about these and other things in future posts.