The federal student financial assistance program that evolved in the 1960s and grew rapidly after 1980 is by almost any measure something of a disaster. When conceived, the primary purpose of the Byzantine system of subsidized federal student loans, Pell Grants, income tax credits, work study, and parental PLUS loans was to make college more accessible to lower income kids having difficulties financing college. Yet the proportion of recent college graduates from the bottom quartile of the income distribution is lower today than in 1970 when these programs were in their infancy. Although a subject for another day, a case can also be made that federal student financial assistance has lowered educational quality, increased drop-out rates, led to reduced home buying, lowered national savings, and even lowered fertility.

Over the years, student loan debt has risen to nearly $1.5 trillion, more than borrowing for cars, home equity loans, or credit cards. The average student loan debt is over $30,000, but there are large variations around that average. The vast majority of it is federally financed: in effect, the federal government borrows (because of its budget deficits) money from bond purchasers, many of them foreign investors or governments like China. This is in itself potentially worrisome, as federal deficits are expected by the Congressional Budget Office to approach an extremely high 5% of national output in the next fiscal year starting in October.

The biggest problem is that the current student loan system encourages colleges to increase tuition fees—student borrowing is based on “cost of attendance,” so colleges have raised fees vigorously. The rate of increase in posted tuition fees has roughly doubled in the era of large federal sanctioned student lending. Changing the formula governing providing loans, as suggested by my erstwhile student and colleague Andrew Gillen, could help a good deal, but perhaps even more fundamental reform is desirable.

One intriguing idea: rather sharply reduce federal student loan availability (and ultimately perhaps end it), by further restricting the amounts and number of years of permitted borrowing, stop lending for vocationally oriented graduate programs such as M.B.A. degrees, and even impose some minimal academic performance expectations on borrowers. Simultaneously, make it clear that a new form of financing, private Income Share Agreements, is available and that these agreements are enforceable under federal law.

What are Income Share Agreements (ISAs)? They are agreements between a private investor and a potential student, whereby the investor pays a fixed amount of college costs for the student in return for a share of the student’s earnings after college completion. Whereas student loans are a debt instrument or IOU, ISAs are equity—people are selling a percentage of their “human capital.” If the student does extremely well financially post-graduation, the investor will make a good profit, but if the student does poorly, the investor probably will lose money. Purdue University has endorsed the concept by investing in ISAs for its students.

The risks of financing college are shifted from generally uninformed and financially inexperienced students to relatively savvy investors with deep pockets. But there are other, huge advantages of ISAs unless they are nullified by federal regulation interfering with contract terms. ISA terms would vary with the quality of educational institution the student attends, major subject, and expected student performance.

A student graduating at the top of her high school class majoring in mechanical engineering and attending M.I.T. probably could get $75,000 of financial assistance with a contract requiring payment perhaps 8 percent of her income for 10 years after graduation, while a below average high school student attending a non-selective admission university and majoring in gender studies might be able to obtain only $60,000, and then have to pay 15 percent of her earnings for 25 years. The differential investor assessment of varying majors, schools, and levels of high school performance would provide useful information to students.

To be sure, ISA-generated information might lead to plummeting applicants at some colleges or for some majors, even discouraging some kids from even going on to college. But arguably this is not all bad: as stated in an April 5 post, perhaps some “creative destruction” would be useful to higher education, eliminating some relatively ineffective schools and reducing underemployment of graduates in some academic disciplines. At a time of huge federal deficits, downsizing federal financial obligations toward college students is probably a prudent move fiscally. ISAs are a way to do it without denying access to students with strong potential for contributing importantly to future American life.