The following is chapter 21 from the book The Not-So Great Society, edited by Lindsey M. Burke and Jonathan Butcher (Heritage Foundation, 2019)

For more than three centuries—from 1636 to 1944—the federal, or earlier colonial, government did virtually nothing to provide direct financial assistance to students attending college.[1] Yet the nation’s system of colleges and universities grew to arguably the best in the world, and millions of Americans, many of modest means, graduated. The Serviceman’s Adjustment Act of 1944 (the GI Bill) was designed as a form of deferred compensation for veterans, and was followed in the next decade by the 1958 National Defense Education Act, providing some financial assistance, especially for those pursuing science and engineering careers. But even in 1965, federal student loans totaled a modest $159.2 million—which, after adjusting for inflation, is equal to about one billion dollars in 2019.

The modern era of federal student financial assistance really begins with the Higher Education Act of 1965 and later extensions. The Middle Income Student Assistance Act of 1978 greatly extended grants (later renamed Pell Grants), and especially student loans, to a large number of those wanting to attend college. President Jimmy Carter bragged that “an additional 1.5 million students from middle-income families will be eligible for the Basic Grants program.”[2] In today’s dollars, the 1978 legislation provided some form of assistance for persons with incomes up to about $100,000, well above the average. Legislation initially designed to help poor people became an entitlement available for the large majority of Americans.

The administration of loans has varied over time, with large private-financial-firm participation until 2010, when the loan programs were made purely public.

These post-1965 programs are an example of the law of unintended consequences. They have had negative effects that were not completely envisioned by their proponents. Student loan debt at over $1.5 trillion in 2019 exceeds debt for credit cards or car purchases. Nearly 45 million Americans have student loan debt, with about eight million of those borrowers either in default or in “forbearance”—temporarily relieved of making timely payments. These programs need to be fundamentally revised, preferably by ending federal loans and replacing them with private forms of assistance, but at the minimum by curtailing their magnitude and pernicious impact.

Nine Problems with the Federal Aid Programs

There are at least nine problems with the U.S. government’s programs to financially assist college students.[3]

First, by far the biggest problem is that the federal programs have dramatically increased the price of college, validating a hypothesis first advanced by then–Education Secretary William Bennett more than 30 years ago.[4] Before the king-size expansion in 1978 of the modest early federal programs, college tuition for decades was rising by approximately 1 percent annually, adjusting for the overall inflation rate. Many attributed that inflationary increase to the labor-intensive nature of higher education, which is a service industry.[5] However, over the next 40 years, the tuition inflation rate rose to 3 percent annually.

The difference between 1 percent and 3 percent tuition price inflation is profound when compounded over decades. Tuition fees today would be less than half of what they are if they had simply continued to rise by 1 percent a year—which is less than income growth has been—over the past 40 years.[6] With these lower fees, the burden of college attendance, which rose as fee growth outpaced income increases, would have fallen, and there would be much less student debt today.

Colleges have used rising tuition revenues to finance a plethora of unproductive things, including a huge increase in administrative bureaucracies. It is not a coincidence, I think, that when the federal loan programs began, there were typically about two professors for every administrative staffer at the typical American university, yet today, the numbers of administrators outnumber the professors. In general, university staffers have benefited more from federal student financial assistance than have the students; students in a sense are a pawn in an elaborate academic rent-seeking exercise.

Second, the programs have failed miserably in achieving their single most important goal: increased access to college for lower-income Americans. Around 1970, when federal financial assistance programs for students were small, about 12 percent of recent college graduates came from families with incomes in the bottom quartile of the income distribution; that proportion is slightly lower today.[7] High tuition has scared price-sensitive lower-income students away from college more than affluent individuals, who view college as a necessity for future vocational success. As of this writing, some 63 percent of student loan debt of adults over 25 is owed by persons with above-average incomes. Nearly half that debt is not for undergraduate training, but for those pursuing advanced degrees.[8]

Third, the programs are Byzantine in their complexity. There are well over a dozen different loan, grant, tuition-tax-credit, and student work-study programs. To even apply for federal aid, students must complete a questionnaire—the Free Application for Federal Student Aid (FAFSA) form—of well over 100 questions—a significant deterrent to many lower-income students.[9]

Fourth, student financial assistance programs, have sharply increased the demand for higher education. Increased college enrollments have raised the portion of adult Americans with college degrees, aggravating the college-underemployment problem. An increasing number of recent college graduates are either unemployed (relatively rare) or, more likely, holding lower-paying jobs generally held by high school graduates. The New York Federal Reserve estimated the underemployment rate among recent college graduates to be over 40 percent at the end of 2018.

Fifth, the great irony is that a program aimed at easing financial problems for young Americans attending college has accentuated those problems, as federal student loan debt passes $1.5 trillion, far more than credit card debt. Because of high default rates, an honest accounting shows that these federal loan programs are a burden on taxpayers, as are, of course such grant programs as Pell Grants and income tax credits.

Sixth, rising enrollments encouraged by student loans has meant that a growing proportion of college students are not strong academic performers, often coming from the lower half of their high school graduating class and having mediocre college preparatory test scores. In order to avoid massive reputation-destroying dropout rates, colleges have engaged in aggressive grade inflation, reducing incentives for students to work hard in school. Why study much if you are going to get a grade of “B” almost no matter what you do? The typical student today spends perhaps 27 hours weekly on all academic efforts, down from about 40 hours in the middle of the past century.[10] Indirectly, the federal student assistance programs have contributed to a decline in academic standards.

Seventh, there is growing evidence that the rise in college costs as manifested in growing student loan debt has affected family formation and house purchases. There has been a decline in the purchase of homes by young Americans, who are pinched paying off large student loan debts. Likewise, the deferral of having children because of these financial considerations has no doubt aggravated an already significant decline in marriage rates and fertility rates, ironically leading to enrollment declines at colleges beginning in the next decade or so.[11]

Eighth, student loan programs are operated on a noncommercial basis, meaning that interest rates charged are not closely related to the associated risks. As a consequence, delinquency and default rates on loans are high by commercial banking standards, hovering around 10 percent today.[12] Students with a high prospect for repaying their loan, such as honors engineering graduates from top-flight universities, pay the same interest rate as mediocre students majoring in subjects with little vocational demand. Universities have no “skin in the game,” having no incentive to reject applicants whom they suspect to be poor prospects for graduation and vocational success.

Ninth, although the evidence is not definitive, an excellent case can be made that the personal savings of households has declined in the U.S. in response to the growth in student loan programs.[13] While individuals typically saved about 12 percent of their disposable personal income in the era before large federal student financial assistance, that percentage has fallen sharply, to about 6 percent or 7 percent in the 21st century, when these programs have played a major role. Reliance on student loans has led some families to not save for college as they once would have. Lower savings, other things being equal, reduce the rate of capital formation and the rate of economic growth.

What to Do? Reforming Federal Student Financial Assistance

If the federal college student assistance programs have the nine adverse consequences just stated, would ending them not alleviate those consequences, even reverse them? In general, it seems the answer is “yes,” although there are other factors affecting the economy as well, so the precise impact of an elimination of student financial assistance programs is hard to predict.

For example, would ending the federal student loan programs stop the rise in tuition fees, perhaps even reverse that trend, leading to lower fees? Certainly, the impact would be in that direction. Already, a growing consumer resistance to fee increases has led to tuition fees rising more slowly. Yet it is politically inconceivable under any realistic scenario that the federal programs would end overnight—indeed, the current political environment points to even greater federal provision of federal financial assistance to students. All of this, of course, is occurring in an era of very substantial federal government budget deficits despite overall very low unemployment and high levels of prosperity. Near-trillion-dollar annual deficits are almost certainly not sustainable long term, particularly as the unfunded liabilities associated with federal entitlements become much larger with an aging population.

In an ideal world, the government would announce a phasing out of new loans over the next several years; existing borrowers still in school could continue to borrow until they graduate, or until they drop out. Over time, the number and magnitude of student loan debt would fall toward zero. What, however, would happen to potential future college students? Some would take on private student loans, which would almost certainly increase substantially from the modest levels prevailing now. Banks lend money for the purchase of houses, cars, and other big ticket items—why not for college?

Another possibility would be for the growth in income-share agreements (ISAs). Students would sell the equivalent of equity in themselves, instead of incurring debt as at present. For example, a student might accept $50,000 in college funding from a private investor (probably a firm in the financial service industry, although possibly universities themselves) in return for 10 percent of her or his earnings for eight years after completing the education financed by the ISA. ISAs are relatively uncommon in the United States, but have begun to be used in academic settings, most notably at Purdue University, but also at some smaller schools and coding academies.

ISAs have several advantages over current federal student loan programs. Most importantly, the risk of financing college shifts from young, financially inexperienced students to seasoned but risk-taking investment experts. Second, ISAs, if not constrained by all sorts of government regulations, would provide valuable information to future students, helping them make realistic college choices. Compare two students needing $60,000 in college support, one majoring in electrical engineering contemplating going to Carnegie Mellon or Georgia Tech, and the second wanting to major in gender studies and attending Slippery Rock University. The engineering student might face having to pay 9 percent of his or her income for 10 years, while the gender studies major might have to pay much more, say 15 percent of earnings for 18 years.

Students would learn that their ability to finance college varies directly with the quality of the institution attended, the major study area, as well as the perceived probability of success based on high school experiences and extracurricular activities. Collegiate choices, such as school attended and subject studied, have major financial consequences, and ISA contract terms would spell that out in ways not presently observable to prospective college students and their parents.

All of this potentially could have a host of secondary effects. The “flight to quality” in higher education probably would accelerate—top schools would become more desirable than ever. A number of schools perceived to be mediocre would close—Schumpeterian “creative destruction,” so vital to the dynamic reallocation of resources in a capitalist society, would come to higher education, now largely shielded from it by heavy public subsidies and inadequate consumer information about educational alternatives. Politically fashionable subjects, or those tainted with ideological indoctrination offered as majors, would be subject to a market test, with a probable loss of support. Schools threatened with extinction might adopt innovative cost-saving approaches, such as three-year bachelor degree offerings (reducing the need for a large private loan or ISA), or combined online and classroom instruction programs that are less expensive than traditional residential degree programs.

A Less Radical Approach. The approach of eliminating all federal assistance described above may be theoretically appealing, but it is politically challenging. There are ways to improve the system, reduce federal involvement while maintaining support for those with the greatest needs, and improve institutional behavior. Some elements of a revised federal student financial assistance program would include:

  1. Simplification. The government should offer one federal loan program, and one grant program, period. The government should eliminate tuition tax credits, loans to parents (PLUS loans), work-study programs, and multiple types of loan-based assistance. The government should also eliminate the FAFSA form, or limit it to fewer than 10 questions, such as, for students who are dependent on their parents: What is your family income? What are the family total net assets? How many other siblings are attending college?
  2. Performance standards. Students should not be allowed to receive aid for full-time study for more than five years of undergraduate training. Students who graduate early (such as taking only three years for a bachelor’s degree) should receive a bonus for their diligence.
  3. Changing eligibility standards. Should the government finance expensive graduate or professional training? For example, should persons attending the Yale Law School to become a corporate lawyer, or Northwestern to get an MBA, receive federal financial assistance, when shortly after graduating they will be earning six-digit incomes? A lifetime loan limit of, say, $100,000, could be imposed. On the other hand, why should those wishing to earn non-degree certificates to become, say, welders or plumbers, not receive at least modest support?
  4. Skin in the game. By their admission decisions, colleges affect the magnitude of student lending. Some colleges accept students despite the knowledge, based on high school performance and test indicators, that they have low probability of success. Most of these students end up dropping out and gain little from their schooling other than a sizable debt, humiliation, and a reduced sense of self-worth. The colleges, however, gain tuition revenues, sometimes additional state subsidies, and suffer no consequences if the student drops out. If colleges were on the hook financially for an excessive rate of failure of students to repay loans, they would revise upward admission standards, lowering dropout rates.
  5. Revised Pell Grants. Very low-income students (typically well below the median household income level) could still receive some scholarship assistance provided they meet minimal academic standards (not a current provision). It would be wise to move Pell Grant funds away from college financial aid offices and give them directly to the students in the form of vouchers (scholarships). The students then would have more clout in dealing with universities, making them more responsive to student needs.

This author’s estimate is that such a revised system could reduce federal student financial assistance commitments by at least 30 percent to 40 percent, yet still maintain the principle that no students capable of postsecondary work will be denied that opportunity because of financial condition. There would be fewer college students, and probably fewer drop outs, and far less financial angst among those participating in federal programs.


The current system of federal student financial aid is, by any standards, a disaster. Either of the two approaches outlined above—complete privatization, or downsizing and restructuring—would yield a more efficient system at a lower burden to taxpayers. It is true that total college enrollments would likely fall, but given the fact that 40 percent of those entering college end up dropping out, and that also about 40 percent of recent college graduates are underemployed, this would be a good thing. Some students would be better served in non-collegiate forms of training or by working than by attending schooling for which they are ill prepared.

[1]. To be sure, beginning with the Morrill Act in 1862, the federal government began to become involved in higher education, but the pre–World War II impact of that legislation has been vastly overstated. See my “The Morrill Land-Grant Act: Fact and Mythology,” in Todd Zywicki and Neal McCluskey, eds., Unprofitable Schooling (Washington, DC: Cato Institute, 2019), pp. 31–63.

[2]. The American Presidency Project, “Education Amendments of 1978 and the Middle Income Student Assistance Act Statement on Signing H.R. 15 and S. 2539 Into Law,” (accessed August 27, 2019).

[3]. The analysis here draws directly and extensively on Chapter 8 of the author’s book Restoring the Promise: Higher Education in America (Oakland, CA: Independent Institute, 2019).

[4]. William J. Bennett, “Our Greedy Colleges,” The New York Times, February 18, 1987.

[5]. Many call this the Baumol thesis. See William J. Baumol and William G. Bowen, Performing Arts: The Economic Dilemma (New York: Twentieth Century Fund, 1966). For a modern perspective supporting this view, see Robert B. Archibald and David H. Feldman, Why Does College Cost So Much? (New York: Oxford University Press, 2011).

[6]. A variety of studies empirically support the Bennett Hypothesis and its role in rapidly rising tuition fees. To name two widely cited ones, see David O. Lucca, Taylor Nadauld, and Karen Shen, “Credit Supply and the Rise in College Tuition: Evidence from the Expansion in Federal Student Aid Programs,” Federal Reserve Bank of New York Staff Report No. 733, revised March 2016, and Grey Gordon and Aaron Hedlund, “Accounting for the Rise in College Tuition,” National Bureau of Economic Research Working Paper No. 21967, February 2016.

[7]. This is based on federal data analyzed and collected by Postsecondary Educational Opportunity, a publication of the Pell Institute.

[8]. Scott Jaschik, “Widely Cited Debt Statistic Was Wrong,” Inside Higher Education, April 29, 2019. (This article is based on an analysis by the Urban Institute.)

[9]. For a discussion of how FAFSA is an obstacle for low-income students, see Susan Dynarski and Judith Scott-Clayton, “Financial Aid Policy: Lessons from Research,” Future Child (Spring 2013), (accessed August 27, 2019).

[10]. Philip Babcock and Mindy Marks, “The Falling Time Cost of College: Evidence from a Half Century of Time Use Data,” Review of Economics and Statistics, Vol. 93 (May 2011), pp. 468–478.

[11]. Aggregate enrollments have been declining since 2011, probably the longest relative peacetime enrollment drop in U.S. history. In the short run, student loans enhanced enrollments, but in the longer run, demographic trends and increased sensitivity to tuition increases are starting to reduce them.

[12]. Ashwini Sankar, “A Rising Mountain of Student Debt,” Federal Reserve Bank of Minneapolis FedGazette, April 4, 2018, (accessed August 27, 2019).

[13]. For historical savings rate data, see The White House, “Economic Report of the President,” March 2019, p. 656, (accessed June 17, 2019). The calculation of the savings rate has varied considerably over time with changes in methodology, but 21st-century savings rates are clearly far below levels prevailing in the 1960s or early 1970s, when federal student financial aid programs were in their infancy.