Student Loans Aren't Working
But free college won't fix this slow-moving catastrophe.
This article is part of a feature package on how the American Dream became unaffordable for millions of working-class and middle-class Americans. For more on Reason's autopsy of how things veered off track, read "How the American Dream Became Unaffordable" or the other two features in the package, "How Doctors Broke Health Care" and "Can't Afford Your Rent? Blame Herbert Hoover."
On the day he signed the Higher Education Act of 1965, President Lyndon B. Johnson declared that the law would "swing open a new door for the young people of America" and provide "a way to deeper personal fulfillment, greater personal productivity, and increased personal reward." Johnson wanted Americans to know that his government would do whatever it could to help "every child born in these borders to receive all the education that he can take."
Signed at Southwest Texas State College, Johnson's alma mater, the Higher Education Act authorized federal scholarships and federally funded part-time jobs for students who could get into college but couldn't pay for it. But the real catalyst for increasing college enrollment was a provision that allowed the government to directly lend students money for tuition and to guarantee loans made by other entities. This new lending authority, Johnson said at the signing ceremony, would allow the federal government to issue to "worthy, deserving, capable students" loans "free of interest and free of any payment schedule until after you graduate."
The modern approach by which students finance higher education grew on and around these three policies—federal scholarships, federal work-study funding, and federally guaranteed loans—like vines around a trellis. Yet half a century later, many young Americans feel that Johnson's signature education initiative has saddled them with excess debt and delayed their graduation into middle-class adulthood.
The Occupy Wall Street protests of 2011 started as a cacophony of disparate complaints against the financial sector but eventually coalesced around college debt. The idea of forgiving student loans has since moved from poster boards on the streets of Manhattan and D.C. to the campaign websites of Democratic presidential candidates, with Sen. Bernie Sanders (I–Vt.) promising to wipe out student debt for all borrowers and former Vice President Joe Biden pledging to expand existing loan forgiveness programs.
Johnson wanted future generations to think of the Higher Education Act of 1965 as a promise from the federal government: "Tell them that we have opened the road and we have pulled the gates down and the way is open, and we expect them to travel it." He and the 89th Congress paved that road with the best intentions, and many millions of young people have indeed traveled along it. Why, then, do so many Americans who have participated in our system of financing higher education feel like they've been ripped off?
Because many of them were.
The first student loan crisis occurred more than 20 years before Occupy Wall Street protesters set up camp in Zuccotti Park. Yet it "followed a strikingly similar path to the more recent experience," the Brookings Institution's Adam Looney and the University of Chicago's Constantine Yannelis wrote in a January 2019 working paper titled "The Consequences of Student Loan Credit Expansions: Evidence From Three Decades of Default Cycles."
In the 1980s, an alarming number of student loan borrowers began defaulting on their payments. In 1985, the default rate for federally guaranteed student loans—the percentage of borrowers who had failed to make a payment within 180 days of the repayment period—jumped to 11.7 percent from 10.7 percent the year before. The federal government's payments to lenders who couldn't collect from student borrowers rose from $749 million to over $1 billion. "The financial implications…are staggering," U.S. Education Secretary William Bennett said in August 1985. Without major policy reforms, he projected the default rate would increase to 13.6 percent by 1990.
He was wrong. The student loan default rate for all borrowers was roughly 22.4 percent in 1990. By 1992, the federal government's annual cost for default payments was nearly five times what it had been in 1985. The agency then known as the General Accounting Office (GAO) estimated in a 1993 report that the Department of Education the year prior had "paid about $5 billion in default claims and interest subsidies."
But the student loan default explosion was not occurring across the entire education sector. The Department of Education's data suggested, and the GAO and the Office of Management and Budget would later confirm, that students of for-profit or "proprietary" schools—then and now the largest providers of technical and vocational training to American workers—were defaulting at much higher rates than other types of students, largely due to bad federal incentives.
"While proprietary school students comprised about 22 percent of all Government Student Loan Program borrowers who received their last loan in academic year 1983, they accounted for 44 percent of defaulters as of September 30, 1987," read a bipartisan 1990 report from the Senate Permanent Subcommittee on Investigations. "Over that period, the student default rate for proprietary schools was 39 percent, as contrasted to a 10 percent rate for four-year public and private schools."
The rate of default was essentially eating the Government Student Loan Program (GSLP) program from the inside. "The cost of defaults, as a percentage of all GSLP program costs, rose from about 10 percent in [fiscal year] 1980 to 36 percent in [fiscal year] 1989, and to more than 50 percent in [fiscal year] 1990," the Senate report said. "In other words, currently more than half of the government's GSLP cost is going to pay for defaulted loans from the past rather than to subsidize education and training for today's students."
In their paper, Looney and Yannelis trace the default crisis of the late 1980s and early 1990s to Congress' 1972 reauthorization of the Higher Education Act, in which it allowed for-profit or proprietary colleges to participate in the GSLP. When it reauthorized the act in 1976, Congress extended eligibility to applicants who had not completed high school and encouraged states that hadn't done so already to create their own "guaranty agencies," which would partner with the federal government to initiate and guarantee student loans for low-income students.
In hindsight, allowing high school dropouts to borrow federally guaranteed money in order to attend for-profit colleges was disastrous for a large number of borrowers. But you can see the nobility of the idea if you squint. The federal government was already subsidizing the education of future teachers, lawyers, doctors, and accountants; what about mechanics, electricians, plumbers, and welders? Didn't they need training, and didn't that training cost money?
Bennett, who served as education secretary from 1985 to 1988, did not mince words about for-profit schools, telling Congress in 1988 that his department found "serious, and in some cases pervasive, structural problems in the governance, operation, and delivery of postsecondary vocational-technical education."
But he also trained his criticism on state guaranty agencies, which were intended to serve as paternal co-signers for students but ended up being something else entirely.
First authorized by Johnson's Higher Education Act of 1965, such agencies were supposed to "partner with the federal government to co-sign the bank loans" and serve "as ambassadors at the local level, educating high school students about college opportunity and the availability of loans," the progressive Century Foundation's Robert Shireman and Tariq Habash wrote in a 2016 report on the legacy of guaranty agencies. These supposedly benevolent bodies would not only share the risk of default with the federal government; they would provide the Department of Education with a network of partner organizations that could help implement the Higher Education Act on the front lines.
"The idea," Shireman and Habash wrote, "was that by putting their own donated resources on the table, guaranty agencies would have a stake in a humane and successful loan program, helping low-income students attend quality colleges." Ideally, "they would operate as charities do, with an approach that hinged on more than just the bottom line: when borrowers did default, rather than immediately engaging in aggressive collection tactics, the agencies could assess the situation and provide assistance and advice as appropriate."
But states were not eager to take on the risk of loan defaults, as evidenced by the fact that only half of them formed any kind of guaranty organization after the passage of the Higher Education Act of 1965. To incentivize the formation of guaranty agencies in every state, Congress in 1976 made the federal government fully responsible for reimbursing lenders in the event of default.
"In essence, the federal government was issuing a blank check to cover the cost of operational expansion by any guaranty agency that decided to take advantage of that opportunity," Shireman and Habash wrote. "The system boomed, but rather than having risk-sharing partners, the federal government instead had a set of guaranty agencies that, like a sole-source contractor, earned more money with every loan they guaranteed rather than contributing anything at all."
In its 1993 report, the GAO noted that the structure of state guaranty organizations was fundamentally problematic. They operated with a federal charter as a co-guarantor of student loans, yet they "assume[d] little financial risk and are not compensated in a way that provides sufficient incentives to prevent defaults."
In fact, guaranty organizations were compensated by the federal government in a way that encouraged risky lending, according to the 1990 report from the Permanent Subcommittee on Investigations. To cover their operating costs, guaranty agencies were allowed to charge both student borrowers and the Education Department a percentage of loan values. When students defaulted or died or declared bankruptcy, the Education Department reimbursed lenders for 100 percent of the default amount, with state guaranty agencies acting essentially as a conduit. Yet when guaranty agencies managed to collect defaulted payments from borrowers, they had to give the Department of Education only "65 percent or 70 percent of any monies collected." That combination of incentives resulted in assembly line lending, not prudence.
By the late '80s, the federal government's efforts to educate more low-income students had resulted in a mad dash to lend as many poor people as much money as possible. Bennett sought to reduce taxpayer exposure to the federally guaranteed student loan market by submitting a host of reforms to Congress, many of which were later enacted. New rules and regulations passed by Congress in 1989 and 1990 capped the amount of revenue proprietary colleges could derive from federal student aid programs at 85 percent; the guaranty agencies were tasked with paying a larger share of default costs; and schools whose students had default rates above 30 percent were barred from receiving federal aid.
These policy changes shrunk both the for-profit college sector and the student loan default rate, according to Looney and Yannelis. When Congress loosened the rules in the mid-1990s and again in the early 2000s, the number of for-profit colleges increased, as did the default rate.
All told, Looney and Yannelis estimate that this process—federal credit expansion, default rate increase; federal credit contraction, default rate decrease—has happened three times, with the most recent credit contraction cycle occurring just as Occupy Wall Street was finding its footing. Considering that we're now in a period of relatively tight credit for higher education, it should come as no surprise that default rates are down.
The unintended consequences of federally guaranteed loans to students of for-profit institutions has been a major policy topic for decades now. But what about the attendees of traditional academic institutions, such as public colleges and nonprofit private universities? Have they been fleeced as well?
Bennett argued in a 1987 New York Times op-ed, "Our Greedy Colleges," that federal aid was actually making all forms of college more expensive. "In 1978, subsidies became available to a greatly expanded number of students," he wrote. "In 1980, college tuitions began rising year after year at a rate that exceeded inflation. Federal student aid policies do not cause college price inflation, but there is little doubt that they help make it possible."
The "Bennett Hypothesis" has been a source of conflict among higher education policy makers and economists, largely because it's been so difficult to test. What's not in dispute is that college costs have increased dramatically over the last three decades. Between 1981 and 1995, the average tuition and gross fees for full-time undergraduate students across all institution types increased 88.7 percent; between 1996 and 2014, the increase was 70 percent. In short, every type of institution is significantly more expensive now than it was three decades ago in today's dollars, though public institutions remain the most affordable.
How much of that cost increase was driven by federal aid? It's difficult to determine. The current consensus is that private colleges, both nonprofit and for-profit, respond to federal aid by reducing the discounts they offer to students. Less federal aid, bigger discount; more federal aid, smaller discount.
Proving causation is tougher. David O. Lucca, Taylor Nadauld, and Karen Shen noted in a 2015 report for the Federal Reserve Bank of New York that researchers "only have reliable time series data on the sticker-price of tuition rather than the net tuition paid by students after accounting for scholarships or discounts to lower-income students." Their paper, which looked at the impact of credit expansions on tuition increases, found that "expensive, private, or sub-four-year programs are associated with larger tuition responses to loan maximum changes."
The difficulties of proving the Bennett Hypothesis reflect the complexity of figuring out exactly how much college costs, particularly at private nonprofit institutions. As Lucca et al. note, the actual cost of college is often not the same as the advertised price of college. This phenomenon is also a result of federal intervention in the higher education market.
To determine the cost of attending any particular school, students can't just look at the coming academic year's "sticker price," which at private nonprofit institutions can exceed the median U.S. household income, and which very few students pay in full. Instead, applicants must fill out a Free Application for Federal Student Aid (FAFSA), which the federal government uses to determine a student's "expected family contribution" as well as what types of federal aid the student is eligible for. That information is shared with the schools that the student designated in her FAFSA application. The schools then determine how much "institutional aid"—essentially, merit- and need-based discounts off the sticker price—to offer the person. Eventually, students receive a tailored aid package from their selected schools informing them how much it will actually cost them to attend.
In 2017, economist Lesley J. Turner looked at the impact of need-based Pell Grants on the final cost students paid. Across the entire higher education sector, she estimated that between 11 and 20 percent of Pell Grant aid is "passed through" to schools, meaning that the schools find a way to capture a portion of the aid without charging students any less out-of-pocket for their educations. While Bennett claimed that all types of schools capture aid without lowering costs, Turner found that the phenomenon occurs most notably at private nonprofit institutions. Some selective private nonprofit schools, Turner found, manage to capture as much as 75 percent of Pell Grants. In short, needy students attending these schools are unable to increase their purchasing power by the total amount of the Pell Grants they receive.
My own experience is instructive. After I filled out and submitted my FAFSA in spring 2004, the private nonprofit university I ended up attending informed me that I would be receiving a merit-based and a need-based scholarship from the university, a merit-based scholarship from the state of Florida, and a need-based Pell Grant from the federal government. The remaining amount I owed in tuition and fees that first year was roughly as much as I was eligible to borrow in federal loans as a first-year college student. The following year, my alma mater increased its tuition and fees, but my Pell Grant dropped slightly with the federal schedule. Rather than increase my discount, my university kept my discount the same and informed me that I owed slightly more—but still roughly as much as I was eligible to borrow in federal Stafford loans as a second-year college student. This process repeated in my third and fourth years, when, due to tuition increases, I also became eligible for an additional need-based federal Perkins loan.
There are, of course, alternative theories to explain why the increase in college costs has far outpaced inflation. In a 2014 report, the Congressional Research Service suggested that colleges may have "ineffective centralized control of costs, suffer from various types of productivity issues, and have institutional orientations and incentives targeted toward raising and spending considerable amounts to enhance students' experiences as opposed to orientations toward using resources efficiently." But even if other factors are driving up costs, federal aid makes those cost increases possible. Which means that financial aid for private nonprofit university students looks a lot like health care, wherein the customer, the third-party payer, and the service provider all have varying amounts of information and the sticker price is never the actual price.
Just as expensive health care keeps you alive, expensive college increases your earning potential. In 2015, researchers Christopher R. Tamborini, ChangHwan Kim, and Arthur Sakamoto published a paper in Demography that measured the 50-year lifetime earnings gap between high school graduates and bachelor's degree holders at $896,000 for men and $630,000 for women. Considering that only 2 percent of students borrow more than $50,000 for an undergraduate degree, most of us are getting a good return on our education.
But what if you borrowed money for college and did not get a degree? What if you borrowed more than you can afford to pay back? In other arenas, people who lose everything on a bad bet have the option of discharging their debts and starting over by filing for bankruptcy. Student loan debts, however, are "presumptively nondischargeable." That means that it's possible to get rid of them by declaring bankruptcy, but most people can't.
The Bankruptcy Reform Act of 1978 prohibited borrowers from discharging student loans in bankruptcy for the first five years of repayment. Later amendments changed that to seven years, and then to the entire life of the loan. In 2005, the Higher Education Reconciliation Act made even private loans nondischargeable in bankruptcy.
In a 2012 piece for Reuters, the progressive writer Maureen Tkacik highlighted fearmongering in the 1970s, when Los Angeles Times reporter Linda Mathews wrote about "underground newspapers urging students to use bankruptcy to avoid paying loans." The Wall Street Journal and The New York Times also published pieces warning of a looming generation of college-educated deadbeats who would rather plead poor before a bankruptcy judge than pay a penny for their own education.
A GAO report commissioned by Congress supported the claim that student loan borrowers were not abusing bankruptcy, but "the evidence of a lower than 1% discharge rate of federally insured student loans in bankruptcy did not block the nondischargeability provision from entering the Bankruptcy Code," University of Michigan law professor John A. E. Pottow wrote in 2005. After all, the fact that borrowers were not then fleecing the federal government did not mean they wouldn't eventually try. Rep. Allen E. Ertel, a Democrat from Pennsylvania, claimed during the 1970s bankruptcy debate that the effort to preserve dischargeability was "almost specifically designed to encourage fraud."
So why has Congress continued to make student loan debt presumptively nondischargeable? The two main objectives, according to a 2019 report from the Congressional Research Service, are protecting the availability of student loans for future generations and protecting the "public fisc," i.e., the federal purse, from the forgiveness of debt owed to, or guaranteed by, the government. What's more, there's little to no collateral for creditors to liquidate in the event of bankruptcy. My diploma is not worth what it cost to print it, and my degree is nontransferable. Human capital is not like other kinds of capital—think of a factory or even a patent, which a bank could seize and sell if the borrower failed to pay.
But there are two very good reasons to reinstate dischargeability. The most libertarian reason is that bankruptcy is a market signal, and the higher education sector badly needs more of those. Treating student loan debt like any other kind of debt would make lenders more cautious; schools would need to demonstrate to both banks and potential students that they can equip the latter to repay the former, making it harder for students to indebt themselves.
That said, if we did require banks to assess a borrower's credit risk and an institution's ability to help students find work that will allow them to meet their financial obligations, it would likely result in restricted credit access for low-income and minority borrowers. Such a market might offer them no credit at all or interest rates significantly higher than those available to their less risky peers. That's not an ideal future.
But the same students who might lose access to funding in an exclusively private credit market are also the ones who suffer most under the current system. As University of Michigan economist Susan Dynarski noted in a 2015 piece for The New York Times, students who have the smallest debt loads also have the highest default rates. "Defaults are concentrated among the millions of students who drop out without a degree, and they tend to have smaller debts," she wrote. "That is where the serious problem with student debt is. Students who attended a two- or four-year college without earning a degree are struggling to find well-paying work to pay off the debt they accumulated."
Preventing those borrowers from discharging their loans in bankruptcy is not helping the American economy, but neither is lending them money to finance degrees they can't complete. Those students need a different route to increasing their earning potential.
The gainfully employed college graduate who has to delay buying a home or starting a family by a few years because of student loans from undergraduate school may not seem like a sympathetic character. He has a job, after all, and is likely making more money than he would be if he hadn't attended college. The working adult who borrowed money to attend a sham for-profit school is also not blameless. But the status quo will not hold forever, and the alternative to what we have now is not necessarily a free market for higher ed.
If elected president, Sanders would "cancel the entire $1.6 trillion in outstanding student debt for the 45 million borrowers who are weighed down by the crushing burden of student debt," according to his campaign website. His plan calls for canceling all debt currently held or guaranteed by the federal government in addition to purchasing and canceling all private student loan debt. Biden has proposed a less ambitious debt forgiveness plan and a doubling of the maximum value of the Pell Grant per student. Both candidates want to use federal funds to provide two years (Biden) or four years (Sanders) of free public college.
Voters seem open to further federalizing higher education funding. A Quinnipiac University national poll released in April 2019 found that 57 percent support a debt forgiveness plan proposed by Sen. Elizabeth Warren (D–Mass.) that is more radical than Biden's but less radical than Sanders'. (Support fell to 44 percent when pollsters asked about funding the Warren plan with "a new tax on the wealthy.")
The Democratic Party, meanwhile, is highly receptive. In a 2019 survey from the New America Foundation, only 17 percent of Democrats strongly agreed and only 28 percent somewhat agreed with the statement, "Americans can get a high-quality education after high school that is also affordable." When asked "who should be more responsible for funding higher education," 80 percent of Democrats chose "the government, because it is good for society." Only 19 percent chose "students, because they personally benefit."
American colleges are not going to endorse reforms that would require them to do more with less. Despite the mountain of evidence that its programs have made college more expensive for the middle class and jeopardized the financial security of nontraditional students and the working poor, the federal government is not going to cede its role atop the system. But current and future debtors are up for grabs. And convincing them to endorse a radical market-based proposal over a radical socialist one starts with acknowledging that our current system did them dirty.