Thank you for inviting me to speak with you today. My name is Neal McCluskey and I am the director of the Center for Educational Freedom at the Cato Institute, a nonprofit, non-partisan public policy research organization. My comments are my own, and do not represent any position of the institute.
Introduction
The federal student loan programs are highly problematic, and it is crucial to address both their current impacts and to fix them in the long term. They are, indeed, inflicting pain on too many borrowers and on society. But we must also look at the situation with clear eyes, recognizing that student debt is often very manageable because higher education typically brings access to greatly enhanced lifetime earnings, and because efforts to make things easier on current debtors can have major, negative, unintended consequences.
Constitutionality
Before tackling possible reforms of federal student loans, it is important to note that, constitutionally, they should not exist. The federal government has only specific, enumerated powers, primarily found in Article 1, Section 8 of the Constitution, and authority to spend money on education outside of federal lands or territories is not among them. Importantly, the “general welfare” clause does not authorize such expenditures. As James Madison explained in Federalist no. 41:
For what purpose could the enumeration of particular powers be inserted, if these and all others were meant to be included in the preceding general power? Nothing is more natural nor common than first to use a general phrase, and then to explain and qualify it by a recital of particulars.
Similarly, Alexander Hamilton, writing about the taxation and “necessary and proper” clauses in Federalist no. 33, noted that the federal government is only given specific powers:
[I]t may be affirmed with perfect confidence that the constitutional operation of the intended government would be precisely the same, if the clauses were entirely obliterated, as if they were repeated in every article. They are only declaratory of a truth which would have resulted by necessary and unavoidable implication from the very act of constituting a federal government, and vesting it with certain specified powers [italics added]. This is so clear a proposition, that moderation itself can scarcely listen to the railings which have been so copiously vented against this part of the plan, without emotions that disturb its equanimity.
Finally, records of the 1787 constitutional convention include almost no mention of education save for some discussion of authorizing creation of a national university. Of course, the authority to create a national university has never been among the enumerated powers, but not due to an assumption that it would fall under the “general welfare” clause. No, because a different enumerated power could cover it. As James Madison recorded Gouverneur Morris of Pennsylvania explaining, enumerating a power to create a national university “is not necessary. The exclusive power at the Seat of Government, will reach the object.”1
Student Loan Servicing is a Symptom, Not the Disease
Complaints about the failures of loan servicers hired by the U.S. Department of Education (ED) are many and well documented, both by government entities such as the Consumer Financial Protection Bureau (CFPB)2 and media outlets such as National Public Radio (NPR).3 There have been, no doubt, errors committed by servicers over time, instances of poor customer service, and other problems. But the servicers are not the root problem in student lending. They are symptoms of the root problem: federal student lending itself.
Servicers are, of course, hired by ED to do a job, and if they do it poorly it is incumbent on ED to get the problems fixed. But it has been widely reported, and acknowledged by ED, that the department has long failed to do that adequately. As an ED spokesperson told NPR concerning widespread repayment tracking failures in Income-Driven Repayment (IDR), “Borrowers place their trust in us to make sure these plans work the way they were intended to, and we intend to honor that trust. We are aware of historical issues with prior processes that had undermined accurate tracking of eligible payments. The current situation is unacceptable and we are committed to addressing those issues.“4 The Government Accountability Office (GAO) reported last month that ED does not require servicers to do many valuable things for borrowers in IDR, such as provide sufficient information about how one qualifies for loan forgiveness, or regular updates of where borrowers are on their payment schedule toward forgiveness.5
ED, however, has not just failed to ensure efficient and accurate work by servicers. It has failed to conduct proper tracking itself of data that is crucial for the workings of IDR. Reported the GAO:
Education approved IDR forgiveness for 157 loans under IDR plans as of June 1, 2021, the most recent data available at the time of our review. These loans were held by 132 borrowers who collectively received forgiveness of about $4.4 million in loan balances, an average of about $34,000 per borrower. However, Education’s data do not provide enough information to definitively determine why thousands of additional loans were not forgiven. We found that about 7,700 loans (approximately 11 percent of loans analyzed) potentially had enough qualifying payments to qualify for forgiveness as of September 1, 2020, after subtracting months in nonqualifying statuses, such as delinquency, default, certain deferments, and forbearance…
Given gaps in Education’s data, we were not able to determine whether any of the 7,700 loans we identified are eligible for IDR forgiveness. According to Education officials, prior to 2014, NSLDS did not contain complete and reliable data on delinquencies or information on repayment plans, which is needed to determine whether a loan is on a plan that could qualify for IDR forgiveness. Education took steps to address these data limitations starting in 2014; however, the changes do not retroactively apply to older data according to Education officials. Further, while NSLDS also includes cumulative counts of qualifying payments toward IDR forgiveness, officials advised against using them in our analysis because they are not sufficiently accurate to determine when a loan is eligible for forgiveness. These data limitations make it difficult for Education and loan servicers to systematically track the number of qualifying payments made by borrowers, raising the risk that some of the approximately 7,700 loans we identified were eligible for but did not receive forgiveness as of September 1, 2020.6
The GAO also reported major errors in repayment tracking when accounts moved from one servicer to another, but wrote, “Despite the known issues with older qualifying payments, Education advises servicers to consider previous servicers’ counts as accurate.”7
ED is in charge of administering the student loan system, and appears to have performed poorly at the basics, and to have done so for a long time. But this should not be a surprise. The federal government has for decades been the dominant force in student lending, first through guaranteed loans, then through Direct Lending. In the period the College Board has tracked student aid, including non-federal loans — the 1995–96 to 2020–21 academic years — federal loans have never accounted for less than 75 percent of total student loan volume, and has averaged 88 percent.8 Essentially, ED is running the country’s entire student loan industry, and a federal department is likely not capable of efficiently handling such a huge and complex task.
But this is not simply a failure of bureaucracy or the entities it hires. The student loan system itself, created by Congress in a long evolution of well-intended but poorly thought-out programs and patches, is a byzantine mess that is inscrutable to borrowers, servicers, and quite possibly ED itself. It features a dizzying array of loans and repayment plans, often festooned in arcane lending terms, the sheer volume of which alone is a great deal to wrap one’s head around. There are “direct” loans, “guaranteed” loans, “subsidized” loans, “unsubsidized” loans, “consolidation” loans, “Grad PLUS” loans and “Parent PLUS” loans. For repayment there is “standard” repayment but also “graduated,” “Pay As You Earn (PAYE),” “Revised Pay As You Earn (REPAYE),” “Income-Based Repayment Plan (IBR),” “Income-Contingent Repayment Plan (ICR),” “Income-Sensitive Repayment,” and “Public Service Loan Forgiveness,” not to mention “deferment,” “forbearance,” and “default.” And the loan and repayment programs interact with each other in myriad ways, creating a dizzying array of possibilities.
Given this extreme complexity, and all the rules and regulations that go with all of these programs, it is not a surprise that many complaints about servicers are actually complaints about the workings of federal student aid itself. As Jason Delisle and Lexi West at the American Enterprise Institute found in a 2019 analysis of 1,200 randomly selected complaints about servicers in the CFPB complaint database, only 44 percent of complaints that included explicating narratives concerned things under servicers’ control, while 35 percent were about the rules of federal programs, such as how payments are applied to interest or principle, the size of monthly payments, and the workings of graduated repayment.9 PSLF has been similarly plagued by confusion about what types of loans qualified, what employers were sufficiently focused on public service, and more.10
Federal Student Aid, Especially Lending, Has Massive, Negative, Unintended Consequences
Well beyond the direct problems with federal student loans, they, along with other aid programs, fuel major problems that may, in fact, render them self-defeating, or even negative in their overall impact.
The first negative effect, which has been well established at this point, is fueling skyrocketing college prices. Colleges using federal loan dollars as funding streams to pay for things that benefit them, such as higher professor salaries, was feared in the earliest days of federal funding of student loans under the National Defense Education Act in 1958.11 In 1987, U.S. Secretary of Education William Bennett argued in the New York Times that federal student aid enabled colleges “blithely to raise their tuitions, confident that Federal loan subsidies would help cushion the increase.”12 And in 2015, in addition to numerous other studies finding student aid programs having inflationary effects, a Federal Reserve Bank of New York report found that for every one-dollar increase in federal subsidized loan maximums, prices rose 60 cents.13
Fueling price inflation, of course, greatly diminishes the value of aid to students, though it is of significant value to the institutions that enroll the students. The schools get the money and bear no risk in doing so. Nonetheless, if the goal of federal student lending is to increase enrollment in higher education, it appears to have succeeded. In 1910, only 2.7 percent of Americans ages 25 and above had a bachelor’s degree or higher. 50 years later that number was only up to 7.7 percent. But by 1980 — major expansion of federal lending happened between 1965 and 1980 — that number was up to 17.7 percent, and by 1990 it was 21.3 percent. Today it sits at almost 38 percent.14
Of course, the ultimate goal of funding education is not to get more people into places called “college” or “university,” but to increase skills and knowledge. And no doubt there has been increased amounts of skills and knowledge in society since the 1960s, much of it gained by people attending college. But the evidence suggests that there have been greatly diminishing returns when it has come to learning as enrollment has grown.
We do not have comprehensive data for learning in college as there is no widely agreed upon, national measure of the “right” amount of postsecondary learning, but what we do have is troubling. First, looking at what students actually do in college, sociologists Richard Arum and Josipa Roksa, in their book Academically Adrift, report major decreases in the time college students have spent engaged in academic work. In 1961, full-time students spent 25 hours per-week studying, in 1980 it was 20 hours, and by 2003 it had fallen to 13 hours.15
Assessments of literacy, both reading and computational, also show declines over time for adults with degrees. The first assessment is the National Assessment of Adult Literacy (NAAL), which tested adults’ ability to comprehend prose such as newspaper articles or brochures, documents such as tax forms, and quantitative literacy.16 It was administered in 1992 and 2003, during which time the share of Americans ages 25 and above with bachelor’s degrees rose from 21.4 percent to 27.2 percent.17 The share of adults who ended their formal education with a bachelor’s degree who were proficient prose readers dropped from 40 percent to 31 percent between 1992 and 2003. For document literacy the shares dropped from 37 percent to 25 percent. In quantitative, results were unchanged. The direction for adults with advanced degrees was also negative, with those who were prose proficient dropping from 51 percent to 41 percent, document from 45 percent to 31 percent, and quantitative from 39 to 36 percent, though the latter was not statistically significant.
The second exam, which has essentially replaced the NAAL, is the Program for the International Assessment of Adult Competencies (PIAAC), which was administered in 2012/14 and 2017. It is not directly comparable to NAAL, including that it does not designate “proficiency,” but it does report performance levels.
In terms of enrollment, between 2012 and 2017 the share of Americans 25 and older with a bachelor’s degree rose from 30.9 percent to 34.2 percent. The literacy trend moved in the opposite direction. For U.S. households with members ages 16 to 65 years old, in the 2012/14 administration 68 percent of people with more than a high school education scored in the third literacy level or above. In 2017 only 64 percent did.18 In numeracy the drop was from 57 to 53 percent. More movement in the wrong direction, though the PIAAC drops fell short of statistical significance.
A likely major reason we are seeing decreasing skills for the average degree holder is that employers are increasingly seeking degrees for jobs that did not previously require them. Why? Because the more degrees there are, often funded by federal loans, the more that employers can easily demand such signals of basic perseverance and maybe intelligence, and the more not having one makes an applicant look deficient. But if the demands of the job are unchanged, that basically puts workers on a progressively speeding up credential treadmill — they have to run faster educationally just to stay in one place.
This was illustrated in 2014 by the job market data firm Burning Glass Technologies,19 and was analyzed in 2017 by researchers at the Harvard Business School. Analyzing more than 26 million job openings and current occupants of “middle-skill” jobs, the latter researchers found massive credential inflation, which they called “degree inflation.”20 The most striking example was supervisors of production workers, with 67 percent of job postings calling for at least a bachelor’s degree, but only 16 percent of current occupants possessing such a credential, a 51 percent gap. Other large gaps in job categories with many people employed in them were 47 percent for executive secretaries and executive administrative assistants, and 44 percent for supervisors of construction trades and extraction workers.
Even as we have seen credential inflation enabled by greatly increased college enrollment, we have seen a persistent glut in degrees, at least as measured by individuals with degrees who are in not just jobs, but career tracks, that do not require them. According to data from the Federal Reserve Bank of New York, roughly a third of non-recent graduate, four-year degree holders are in jobs that do not require their credential.21
Finally, aid does not appear to have coincided with an increase in the share of four-year degree holders who are low-income. Data from the Pell Institute at the University of Pennsylvania going back to 1970 show that the share of all bachelor’s degrees held by dependent family members ages 18 to 24 from families in the lowest income quartile has hovered around 10 percent throughout the period, and actually dropped a bit from beginning to end.22 Obviously, this is not all people, but at the very least there has been no increase in people from the lowest-income families getting degrees in “traditional” timeframes.
To summarize, while student aid, including loans, may have helped to increase college enrollment — we do not know how enrollment would have looked in the absence of federal aid — the data on learning, and unintended consequences, suggest the effect of aid may well be a net negative. We have seen college prices skyrocket, declines in human capital per degree, more demand for those degrees, an overabundance of graduates, and no relative increase in degree attainment by the lowest-income Americans — presumably the people most in need of help. And, of course, we have seen a huge increase in federal student debt, from $516 billion in 2007 to $1.6 trillion today.23
Fixes
The ideal solution to the student aid problem going forward, one which is consistent with the Constitution and that would eliminate the sundry negative effects of student loans, is to phase out the federal loan programs. The federal government does not have constitutional authority to run such programs, and the results we have seen are testament to the Founders’ wisdom. But what to do about the current situation?
First, it is time to end the freezing of student debt repayments started in the early days of the COVID-19 pandemic. That was intended to be a short-term response to the very dramatic economic uncertainty created by the pandemic and lockdowns to slow it, and that economic shock was overcome by the first quarter of 2021.24 During that time, college graduates tended to suffer the least in the workforce, often occupying jobs that could be done remotely.25 Freeze extensions have been coupled with seemingly ad hoc changes to student loan repayment such as IDR and PSLF, including retroactively changing rules to qualify for relief.26 This has been a very unstable and expensive state of affairs, likely costing taxpayers more than $100 billion27 and yo-yoing borrowers and student loan servicers.
The second answer is to do nothing extreme, especially massive debt cancellation.
First and foremost, mass cancellation is a poor idea because despite all of the major problems aid has created, getting a degree is still, on average, a good investment. Data from the Georgetown University Center on Education and the Workforce show that the average person possessing a four-year degree will earn approximately $1.2 million more over their lifetime than someone with just a high school diploma. Someone with a professional degree such as in law or medicine — the top earners — will make about $3.1 million more.28 As a baseline, taxpayers — roughly two-thirds of whom do not have four-year degrees29 — should not be forced to fund loan cancellation for people who will, on average, see major financial payoffs.
Because of this large lifetime earnings increase, paying off student debt is not especially burdensome for most borrowers. In 2019, economist Beth Akers noted that, “Typical four-year-degree graduates who borrow will accumulate $28,500 in debt over the course of their enrollment. That can be paid back with monthly payments of less than $200, which is a relatively small share (4%) of the average monthly earnings for this population ($4,717).”30
Looking at repayment of federal direct student loans in the fourth quarter of 2019 — before the COVID-19 repayment freeze — 70 percent of loans were either being repaid or were for students still in school or in their grace period.31 The remaining 29 percent — 1 percent is for “other” — represent borrowers who might be struggling to repay. Nine percent were in deferment, in which borrowers can pause repayment, and those with “subsidized” loans can do so without accruing interest. Qualifications include “economic hardship,” but also undergoing cancer treatments, entering military service, and more.32 Seven percent were in forbearance, which accumulates interest, and can be granted for many things, including serving in AmeriCorps, being in the National Guard and activated by a governor, and more.33 Servicers also put lenders seeking to enter IDR into “administrative forbearance” as they transition, which is supposed to end after 60 days.34 Finally, 14 percent were in default, meaning they were not repaying and were incurring penalties. So while some debt-holders may truly struggle — which must be more than simply saying one feels their loan repayment is a burden — most likely do not, at least on a lasting basis.
Importantly, debt is disproportionately incurred by wealthier people. The Urban Institute found that in 2016, households ages 25 and older in the highest income quartile held 34 percent of all student debt, and the highest decile held the highest average level of student debt at $46,700. The lowest quartile held only 12 percent of all debt, with an average debt load of $26,000.35
This appreciable skew of degrees and debt to wealthier Americans is why numerous analyses of debt forgiveness proposal have found they would typically be regressive, or at least poorly targeted to those most in need. Analysts at the New York Fed, for instance, estimated that $50,000 of federal loan forgiveness with no income cap would deliver 30.1 percent of the benefits to people living in neighborhoods in the top-third of income and only 24.2 percent to the lowest third.36$10,000 of forgiveness and a $75,000 income cap — the most restrictive scenario — would still see 17.4 percent of the benefits accrue to people in the income top-third. Economists Sylvain Catherine and Constantine Yannelis estimate that cancelling $50,000 per borrower would result in $4,223 in forgiveness for people in the top income decile and only $866 in the bottom. Cancelling $10,000 would result in $1,391 for people in the top decile and only $386 in the bottom.37
Analyses of distribution by race find a skew toward white people, not African Americans who have particularly high debt levels thanks to low wealth accumulation at least partially attributable to centuries of slavery and discrimination. The Fed analysis found that $50,000 with no income cap would provide 66.5 percent of the benefit to people in majority white neighborhoods and only 32.9 percent to majority-minority neighborhoods. $10,000 with a $75,000 income cap would deliver 61.9 percent of the benefit to majority white neighborhoods and 37.3 percent to majority-minority. With $50,000 in forgiveness Catherine and Yannelis find lower average forgiveness for Black than White borrowers, but it is roughly even at $10,000.
Importantly, the relative impact of forgiveness on a low-income person or racial minority might be greater than someone with higher income or wealth. $10,000 for someone struggling to repay $15,000 in debt is more meaningful for that person than for someone comfortably handling $50,000. But that does not necessarily justify special cancellation for any borrower who agreed to repay money received from taxpayers, and it certainly does not justify also sending huge amounts of money to those who are not struggling.
Of course, mass cancellation of loans would have many other major problems than giving taxpayer money to people who do not need it.
Barring major reforms of the student loan system, it would encourage more borrowing and even worse price inflation than we have seen for the last several decades, as neither borrowers nor schools would expect loans to really have to be repaid. That would also exacerbate already painful credential inflation. And it would be a de facto penalty against the many Americans who economized on their education in order to avoid debt, or who economized after graduation to pay it off. And if mass cancellation were initiated by executive action, it would be both illegal — there is no compelling argument that the president has power to execute mass cancellation, as President Biden has himself acknowledged38 — and even if mass executive cancellation were somehow interpreted as in accordance with law, it would still truly violate the separation of powers: Congress, not the President, has the power of the purse, which surely applies to the disposition of $1.6 trillion in federal student loans, which is more than twice the size of the Department of Defense’s budget.39
While the ultimate solution to the student lending problem is for Washington to remove itself from student loans, there are things it can do to improve the situation for current debtors and future borrowers without using the very blunt and perhaps unfair instrument of mass cancellation:
- Income-Driven Repayment: The federal student loan system already has a way to help people struggling with repayment: income-driven repayment. The problem has been that it is too complicated for many borrowers, servicers, and the U.S. Department of Education. As quickly as possible, the department needs to correct all borrower records, and anyone who has qualified for IDR must be put on an accurate path and given the forgiveness to which they are entitled. Terms should not be changed, as it appears the current department will do for IDR and PSLF, but borrowers must be treated consistently with the original terms of the program. Going forward, IDR should be simplified, perhaps consolidating it into just one program in which borrowers are automatically enrolled unless they choose a standard payment. For terms, Congress might consider, say, a standard IDR of 15 percent of income above 150 percent of poverty and forgiveness after 20 years for undergraduate borrowers, and after 30 years for graduate borrowers.40
- End PLUS: Both parent and graduate PLUS loans should be immediately eliminated. Parents are often saddled with debt they cannot repay under PLUS, which unlike “normal” undergraduate loans are not capped, while graduates in programs that train for well-paying jobs should not need taxpayer assistance. Those pursuing other studies have become cash-cows for many colleges and universities.41
- Bankruptcy: It has become more difficult over the last few decades to declare bankruptcy and have debts discharged, and student loans are even more difficult to discharge for fear of borrowers graduating and then quickly declaring bankruptcy to clear their debts. It is worth examining making it easier to discharge federal student loans in bankruptcy, but with an important reminder that taxpayers had no say in any given student loan but would bear the cost of whatever amount is discharged. It may be that the “undue hardship” standard currently in place to discharge student debt is sufficient.
- “Skin in the game”: Colleges are not primarily at fault for excessive debt and noncompletion. Washington, which created student loans despite highly predictable negative effects, is. It created the incentives for schools to charge high prices and enroll many students with poor likelihoods of completing programs in adequately paying fields when it decided to give out bundles of taxpayer money with little assessment of prospective borrowers’ ability to complete programs in in-demand fields. That said, the no-risk beneficiaries of that aid have been colleges, which win financially whether their students succeed or not. In light of this, holding institutions responsible for a share of loan defaults by students they enroll may be a wise policy even if it distracts from the true source of the problem: the federal government. Schools might be required to pay for 20 percent of default costs after, say, 10 percent of their enrollees have defaulted.42
Conclusion
Federal loan servicers are not the problem in student lending. The problem is the massively distorting, incredibly complex, student loan labyrinth that the federal government has created over the last roughly 60 years. It has not created so terrible a problem that it justifies extreme acts such as mass loan cancellation — higher education is still a major benefit to most people who complete it — but it is urgent to fix the existing programs intended to help struggling borrowers, and in the long run to shrink federal lending to a much smaller size, including eliminating it completely.
Thank you again for this opportunity to testify. I look forward to addressing your questions.
