Quality Assurance: How Executive Authority Can Hold Colleges Accountable for Outcomes

Key Points

• The federal government spends tens of billions of dollars subsidizing colleges and programs of study that typically leave students worse off.
• The Higher Education Act allows the secretary of education to implement a “quality assurance” system for colleges participating in the federal Direct Loan program. This issue brief establishes a framework for the use of this authority.
• The proposed Quality Assurance rule could protect over two million students who would otherwise take on excessive debt for degrees that produce insufficient economic value.
• If implemented, the Quality Assurance rule could save taxpayers $100 billion over a decade.

Introduction

Higher education is supposed to be a path to the middle class. But new research from FREOPP shows that millions of college students are pursuing degrees that are not likely to boost their earnings enough to justify the cost of college. Low-quality higher education includes everything from beauty schools to doctoral programs, but much of it has one thing in common: the federal government pays for it. Between 2017 and 2022, $86 billion in federal student loan funding went to programs that leave their students with no return on their investment.

Substandard degrees and overpriced colleges are actively hurting students who must deal with mediocre earnings and high debt burdens after they graduate, if they graduate at all. This problem has created the political demand for student loan cancellation. Even if President Biden’s plans to cancel debt run get struck down in court, taxpayers are still likely to pick up some of the bill. Some students will default outright on their loans, while others won’t fully repay what they borrowed.

Budget scorekeepers expect the federal government’s losses on the student loan program over the next decade to be $249 billion. The primary beneficiaries of this excess are not students, but colleges. Federal student loan subsidies enable universities to charge high tuition and maintain degree programs with almost no labor market value

Historically, the federal government has relied on third-party agencies called accreditors to ensure that the colleges it funds uphold high standards of quality. But accreditors, which are voluntary membership organizations made up largely by representatives of the colleges they oversee, were never suited to the task of gatekeeping access to taxpayer dollars, given the natural conflicts of interest. Moreover, FREOPP has found that all major accreditors have allowed hundreds of nonperforming programs to thrive. Accreditors rarely sanction colleges for poor student outcomes.

The federal government has introduced outcomes-based accountability tests that colleges must satisfy to maintain access to federal aid. But these have critical shortcomings. The first such measure is called the Cohort Default Rate (CDR), which cuts off aid to colleges where a high proportion of students default on their loans. But CDR has become subject to manipulation thanks to widespread use of loan forbearance, which allows students to avoid repayment without defaulting.

More recently, the Biden administration introduced the Gainful Employment (GE) rule. This change is somewhat stronger: the GE rule cuts off aid to individual programs of study where students’ earnings after graduation are too low relative to their debts. These metrics are less prone to manipulation than CDR. However, GE exempts degree programs at public and private nonprofit colleges, which enroll 85 percent of students.

Ideally, Congress would enact common-sense reforms to federal student aid programs to demand accountability from taxpayer-subsidized colleges. FREOPP has advanced several proposals: limits on how much students can borrow; a requirement that colleges face financial consequences when students don’t pay back their loans; and rewards for schools that place low-income students into high-wage jobs. Unfortunately, the pace of legislative movement in Congress has slowed to a crawl.

But the Higher Education Act (HEA), the principal law governing federal student aid, empowers the secretary of education to take action on his own. Indeed, the law requires the secretary to ensure institutions using federal loans are upholding high standards of quality. Previous secretaries have neglected this critical duty.

Section 454 of the HEA states that “an agreement with any institution of higher education for participation in the direct student loan program under this part shall … provide for the implementation of a quality assurance system, as established by the secretary and developed in consultation with institutions of higher education, to ensure that the institution is complying with program requirements and meeting program objectives.” (emphasis added)

This issue brief lays out a framework for the secretary of education to use the quality assurance authority to ensure that colleges participating in the student loan program produce satisfactory outcomes. The basis for the framework is the Biden administration’s GE rule, but the proposed quality assurance framework strengthens GE in certain ways and — critically — applies its standards to all schools, including public and private nonprofit colleges.

Specifically, the quality assurance framework requires that institutions receiving federal student loan funds ensure that the median student’s estimated loan payments consume no more than eight percent of her income after graduation. In addition, institutions would need to ensure that their students’ earnings exceed those of a typical high school diploma holder (for undergraduate programs) or a typical bachelor’s degree holder (for graduate programs). Schools would need to meet these requirements for each program of study and on an institution-wide basis.

If implemented, every year the quality assurance framework would stop 2.2 million students from borrowing federal loans for programs that are unlikely to produce adequate labor market returns. I estimate that such a policy would save taxpayers $10 billion per year, which would erase nearly half the government’s annual losses on the student loan program. Moreover, while Congressional intervention is optimal, a Quality Assurance rule could be implemented without the need for Congress to pass a new law.

The Gainful Employment rule and its shortcomings

The persistent failure of the Cohort Default Rate to hold low-value colleges accountable led the Obama administration to issue the first Gainful Employment rule, which the Trump administration overturned before any sanctions could take effect. Last year, the Biden administration finalized a new version of GE. The Biden iteration of GE has not existed long enough to “bite” — no college has yet faced punishment under the rule — and the earliest year any school could face sanction is 2026.

GE requires programs of study participating in the Title IV federal student aid programs — mostly student loans and Pell Grants — to maintain a reasonable debt-to-earnings ratio for graduates and ensure that graduates’ median earnings are higher than those of the typical early-career high school diploma holder. Programs which fail either metric for two of three consecutive years lose access to Title IV aid. The test is applied at the program level, so an institution with both passing and failing programs may continue to operate programs that pass GE even after its failing programs lose access to Title IV.

But GE has several critical shortcomings. The rules only apply to for-profit colleges and certificate programs. Degree programs at public and private nonprofit schools — which enroll most college students — are exempt. Even though the Department of Education (ED) collects debt and earnings data on exempt programs, it does not even require schools to notify students that these programs’ outcomes are unsatisfactory.

ED claims its hands are tied under the statutory text of the HEA, which requires that for-profit colleges and vocational institutions prepare students for “gainful employment in a recognized occupation.” No such language applies to degree programs at public and private nonprofit schools.

Nevertheless, this division of higher education into GE and GE-exempt programs creates arbitrary distinctions. The University of Phoenix, a for-profit school, offers 13 degree programs enrolling more than 20,000 students that are likely to fail GE due to high debt. Those programs should lose access to Title IV aid. But the University of Idaho, a public nonprofit institution, is currently seeking to acquire Phoenix and reorganize it as a nonprofit. If Idaho succeeds, those 13 programs will maintain access to Title IV, even if students’ debt burdens remain excessive.

In addition to these broad exemptions, GE only calculates debt and earnings measures for students who complete their programs. Also, there must be at least 30 students in a cohort of completers for ED to calculate these measures. If a program does not have enough graduates, it automatically passes and may continue to draw on Title IV aid. Colleges can therefore avoid accountability if they keep program sizes small, spread students across programs, or maintain low graduation rates to keep the number of completers small relative to the overall number of students.

Altogether, GE’s principal shortcoming is its limited coverage. Most college students using Title IV aid are enrolled in a degree program at a public or private nonprofit school, and thus are not protected. Even among students whom GE theoretically protects, a large proportion are enrolled in programs with too few graduates to calculate GE metrics, and so those students are effectively unprotected. Moreover, the debt and earnings data of non-completers are not considered in the GE metrics, even though non-completers experience some of the worst outcomes from higher education, including the highest loan default rates.

Because of these constraints, less than 10 percent of college students who used Title IV aid in the 2021–22 academic year are enrolled in a program that GE is likely to protect, according to data ED released on the GE rule’s likely effects.¹ The other 90 percent — either because they are enrolled in a degree program at a nonprofit school or are in a cohort with too few completers to calculate GE data — do not enjoy the protections of the rule. Their schools are free to load them up with excessive debt backed by the taxpayer, and the schools will face effectively no consequences.

The Biden administration claims that the possible scope of GE is constrained by the breadth of the gainful employment language in the HEA. But the quality assurance authority in the same law contains no such constraints. ED could use this authority to fill the enormous gaps in student protections that the GE rule leaves behind.

The quality assurance authority

The HEA sets out certain requirements for institutions that participate in the federal Direct Loan program. Institutions must enter into agreements with ED in order to use direct loans, and the HEA sets forth conditions that must be part of these agreements. Section 454 of the HEA states:

An agreement with any institution of higher education for participation in the direct student loan program under this part shall — … provide for the implementation of a quality assurance system, as established by the secretary and developed in consultation with institutions of higher education, to ensure that the institution is complying with program requirements and meeting program objectives; … [and] include such other provisions as the secretary determines are necessary to protect the interests of the United States and to promote the purposes of this part.

In other words, the HEA seems to require the secretary of education to develop a “quality assurance” system to ensure that institutions of higher education using direct loans are meeting the objectives of the Direct Loan program. The last clause, which allows the secretary to include other criteria in these agreements to “protect the interests of the United States” and “promote the purposes of this part,” also seems to give the secretary broad authority to impose conditions on institutions.

The language here applies only to the Direct Loan program, not the Pell Grant program or the now-defunct guaranteed student loan (FFEL) program. The quality assurance provision has been in the HEA since the genesis of the Direct Loan program in 1993.

Currently, the agreements that institutions sign to participate in the Direct Loan program (example) place the onus for quality assurance on the institution. A straightforward reading of these agreements suggests that quality assurance can be whatever the institution wants it to be. But there’s no reason things have to operate this way. Indeed, the language of the statute suggests that the secretary of education, not institutions, should be the one to establish a quality assurance system.

In a 2020 report for the Brookings Institution, legal scholars Aaron Ament and Dan Zibel explain how the quality assurance authority could be deployed for the purposes of holding colleges accountable for student outcomes. Ament and Zibel note that the authority has “never been invoked by the department as authority to promulgate new policies.” Nevertheless, it gives the secretary discretion to attach performance requirements to agreements with institutions.

The authors write:

Developing a system of “quality assurance” to “ensure that the institution is . . . meeting program objectives” must, of course, consider what it means for an institution to “meet[] program objectives.” In this regard, a core “program objective” of the Direct Loan program is to ensure not only that students have access to higher education, but also to ensure that federally issued loans are repaid.

Ament and Zibel take an expansive view of the power this gives the secretary, especially when considering the quality assurance authority together with the authorities to “protect the interests of the United States” and to “promote the purposes of this part.” The authors contend that this language allows the secretary to make institutions’ right to participate in the Direct Loan program conditional on good student outcomes. These conditions could apply at the institutional or programmatic level.

Indeed, an accountability regulation based on the quality assurance authority is likely on stronger legal footing than a regulation based on the gainful employment language. The HEA stipulates that for-profit colleges and vocational institutions using federal aid must “prepare students for gainful employment in a recognized occupation,” but does not require the secretary to develop criteria to determine what constitutes gainful employment. By contrast, the HEA explicitly requires the secretary to establish a quality assurance system.

The secretary of education could therefore develop a regulation requiring that, as a condition of participation in the Direct Loan program, institutions and programs of study meet certain benchmarks with respect to student outcomes.

The HEA does require the secretary to develop such a quality assurance system “in consultation with institutions.” The secretary should therefore implement this system through the formal rulemaking process, which will require ED to hold negotiated rulemaking sessions in committees that include representatives of various stakeholders in the higher education system, including institutions. In the past, negotiated rulemaking committees have included representatives from public, private nonprofit, and private for-profit institutions, often alongside representatives from groups such as community colleges and minority-serving institutions. Unless a negotiated rulemaking committee reaches consensus on a proposed regulation — unlikely in the case of a quality assurance rule — its recommendations are not binding. If the committee does not reach consensus, the secretary of education is free to move forward with a proposal of his own design.

While a negotiated rulemaking with representatives of institutions should satisfy the consultation requirement, the secretary could go a step further and host additional convenings with representatives of institutions. This could place the rule on even sounder legal footing.

Accountability for higher education: a Quality Assurance rule

The secretary of education could use his authority under the quality assurance clause of the HEA to issue a modified version of the Gainful Employment rule that applies to all programs of study, regardless of the credential type or the sector of the institution.

The Quality Assurance rule (QA) should require that each program of study with sufficient data pass two tests in order to maintain access to direct loans:

• Debt-to-earnings test: The estimated annual loan payment for the median program graduate should not exceed eight percent of the median graduate’s annual earnings three years after completion. Annual loan payments should be calculated with current interest rates and a loan amortization period of 10 years for most credentials, 15 years for bachelor’s degrees, and 25 years for doctoral and professional degrees. Annual loan payments should be calculated based on the sum of the student’s loans and any Parent PLUS loans borrowed on the student’s behalf.
• Earnings premium test: The annual earnings of the median graduate three years after completion should be no less than the annual earnings of the median high school diploma holder ages 25 to 34 in the same state (for undergraduate programs) or the annual earnings of the median bachelor’s degree holder ages 25 to 34 in the same state (for graduate programs).

If a program of study fails either of the two tests in two of three consecutive years, that institution will no longer be allowed to enroll students in that program using federal Direct Loans. The institution would also be barred from offering a substantially similar program using Direct Loans. However, because the quality assurance provision of the HEA applies only to Direct Loans, the institution may continue to operate the program using Pell Grants if it chooses.

The proposed QA applies these tests at the program level because there is considerable variation in student outcomes across programs, even within an institution. A Master of Business Administration may pass QA while a Master of Fine Arts fails. If one program fails, the institution may continue to operate passing programs. This will allow institutions to reallocate resources from low-quality programs to high-quality programs, so those high-quality programs can enroll more students and improve aggregate outcomes across higher education.

The proposed QA is similar to GE, but modifies the GE structure in important ways. The measure of student debt includes Parent PLUS loans borrowed on the student’s behalf, as these are reflective of the cost of the student’s education, and students are often expected to help repay them. The QA enhances the earnings premium test by requiring graduate programs to boost earnings beyond the typical income of a bachelor’s degree holder, as this is a more appropriate baseline for the earnings graduate students should be able to expect from higher education. The QA rule also closes a loophole in GE’s debt-to-earnings test that allows certain high-debt programs to escape accountability, and modifies the amortization periods GE uses to calculate loan payments.

I estimate that almost two million students — 12 percent of all students receiving federal student aid funding in the 2021–22 academic year — are currently enrolled in a program that would fail QA, were it in effect. Failing programs are especially concentrated below the baccalaureate level — where earnings can be quite low — and among graduate programs, where student debt burdens are high.

The QA rule effectively targets programs that tend to leave their students in a worse financial position. FREOPP has estimated the return on investment (ROI) for tens of thousands of degree and certificate programs nationwide. ROI is a measure of how much college increases the typical student’s lifetime earnings, after subtracting the costs of college. Positive-ROI programs tend to make students better off; negative-ROI programs usually leave students worse off.

The proposed QA is decently calibrated: 61 percent of programs where estimated ROI is negative would fail, while 89 percent of programs where estimated ROI is positive would pass. Overall, QA classifies 85 percent of programs correctly, compared to just 76 percent for GE. QA would help steer students towards programs that provide better financial value and dissuade institutions from charging too much. Both mechanisms would improve the economic return students enjoy from higher education.

The institutional failsafe

Applying accountability at the program level carries one major drawback: it divides cohorts of students into smaller groups. As a result, each cohort is less likely to have enough students to calculate debt and earnings measures while respecting students’ privacy and maintaining sufficient sample sizes to yield actionable data.

Moreover, students need to complete a program to count in a program cohort. However, non-completion is a major issue in higher education. Nearly 40 percent of students fail to earn a degree within six years, and some schools have far lower completion rates. Non-completion tends to drive many of the worst outcomes in higher education: low earnings, high loan default rates, and taxpayer losses on loans.

Institutions with lots of small programs and institutions with low completion rates would not face sufficient accountability under a purely program-level framework. The QA rule could therefore add a “failsafe” provision that applies modified versions of the QA tests on an institution-wide basis. These tests are:

• Debt-to-earnings test: The estimated annual loan payment for the median student upon entering repayment should not exceed eight percent of the median student’s annual earnings six years after initial enrollment (for less-than-four-year colleges) or eight years after initial enrollment (for four-year colleges). Both completers and dropouts are included in the cohorts used to calculate debt and earnings figures. Annual loan payments should be calculated based on the sum of the student’s loans and any Parent PLUS loans borrowed on the student’s behalf.
• Earnings premium test: The median student’s annual earnings six years after initial enrollment (for less-than-four -year colleges) or eight years after initial enrollment (for four-year colleges) should be no less than the annual earnings of the median high school diploma holder ages 25 to 34 in the same state. For graduate students, median earnings should be no less than the annual earnings of the median bachelor’s degree holder ages 25 to 34 in the same state.

If an institution fails either of the two tests for two out of three consecutive years, that institution would no longer be allowed to participate in the Direct Loan program. The school would lose eligibility for Direct Loans even if some of its programs pass the program-level QA tests.

There is one exception: For institutions that enroll both undergraduate and graduate students, the institutional failsafe tests should be applied separately for the undergraduate and graduate student populations. If an institution fails at the graduate level but passes at the undergraduate level, it could maintain access to Direct Loans for undergraduates.

The failsafe tremendously increases the share of the student population protected under QA. I estimate that around 61 percent of aided students are enrolled in a program with sufficient data to calculate debt and earnings metrics. But when combined with the institution-level tests, 99 percent of students receiving Title IV aid are enrolled in a program or institution where debt and earnings outcomes may be measured.

Incorporating the failsafe increases the share of students in programs that would fail QA only modestly, from 12 percent to 13 percent. However, the failsafe provides a check against institutions deliberately arranging their programs to keep them small and below the radar of the QA tests, since the institutions will be accountable for the outcomes of their whole student body no matter what.

The failsafe also protects non-completers, since the cohorts used to generate the debt and earnings figures in the QA test include dropouts. Under the program-level tests, a school with decent outcomes for completers but a high dropout rate could skate by. However, the failsafe ensures that the school is accountable for the outcomes of non-completers as well.

Overall, 2.2 million students are enrolled in programs or institutions that would fail QA, including the effects of the failsafe. The failure rates are relatively low for bachelor’s degree programs, where just six percent fail the QA rule. Failure rates are highest among subbaccalaureate programs and graduate programs, and in the private for-profit sector.

The QA rule goes a long way towards protecting students who would otherwise leave school with excessive debt burdens or insufficient earnings. But the rule also protects taxpayers, who would no longer make quite so many loans that students are destined never to repay.

The fiscal impact of quality assurance

If QA had been in effect for the 2021–22 academic year, I estimate that taxpayers would have avoided lending $18 billion to students at failing institutions and programs — more than one fifth of all federal student loans issued that year. Most ($13 billion) of those avoided loans would have been loans to graduate students, who account for an outsized share of student debt because their balances are typically much higher.

The $18 billion in avoided loans does not translate directly to fiscal savings. Some students who would have attended a failing program will switch to a passing program instead, where they may still borrow. Students attending passing programs will probably repay more of their loans. But thanks to ultra-generous loan repayment plans the Biden administration introduced last year, most students will not need to repay their loans in full, even those in passing programs.

The role of Pell Grants also complicates the analysis. The quality assurance authority applies only to the Direct Loan program, but many institutions will find operating programs without access to loans untenable. If they close failing programs, taxpayers will save money on Pell Grants as well. However, schools may continue to operate failing programs using Pell Grants if they so choose. Many community colleges, where borrowing rates are low, will find this option feasible and desirable.

Taking these factors into account and imposing some basic assumptions,² I estimate that if QA were in effect in 2021–22, taxpayers would have saved slightly over $10 billion. The federal government would have distributed less Title IV aid to begin with, while some students in failing programs will switch to courses of study where their repayment rates will be higher. Over a decade, the cumulative savings from QA could reach $100 billion. This would erase roughly 40 percent of the losses that taxpayers are expected to bear on student loans over the next 10 years.

One wild card is the Biden administration’s new student loan cancellation regulation, proposed in April, that will allow the secretary of education to cancel the debts of students who attended a program that failed a federal outcomes-based accountability rule. While the regulation was obviously written with GE in mind, the language would encompass any potential QA rule.

If the loan cancellation regulation and the QA rule both go into effect, students in failing QA programs would have their debts cancelled in full. Though it is likely that cutting off loans to failing QA programs going forward would save money in the long run, the QA rule’s interaction with the loan cancellation regulation represents a significant short-run cost. The drafters of a potential QA rule may wish to take this interaction into consideration as they craft the policy.

Conclusion

The federal government is the nation’s biggest funder of low-quality higher education, which leaves both students and taxpayers worse off. Fortunately, the secretary of education has the authority to stop the flow of federal student loans to colleges that leave students with excessive debt burdens or unacceptably low earnings. The Quality Assurance rule proposed in this issue brief would protect millions of students from low-quality higher education and could save the government $100 billion over a decade.

Congress has given the secretary of education broad authority to safeguard the interests of students and taxpayers. The QA framework illuminates a way to deploy it.

Endnotes

1. These figures are based on GE informational data that ED released in May 2023. The true proportion of students in a program covered under GE will be slightly higher in the long run, because the GE rule allows ED to combine up to four graduating classes into a single cohort; the GE informational data only combines two graduating classes.
2. These assumptions are: Students in failing programs are currently on track to repay just 25 percent of the amount they borrow. Programs will continue to operate if they receive more than 70 percent of their Title IV revenues from Pell Grants. Half of students in failing programs that close will not pursue higher education elsewhere. The other half will continue to draw on Pell Grants and Direct Loans, but the portion of the amount they borrow that will be repaid will rise to 50 percent.