The Metrics: Loan Repayment Rate, Pell Access & Success Rate, and Completion Rate

In terms of measuring risk for the federal student loan program, the loan repayment metric is widely acknowledged as being superior to CDR and should replace it, although further refinements are needed especially in the interactivity with repayment programs. A new federal accountability system would use an institutional loan repayment rate, which serves as a proxy for value, completion, and labor market outcomes, to hold schools accountable through both a bright line for eligibility loss and a risk-sharing system of graduated sanctions for low performers—who would be required to repay some of the under-paid amounts to the federal government as a penalty for leaving students not well off enough to comfortably repay their loans. Additionally, a program-level loan repayment rate would be used in GE accountability to hold the highest-risk programs accountable. One feature of the CDR that should be maintained is the mechanism for assessing risk based on the percentage of students who borrow loans. An institution with a higher borrowing rate presents a higher risk to the students it enrolls, and vice versa, and this must be reflected in the new system. A proposal put forth by the Institute for College Access and Success in 2016 for a Student Non-Repayment Risk Indicator (SNRI) demonstrates how this feature can be incorporated.¹ This metric would serve the objectives of protecting the federal investment and promoting completion.

In regards to the second largest federal student aid program, the Pell Grant, institutions must be held accountable for enrolling and graduating low-income students through the use of an access and success metric that compares their completion rate with the college’s overall completion rate, to measure the achievement gap. This metric would not be used to draw a bright line for eligibility, but would be used for a sliding scale of rewards for institutional performance. Schools with high shares of and graduation rates for Pell recipients would receive increasingly sizable rewards, which would serve as an incentive for ineligible schools to become eligible and for eligible schools to gain additional rewards. To avoid gaming, perhaps higher thresholds could be established for shorter credentials, but this will need to be balanced against the need for simplicity. This metric would serve the objectives of completion and equity.

However, the two metrics that apply to federal student loan borrowers and Pell Grant recipients fail to capture the remaining college student population. An overall completion rate metric is essential for holding all federally aided schools accountable for serving all students well and working hard toward their graduation. Here, similar to the loan repayment rate, there would be an eligibility threshold for all federal student aid. Above that baseline threshold to further incent improvement, Congress should establish a system of sanctions that will require corrective action through improvement plans, with the loss of access to campus-based aid and other HEA institutional grant programs for persistently underperforming institutions.

These three metrics (see Table 4) would form the basis for institutional consequential accountability and advance the federal government’s objectives in protecting students and taxpayers against risk and promote completion and equity. While the devil is in the details and any system will be admittedly imperfect, the combined effect of these metrics will be a major improvement over the current system.

Table 4: Consequential Accountability Metrics

In terms of technical feasibility, the proposed metrics rank high, as they are all already collected and reported. While there are various proposals on the exact metrics, especially on the loan repayment rate, the differences are not insurmountable and consensus is within reach. Introduced legislation in both chambers of Congress already provide a menu of options that can be further refined and negotiated.

However, the real challenges center on the political feasibility and the need for both sides of the aisle to recognize that accountability necessarily will involve loss of eligibility for the lowest-performing institutions. One potential way to ameliorate the immediate impact and allow for institutions to plan accordingly would be—similar to the GE experience—to publish rates and provide a delayed or phased-in implementation; which would include a disclosure period followed by sanctions kicking in, thus providing enough time for adjustments. Another important element is the need to recognize up front that the metrics, and their thresholds, will also need to be revisited and adjusted in the future. Just as we now need to replace the CDR with a new metric, legislation could incorporate periodic reviews through special reports and/or commissions that monitor and examine the impact of the accountability system.

A quick note on the often-cited problem (or excuse) of unintended consequences: If there is one thing GE clearly demonstrated, it is that even the prospect of accountability changes behavior. Institutions responded to the data releases by discontinuing failing programs and making improvements in their program offerings. Any analysis that simply applies a proposed accountability framework while assuming unchanged behavior misses the mark. The “sky is falling” prognoses have been proven wrong in several instances of federal higher education legislation and regulation. While scare tactics might be an easy way to oppose change, policymakers should instead anticipate that institutions will respond to accountability by seeking to improve in order to avoid sanctions and/or obtain rewards.