How to Get Student Loan Repayment Rates Right for Accountability



By Allie Bidwell, NASFAA Senior Reporter

With a new Congress poised to renew efforts to reauthorize the Higher Education Act (HEA) in the new year, conversations around institutional accountability and student outcomes will likely resurface.

A key issue at hand is how to better hold institutions accountable for how students fare in repaying their loans. While lawmakers on both sides of the aisle have expressed concern with the current method for accountability—the cohort default rate (CDR)—they have yet to come to an agreement on how to move forward with a seemingly popular idea to supplement or replace the CDR metric through institutional or programmatic repayment rates.

A new paper written by Robert Kelchen, an assistant professor at Seton Hall University, and published by Higher Learning Advocates, identifies three central questions and potential roadblocks to moving forward with a repayment rate metric, such as how to define and calculate a repayment rate, how to include more students in the calculation, and identifying a time period for measuring the repayment rate.

Lawmakers and think tanks have criticized the CDR metric because it’s easy for institutions to game the system and give an inaccurate picture of just how many borrowers may be struggling to repay their loans. The most recent data from the Department of Education (ED) found that 11.5 percent of borrowers in the latest three-year cohort had defaulted on their loans. But some have questioned whether the measure gives an accurate picture of how borrowers fare over a longer period of time, whether some may be pressured into entering deferment or forbearance in order to avoid being counted in the default metric, or avoid default but fail to make progress toward paying down their principal balance.

Kelchen noted in the paper that new research has shown more than a quarter of borrowers who began college in 2003-04 had defaulted on a federal loan more than 10 years later, in 2015. That research, conducted by Judith Scott-Clayton, an associate professor of economics and education at Columbia University’s Teachers College, also found the default rate for that cohort could increase to as much as 40 percent by 2023.

There has been interest in finding another way to determine how well institutions are serving their students. The PROSPER Act, House Republicans’ bill to reauthorize the HEA, included a provision to link federal financial aid eligibility to program-level repayment rate metrics. Despite the growing interest, many questions about how to structure such a metric remain unanswered.

Policymakers will have to determine, for example, whether to use a repayment rate based on the percentage of borrowers paying down their loans, or one based on the number of dollars repaid. One would be more appealing to students and families, Kelchen wrote, while the other might be preferable to “taxpayers and budget-minding policymakers.” The latter option, he said, could give institutions an incentive to focus on helping borrowers with the largest outstanding debt amounts, “which can raise equity concerns given the strong association between outstanding student loan debt and future earnings.”

Kelchen also said in the paper that policymakers will need to discuss how to define a “program” at an institution, and how detailed of a classification would be necessary or acceptable.

“One possibility to increase the share of borrowers captured by program-level loan repayment measures would be to use a less-nuanced definition of program than the six-digit CIP code used in the gainful employment regulations,” he suggested.

Given the numerous and complex issues to consider, Kelchen suggested policymakers begin with an institution-level repayment rate metric for accountability purposes, and make program-level repayment rates for graduates available to the public—before eventually moving to a program-level accountability metric once more concerns are addressed.

“While students and their families will arguably find program-level repayment rates more useful as they choose which college and program to attend, there are a number of difficulties in measuring repayment rates beyond a potentially small percentage of students who actually graduate,” he wrote.

The two-tiered solution he suggested “would sidestep many of the data concerns … while still allowing students, families, institutions, and accrediting bodies to make their own judgments about the quality of the program based on available data.”