The Debt Gets Canceled. The Degree Rarely Does.

Federal policy has exactly one built-in answer for a student whose college shuts down mid-degree: cancel the loan.▸ It is called the closed-school discharge, and on paper it sounds like the system working — offenders, in this case defunct colleges, don't get to leave the debt behind on someone who never got a diploma. In practice, the debt is the smallest part of what a closure takes, and even the debt relief itself only reaches a fraction of the people it's supposed to.

The Government Accountability Office counted 1,106 colleges that closed between 2010 and 2020, leaving about 246,000 borrowers holding roughly $4 billion in federal student loans between them.▸ Education's own discharge program reached less than a third of that money: about $1.1 billion, for borrowers connected to 682 of those closures, going to just over 80,000 people. The math the program was designed around — cancel the debt of anyone whose education was cut short through no fault of their own — never got applied to most of the debt it was designed for.

The gap has a design reason, not just a funding one.▸ For most of the period GAO studied, borrowers had two paths to a discharge: apply, or wait — Education would automatically cancel the loan three years after a closure if the borrower never reenrolled elsewhere. That automatic path mattered because it caught people who never knew they qualified: GAO found more than 70% of borrowers who got an automatic discharge were already in default or seriously past due, having defaulted at roughly five times the national average before the relief ever arrived. For closures between November 2013 and April 2018 — the window GAO could compare directly — the automatic process discharged $360 million for 27,600-plus borrowers who never filed a form, versus $526 million to 38,700-plus who applied. Then, in 2020, Education eliminated the automatic process entirely. Every borrower affected by a closure since July 1, 2020 has to know they qualify, and apply, or get nothing.

Money owed is not the injury.▸ The injury is the credential a student never finished. SHEEO and the National Student Clearinghouse Research Center tracked 143,215 students who lived through a closure at 467 institutions between July 2004 and June 2020 — a different, longer window than GAO's, and a different agency entirely, which is exactly why the two findings corroborate rather than repeat each other. Seventy percent of those students got an abrupt closure: no teach-out agreement, no advance warning, no plan for finishing what they'd started. Less than half, 47.1%, ever reenrolled anywhere. Of those who did, only 36.8% went on to earn a credential.

Chain those two rates together and the result is the number that should anchor this story, not the debt figure: about 17% of everyone who lived through a closure ever came out the other side with a credential. Canceling their loan didn't touch that number.▸ A discharge zeroes out what a student owes for an education they didn't get; it does nothing to give them the education back.

The wave GAO and SHEEO studied already happened.▸ The next one is being modeled now, and by a source with nothing to do with either of them: the Federal Reserve Bank of Philadelphia, whose researchers built a machine-learning model of closure risk across U.S. private nonprofit and for-profit colleges from 2002 to 2023. Run against the "demographic cliff" — the real, already-arriving drop in 18-year-olds — their worst-case, one-time scenario puts 80 additional colleges at risk of closing, affecting 101,040 more students and 20,880 more staff, with $1,231 million in institutional expenses stranded. Even their gradual scenario, in which the cliff phases in over five years instead of hitting all at once, still adds an estimated 4.6 closures, 7,337 students, and 1,200 staff every single year on top of the historical baseline.

None of those future students will have access to the automatic discharge that caught the 27,600 borrowers Education never had to chase down.▸ They will have to know their college failed them in the specific legal sense that triggers a discharge, find the form, and file it — the same requirement GAO already found leaves people defaulting for years before anyone helps them, just applied now to a population the Fed's own model says could be twice as large as anything in the historical record.

Figures come from two federal-closure eras and one forward-looking model: GAO's 2010–2020 borrower analysis, SHEEO/NSCRC's 2004–2020 student outcomes study, and the Philadelphia Fed's 2002–2023-trained prediction model. Reenrollment and credential counts in the second chart are computed by applying SHEEO's published rates to its published sample size; SHEEO reports the rates, not the raw counts.