Journal / Essay · Debt

The Math of the For-Profit College

For-profit colleges post a 14.7% student loan default rate, more than double the nonprofit college average.

Another piece in this journal covered college ROI by major, every sector mixed together. This one isolates for-profit schools specifically: a default rate nearly 60% higher than the public-college average, a federal accountability rule that's been law, not law, and law again three times since 2014, and the sudden closures that leave people holding debt for a credential that no longer exists.

And Then You're Dead · August 2026 · 8 min read

Every college sells the same basic promise: pay now, earn more later. At a for-profit college, that promise comes from a school that also has shareholders or an owner to pay, and it has to extract a profit margin from the same tuition dollar a public or nonprofit school doesn't need to. That difference alone doesn't make the degree worthless. But three separate bodies of evidence — who defaults, how the federal government tries to regulate the worst outcomes, and what happens when a for-profit school runs out of money — all point the same direction.

The Default Gap

The government's primary yardstick for loan outcomes by school is the cohort default rate: the share of a school's borrowers who default within three years of entering repayment. For-profit colleges have topped that ranking for as long as the metric has existed.

For-profit borrowers default at a rate nearly 60% higher than the public-college average, and better than double the nonprofit rate. That gap holds even though for-profit schools enroll a minority of all students, and even though for-profit programs tend to run shorter and cheaper than a four-year degree — the kind of program that should, in theory, carry less risk.

The comparison above uses the last cohort year before the numbers stopped meaning anything. The federal student loan payment pause that started in March 2020 ran straight through the measurement window for the fiscal year 2022 cohort, so no borrower with an Education Department-held loan could default during that period no matter which school they attended. The result: the official national cohort default rate the government released for that cohort was 0.0%, across every sector, for the first time on record. The metric gainful employment and every other accountability rule leans on was, for that cycle, broken by design rather than by improvement.

A Rule That Can't Hold Still

The federal government has tried, four separate times since 2014, to write a rule that cuts off student aid to programs whose graduates can't earn enough to pay back what they borrowed. None of the four versions has survived past a change in administration.

Gainful employment has been law, then not law, then law again, then rewritten a fourth time — and the newest version doesn't have real teeth until 2028.

The Obama administration finalized the original gainful employment rule in 2014, testing for-profit and non-degree programs against a debt-to-earnings ratio and cutting off federal aid to programs that failed it repeatedly. The Trump administration's first Education Department, under Betsy DeVos, rescinded that rule outright, effective July 2019. The Biden administration brought a version of it back in 2023, paired with a new Financial Value Transparency framework, effective July 2024. The second Trump administration then replaced that framework in 2026: the Working Families Tax Cuts Act — the reconciliation law also known as the One Big Beautiful Bill Act — directed the Education Department to drop the old debt-to-earnings test for a single "earnings premium" metric, applied for the first time to every federally funded program at every college, not just for-profit and certificate programs.

The final rule implementing that new framework was published June 29, 2026. Most of its provisions take effect July 1, 2027, with early adoption allowed a year sooner. The first earnings-premium calculations aren't due until July 2027, and a program only loses eligibility after failing the test in two of three consecutive years — which puts the earliest possible cutoff at July 2028. Whatever the merits of a broader test that covers every school instead of singling out for-profits, the practical effect is a multi-year gap in enforcement while the new standard phases in, on top of the years the prior standard spent rescinded entirely.

Closing With No Warning

A default rate is a statistic that shows up on a spreadsheet years later. A closure is a phone call, or an email, or a locked door — and for-profit schools close suddenly far more often than the rest of higher education.

Nearly 300 degree-granting colleges and universities closed between 2008 and 2023. For-profit operators ran more than 60% of them, despite enrolling a much smaller share of all U.S. college students. Closure itself isn't unique to the sector — plenty of small nonprofit colleges have shut down too — but how the closure happens is different.

ITT Tech's closure came after the Education Department cut off its access to federal aid over accreditation and financial-stability concerns; the company gave students effectively no runway. Corinthian Colleges collapsed the year before, in 2015, and roughly 560,000 of its former students were later approved for a combined $5.8 billion in automatic loan cancellation, on the finding that the company had misrepresented job placement rates and program quality. The Art Institutes chain closed its eight remaining campuses on September 30, 2023, giving its roughly 1,700 remaining students about a week's notice by email; the Education Department later approved $6 billion in loan discharges for borrowers who'd attended. In July 2024, two more for-profit schools — the American Academy of Art and Northwestern College, both in the Chicago area — shut down "effective immediately" with no advance warning, leaving around 600 students combined with debt and no diploma; Northwestern had also told some students their credits would transfer to partner schools it had never actually contacted.

A closed-school discharge exists for exactly this situation — a borrower who was enrolled when the school shut down, or withdrew shortly before, can have the associated federal loans canceled. It works, eventually, for people who know the program exists and file for it. It does not un-spend the years already sunk, the credits that don't transfer, or the job the unfinished credential was supposed to lead to.

The Loophole That Let Schools Chase Veterans

For-profit colleges have had a federal-dependence limit on the books since the Higher Education Amendments of 1992 set it at 85/15: no more than 85% of a school's tuition and fee revenue could come from federal student aid. Congress loosened that to 90/10 in 1998. For most of the rule's history, though, one category of federal money didn't count toward the cap at all — Post-9/11 GI Bill benefits and Department of Defense Tuition Assistance were treated as "non-federal" for this specific calculation, even though both are, in fact, federal dollars.

The incentive that created is straightforward: a school could enroll almost entirely veterans and military-connected students, run on federal money nearly exclusively, and still post a clean compliance number, because the dollars that made up the bulk of its revenue simply weren't in the count. The Congressional Research Service, tracking the pattern, found Post-9/11 GI Bill spending flowing to schools overall grew from about $2.8 billion in fiscal year 2008 to $9.8 billion by fiscal year 2011, and as of fiscal year 2020, roughly $1.2 billion in GI Bill benefits was still going to for-profit institutions specifically — money that funded a large share of some schools' operations without appearing anywhere near their official 90/10 math.

The American Rescue Plan Act closed the loophole in 2021, amending the law so GI Bill and Tuition Assistance dollars count toward the same 90% ceiling as every other federal aid dollar, effective for institutional fiscal years beginning on or after January 1, 2023. That a sector-wide compliance rate under 1% could coexist with billions of dollars a year in veteran-specific federal funding flowing to the same schools, unmeasured, says less about how well-behaved for-profit colleges actually were and more about how much a number can hide depending on what's allowed to sit outside it.

And Then You're Dead

None of this means every for-profit credential is worthless, or that every public and nonprofit school gets outcomes right — plenty don't. It means the sector where a student is statistically most likely to default, and most likely to have the school vanish out from under them with a week's notice or less, is also the sector where the federal rule built to catch bad outcomes early keeps getting rewritten, repealed, and pushed further into the future every time enforcement gets close. The math isn't that for-profit college is a scam. It's that the systems meant to catch the ones that are keep losing their footing right when they'd start to matter.

That's it. That's the whole thing.

Read about degree ROI by major · Read about the broader default crisis · or see every essay