Student housing's $1.25 billion problem is a clock, not a default wave

Trepp's new read on securitized student housing opens with a number that invites a shrug, a 1.4% balance-weighted non-performing rate across the $29.26 billion market, but the number that matters sits further down the page. Some $1.25 billion of outstanding balance is current on its payments while reporting a debt service coverage ratio below 1.00x, the ratio that says the cash flow a property reports does not cover the debt it carries. Trepp is precise about what keeps those loans performing — borrower support or interest reserves built in at origination — and either way the payment is not coming from the rent roll.

For a lender pricing student housing debt off the sector's non-performing rate, that distinction is the whole exercise: a loan can be current and short of coverage at the same time, and the two measures answer different questions — whether someone is writing a check, and whether the asset is. Trepp's finding is that $1.25 billion of the sector sits on the wrong side of the second question while sitting on the right side of the first, which is what a problem looks like before it becomes anyone's loss.

Trepp splits the market by capital source: agency lenders — Fannie Mae, Freddie Mac, Ginnie Mae — hold most of the balance and run the most conservative book in it, stabilized buildings, low leverage, a 1.74x median DSCR on net cash flow, and a median debt yield of 12.2%, with only 3.9% of agency balance below breakeven coverage and a 0.5% non-performing rate. That is a business being paid to take almost no coverage risk, and the debt yield is the number that explains it — a cushion wide enough that a soft leasing year does not immediately show up in coverage.

Private-label CMBS holds 18.4% of the balance and 56.8% of all non-performing balance, a 4.3% non-performing rate that Trepp identifies as the largest share of realized distress among the three capital sources. The instructive pair is $236.95 million below a 1.00x DSCR against $230.18 million already non-performing, which means nearly the entire weak cohort has already defaulted; Trepp treats the elevated rate as history rather than forecast. The wider gap between the books — 12.2% median debt yield at the agencies against 8.1% for private-label CMBS — is the underwriting difference between stabilized cash flow at low leverage and a business plan at high leverage, and the clearest single measure of how differently the two books were built.

That leaves the CRE CLO book, where the structure does something specific: CLOs finance transitional properties, and interest reserves can fund debt service while a business plan is executed, so the borrower's ability to execute that plan is what will drive refinancing prospects. In its dataset, 20 of the 76 CLO-financed properties report sub-breakeven DSCRs, and 97.3% of CLO balance remains current on payments. More than a quarter of the collateral is not covering its debt while almost none of the loans have stopped paying, the distance between those two facts funded by reserves or by the sponsor. Private-label CMBS accounts for $236.95 million of the sub-1.0x balance; the data does not break out how the remainder splits between the agency and CRE CLO books, which is the gap a lender would most want closed.

None of that makes student housing a bad credit. Trepp's own conclusion is that the realized distress sits where a high-leverage book would put it, in loans that have already been worked through, while the performing balance is overwhelmingly current. A lender pricing new student housing exposure off a single sector average is blending a 0.5% non-performing agency book, an 8.1% debt-yield CMBS book, and a CLO pool where a quarter of the collateral is short of coverage — three borrowers with that little in common do not share a spread.

The unresolved credit in student housing now sits in the transitional book: private-label CMBS has already printed its losses, the loans that were going to default have defaulted, and what remains there is workout timing. The transitional book is different, because an interest reserve is a clock: every quarter a property misses coverage draws it down, and the lender holding one of those 20 loans is waiting on a lease-up, a rent reset, or an enrollment number rather than on the market to reprice the asset.

As this publication argued in September, student housing's risk is, at the long end, a 2029 problem: $5.29 billion of sub-8% debt-yield paper maturing in 2029 and 2030, straight into a declining enrollment curve. Trepp's sub-1.0x balance is the front end of the same story: a loan living on reserves until 2029 has to clear a takeout underwritten on the coverage it shows then, and the current rate cycle has been a long demonstration of how much of the sector's coverage depends on a business plan landing on schedule.

The test over the next four quarters is narrow. If that $1.25 billion is still reporting current a year from now, the reserves and the sponsors held, and student housing's low realized default rate was earned. If pieces of it migrate into the non-performing column, the figure lenders have been quoting for the asset class will turn out to have been a blend, and the CLO cohort is where the migration starts — 20 properties that need a refinance on an executed business plan, sitting in a maturity queue where agency capital and private credit are repricing the same assets at different speeds.