The $2 Trillion Question in Commercial Real Estate

The figure most often cited is two trillion dollars, and like most round numbers describing the commercial real estate maturity wall, it is less a measurement than a consensus. The Mortgage Bankers Association, which surveys outstanding balances after paydowns and modifications, put 2026 maturities at $875 billion—seventeen percent of roughly $5.0 trillion in outstanding commercial mortgages—with another $652 billion scheduled for 2027.¹ S&P Global Market Intelligence, working from nationwide property records rather than survey response, estimates a considerably larger wall - about $1.148 trillion in 2026, $1.257 trillion in 2027, and $1.138 trillion in 2028.² Two trillion over three years sits comfortably between the two methodologies. The precise number is contestable. The direction is not.

What follows from that number, in most of the commentary, is a demand story. A very large volume of debt must be refinanced, extended, recapitalized, sold, or resolved, and the market will therefore need an enormous amount of capital to absorb it. That framing is correct as far as it goes, and it is the framing under which a great deal of dry powder has been raised. It is also, from where I sit, the least interesting part of the problem.

Capital is not what is scarce.

Skilled nursing, assisted living, and memory care assets carry a maturity profile broadly consistent with the rest of commercial real estate—the MBA reported that fifteen percent of balances backed by healthcare properties were scheduled to come due in 2026, a share sitting just above multifamily and just below office.³ Those assets have a designated federal takeout vehicle in the Section 232 program, and in fiscal year 2025 that vehicle demonstrated, emphatically, that it can move. FHA issued a record $8.1 billion in Section 232 firm commitments and closed roughly $6 billion in mortgages across 337 loans, nearly doubling the $4.1 billion in commitments approved the prior year, while cutting its outstanding application backlog from 130 at the start of the fiscal year to 29 by September 30.⁴ The Office of Residential Care Facilities did this in the same period it launched the Express Lane, which grants queue priority to lower-risk 232/223(f) refinances and has produced firm commitments in as little as seven to ten days on transactions that historically took up to 150.⁵

That is a program operating near the top of its demonstrated capacity, and HUD's own budget projections anticipate roughly $6 billion in insured healthcare volume continuing forward.⁶ The money is available. The processing capacity, having been the industry's chief complaint for a decade, is materially better than it was.

And yet a substantial share of the bridge debt written into these assets between 2021 and 2023—much of it originated expressly to exit to HUD—will not clear the takeout on the timeline its sponsors assumed. The reason has almost nothing to do with the availability of capital and almost everything to do with whether a given project can prove it belongs in the program.

Section 232 did not become faster by becoming easier. It became faster by shifting the burden of completeness onto the front end. The Express Lane criteria are explicit on this point: the operator must have been in place at the facility for two or more years, the mortgage amount must not exceed $50 million outside the greater New York market, and—the operative requirement—the application must be ready to underwrite and capable of supporting a firm commitment decision immediately upon submittal.⁷ ORCF revised those criteria in January 2026 to further delineate its expectations on loan-to-value, requiring that the ratio be supported by in-place, trailing-twelve normalized net operating income, with normalization limited to a narrow set of adjustments including Medicaid rate increases already in effect and evidenced by a rate letter.⁸ Read that as a credit standard and it is unremarkable. Read it as what it actually is—a documentary standard—and it explains why two projects with identical economics can end the year in entirely different places.

The eligibility architecture underneath all of this predates the current cycle and has not softened for it. The statutory ceiling on refinancing existing indebtedness runs through 12 U.S.C. § 1715n(f), and the Section 232 Handbook's seasoning and eligible-debt provisions determine which dollars of an existing bridge facility HUD will actually recognize.⁹ The relaxed seasoning rules HUD extended to conventional 223(f) multifamily under Notice H 20-03 carved out healthcare properties, so the three-year rule continues to govern residential care facilities regardless of how routine the takeout appears.¹⁰ A bridge loan sized in 2022 against a going-in basis and an underwritten stabilization curve does not automatically produce a HUD-recognizable debt figure in 2026. It produces a number that must be reconstructed, dollar by dollar, from the bridge closing statement and the uses that followed.

There is also a portfolio overlay that borrowers consistently underestimate. ORCF has said repeatedly, and said again recently, that it expects operators to demonstrate a strong track record of care and sustainable positive cash flow, and that lenders are expected to evaluate the applicant's entire FHA-insured portfolio before submitting—confirming that payments are current and that other insured projects are not exhibiting substandard quality metrics or poor financial performance.¹¹ The practical consequence is that a single distressed facility three states away can slow a refinance that is, on its own numbers, immaculate. That posture is unlikely to loosen. An April 2026 HUD Office of Inspector General audit reviewed four portfolios comprising 70 properties and 84 loans with an unpaid balance exceeding $410.6 million, and concluded that ORCF had not consistently acted on financial risks already disclosed in borrowers' audited statements—withdrawals where no surplus cash was available, cash deficiencies, and cash flow insufficient to service current debt.¹² Agencies respond to that kind of finding by tightening at both ends of the file.

None of this is an argument that the wall will go unmet. It is an argument about where the work actually happens.

The capital raised against the two-trillion-dollar figure is real, and it will find deals. But it will find them one asset at a time, through underwriting boxes that were built for a slower market and have not widened to accommodate a faster one. In a market where money is fungible and eligibility is not, the advantage does not go to the borrower with the best relationship or the lender with the most dry powder. It goes to whoever did the documentary work early enough that, when the queue slot opens, there is nothing left to explain.