A 5% Ten-Year Did Not Halt A $400 Million Skilled Nursing Trade. The Spread Did The Work.

The week the benchmark broke 5%

On September 14, 2026 the yield on the 10-year Treasury note crossed 5.00% for the first time in nearly three years, a level briefly touched in 2023 and otherwise not seen since 2007 (CNN Business, September 14, 2026). It held there. By early Wednesday September 16 the benchmark was flat at 5.004%, having added seven basis points to 5.029% earlier in the session (CNBC, September 16, 2026). The drivers were hot inflation data pressing on the long end of the curve and crude oil pushing back above $100 a barrel, which clouds the inflation path further (CNBC, September 16, 2026).

For commercial real estate this is normally the part of the story where transaction volume halts. It did not. In the same week, CareTrust REIT announced it had closed a $400 million acquisition of 2,622 licensed skilled nursing beds, effective September 1, 2026 (CareTrust REIT press release, September 15, 2026). That is the largest single print on our transaction wire. Understanding why it happened anyway is more useful than noting that it did.

The spread, not the rate

The number that explains the trade is not 5.004%. It is the distance above it.

CareTrust expects the portfolio to generate a stabilized yield of approximately 8.6% (CareTrust REIT press release, September 15, 2026). Against a 10-year Treasury at 5.004% on September 16, that is a spread of roughly 360 basis points — computed here by subtracting the benchmark from the disclosed stabilized yield, not a published figure. Across everything CareTrust has bought this year the picture is the same: a 2026 year-to-date investment total of $1.9 billion across two dozen transactions at a blended stabilized yield of approximately 8.7% (CareTrust REIT press release, September 15, 2026).

A 360 basis point spread is a workable risk premium. It is not a heroic one, and that is precisely the point. Seniors housing and skilled nursing did not need the rate to fall; it needed the going-in yield to sit far enough above the risk-free rate that the trade still clears. At a 6.5% cap rate against a 5% ten-year, nothing clears. At 8.6%, a buyer with an in-place operator and a triple-net lease can still underwrite it.

Irving Levin Associates made the same observation from the deal-flow side that week, noting that as rates rose through the year the seniors housing and care M&A market remained remarkably insulated in both activity and dealmaking (The SeniorCare Investor, 60 Seconds with Swett, September 15, 2026).

What the structure was doing

Read the CareTrust disclosure closely and the resilience looks less like optimism and more like engineering.

The portfolio is triple-net leased back to the existing operator under a long-term lease with annual inflation-based rent escalators and multiple renewal options (CareTrust REIT press release, September 15, 2026). Every element of that sentence is a rate hedge. Inflation-based escalators are the direct answer to a long end selling off on inflation data — the landlord's income line moves with the same index that moved the yield. Multiple renewal options extend duration without re-underwriting. A known operator with, in CareTrust's words, established scale and deep operating history in those markets removes the re-tenanting risk that would otherwise demand a wider spread.

The opportunity was also sourced off-market, and it closed through a joint venture in which CareTrust invested approximately $380 million of the $400 million total (CareTrust REIT press release, September 15, 2026). A bilateral negotiation with a counterparty you already know does not require the price discovery that a marketed process does — which is exactly what a volatile rate week makes expensive.

The recapitalization tell

The second transaction of the week is the more honest read on where the market actually hurts.

Senior Living Transformation Company closed a $300 million recapitalization of the Prime Care Portfolio — 957 assisted living and memory care units across nine assets in Connecticut, New Jersey, Maryland and North Carolina, operated by Sunrise Senior Living. The transaction was sourced off-market, no partners exited, ownership shares were reshuffled, and SLTC came in with preferred equity (The SeniorCare Investor, September 15, 2026).

Nothing was sold. That is the signal. The assets are described as older vintage in strong, affluent markets — product with real underlying demand and a capital stack that no longer works at a 5% benchmark. When the senior debt math turns against a sponsor but operations do not justify a sale, the market clears with preferred equity instead of a transfer of title. A $300 million recap total is not a price and should never be lifted as one; dividing it by 957 units would produce roughly $313,000 per unit and that figure would describe a rebuilt capital stack, not the value of a fee interest.

Two deals, one week, one rate environment: assets with an in-place operator and inflation-linked income trade; assets with a stressed stack get refinanced rather than sold.

Where leverage ceases to be the variable

Rising rates compress what debt can do, and in seniors housing the ceiling is published rather than negotiated. HUD's Section 232/223(f) Express Lane caps proceeds at 70% loan-to-value supported by in-place trailing-twelve normalized NOI, with minimum debt service coverage of 2.0x for skilled nursing and 1.6x for non-skilled portions (HUD Office of Residential Care Facilities, announced and effective June 11, 2025, LTV requirements revised March 31, 2026).

A DSCR test is a rate test. At a higher coupon, the same NOI supports less principal, and the binding constraint silently moves from the LTV cap to the coverage cap. Sponsors who underwrote to 70% LTV in a 4% world are discovering that coverage, not value, now sets proceeds. That is the mechanism behind the recapitalization wave — not a collapse in asset values, but a shrinkage in what the same asset can carry.

It is also why the life-company execution on Allegro Harrington Park that same week is worth noting: JLL arranged a five-year acquisition loan from a life insurance company for AEW Capital Management and Foundry Commercial on a 173-unit Class-A community at 200 Old Hook Road in Bergen County, New Jersey (Real Estate NJ, reporting JLL Capital Markets, September 2026). Terms were not disclosed. Life-company paper competing for stabilized seniors housing at this point in the cycle means the lender is underwriting durable in-place cash flow, not a lease-up story.

What this means for underwriting now

Three practical conclusions, each of which follows from the disclosed figures above rather than from a forecast.

First, underwrite to a spread rather than to a rate. The transactions clearing in September 2026 are clearing at roughly 360 basis points over the benchmark. If a deal does not produce that on in-place numbers, the rate is not the problem.

Second, treat inflation-linked escalators as a valuation input, not boilerplate. In a market where the long end is moving on inflation prints, a lease whose rent tracks the same index is structurally different from a lease with fixed 2.5% bumps, and it should not be capitalized identically.

Third, expect the tape to fill with recapitalizations rather than sales, and refuse to read them as price discovery. A recap total divided by a unit count is not a comparable. Our own wire carries several of these now, and the price field on every one of them is deliberately null.

The demand side has not changed through any of this. The defensible Census-derived figure is roughly 2,200 net daily additions to the 80-plus population — a demographic that arrives regardless of where the ten-year settles. What changes with a 5% benchmark is not whether the resident shows up. It is who can afford to own the building when they do.