SOFR closed at 3.62% on June 4. A $20M hotel loan at SOFR + 250 bps is running $1.224M annually in interest. Futures are pricing the policy rate near 3.8% by December. That's not a cut cycle. That's a drift higher... and the math on floating-rate hotel debt just shifted from "manageable" to "actively corrosive."

Let's decompose what a 30-basis-point move actually does. On that same $20M loan, annual debt service increases by $60,000. Sounds modest. Apply a 7% cap rate and you've lost $857,142 in asset value. At 8%, it's $750,000. Neither number is modest. Neither number shows up in a press release about the Fed holding steady. But both numbers show up in a disposition model, a refinancing appraisal, and an owner's equity position. The headline says "no change." The balance sheet says otherwise.

The refinancing wall makes this worse. There's roughly $48 billion in CMBS hotel loan maturities hitting between 2025 and 2026. Owners who locked in at legacy rates near 4.5% are now facing refi environments at 6.25-7%. That's a 40% jump in servicing costs. Debt service coverage ratios across the sector have compressed by 217 basis points since Q1 2024. I've seen this compression pattern before at a REIT I worked at... properties that looked healthy on a trailing-twelve NOI basis suddenly couldn't cover debt service under the new rate, and the conversation shifted from "refinance" to "extend and pray" to "sell." The sequence happens faster than most owners expect.

The construction side confirms the thesis. Hotel rooms under construction hit their lowest level since August 2022 as of late 2024. Q1 2026 completions were the lowest quarterly total CBRE has ever tracked. CBRE's forecast of $562 billion in commercial real estate investment this year is almost entirely capital chasing existing assets, not new builds, because new construction pro formas don't pencil at current rates. That's good news if you own a stabilized asset with fixed-rate debt (less future supply competition). It's irrelevant if your floating-rate loan is repricing upward while your NOI stays flat.

The owners who assumed 2026 would bring rate relief are out of runway. Friday's jobs report pushed Fed Fund futures to their highest level since February 2025... a rate hike is now priced in by year-end, not a cut. Every month an owner waits to address a floating-rate exposure is a month where the refi economics get worse, not better. The spread between "I should have locked in" and "I can still lock in" is widening. At some point it becomes "I can't lock in at a rate that covers my debt service." That's the point where refinancing becomes recapitalization... or disposition.

$48 billion in CMBS hotel loan maturities hit between 2025 and 2026. That is the largest concentration of any commercial property type. Hotel mortgage spreads already widened to 375 basis points over comparable treasuries in Q4 2025 (a 125-150 basis point premium over multifamily and industrial). The Fed held rates last week. The market is now pricing a hike.

Four FOMC dissents. First time that's happened since October 1992. Three regional presidents argued the committee's easing bias was wrong... that the next move could be up, not down. A fourth wanted a cut. That's not consensus. That's a committee that doesn't agree on direction, which means the rate path everyone underwrote in 2022 (originate floating, refi when rates drop, capture the spread) is broken. Rates didn't drop. They might rise. And 30% of hotel mortgage balances mature this year.

Let me decompose what a hike means at property level. A 25-basis-point increase on a $20 million floating-rate loan is $50,000 in annual debt service. The source article equates that to 3-6 lost room nights per month at a 300-room hotel running 70% occupancy and $150 ADR. Check again. $50,000 divided by 12 months is $4,167. Divided by $150 ADR, that's 28 room nights per month. Not 3-6. Twenty-eight. At 50 basis points, it's 56 room nights per month. That's the real number, and it changes the severity of this story considerably. (I flag math errors because math errors in debt analysis get people into trouble. Ask anyone who trusted a franchise sales projection without checking the denominator.)

The squeeze isn't just debt service. CPI printed 3.3% in March. PCE ran 4.5% in Q1. Labor, insurance, F&B, utilities... all inflating. RevPAR has to outrun both operating cost inflation and rising debt service simultaneously. For a property that underwrote 5% annual RevPAR growth and got 2%, the gap between the pro forma and reality is now wide enough to trip a debt service coverage covenant. I've audited portfolios where the DSCR cushion looked comfortable at origination and evaporated within 18 months when two assumptions moved against the owner at once. Two assumptions are moving right now.

One more variable. Jerome Powell's term as chair ends May 15. Kevin Warsh, the incoming nominee, has advanced through the Senate Banking Committee. A leadership transition at the Fed during a period of internal disagreement adds uncertainty to the rate path that no pro forma can model. Owners with loans maturing in the next 18 months are refinancing into a market where spreads are already elevated, the benchmark rate may rise, and the new chair's policy stance is untested. That is not a "watch and wait" situation. That is a "call your lender this week" situation.

SOFR closed at 3.63% on May 4. The Fed held steady on April 29. Three FOMC members dissented against the statement's easing language. Kevin Warsh, widely regarded as more hawkish than Powell, takes the chair in mid-June. The direction of the next move just shifted, and for hotel owners carrying floating-rate debt, the shift reprices their entire capital structure.

Let's decompose the exposure. A 25-basis-point increase on a $15M floating-rate loan adds roughly $37,500 in annual debt service. On a $40M full-service asset, that's $100,000. These aren't hypothetical numbers pulled from a model... they're arithmetic applied to loan balances that exist on real balance sheets right now. A significant volume of hotel debt originated or refinanced between 2022 and 2024 is floating-rate, often SOFR-based, because that's what the debt funds and transitional lenders were underwriting during the rate run-up. Owners who took that paper expecting rate relief by 2026 are now facing the possibility of rate expansion instead. The spread between expectation and reality is where defaults live.

The commercial real estate delinquency data confirms this isn't theoretical risk. Overall CRE mortgage delinquencies hit 4.02% in Q1 2026, up from 3.86% the prior quarter, with lodging among the sectors posting increases. Office CMBS delinquencies reached 12.34% in January before easing to 11.4% in February. Office is the headline, but the mechanism is identical for hotels: owners can't refinance maturing debt at rates that preserve positive leverage, covenant headroom erodes, and the workout conversation starts. Hotels in secondary markets running 1%-1.5% RevPAR growth against 25-50 basis points of potential debt service increase are staring at margin compression that no operational efficiency can offset.

There's a structural irony here that's worth stating plainly. The same rate environment that pressures existing owners also suppresses new construction (the U.S. hotel pipeline contracted roughly 5% year-over-year in Q1 2026). Fewer new rooms means less supply competition for properties that survive the refinancing gauntlet. The owners who can service their debt through this cycle inherit a better competitive position on the other side. The owners who can't... don't get to participate in that upside. The market is selecting for balance sheet strength, not operating quality. I've seen this pattern in prior cycles. The best-run hotel in a submarket can still lose to a mediocre property with better capitalization if the debt structure breaks first.

The immediate action isn't strategic. It's mechanical. Pull your loan documents. Confirm whether you're floating or fixed. Check your rate cap expiration (a surprising number of caps purchased in 2022-2023 are expiring or have expired without replacement). Model 25 and 50 basis points of upside on your current debt service and compare that to trailing NOI after reserves. If the coverage ratio drops below 1.25x, you're in lender conversation territory whether you initiate it or not. Better to initiate it.

The federal funds rate sits at 3.5%-3.75%. The 90-day SOFR average is 3.67%. March CPI came in at 3.3% year-over-year, up from 2.4% in February, driven largely by a 21.2% spike in gasoline prices. CME FedWatch shows a 78% probability of zero cuts through 2026. JPMorgan's chief U.S. economist is forecasting a potential 25 basis point hike in Q3 2027. That is the current rate environment. Now go pull the underwriting assumptions on every hotel deal signed in 2023.

I'll tell you what those assumptions said, because I've audited enough of them. They assumed SOFR in the low 2s by mid-2026. They assumed a refinancing window that would let sponsors term out floating-rate construction debt at materially lower spreads. They assumed cap rate compression on exit... 7%, maybe 7.25%, because "the cycle is turning." Approximately 30% of all loans backed by hotel properties are scheduled to mature this year alone. The Mortgage Bankers Association puts the combined 2026-2027 commercial mortgage maturity wall at $1.5 trillion, with an estimated $114 billion in hotel-specific debt. The sponsors who underwrote those deals aren't getting the rate environment they modeled. They're getting SOFR at 3.67% and lenders who just watched the office sector implode and decided to tighten standards across every CRE property type.

Let's decompose what this means per key. A 200-room select-service project financed with floating-rate construction debt in 2022, assuming a 2026 takeout at SOFR plus 200 basis points, probably modeled permanent debt at roughly 4.5%. The actual refinancing rate today is closer to 6%-6.25%. On a $30M loan, that's approximately $450K-$525K in additional annual debt service. That's $2,250-$2,625 per key per year in carrying cost the original pro forma didn't account for. Run that against trailing NOI. If the DSCR was modeled at 1.35x and actual NOI hasn't moved, it's now sitting at or below 1.10x. That's covenant territory. That's the lender calling you, not the other way around.

The development pipeline isn't dead but it's repricing in real time. Limited-service hotels in secondary markets are running $245K per key in total development cost. Exit cap rate expectations have moved from the low 7s to 8%-8.5%, with some brokers quoting 9.5% for upscale product. That's a 150-200 basis point shift in assumed exit value on the same NOI. A portfolio I analyzed last year had three development deals in the pipeline, all approved at a 7.2% exit cap. The sponsor hasn't broken ground on two of them. They won't, at current pricing. The equity check to make those deals work at an 8.5% exit cap is a fundamentally different conversation with investors... and most sponsors haven't had that conversation yet.

Extension requests are where this gets real. Twelve months ago, a borrower with decent trailing performance could get a 12-month extension with a phone call and a small fee. That environment is gone. Lenders now want current TTM NOI (not the NOI from your original underwriting package), updated appraisals (which are coming in lower because cap rates expanded), and evidence of demand stability in the specific market. I've seen three extension requests in the past 60 days that required fresh equity from the sponsor just to maintain covenant compliance. That's not refinancing. That's recapitalization at the worst possible time. The sponsors who haven't stress-tested their entire portfolio against a flat-to-higher rate environment through Q4 2027 are making a bet they don't realize they're making.

The federal funds rate holds at 3.5%-3.75%, and J.P. Morgan now expects it to stay there for the rest of 2026. That's not a forecast revision. That's a repricing of every hotel deal underwritten in the last 18 months on the assumption that relief was six months away. It wasn't. It isn't.

Let's decompose what "holds steady" actually costs. A 200-key select-service property carrying $18M in floating-rate debt at SOFR plus 400 basis points is paying roughly 7.8% today. The owner who penciled a 2026 refi at 6.5% (assuming two 25-basis-point cuts) just lost $234,000 in annual debt service savings that were already baked into the hold model. That's not a rounding error. That's the difference between a property that cash-flows and one that doesn't. And the Feb jobs report (negative 92,000 payrolls, unemployment at 4.4%) suggests the revenue side isn't coming to the rescue either.

The PIP math is worse. Bank construction loan rates for hospitality sit at 7.33% to 8.33% right now. An owner facing a $4M brand-mandated renovation is financing that at roughly $330,000 in annual interest alone before a single wall gets touched. I audited a management company once that ran a portfolio-wide PIP analysis assuming "normalized" financing costs of 5.5%. Every property in the model showed positive ROI. At actual rates, eleven of fourteen were underwater. The spreadsheet was beautiful. The assumptions were fiction. That's the gap I keep finding... the model that "works" versus the model that reflects what the lender actually quotes.

The development pipeline is where the math gets interesting (and by interesting I mean it doesn't close). Ground-up hotel construction requires cap rate compression or revenue growth to justify current financing costs, and neither is appearing. Average hotel cap rates ran 9.5% in 2025. A developer borrowing at 8% on a construction loan and targeting a 9.5% exit cap has roughly 150 basis points of spread to absorb all construction risk, lease-up risk, and timing risk. That's not a deal. That's a prayer. The secondary story here is adaptive reuse... converting distressed office and retail into hotels at 60-70% of ground-up cost, with faster timelines. Oil at $96 a barrel (up 44% this month alone on the Iran conflict) is pushing construction material costs higher, which only widens the gap between conversion economics and new-build economics.

One more number, because it matters. Core PCE inflation printed 3.1% in January. The Fed's target is 2%. Until that gap closes, rate cuts aren't a debate... they're a fantasy. Every owner, asset manager, and developer reading this should update their models today with one assumption: 3.5%-3.75% through December 2026. If you're still running scenarios with H2 rate relief, you're not modeling. You're hoping. Check again.

$48 billion in CMBS hotel loans mature across 2025-2026, and refinancing costs are jumping roughly 40% from where they were at origination. That's the real number in this Fed hold. Not the rate itself. The refinancing gap.

Construction loan rates sit between 5.50% and 8.75% as of February. Compare that to what developers underwrote three years ago. A select-service project penciled at a 6.2% unlevered yield with 4% debt looked like a solid spread. That same project at 7.5% debt doesn't pencil at all. The yield didn't change. The cost of capital did. And the margin between "viable" and "dead" in select-service development is maybe 150 basis points on a good day. We blew past that threshold 18 months ago and haven't come back.

Prediction markets put the probability of a March hold at 99%. The January FOMC minutes showed two members dissenting in favor of a 25-basis-point cut, which means the committee isn't unanimous, but it's close. Boston Fed President Collins said last week she sees no urgency for cuts until inflation returns to 2%. Core PCE came in at 4.3% annualized in December. That's not close to 2%. The American Bankers Association projects inflation stays above target for the next eight quarters. Eight. If that holds, we're looking at late 2026 for the first meaningful relief (and even Goldman's optimistic forecast only gets you to 3.00%-3.25% by year-end, which still leaves construction debt expensive by any historical standard).

Here's what the headline doesn't tell you. The distress isn't evenly distributed. An owner who locked a 10-year fixed rate in 2018 at 4.2% is fine. An owner who took a 5-year floating-rate construction loan in 2021 at SOFR plus 250 is staring at a refi that could push debt service above NOI. I analyzed a portfolio last year where three of seven assets had loan maturities within 18 months. Two of the three couldn't cover projected debt service at current rates. The ownership group's options were inject equity, sell at a discount, or hand back the keys. That's not a hypothetical. That's the math for a meaningful percentage of the $48 billion in maturities. REITs and institutional buyers with undrawn credit facilities and sub-4% weighted average cost of capital are building acquisition teams right now. They should be.

HVS projects 2.2% RevPAR growth for 2026. Modest. But pair that with supply growth slowing (because nobody's breaking ground at 8% construction financing), and existing assets in good physical condition get a tailwind. The owners who renovated in 2019-2021 when capital was cheap are sitting on a competitive advantage they didn't plan for. The owners who deferred CapEx hoping rates would drop are now deferring into a market where their comp set is pulling ahead. RevPAR growth without margin improvement is a treadmill. But RevPAR growth with suppressed new supply and a recently renovated product... that's the rare scenario where the math actually works for the operator.