The US 10-year Treasury touched 5.36% last week, its highest level since 2002, in the same week the odds of an October Fed hike fell to one in five. The long end of the bond market is now setting its own price. Every private asset underwritten against a 4% Treasury is being marked to the wrong number.
The number
On Wednesday 7 October the 10-year US Treasury yield reached 5.36% intraday, a level last seen in April 2002. The 30-year bond traded above 5.7%, also a 24-year high. The 10-year has risen around 60 basis points since the end of July, and the third quarter was the biggest quarterly rise in the global benchmark this century.
The Treasury sold $39 billion of 10-year notes into that market the same afternoon. Buyers turned up. That matters: it tells you 5.3% is a level investors are prepared to own, not a liquidity accident that corrects by the end of the week.
Who is not driving this
Last month we wrote that the Federal Reserve had hiked into a refinancing wall. On 16 September the Fed raised rates by 25 basis points to 3.75% to 4.00%, its first increase since July 2023, on a unanimous vote.
Since then the data has gone soft. September payrolls missed, August PCE came in cooler than expected, and the market’s probability of a second hike at the 28 October meeting fell from roughly 70% to about 20%. Two Fed officials, including the New York Fed president, have said there is no need to rush.
And the 10-year went up anyway.
That is the story. The front end has stopped pulling the long end. The 2-year sits near 4.8%, the 10-year near 5.35%, and the gap between the Fed’s policy rate and the 10-year has widened to roughly 140 basis points. Investors are no longer pricing what the Fed will do next month. They are pricing what it costs to lend to a government for a decade, and they want more for it: a term premium for deficits that are not shrinking, for record issuance, for corporate borrowing that is competing for the same capital, and for oil near $100 keeping inflation expectations from settling.
The cleanest way to see it is the real yield. The 10-year inflation-protected Treasury now yields 2.94%, more than a full percentage point higher than a year ago. A near-3% real return, risk-free, is the number every private asset in the world now has to beat.
This is not a US story
The UK 30-year gilt yield crossed 6% two weeks ago, its first time above that level since 1998. France’s 10-year is approaching 5%, with the spread over German Bunds stretching toward levels last seen in the eurozone debt crisis. Japan’s 10-year is at 3.11%, a three-decade high, as the Bank of Japan keeps raising rates. Goldman Sachs’ international chief has publicly called for government spending cuts in response to the bond market.
This is a global repricing of duration. Long-term money has become expensive everywhere at the same time, and the countries with the largest borrowing needs are paying the most for it.
What a 5.35% ten-year does to private markets
Real estate feels it first because the maths is unforgiving. Earlier this year the US all-asset transaction cap rate was about 6.3% against a 10-year averaging 4.4%, a spread of under two percentage points and historically narrow. If cap rates have not moved since then, the spread against a 5.35% Treasury is now roughly one percentage point. Historically, that spread has been considerably wider.
That gap closes one of three ways. Treasury yields fall. Cap rates rise toward 8% and property values fall with them. Or buyers accept being paid almost nothing for illiquidity, leasing risk and leverage. The residential side has already repriced: the 30-year US mortgage averaged 7.28% at the start of October, up from 6.34% a year ago. In early September we wrote that nearly $1 trillion of commercial real estate debt matures in 2026. Every refinancing assumption behind that wall used a lower 10-year than the one on the screen today.
Private credit is mostly floating rate, priced over SOFR, so the long end does not reprice its coupons directly. It reprices everything around them. Borrowers who planned to term out floating debt into fixed-rate markets now face worse pricing, so leverage stays floating for longer. The real estate and infrastructure collateral behind the loans is worth less at a higher discount rate. And the investor who was willing to sit in a gated semi-liquid fund for a 9% target when Treasuries paid 4% is doing a different calculation at 5.35% nominal and 2.9% real. Last month US private credit defaults set a record while the largest funds rationed withdrawals. The risk-free alternative just got better.
Private equity is the slowest to reprice and that is the problem. Terminal values that were set with a 4% discount rate do not survive a 5.35% one. Buyers using debt at 9% to 10% all-in cannot pay 2024 multiples, so exits either clear lower or do not clear. Marks that have not moved in two years are now a long way from where a transaction would actually print.
Where the capital goes
The winners in this repricing are not complicated. Owners of cash flow with long-term fixed-rate debt locked in before 2022. Lenders with term funding who can write new loans at today’s rates against assets priced at yesterday’s. Unleveraged capital, which in this region means sovereign and family balance sheets, that can buy from sellers who have to refinance and cannot. And short-duration credit, where a 5% real-plus return no longer requires taking duration risk at all.
The losers are anything whose return was the spread between a 4% Treasury and a 6% asset yield. That spread was always thin. It is now close to zero.
Bonds reprice in a day. Private markets reprice over quarters. The gap between the two is where the next twelve months of transactions will happen, and the sellers will be the ones who waited for the Fed to decide what the long end had already decided without it.
Last month the question was whether banks could afford to refinance the wall. Last week the bond market answered a different one: what the assets behind the wall are actually worth.
Sources: US Treasury and Tradeweb yield data via CNBC and Bloomberg (7 October 2026), Federal Reserve September FOMC statement and minutes, Freddie Mac Primary Mortgage Market Survey (1 October 2026), First American Potential Cap Rate Model (Q1 2026), CME FedWatch.
