EstateVision

Market

One energy shock, three mortgage designs: who pays the higher rate depends on the loan

The US 30-year fixed rate went from 6.76% to 7.40% in four weeks and has since eased to 7.18%. UK quoted fixes sit about 95 basis points above pre-conflict levels, and the euro-area composite is 3.62%. The loan design decides who feels it.

02.004.006.008.0010 Sep17 Sep1 Oct8 Oct29 Oct5 Nov%
US 30-year fixed mortgage rate, weekly average, autumn 2026. Source: freddiemac.com · United States.

The average US 30-year fixed mortgage rate rose from 6.76% on 10 September 2026 to 7.40% on 8 October, then eased to 7.18% on 5 November, according to Freddie Mac's weekly survey. A year earlier it was 6.22%. Over the same autumn the Bank of England put the quoted two-year fixed mortgage rate in the UK about 95 basis points above its pre-conflict level, and the ECB reported the euro-area composite rate on new house-purchase loans at 3.62% in September, up from 3.44% in April. These numbers are not like for like: one is a weekly survey average, one a quoted product rate, one a weighted composite, and the windows differ. They still show how one shock reaches borrowers through three different loan designs.

The common cause has an official source. The Bank of England's September minutes say UK financial conditions tightened because short-term overnight index swap rates rose, with similar moves in the United States and the euro area, which it attributes to the global energy supply shock. The same minutes record Brent crude at $106 a barrel and UK wholesale gas at 207 pence per therm on 14 September.

The US rate is a bond price, so it reprices within days

A US borrower typically takes a 30-year fixed loan that lenders fund by selling mortgage bonds. The quoted rate therefore tracks long-term bond yields, not the central bank's overnight rate. When yields jump, quotes jump with them. Freddie Mac's survey draws on purchase applications submitted that week, so it records the move almost as it happens.

The weekly path in the chart below shows it: 6.76% on 10 September, 6.95% on 17 September, 7.28% on 1 October, 7.40% on 8 October, 7.24% on 29 October and 7.18% on 5 November. The values come from Freddie Mac's survey and its dated weekly releases, which also report the prior week's and the year-earlier reading. The rise and the partial retreat each took roughly a month.

For a buyer the effect is large. On a 30-year amortising loan, 7.18% instead of 6.22% lifts the monthly payment per $100,000 borrowed from about $614 to about $677, or 10.4%. A household with a fixed monthly budget can borrow 9.4% less. These are my calculations from the two survey rates, ignoring taxes and insurance.

Existing borrowers with fixed loans feel none of it. The shock lands on buyers and refinancers, which makes US pass-through quick but narrow.

The UK moved on swap rates while Bank Rate stood still

UK lenders price two- and five-year fixed deals off swap rates, which follow expectations for Bank Rate. The Monetary Policy Committee voted 6 to 3 at its meeting ending 16 September to hold Bank Rate at 3.75%. Three members wanted a rise of 0.25 percentage points. The market curve for short-term rates sloped upward and peaked around 4.9% by end-2027.

The minutes describe full and fast pass-through from short-term market rates to key lending rates for households and businesses, and put the quoted two-year fix about 95 basis points above its pre-conflict level. A held policy rate and a sharply higher mortgage rate can coexist because mortgages follow the expected path, not today's setting. UK CPI inflation was 3.1% in August, and the Committee judged the risks to inflation tilted to the upside.

The UK's cost arrives in instalments. Each month some borrowers reach the end of a fix and roll onto the current quote, so the stock of mortgages reprices slowly even when new quotes jump. I have not used the Bank's series for new and outstanding mortgage rates here, because I could not tie them to a page I read for this note. They are the data that would measure the lag directly.

The euro area blends both designs, and the aggregate hides the exposed borrowers

The ECB's composite cost-of-borrowing figure averages countries with different mortgage cultures. Some markets lean on floating or short-fixation loans, others on long fixes. In its September release, published on 2 November, the ECB gives the composite for new house-purchase loans at 3.62%, which it describes as broadly unchanged from 3.60% in August. April's figure was 3.44%.

The segments diverge. The rate on loans with a floating rate or an initial fixation of up to one year went from 3.56% in April to 3.89% in September, up 33 basis points. Loans fixed for five to ten years went from 3.61% to 3.82%, up 21 basis points. On a 25-year loan, the composite move lifts the payment by about 1.9%, the floating segment's by about 3.6% and the five-to-ten-year segment's by about 2.3%. These are again my calculations.

The sources I read give no country split and no share of floating-rate lending, so the exposed group is visible in the rates but not sized. The aggregate is a weak guide to who is paying more.

Who bears the shock matters more than how fast it arrives

I do not claim that product design explains the size of the gaps between these numbers. The windows run from four weeks to five months, and the definitions differ. What the evidence does support is narrower. The shock is common, and the design decides the incidence.

In the US, new buyers carry all of it, and owners with older, cheaper loans have little reason to sell. In the UK, the cost is spread across refinancing cohorts, so affordability pressure can keep building after quotes stop rising. In the euro area, floating-rate and short-fixation borrowers carry the adjustment first, which points to differences between countries rather than one euro-area outcome.

Two uncertainties cut across this. A Bank Rate rise would feed straight into swap-priced UK fixes. And the euro-area figures describe the market only through September, so anything that has happened to bank pricing since is not in them.

What to watch:

  • Freddie Mac's weekly survey, published Thursdays. A reading below 6.95%, the 17 September level, would show the autumn spike fully reversed. A return to 7.40% would show the retreat was a pause.
  • The next Bank of England decision. A move in Bank Rate, or a vote count shifting further toward a rise, would set the path for UK swap-priced fixes.
  • The ECB's next monthly bank interest rate statistics, covering October. The September figures came out on 2 November, so on a monthly cadence the October release should follow about a month later. A composite well above 3.62%, with the floating and up-to-one-year rate beyond 3.89%, would confirm that short-reset borrowers are carrying the adjustment.

Questions

Why can UK mortgage rates rise while Bank Rate is held?
UK fixed deals are priced off swap rates, which follow expectations for Bank Rate. The Bank of England held Bank Rate at 3.75% in September 2026 while the quoted two-year fix was about 95 basis points above its pre-conflict level.
Are the US, UK and euro-area rates directly comparable?
No. They are a weekly survey average, a quoted product rate and a weighted composite, over different windows. The comparison is of mechanism and direction, not of speed or level.
Who feels a US mortgage-rate rise?
Mainly buyers and refinancers. Holders of existing fixed-rate loans keep their rate, so pass-through is fast but narrow.

Sources

  1. Freddie Mac, Primary Mortgage Market Survey (as of 5 November 2026)
  2. Freddie Mac, Primary Mortgage Market Survey archive (weekly series)
  3. Freddie Mac: Mortgage Rates Average 6.95% (17 September 2026)
  4. Freddie Mac: Mortgage Rates Average 7.40% (8 October 2026)
  5. Bank of England, Monetary Policy Summary and Minutes, September 2026
  6. ECB, Euro area bank interest rate statistics: September 2026
  7. ECB, Euro area bank interest rate statistics: August 2026
  8. ECB, Euro area bank interest rate statistics: April 2026

How we write: each note starts from published data and the sources listed above, is drafted with AI and passes automatic checks on every figure, link and claim before it goes out. Analysis, not investment advice.

More research notes