Supplement 27 Sep 26 - The Lettuce-versary, Vistry Goes Small, and the Banks Mind the Gap
"Victory has a hundred fathers and defeat is an orphan." - John F. Kennedy, press conference, 21st April 1961
The quote pertains to the Deep Dive, and to an anniversary that nobody in Downing Street or Threadneedle Street was keen to mark. Kennedy was repeating an old saying a few days after the Bay of Pigs, and the bit people forget is that he used it to take the blame himself. Four years ago this week the gilt market had the worst fortnight of my working life, and in the four years since, that particular defeat has been an orphan in the most British way imaginable. The Treasury of the day blames the Bank. The Bank points politely back at the Treasury. The pension schemes point at their consultants, and the lettuce, which outlasted a Prime Minister, keeps its own counsel. This week I went back to the primary record to find out who the father actually was, and with gilt yields making higher highs again, it is not a history lesson.
As the gilt market sets fresh highs and the party conference season opens, our next Property Business Workshop is now four days away. Thursday 1st October, the Hyatt Regency in Manchester, with myself and Rod Turner, powered by Roma Finance, who do bridging, development and term lending and who have been very good company on this journey. JVs and M&A is the subject. When the cost of debt moves this far this fast, knowing how to bring in a partner with cash, or how to buy well from somebody who needs to get smaller, stops being an advanced topic and becomes the whole game. Read the Deep Dive and you will see a listed housebuilder doing the shrinking in public this week. Book in at www.tinyurl.com/pbwoct26
Welcome back to Trumpwatch. Last week the President's own Fed chairman put interest rates up. This week the President flew to New York, stood at the podium of the United Nations General Assembly, and asked the room a question out loud. Would he make a deal with Iran, or would he “annihilate” the Islamic Republic? His word, not mine. He then told the same room that he believes a deal will come right after the election, which is either a negotiating position or a campaign diary. Possibly both.
Let me be careful here, because it is very easy to write this section as a sneer and I do not think that helps anybody with a portfolio to run. Underneath the theatre, the substance moved this week, in both directions. On Tuesday, the same day as the speech, his envoys Steve Witkoff and Jared Kushner met Iranian officials on the sidelines. The President's review was that it was a very good meeting with people who were not the highest level, "but it's something." Heh. By Thursday Iran's foreign minister, Abbas Araghchi, was telling reporters that Tehran had sent Washington, via Qatar, a seven-day plan to reopen the Strait of Hormuz, built on the Islamabad memorandum that fell apart in early July. Reports on Thursday and Friday had negotiators exploring a phased version: Iran reopens the Strait, America lifts its blockade of Iranian ports. And then on Saturday morning came reports, citing unnamed US officials, that the President had rejected the Iranian proposal.
So where are we? Roughly where we were a fortnight ago, with more people in the room. I suspect the phased idea survives the rejection of this particular draft, because both sides need a version of it and nobody has come up with a better one, but I would not put money on the timing. Back home, the Senate narrowly rejected a resolution that would have forced the President to seek congressional authorisation for the war, a week after the House passed a separate one of its own. Standard Fare. Likely outcome before the midterms on 3rd November - more talking, some bombing, and no Strait.
Now the bit of Trumpwatch that could end up in your builder's van. American diesel hit a record $6.53 a gallon on Tuesday, and the President said he had called internally for a ban on US diesel exports. Treasury Secretary Scott Bessent said the administration was looking at whether a full or partial ban was practical. His own Energy Secretary, Chris Wright, and Interior Secretary, Doug Burgum, are against it. The American Petroleum Institute warned it would backfire and pointed out that the United States supplies roughly a fifth of the diesel traded by sea each day, and the energy economist Philip Verleger reckons a ban could push world diesel prices up by as much as 100%. Why does a policy argument in Washington matter in Walsall? Because we are a net importer of diesel, and our pumps are already close to the edge. The RAC had the UK average at 198.32p a litre on Friday, within a penny of the 199.09p record set in 2022, with petrol at 173.6p, the highest in more than four years. Every trade, every delivery, every contractor quoting you for a refurb runs on the stuff. If America stops selling it, Europe bids for what is left, and that lands in our inflation number in about a month.
Which brings us to the bond market, and this was the week the American one had its own go at higher highs. The ten-year Treasury yield jumped 15 basis points on Wednesday to 5.11%, the highest since 2007 and the biggest one-day rise since April last year, and touched 5.18% on Thursday. The thirty-year went as high as 5.47%, a level not seen for 22 years. Across the week the ten-year was up about 22 basis points. The average thirty-year fixed mortgage in America hit 7.26% on Wednesday, the highest of this President's second term. Meanwhile the flash composite PMI for the US came in at 58.4, a 54-month high, so the economy is running hot at exactly the moment its energy bill is going through the roof. Put those together and you can see why the Fed hiked last week and why markets are asking whether it has to go again. A reminder that the long fixes are great when the market is cheap, but punishing when the market is expensive. You think 5% or 6% debt is bad…….
Is there a world in which the President who wants 1% interest rates blames the man he picked for 7.26% mortgages? There is. Not yet. Last week the target was a "hostile board". I said then that the volte face was a matter of when, not if, and a mortgage rate with a seven in front of it five weeks before the midterms is the sort of thing that brings a when forward.
So why does any of this matter to somebody with six houses in Derby? Two reasons, and they both land on the gilts.
First, contagion. Our ten-year rose about 6 basis points this week while theirs rose 22. For once we were the calm one, and I would like to think that is partly because the Bank of England stopped selling long gilts last Thursday, which took one source of supply pressure out of our market that the Americans still have in theirs. I might be flattering us. A week is not a trend, and when Treasuries go on a run, gilts have a habit of catching up.
Second, the rate gap I flagged last week. The top of the Fed's range is 4%. Our Bank Rate is 3.75%. Sterling closed the week at about $1.325, and a weaker pound makes every barrel of oil and every tonne of diesel dearer in pounds. That is precisely the transmission the Bank's Deputy Governors spent this week talking about, and it is most of the Deep Dive. My feeling now is that it is like the last hiking cycle - we are moving too slowly.
And the five thousand dollar cheque? Still promised, still uncosted, still conditional on the midterms. Onwards.
Phew. Back we go to the comfort blanket of the real time UK property market. As is customary, Chris Watkin has been relentlessly crunching the portal numbers and publishing them at Property Industry Eye, and his analysis for Week 37 of 2026 is where it is at. This week he is joined by Alice Bullard, boss of Nested, and the conversation goes where you would hope: overvaluing, cheap fees, and whether winning an instruction means anything at all if the homeowner never moves. If you want to know how the macroeconomic weather translates to the local high street, look no further.
Supply first. 35.9k new listings in Week 37, down a touch from 36.7k the week before, against a 2026 weekly average of 36k and a ten-year Week 37 average of 35.6k. Normal, in other words. Year to date that is 1.336 million homes brought to market: 0.5% below last year's 1.342 million, 2.4% ahead of 2024, 10% ahead of 2023, and 10% higher than the 2017 to 2019 average of 1.213 million. My ready reckoner of "10% more stock than a normal market" is sitting on 10% to the decimal point. I declared it broken twice this year, parked it altogether a fortnight ago, and it has come back to sit exactly where I left it. I am tempted to have it framed.
Demand. 22.8k homes sold subject to contract, against 23k the week before. The ten-year Week 37 average is 25.2k, so this week ran at about 90% of normal, a step down from last week's 94%. The 2026 weekly average is 24.2k. Year to date 897k sales agreed: 7.4% down on last year's 969k, 1.9% down on 2024, 11% up on 2023, 5.1% above the pre-Covid years and 0.9% below the decade average of 905k. Net sales, which is gross sales minus fall-throughs, came in at 17.1k against a ten-year Week 37 average of 19.1k, and 692k for the year so far: 6.2% below 2025, 1.4% below 2024, 14.1% ahead of 2023 and 3.2% above 2017 to 2019. The Five Ds keep turning up regardless. Death, debt, divorce, downsizing and diddy ones do not read the gilt market. Sales continue to cool now, and that’s two weeks in a row on that trend - that’s definitely on the watch list.
Now the friction arithmetic for the week, and I have tried to find something new rather than recycle last week's sole agency sum. Divide new listings by sales agreed and you get a crude measure of how many homes are chasing each buyer. This week it is 35.9 divided by 22.8, which is 1.57 listings for every sale agreed. The ten-year average for this week of the year is 35.6 divided by 25.2, which is 1.41. So roughly 11% more homes are competing for each buyer than would normally be the case in late September. Rough working, and it ignores the stock already sitting there (which is at or near a cycle high). But if you are pricing a home this autumn, that 11% is the extra NEW competition you are pricing against, and it is the plainest reason why the first price has to be the right one.
Now the intergalactic number, and it moved. The gap between the average asking price of what came to market and the average asking price of what went under offer was 24.5% in Week 37, down from 29.2% the week before. £453,000 against £363,000. The ten-year norm is 16% to 17%. Same two health warnings as ever: these are different baskets of houses rather than a discount on the same house, and this is not the old intergalactic series, so do not line it up with the spring. But look at where the narrowing came from. The average new listing fell by £3,000, or less than 1%. The average home going under offer rose by £10,000, or nearly 3%. So the gap closed mostly because what was selling got more expensive, not because what was listing got cheaper. I said last week that the September gap was part mix and part fantasy, with the big family homes held back over the summer arriving at July prices. It looks as though some of those bigger homes are now finding buyers. Whether that is buyers stepping up or vendors quietly trimming before they go under offer, I honestly cannot tell you from one week. I would want two more before I say the fantasy half is shrinking. This is fairly typical fare “when September comes” of course - so don’t read a lot into this pattern, but it is worth commenting on.
A note on reductions. Chris's column this week reads 28.1k reductions on 742k homes for sale, which is word for word the same as last week's. I suspect that is the template not being refreshed rather than an identical number - so no conclusions here. What I can give you is a cross-check from elsewhere. Rightmove said this week that 74% of homes sold so far this year were priced right first time and never needed a reduction. Chris had it at about four in five back in August. Two different datasets, the same story: price it right on day one and you are very likely to sell without cutting. Price it wrong and you join the queue.
August's exchanges and withdrawals have been revised up a little as late data came in: 78.4k exchanges and 63.7k withdrawals, so 55.2% of homes leaving agents' books did so because they sold and 44.8% because the owner gave up. The seven-year average is 57.6%. Fall-throughs ran at 24.9% in Week 37 against a decade average of 24.5%. The monthly numbers are unchanged until the next month closes: 742k homes on the market on 1st September, 479k in the sales pipeline against 508k a year earlier, a sell-through rate of 13.3% for August, and £338.41 per square foot on August's agreed sales, 0.3% up on the year and 3.4% off the June record. Chris still reckons the Land Registry will show a fall of around 1.8% to 2.1% between now and March, putting us closer to my 1% growth for the year than 2% (the upper end of my range, and my original number for 2026) and my own 2027 number of 0% to 0.5% nominal growth stands (we will go down before we go back up, a bit - pending interest rates of course).
Rents. The average advertised rent in Week 37 was £1,806, up from £1,794. September so far averages £1,766 against £1,791 a year ago, a fall of 1.4%, and the year to date is £1,764. Available stock is 302k against 357k a year ago, down more than 15%. Less stock and softer asking rents at the same time is still the puzzle it was last week, and my best guess is still the mix, with London new lets softest and dragging the national average down. Affordability bosses the markets, remember - especially London where the affordability ceiling is closest.
The £64k question, then. Is the market rolling over into the autumn? I do not think so, although I can see why the headlines like the idea. Supply is exactly normal plus the reckoner's 10%. Sales are running at nine tenths of a normal September, and net sales are still above pre-Covid for the year. The asking gap narrowed. What I see is a market that has accepted money now costs what it costs, and is transacting at a pace a bit below par, at prices drifting down in real terms and flat to slightly down in nominal terms. That is a buyer's market in the south and a perfectly functional one in the north. Rightmove's regional numbers this week, which we will come to, say the same. I might be wrong about the pace. I doubt I am wrong about the direction. If we approach 15-20% lower in terms of sales (20% is normally the accepted yardstick in these situations) - then it’s time to prepare the humble pie.
Chris - this is my weekly appreciation paragraph. Thanks for what you do! If you want some help positioning yourself as a local market expert - as an estate agent or any form of property professional - give Chris a shout! Either way give his channel www.youtube.com/@christopherwatkin a follow and some love, please! The local focus this week is Shrewsbury, and Chris left a comment under the column that I have been repeating to people all week: "The valuer wins the listing. Then disappears." If you run an agency, that one line is worth the forty minutes.
Dust off the Macroscope, then. Not meat week, but a very telling one. The flash PMIs, my darlings of the real time economy, which softened. GfK's consumer confidence, which did not, and which I am going to keep calling the Burnham bounce until somebody gives me a better name for it. Rightmove's September house price index, with a nod to the official house prices and rents we went through last week. And bringing up the rear, as the contract states, the gilts and the swaps, which did the thing they keep doing.
Onto the PMIs, my darlings of the real time economy, and they had a softer month. The flash composite came in at 51.7 for September, down from 52.5 in August and below the 52.0 that the consensus had pencilled in. Still above the magic 50.0 for a third month running, but the weakest of the three. Services were also 51.7, down from 52.5, a three-month low. Manufacturing is the mixed one. The headline manufacturing PMI rose to 52.0 from 51.7, a three-month high, but the output index inside it fell to 51.4, a six-month low. So the factory headline is being held up by orders, backlogs and hiring rather than by what the factories actually made this month.
The underlying detail is where it gets interesting. Total new work across the private sector fell fractionally in September, after marginal growth in July and August, and that was mostly a renewed downturn in new business at service firms, which respondents put down to weak business and consumer confidence and pressure on discretionary spending. Export sales fell at the fastest pace since June, and manufacturers saw their first fall in export orders since December, mostly to EU customers. Employment fell again, which makes two full years of continuous job losses in this survey, although the latest fall was only marginal and softer than the average in the first half of the year. Factories took on staff for a sixth month, helped by the fastest rise in backlogs of work since January 2022, and firms pointed to AI investment and defence spending as the things holding manufacturing up.
Then the prices. Input cost inflation hit a three-month high, with fuel the most commonly cited reason, followed by energy, labour and raw materials, copper and steel especially. Prices charged rose at the fastest pace since June. Chris Williamson at S&P Global summed it up as "a worrying combination of disappointingly sluggish economic growth and intensifying inflationary pressures", reckoned the surveys are consistent with the economy growing at a mere 0.1% a quarter against 0.4% in the second quarter, and listed the reasons: energy, business costs, geopolitics, higher market borrowing costs and uncertainty over what the Budget will bring. Five weeks before a red box, that last one is always on the list.
What does it mean? A mild dose of the thing central bankers dread, where growth slows while prices speed up. Not the 1970s. Not 2022 either. But it is the combination that makes a Monetary Policy Committee nervous, because the growth half argues for waiting and the price half argues against it, and we will see in the Deep Dive which half is winning. One more thing worth holding next to it. The American flash composite was 58.4, a 54-month high, and the eurozone had its strongest output growth in nearly three and a half years. We are growing, just more slowly than the people we trade with, while paying the same energy bill. Let's hope that flash estimate is as far off as it normally is - the average revision to the composite is only 0.2 of a point, so I would not bank on much. The finals are due on 1st October for manufacturing and 5th October for services.
Next up, the Burnham bounce. GfK's consumer confidence index rose one point to -13 in September. That is the highest reading since August 2024, and the first time since the summer of 2024 that the index has risen three months in a row. The consensus had it falling to -16, so this was a proper beat. Four of the five measures rose. People's view of their own finances over the last year rose three points to -3, and their expectations for the year ahead rose one to +5. Their view of the general economy over the last year rose four points to -36, and expectations for the coming year rose one to -22, which is ten points better than a year ago. The one measure that fell was the Major Purchase Index, down one to -8, although that is still eight points better than last September. And the Savings Index, which is not in the headline, jumped five points to +27.
Neil Bellamy at GfK was the voice of caution, and I think he was right to be. His closing question was "could we soon see consumer sentiment falter?", and he pointed out that the return of higher inflation removes one of the biggest positives of recent months. Why the bounce? I suspect some of it is simply a new government's honeymoon, some of it is July's surprisingly strong GDP number working its way into the conversation, and some of it is the absence of anything dreadful happening at home for a couple of months. But look at the shape of it. People feel better about their finances, and at the same time they are saving more and are slightly less keen on the big purchase. That is not a spending boom. That is a household that feels a bit more secure and is using the security to build a cushion rather than buy a sofa. The Burnham bounce is a mood, not yet a spending pattern - but the household balance sheets are actually very strong, and have been saving because they have been made to feel particularly bearish. If Burnham’s intangible impact can be to instil consumer confidence because he is going to bat for “the people” - that savings rate could come right down and some growth COULD materialise from the consumer end, rather than just expanding the size of the state which is what has broadly been going on since 2024 (not that growth was magnificent before the election, but post-Truss things were looking as healthy as we might have hoped given the runup).
Two health warnings. First, the fieldwork ran from 1st to 14th September. That is before this week's PMIs, before the Governor's comments on Friday, before the ten-year gilt went through 5.4%, and before diesel got within a penny of its record. Next month's reading, due on 23rd October, lands five days before the Budget, and I would not assume the streak survives it. Second, for our world specifically, confidence is not the same as capacity. Rightmove's buyer demand figure this week, which is actual people enquiring about actual homes, was 9% lower than a year ago. People can feel more cheerful and still not be able to afford the mortgage (or they can have broader confidence in consuming, but not in buying houses of course).
OK. House prices. We did the official ONS house prices and rents last week, and the next official prints are not due until October, so this week the Rightmove index, published on Monday 21st, gets the floor. Average asking prices for newly listed homes rose 0.7%, or £2,441, to £367,440. That is the first monthly rise since May and a touch above the ten-year average September rise of 0.5%. It is also 0.8% lower than a year ago and 2.3% below where asking prices were at the start of the summer. Colleen Babcock at Rightmove called it "a modest recovery rather than a major turning point", which I think is exactly right.
The underlying detail says more than the headline. Homes for sale are at a 12-year high for the time of year, and buyer demand is 9% down on last year. Rightmove's average two-year fix is 5.29%, up from 5.09% a month ago and from 4.25% before the war started, and it reckons the average monthly payment on a new mortgage is now around £180 higher than it was at the end of February. The chance of actually selling varies enormously by region: 91% of homes that come to market in Scotland are finding a buyer, 71% in the North West, 56% in the South East and only 42% in London. On asking prices, Scotland is up 4.2% on the year, the North East 2.7%, the North West 1.7% and the West Midlands 1.5%, while London is down 2.6% and the South East 1.8%.
Three things I would draw out. First, the September rise is seasonal and should be read as such. Strip out the usual autumn lift and you have asking prices roughly flat on the month and 0.8% down on the year, which with inflation at 3.1% is a real-terms fall of about 4%. I keep saying this correction is being done by inflation rather than by falling nominal prices, and outside London that is still what is happening. Second, the regional split lines up almost exactly with Chris's regional pack from a fortnight ago, which had Scotland as the natural experiment and inner London as the hardest place in the country to sell a home. Two datasets, same map. Third, and this is the one I would put on the wall, Rightmove's own affordability sums have the monthly cost of a typical first-time buyer home at £1,199 on a 90% mortgage against £1,297 to rent the same sort of home. Even at 5.29%, buying the starter home is cheaper month to month than renting it. The problem for first-time buyers is the deposit and the affordability test, and the stamp duty, not just the monthly payment, and that is a problem for the rental market too, because it keeps would-be buyers renting for longer.
Moving on. Gilty, or not Gilty? Gilty. Again.
Friday's closes first. The five-year closed at 4.909%, more or less exactly where it started the week at 4.914%. But that flat week hides a rise and a fall: it dropped about 7 basis points on Friday alone, which on my arithmetic puts Thursday's close at 4.98%, and intra-day we went above 5% for the second time in a few weeks. It is 73 basis points higher than a year ago. The ten-year closed at 5.355%, up about 6 basis points on last week's 5.297%, and touched 5.423% intraday as the Governor was speaking on Friday. That is a fresh high for this cycle, above the 5.378% of 10th September, and near enough the highest in two decades. It is 60 basis points higher than a year ago. The thirty-year closed at 5.853%, up 10 basis points on last week's 5.756%, although still below the 5.994% intraday high from 10th September. It is 30 basis points up on the year. The two-year closed at 4.692%.
So the higher highs I mentioned at the top are real at the ten-year point and nearly real at the five. At the long end, the thirty-year gave back about two thirds of the 15 basis points it gained on the QT news, and the gap between the five-year and the thirty-year widened from about 84 basis points to about 95. That is the curve steepening again from the long end, a week after the Bank's announcement flattened it. I do not read that as the QT decision failing. I read it as the rest of the world's bond market reasserting itself, because the Americans added 22 basis points at ten years while we added 6. Relative to Treasuries, gilts had a good week. In absolute terms they all had another bad one.
Why? The same three culprits as the Americans had, plus one of our own. Oil stayed above $100 for most of the week. The Deputy Governors and then the Governor spent Thursday and Friday telling anybody who would listen that rates may have to rise, which we will go through properly in the Deep Dive. And the market now prices roughly an 80% chance of a quarter-point rise on 5th November, with slightly more than four quarter-point rises priced over the next twelve months. That would take the Bank Rate from 3.75% to somewhere around 4.75% to 5% by next autumn. Our own contribution was the public finances on Tuesday: £18.3 billion of borrowing in August, the highest August outside the pandemic, £8.1 billion above the OBR's forecast for the year so far, and £8.8 billion of debt interest, the highest August on record. More supply, at the long end, into a market that has just lost its biggest seller and is looking for its next buyer. Thirty-one days to the red box.
For the record, I said last week that I did not see a need for the Bank Rate to go above 4.5% in the next twelve months, and that November was a coin toss. The first half of that I am sticking with, with a caveat I will explain in the Deep Dive. The second half I have to concede has become long odds-on.
The swaps, which is what your lender actually prices off. The five-year SONIA swap was at 4.7% at Thursday's close, setting a new cycle high and 95 basis points higher than a year ago. The Bank's own Deputy Governor said on Thursday that quoted two-year fixed mortgage rates are now more than a percentage point higher than before the conflict, with Bank Rate not having moved at all, and Rightmove has the average two-year fix at 5.29%. My forecast from a fortnight ago stands: limited company five-year money at 6.5% within six months, as a coin toss with a lean. It might be here sooner than that, of course. Nothing this week moves me off it, and the swap back at its high moves me slightly towards it (and with an eye on putting 6.75% into the spreadsheets).
What does it mean in practice? If you have a refinance landing before the Budget, secure the rate rather than surf the curve, because the curve is moving on Sunday newspaper leaks and on Deputy Governors' travel plans. If you are on a tracker, you have been the winner of the last six months, because trackers are priced off 3.75% and fixes are priced off where the market thinks 3.75% is going. That gap - between the rate the Bank has set and the rate the market expects - is the thread that runs through everything that follows.
OK. Here endeth the lesson on current rates - here comes the Deep Dive.
Last week was QT, all of it, four ways, and I promised myself I would leave the subject alone for a month or two. The Bank of England has other ideas - Dave Ramsden is giving a speech on quantitative tightening tomorrow - and so did the gilt market, which made higher highs again in the very week the mini-budget turned four. So this is a hat tip to last week, and then four ways into one question. When borrowing costs move this far, this fast, who breaks, who shrinks, and who gets paid?
Four sources. The Lettuce-versary first: the PropenomAIx Deep Dive this week went back to the primary record on September 2022 to settle who carries the blame. Then Vistry, which published its half-year results and a CEO review on Thursday and announced that it is going to get a great deal smaller on purpose. Then the banks, and specifically Lloyds Banking Group's half-year numbers set against the Bank's own data on what savers are paid, because I wanted to test the claim I keep making in conversation that the banks are doing very nicely out of the gap between Bank Rate and the curve. And finally Clare Lombardelli's speech in Warsaw on Thursday, alongside the Governor and Sarah Breeden, as the Bank spent the week preparing us for a rate rise that the market had already priced four times over.
The thread, if you want it in one sentence, is the gap. In 2022 the gap between how fast yields moved and how fast the pension system could raise cash nearly broke the gilt market. In 2026 the gap between a Bank Rate of 3.75% and a curve sitting well above it at every maturity from two years out is being harvested by the banks, endured by the leveraged, and watched, carefully and slowly, by the Bank. Buckle up.
First up, the Lettuce-versary itself. What follows is the Supplement cut of the PropenomAIx piece, four themes, working from the Bank's own staff paper, the Growth Plan's costing tables, the Deputy Governor's letter to Parliament and two select committee reports, and landing on a verdict I did not entirely expect.
Theme 1: The Tinder, the Match and the Fire
The Summary: Defined benefit pension schemes use liability-driven investment to hedge liabilities valued off gilt yields, many of them with borrowed money through repo and swaps held inside LDI funds. When yields rise, the scheme's solvency improves but the fund must post collateral immediately. The Bank's staff working paper by Gabor Pinter, Emil Siriwardane and Danny Walker, Fire sales of safe assets (July 2024), uses trade-level data covering almost every gilt, repo and swap transaction. It finds that aggregate LDI gearing rose from about 1.9 times at the start of September 2022 to just under 2.4 times before the mini-budget, as gilt prices fell and funds took on about £17 billion of extra repo. LDI selling began immediately after the announcement on 23rd September, gearing peaked at around 2.8 times, and £25.3 billion of gilts were sold over five weeks. The combined LDI and pension sector held about £1.58 trillion of assets with a funding surplus of about £320 billion. Pooled funds sold roughly 13 percentage points more of their holdings than single-client funds. The thirty-year yield rose about 30 basis points on the Friday, 50 on the Monday and 67 on the Tuesday from the morning. The Bank announced purchases on Wednesday 28th September, the thirty-year fell from 4.991% to 3.975% by mid-afternoon, and the Bank bought £19.3 billion in total, stopped on 14th October, and sold the lot by January 2023 at a profit of £3.8 billion.
The Propenomix Perspective: Two numbers from that paragraph are the ones I keep in my head. Gearing went from 1.9 times to 2.8 times. And the sector had a surplus of £320 billion. The pension schemes were not broke. They could not get the cash to the right place on the right day, and in a margin call that is the only thing that matters. Liquidity kills you long before solvency does. I have watched exactly the same thing happen in our own world, on a much smaller scale: the landlord with 40% equity across the portfolio who still ends up selling in a hurry because a bridge came due before the refinance valuation came back. Can you go bust with plenty of equity? Yes, if the equity is in bricks and the demand is for cash by Thursday. And I will allow myself one callback to last week. The Bank bought when everyone else was selling, sold when the panic had passed, and made £3.8 billion. It can be done. It is the only gilt trade it has made a profit on this decade, and it was the one it was forced into.
Theme 2: The Charge Sheet for Downing Street
The Summary: The Growth Plan broke convention rather than law. As the rules stood in 2022 the legal duty was for the OBR to produce two forecasts a financial year, and the fiscal lock in the Budget Responsibility Act 2024 came later, in response to these events. Kwarteng declined to publish preparatory OBR analysis on his first day, promised a forecast before the end of the calendar year, set it for 23rd November and brought it forward to 31st October on 10th October. The Treasury's permanent secretary, Sir Tom Scholar, had been dismissed in early September, and the IMF urged the government on 27th September to re-evaluate the tax cuts. Table 4.2 of the Growth Plan costed the tax measures at £44.8 billion by 2026-27. The largest items cancelled rises not yet in force: keeping corporation tax at 19% rather than 25% (£18.7 billion) and reversing the NICs rise and Health and Social Care Levy (£18.2 billion gross, about £15.3 billion net). Abolishing the 45p rate was costed at about £2.1 billion a year. The energy package was expected to cost about £60 billion over six months. The Debt Management Office's financing requirement for 2022-23 rose from £161.7 billion to £234.1 billion, including additional gilt sales of £62.4 billion. Jeremy Hunt reversed about £32 billion a year of the measures on 17th October. The stamp duty changes survived and reverted on 1st April 2025.
The Propenomix Perspective: I went in believing Truss and Kwarteng carried more of the blame than the Bank, and I came out believing it more strongly. Was it economically illiterate? Parts of it were perfectly respectable. The supply-side case for not raising corporation tax is one I have made myself, and the energy cap actually lowered measured inflation. The illiteracy was in the combination and the timing: £45 billion a year of permanent, unfunded tax cuts, announced with no independent forecast, into CPI of 10.1%, with £62 billion of extra gilt issuance, the morning after the central bank said it would shrink its own gilt holdings by £80 billion. Any one of those on its own probably survives. All of them in the same 24 hours asked the market to fund the stimulus and the Bank's exit at the same time, with nobody vouching for the arithmetic. The thing I find hardest to forgive is the forecast. Refusing the one independent anchor the market was looking for was a choice, and they made it deliberately. I would note that the IMF said it in public within four days, which is quite fast for hindsight.
Theme 3: The Charge Sheet for Threadneedle Street
The Summary: On 22nd September 2022 the Monetary Policy Committee raised Bank Rate by 50 basis points to 2.25% on a three-way split and voted unanimously to reduce its gilt holdings by £80 billion over twelve months, with the first active sales scheduled for the week of 3rd October. In 2018 the Bank had flagged risks from leverage through derivatives in the non-bank financial system, and the Financial Policy Committee tested the largest pension derivative users against a 100 basis point rise in gilt yields; the Work and Pensions Committee heard that well-governed schemes were testing around 200 basis points. The thirty-year yield moved more than 100 basis points in four days. Andrew Bailey later told the Treasury Committee that by looking at the 85% the Bank had left the remaining 15%, the pooled funds, relatively obscure. The Pensions Regulator's then chief executive told MPs it had not been collecting systematic data. Oversight was split: the Pensions Regulator covered schemes, the FCA covered UK managers, most pooled LDI funds were domiciled in Ireland and Luxembourg, the PRA and FCA oversaw the lending banks, and the FPC held the system-wide mandate without direct powers. The Work and Pensions Committee concluded in June 2023 that more should have been done to follow up the risks the Bank identified in 2018.
The Propenomix Perspective: Negligence by omission, shared across four regulators, with the Bank holding the one mandate that was supposed to see the whole board. That is my verdict and I do not think it is a harsh one. The Bank did not miss the risk. It spotted it in 2018, tested it at half the size of what well-run schemes were testing themselves, and then left it sitting on a perimeter that nobody owned. Blindsided on the speed? Yes. Blindsided on the existence? No. I have some sympathy for the people involved, because a regulator with no power over a Dublin fund cannot make a Dublin fund do anything. But the Governor's own phrase about the 15% is the most honest sentence anyone in authority said about the whole episode, and it deserves to be remembered as well as the lettuce. The question it leaves me with is a present-tense one. Where is the leverage now that everybody assumes somebody else is watching? I do not think it is in LDI any more - the Bank has fixed that corner. I suspect a little of it is sitting in our own sector, in portfolios financed at 2.5% that refinance at 6.5%, and I will come back to that with Vistry.
Theme 4: My QT Prior, and What Four Years Changed
The Summary: Active quantitative tightening had not sold a single gilt when the market broke. The Bank postponed the start of sales from 3rd October to 1st November 2022 and began with short and medium gilts only, while the £80 billion target was left unchanged. The Treasury Committee's February 2024 report on QT said it had not received substantial evidence that QT was a causal factor in the LDI episode. The staff paper estimates that gilts in the ten-to-thirty-year bucket fell about 15% after the mini-budget, and puts the most conservative fire-sale discount from LDI selling at 6.87%, roughly half the move, rising to closer to 60% on its pooled-fund estimate, with the remainder attributed to fundamentals, most notably the fiscal changes. On 20th July 2026, Andy Burnham's first day as Prime Minister, the thirty-year gilt reached about 5.75%, above its September 2022 peak, without forced selling or an emergency operation. The FPC now requires LDI funds to withstand a yield shock of at least 250 basis points in addition to day-to-day buffers. This week the ten-year gilt traded as high as 5.40%.
The Propenomix Perspective: Regular readers will know I went into this hoping to find QT's fingerprints on the whole thing. They are not there, and I am not going to pretend otherwise. Active sales had not started, and the committee that looked at it found no causal link. My case against the way the Bank ran QT survives, but it has had to move house: it is a case about cost, which is what last week's edition was about, not a case about September 2022. So, the verdict. Both were in the wrong. Truss and Kwarteng more so, and not narrowly, because the fiscal shock accounts for roughly half the fall and the other half would not have happened without it. The Bank's sin was supervisory rather than monetary. And the lesson I actually take for property is the one in the last two sentences of that Summary. Long gilt yields today are higher than at the peak in 2022, and nothing broke. So the level of yields was never the thing that broke the system. Speed, hidden leverage and a credibility shock, all arriving in the same week, were. Keep that in mind for the next three sources, because the speed has slowed down but the leverage has not gone anywhere.
Second, Vistry. The half-year results and the CEO review landed on Thursday 24th September, and I read the review twice, because it is about as candid a document as a listed housebuilder has put out in a long while. Adam Daniels took over as chief executive in April, and his conclusion is that the business needs to be smaller, more focused and carrying a great deal less debt.
Theme 1: The Numbers Behind the Reset
The Summary: Vistry reported an adjusted loss before tax of £83.3 million for the six months to 30th June 2026, against a profit of £80.6 million a year earlier, reflecting discounting of open-market stock to generate cash and about £50 million of early impacts from the CEO review. The reported loss before tax was £661.3 million against a profit of £40.9 million, including a £475.0 million goodwill impairment and an additional £73.2 million building safety provision. Completions fell 8% to 6,304 and adjusted revenue fell 9% to £1,703.3 million. The adjusted operating margin was minus 2.1% against 6.7%, and return on capital employed was minus 2.7% against 9.6%. Net debt at 30th June was £468.8 million against £293.1 million a year earlier. The company said open-market conditions became more challenging over the summer, with its sales rate slowing to 0.3 reservations per outlet per week. It has redefined its forward order book to include only exchanged or otherwise legally contracted orders; on that basis the order book is £3.3 billion against £3.7 billion a year earlier, with the group 91% forward sold for 2026.
The Propenomix Perspective: Everyone will quote the £661 million, and it is a striking number, but most of it is accounting catching up with reality. The number I would pin to the fridge is 0.3 reservations per outlet per week. That is one sale every three weeks or so from a show home that costs a fortune to staff and heat. It lines up almost perfectly with everything else we have seen this week: Rightmove's buyers down 9% on last year, Chris's sales agreed at nine tenths of normal. And the line about discounting stock to generate cash is the new-build market's version of the price reduction, done quietly on the site rather than publicly on the portal. The detail I liked most is the order book. Vistry has stopped counting deals that were agreed but not exchanged, which is the difference between a salesman's number and a solicitor's number. I have been banging on about the gap between sold subject to contract and actually exchanged for years, and it is quietly reassuring to see a board decide the solicitor's number is the one that counts (unsurprisingly).
Theme 2: Smaller and More Focused
The Summary: The review repositions Vistry as a specialist mixed-tenure housebuilder targeting about 12,000 completions a year over the medium term, against a peak of around 17,000 in 2024 according to Construction Enquirer, with a tenure mix of about 60% partner-funded and 40% open market. The group will increase its exposure to the North, the Midlands and the West. South East operations will move to a fully pre-sold model to eliminate open-market exposure, although joint venture schemes will continue to sell on the open market until they complete. Operating regions will be consolidated from 25 to 12. The owned land bank will be reduced from about 51,000 plots to 36,000, with a target of three years owned and at least 18 months controlled. Overhead savings of £50 million a year have been identified from fewer regions, flatter structures and lower volumes, on top of £25 million already identified from a voluntary exit scheme and a recruitment freeze, with a combined cost in 2026 of about £40 million. The company expects items relating to the South East exit and changes in site strategy to have a full-year impact of around £470 million.
The Propenomix Perspective: Shrink to grow. Walk away from the region where affordability is worst and put the money where the numbers still work. How much is after the horse has bolted? Future years results will tell us, but at this point in time these trends don’t look set for changing anytime soon. Rightmove told us this week that 56% of homes in the South East find a buyer against 71% in the North West, and Chris's regional pack a fortnight ago told us the same thing with different numbers. Vistry has now reached the same conclusion, with a £661 million loss as its tuition fee. What strikes me is the ambition in the smallness. A housebuilder choosing to build 5,000 fewer homes a year than its peak, in the middle of a housing shortage everyone agrees exists, is not a good sign for the national target, and I will say that plainly. But as a business decision for a leveraged company in a market where money costs what it costs now, I think it is the right one. Too cynical to point out that "fully pre-sold" in the South East means the only buyers they trust there are housing associations spending grant money? Perhaps. I would add the usual caution. Turnarounds announced in September are judged in September two years later, and the plan depends on a land market that will buy 15,000 plots at sensible prices.
Theme 3: Leverage, Lenders and a Covenant Waiver
The Summary: Vistry is targeting average daily net debt of about £500 million in 2027, below £400 million in 2028 and about £300 million by 2029, stabilising thereafter, with capital employed falling to about £1.5 billion. Its five-year targets are a return on capital employed above 30% and a 12% operating margin by 2031. It now expects a broadly neutral cash position at the end of 2026, lower than previous guidance, which Construction Enquirer reported as more than £100 million of net cash, because of selective withdrawal from or renegotiation of partner deals and disappointing summer private sales. Its banking group has provided waivers for the interest cover covenants for the 2026 full year and the 2027 half year, which would otherwise have been affected by the charges taken in the review. The company does not anticipate any need to raise equity, will reconsider shareholder distributions once deleveraging has progressed, and expects to appoint a new chief financial officer imminently. Excluding the review items it expects adjusted profit before tax of about £165 million in 2026, and about £185 million in 2027 if market conditions remain broadly stable.
The Propenomix Perspective: Read that covenant waiver sentence twice, because it is the most useful thing in the whole release for anyone reading this with a mortgage book. The lenders looked at the interest cover test, knew it was going to fail, and agreed in advance not to count it. That is what a lender does for a borrower who came to them early, with a plan, before the test failed. It is not what a lender does for somebody who turns up the week after, and it is certainly not what happens if you let them find out from the accounts. The same is true at every scale. If your rental cover is going to fail at 6.5%, the time to ring your relationship manager is now, with a list of what you are going to sell and when. The other lesson is the phrase the review uses for the land bank: managed realisation of value. For a landlord that means selling the weakest yielders deliberately, in an orderly way, into a crowded market while you still have the choice, rather than having the choice made for you. I would rather be the vendor with a plan in a buyer's market than the vendor without one.
Theme 4: The Grant Cheque, and What the Rest of Us Might Copy
The Summary: The first wave of the Social and Affordable Housing Programme for 2026 to 2036 has confirmed funding of £9.58 billion to 33 Strategic Partners, 29 of which Vistry currently works with. Vistry was named a Strategic Partner and awarded a direct grant of £350 million, the largest in this phase, which it says will enable it to deliver more than 3,000 affordable homes directly and on which work has already started. The company expects its second half to benefit from a higher weighting of partner-funded and open-market volumes, higher margins from a richer site mix, land sales as it reshapes the land bank, lower overheads and the conclusion of transactions delayed from the first half. It describes the mixed-tenure model as benefiting from structural demand for affordable and mixed-tenure housing.
The Propenomix Perspective: So the one buyer Vistry trusts in the hardest region is the state, and the state has just written it the largest cheque in the programme. That tells you where the counter-cyclical money is in this market. It is not coming from first-time buyers at 5.29%. It is coming from grants, from housing associations and from institutional rented funds, and the builders who can plug into it will survive the next two years in better shape than the ones who cannot. Is shrinking and focusing the way other leveraged investors go? I suspect yes, and some of the smarter ones already have. Fewer assets, lower loan to value, a tighter patch where the rent-to-price ratio still works, and partner capital rather than more bank debt for anything new. That last one is the workshop's subject on Thursday, and I promise that is a genuine connection rather than a plug. I might be wrong about how many will choose to do it. I do not think I am wrong that the ones who do it by choice will do better than the ones who do it by instruction. Sharing equity is smarter the higher the cost of debt goes - especially if you are already exposed in any way that looks like Vistry were and still are.
Third, the banks. I say it often enough in conversation that I thought I had better test it in print: the banks are doing very nicely out of a Bank Rate that sits well below where the market thinks rates are going. The cleanest evidence I could find is Lloyds Banking Group's half-year results from 30th July, which are two months old but are the most detailed recent disclosure of how a big UK bank makes its money, set against the Bank of England's own effective interest rate data for July. Lloyds is the example here rather than the villain. The other big banks have described the same mechanism in their own results.
Theme 1: The Structural Hedge in One Number
The Summary: Banks hold large balances of current account and low-rate deposits that behave in a stable way, and many hedge that stability by swapping it into fixed rates for several years, rolling a portion each month. As older swaps taken out at low rates mature, they are replaced at prevailing rates, so income from the hedge rises when the curve is higher than it was a few years ago. Lloyds reported structural hedge earnings of £3.4 billion in the first half of 2026, on a notional balance of £246 billion, flat on the quarter. It said it continues to expect hedge income of more than £7 billion in 2026 and more than £8 billion in 2027, with growth to the end of the decade. A secondary summary of the results puts the average hedge yield in the half at 2.7%. The group's net interest income rose because of higher average interest-earning assets and a higher margin, which it attributed to stronger structural hedge income as eligible balances were reinvested into a higher rate environment.
The Propenomix Perspective: This is the gap I talk about, in one line of a results presentation. The hedge rolls at medium-term swap rates - strictly the three to five year part of the curve rather than the ten-year that dominates most debt conversations outside of the Supplement; but the principle is identical - those rates sit well above Bank Rate. The money being hedged is money on which the bank pays very little, because what it pays on current accounts is anchored to Bank Rate or below. So every month a slice of old swaps written at 1% or 2% rolls off and is replaced at somewhere near 4.5% or more, and the income ratchets up. More than £8 billion next year, from one bank. I want to be fair about it. The hedge is prudent, it was designed to protect earnings when rates fall, and it did exactly that for years when rates were on the floor. Nobody sat in a room and plotted this. But when the rate the Bank sets is held below the rate the market expects, the hedge becomes a very efficient machine for turning that difference into profit.
Theme 2: The Margin, and Where It Is Being Squeezed
The Summary: Lloyds reported underlying net interest income of £7.3 billion for the first half of 2026, up 9% on the year, with a banking net interest margin of 3.19%, up 15 basis points year on year and 5 basis points in the second quarter compared with the first. It attributed the increase to strong structural hedge income, volume growth and a 4% rise in average interest-earning banking assets to £475.7 billion, partially offset by asset margin compression, in particular in its UK mortgage portfolio. Statutory profit before tax was £4.3 billion and return on tangible equity 17.1%. The board announced an intention to buy back up to a further £1.0 billion of shares, in addition to the £1.75 billion programme announced with the 2025 results, and reiterated guidance for underlying net interest income of more than £14.9 billion in 2026.
The Propenomix Perspective: Notice the offset, because it is the bit that stops this being a cartoon. Margins on mortgages are being squeezed, not fattened, because lenders are competing hard for your business. The banks are not getting rich by overcharging you on the mortgage. They are getting rich on the other side of the balance sheet, on money that sits in current accounts earning next to nothing, recycled through the hedge. Can we blame them for that? Not really. It is a legal, disclosed and entirely sensible way to run a bank, and a 17% return on equity funds a lot of pension schemes that own the shares. My quarrel is not with the banks. It is with a monetary set-up in which the main lever the Bank controls - the rate on deposits - is out of step with the rate the market charges for everything longer than a few months, and the difference accrues to whoever holds the deposits. At the moment that is the banks, and they are telling their shareholders about it, as they should.
Theme 3: Where the Money Is Made
The Summary: The Bank of England's Money and Credit release for July 2026 reported an effective interest rate on newly drawn mortgages of 4.45%, up from 4.35% in June, and 3.97% on the outstanding stock of mortgages. The effective rate paid on households' new time deposits fell to 4.21% from 4.30%, while the rates on the outstanding stock of time and sight deposits were 3.39% and 1.65% respectively. Bank Rate has been 3.75% since December 2025. The Bank's July Monetary Policy Report noted that quoted rates on sight deposits had been little changed since April, consistent with Bank Rate being unchanged, and that mortgage and corporate credit spreads remained compressed by historical standards. BlueGamma's five-year SONIA swap rate closed at 4.79% on 24th September, and Rightmove's average two-year fixed mortgage rate was 5.29% in September.
The Propenomix Perspective: Put those numbers on one line and the arithmetic does itself. The average instant-access and current account is paying 1.65%. Bank Rate is 3.75%. The five-year swap, which is roughly what the hedge rolls into and what a five-year fix is priced from, is about 4.8%. So for every pound of stable sight deposit that goes into the hedge, the gross spread is somewhere around three percentage points. Rough working, and heavily caveated: not every pound is hedged, the swaps roll gradually rather than all at once, and there are costs, capital and bad debts to pay out of it. But three points on money that costs the bank almost nothing to hold is a very good business. Who pays? Mostly the saver who leaves money in a current account. The saver who fixes gets 4.21%, which is roughly fair value. And for our audience there is a simple practical point hiding in here. If you are sitting on cash for a deposit or a refurbishment in an instant-access account at the average rate, you are funding the bank's margin. Move it, or fix it for the months you do not need it. It is the easiest few hundred pounds you will make this year.
Theme 4: Who Pays, and Whether It Is Wrong
The Summary: Fixed-rate mortgages are priced off swap rates of the matching term plus a lender margin, while trackers and variable rates follow Bank Rate. An analysis by the data provider BlueGamma of Bank of England quoted rates against SONIA swaps puts the average margin of five-year fixes over five-year swaps at about 0.6 percentage points, with swap rises showing in advertised rates within about a month and falls taking two to three. Clare Lombardelli said on 24th September that quoted two-year fixed mortgage rates were more than a percentage point higher than before the Middle East conflict, while Bank Rate had not changed. A reading of the SONIA curve on 22nd September by the broker Mortgage One put Bank Rate at 4.39% in March 2027 and 4.80% in September 2027. Trading Economics reported on 25th September that markets priced slightly more than four quarter-point rises over the next year.
The Propenomix Perspective: So is it wrong? I do not think it is wrong. I think it is the system working exactly as designed when the Bank's rate and the market's rate have drifted apart, and the losers are the people on the wrong side of the drift: the borrower refinancing on the curve and the saver who never moves their money. The gap closes in one of two ways. Either the market is wrong and the curve comes down to meet Bank Rate, or the Bank moves up to meet the curve. Fancy fighting the market? How does that normally work out? Well, Mr Soros knows very well, and currently Mr Bessent is behaving as if he hasn’t been to the school of Soros and Druckenmiller (the latter of whom has been extremely critical of the US Government using the current account as a nice float for a bit of yield curve control). There has been talk of the Chancellor looking at higher taxes on the banks ahead of the Budget, and I understand the temptation, but a windfall tax treats the symptom (although it doesn’t change behaviour afterwards, and can appear to be a largely victimless crime - the bank’s shareholders are the ones that suffer of course). The cause is a Bank Rate set below where the market, the curve and now the Bank's own Deputy Governors think it is heading. For your own borrowing the choice is quite clean. A base rate tracker is priced off 3.75%. A fix is priced off something close to 4.8%. If you believe the Bank will move more slowly and less far than the market expects, the tracker wins. If you believe the market, fix. I have told you where I lean, and I have been wrong before - but base rate trackers will look like exceptional value at this time, if you can afford to ride the lightning on the upside (and please, do be careful and stress test at a Bank rate ABOVE the previous 5.25% cycle high).
Last up, the Bank itself, and a week in which it spent a good deal of time preparing us for a rate rise that the market had already priced in several times over. The centrepiece is Clare Lombardelli's speech, "The outlook for inflation", given at the Sixth Biennial Conference on Macroeconomic Policy in Warsaw on Thursday 24th September. Around it, Sarah Breeden spoke at the National Institute's UK Macro Policy Forum the same day, and the Governor at Oxford's Monetary Economics Conference on Friday.
Theme 1: Four Channels and a Dashboard
The Summary: Lombardelli set out the framework the Committee is using to assess the energy shock. It works through four channels. Direct effects, through fuel and utility prices, act within a year, have come through broadly as expected and are looked through. Indirect effects, through supply chains into other prices, act over up to two years and have so far been smaller or slower than expected. Second-round effects, through inflation expectations and wage bargaining, act over up to three years; it is too early to tell and the response would be to tighten. Demand effects, through lower real incomes, act over up to three years and have so far been smaller than expected. She said firms may be absorbing higher energy costs in their margins or may be more hedged than thought, but that there is a limit to that ability, and that the key judgement is whether emerging evidence suggests inflation may become persistent, a risk she described as cumulative.
The Propenomix Perspective: I like this, and I do not say that about central bank frameworks very often. It is a dashboard with four dials, and it tells you which dial the Committee is staring at. Direct, as expected. Indirect, slower than feared. Demand, stronger than feared. Second-round, unknown. The whole decision rests on the one dial that cannot be read yet, which is uncomfortable but at least honest. Where I would push back is on the indirect dial. The PMIs this week had firms reporting the fastest rise in their own selling prices since June, with fuel the most cited cost. If firms have been absorbing the shock in their margins, the survey suggests that absorption is coming to an end. I suspect the indirect dial moves before Christmas, and that it moves before the second-round one, which is the sequence the framework expects. I might be wrong about the timing. I would be surprised to be wrong about the order. I’m also wary of the “to the nearest year” sort of rhetoric - I’m hoping that’s just for the purposes of the speech, rather than drastically oversimplified BoE modelling - I’m sure it is. Well, sort of sure.
Theme 2: The Numbers in the Speech
The Summary: Lombardelli said Brent crude had risen 26% since the July Monetary Policy Report to around $98 as of the Tuesday before the speech, with gas prices up around 50% in recent weeks and refinery pressures widening crack spreads. Petrol rose from around 132p a litre to around 172p between February and September. The Ofgem cap rises almost 4% in October to £1,723 and is projected to exceed £2,000 in the first quarter of 2027. The Bank's latest near-term forecast has CPI rising from 3.1% to around 3.7% in the fourth quarter of 2026 and around 4.2% in the first quarter of 2027; without last year's Budget measures and the temporary VAT cut on electricity to March 2027, the first-quarter figure would be closer to 4.5%. Food inflation is at a two-year low of 1.3% and is expected to rise to a little under 3% by the end of the year and towards 4% in early 2027. The Bank's Agents report average pay settlements of 3.6% for 2026 against 4% in 2025, and she put wage growth of around 3.25% as consistent with the target in the long run.
The Propenomix Perspective: Two numbers jump out. The first is 4.2%, because that is the inflation rate the Bank expects to be running just as next year's pay rounds are settled, and that is the hawks' whole argument in one figure. The second is the gap between 4.2% and 4.5%. That 0.3 of a point is inflation the government is holding down with a temporary VAT cut on electricity that expires in March 2027. I raised exactly this on 6th September when I was picking a hole in Neil Woodford's inflation call: measured inflation is being managed by the Treasury, and the management has an end date. Now the Bank has put a number on it. When the cut comes off, that 0.3 comes back, in April, the month the uprating is set. Will the Chancellor extend it in the Budget? I would, if I were him, but it is borrowing to buy a lower print - and you can bet diamonds that Burnham wants him to. And the pay settlements tell the other half of the story: 3.6% this year, heading lower, which is why the doves are still doves. Still - there is room for this inflation to die before it starts, and the recent modelling in past years has been all wrong - Woodford’s case is that this time it is wrong on the downside (and it has been so far, in spite of what’s basically been a continued and escalating conflict, save for a few weeks) - and so I’m inclined to agree with him, and would bet the “unders” on the Bank’s inflation forecasts at this time.
Theme 3: The Week's Chorus
The Summary: Lombardelli concluded that policy is increasingly likely to need to tighten if elevated energy prices persist, absent clear evidence of disinflation or weaker activity, while stressing that the Committee should not respond mechanically to energy prices. She said markets had reached a broadly similar conclusion about the balance of risks. Reuters reported Sarah Breeden telling the NIESR forum that it was increasingly appropriate for Bank Rate to respond to rising inflation risks, that she had not been ready to vote for a rise in September, that the question for her now was whether a first move was needed, and that the Bank could not ignore market pricing, after she was asked about markets pricing around 100 basis points of rises over the next year. On Friday the Governor told the Oxford conference that it was going to get harder to maintain the current stance the longer energy prices stayed high, that pass-through had so far been quite subdued, and that policymakers could not wait for full evidence of second-round effects. Swati Dhingra said inflation expectations were not yet a source of concern. Trading Economics reported markets pricing roughly an 80% chance of a rise in November.
The Propenomix Perspective: Last week I counted four of the six members who voted to hold as leaning towards a rise. This week the Governor and both Deputy Governors for monetary policy and financial stability have said so out loud, in public, within 36 hours of each other. Add the three who already voted for 4% in September and the arithmetic for November is not hard to do. Central bankers do not usually trail a move this heavily unless they intend to make it, and the Governor's line about not waiting for full evidence is almost word for word the sentence in the September minutes. So it’s time to move off my coin toss. I now lean towards a quarter-point rise on 5th November, a week after the Budget, with the Budget itself the one thing that could change it - but I doubt it will. All roads point to a disinflationary budget to keep the cost of living down (and tax rises are disinflationary, please remember). But I’m not sure they can do enough to frontrun the November rise that’s priced in so aggressively now. Dhingra is the useful counterweight, with Taylor at her side - but not enough. The chorus has a tune now, and it is not the one the Chancellor would have chosen for the week before his conference speech.
Theme 4: Behind the Curve, Again?
The Summary: Lombardelli said the Bank is in a very different place from the tightening cycle of 2021 and 2022, when Bank Rate started at 0.1% and policy was highly accommodative; today Bank Rate is 3.75% after 150 basis points of cuts from its peak, and overall financial conditions are restrictive, with quoted two-year fixed mortgage rates more than a percentage point higher than before the conflict. She said the option-implied distribution of future rates has widened since July. The September minutes recorded the judgement that, given the lags with which second-round effects appear, it was not appropriate to wait too long for evidence before responding. In November 2021, after the Governor's remarks had led markets to price an increase, the Committee voted 7-2 to hold Bank Rate at 0.1%, as reported by Bloomberg at the time; CPI inflation subsequently peaked at 11.1% in October 2022.
The Propenomix Perspective: Is the Bank behind the curve again? Yes, but not in the way it was in 2021, and the difference matters. Then, the Bank sat at 0.1% while inflation took off, and was so late that it then had to do fourteen rises in a row. Now, the Bank sits at 3.75% while the market has already done a good part of the tightening for it - that was Alan Taylor's point last week, and Lombardelli's this week. The market prices four rises because it thinks the Bank will be late and then have to do more to catch up. My answer is the old one: a stitch in time. That is how I square last week's view that there is no need to go above 4.5% with this week's complaint that the Bank is too slow. A quarter point in September would have made 4.5% look like a ceiling. Every month of waiting makes 4.75% or 5% more plausible, because the market charges a premium for a central bank it thinks will be late. We saw that once before - having to go higher because we were too slow - and that is the same problem recurring, not a different one - albeit it manifests itself in a different way. And that premium is the gap that runs through this whole Deep Dive: it is what the banks are harvesting through their hedges, what Vistry and every leveraged landlord are paying on their refinances, and what, in 2022, moved too fast for the pension schemes to survive without help. The good news is that the speed is different this time.
The Week Ahead. A busy one, with the conference season in full cry and the quarter turning. The Labour Party conference is under way, with the Chancellor speaking on Monday and the Prime Minister later in the week. Listen for any hint of the Budget's shape, and for the spring election chatter that has crept into the Sunday papers. On Monday 28th, Dave Ramsden gives a speech on quantitative tightening, which after last week is the one I will be reading line by line. I can’t leave that car crash alone. On Wednesday 30th the final estimate of second-quarter GDP arrives, and it matters more than a final estimate usually does, because the revised national accounts feed the OBR's Budget forecast ahead of the Blue Book on 30th October. The Bank's money and credit figures for August and one of the monthly house price indices are due around the turn of the month, and the mortgage approvals line will tell us whether August's higher fixes did what Rightmove's buyer numbers suggest. The final manufacturing PMI lands on Thursday 1st October, with services on the 5th. On the other side of the Atlantic, the US inflation measure the Fed prefers, the PCE index, is due on Wednesday and the jobs report on Friday 2nd, and after a week of 5% Treasury yields both are market-moving. Off-calendar: oil and whatever becomes of the seven-day plan; the diesel ban, yes or no; any Monetary Policy Committee member with a microphone; and the Budget countdown, now at 31 days. Every Sunday paper is a market-moving event again until the red box.
As we get towards the end for this week - I can't wait for our next workshop at the Hyatt Regency in Manchester on Thursday 1st October, four days from now, with Rod Turner, powered by Roma Finance. JVs and M&A. If you read the Vistry section and thought "that is a business choosing to be smaller, owe less and bring in partners for the growth", you have already understood half of what we will be doing in that room. The other half is the other side of the trade: how to be the person with the patience and the partner when somebody else has to shrink. Last chance to book in at www.tinyurl.com/pbwoct26 - grab a dinner ticket if there’s still one left when you read this.
Above all - please remember to Keep Calm, ALWAYS listen to or read the Supplement, and Carry On. This is a week to do exactly that. Gilt yields made higher highs, the Bank lined up a rate rise, a housebuilder lost £661 million and a bank is heading for more than £8 billion of hedge income, and all of that sounds a lot more frightening than the market underneath it, which is transacting at nine tenths of a normal September with the reckoner sitting on its number. Four years ago the level of yields was not what broke things. Speed and hidden leverage were. The speed this time is slower, the Bank has fixed the corner that broke, and the leverage is the part each of us controls. The fundamentals have not changed: there remains a shortage of two and three bed terraces and semis, rents are still running ahead of pay, and supply is still leaving the sector the country depends on. Investing in property is not a guaranteed win, and the next year will reward the people who shrank on purpose, fixed on purpose or partnered on purpose over the people who waited to be told. Most of you reading this are in the first group. Have a great week!
KCCO!