"The tax which each individual is bound to pay ought to be certain, and not arbitrary. The time of payment, the manner of payment, the quantity to be paid, ought all to be clear and plain to the contributor, and to every other person." - Adam Smith, The Wealth of Nations (1776)
The quote pertains to the deep dive, as ever - and this week it pertains to rather more than that. It is Smith's second maxim of taxation, the one about certainty, and certainty has had a rough week. The Chairman of the Federal Reserve flew to Wyoming to explain that he will not be telling anyone what he intends to do next, and considers that a feature. Ofgem announced, to the penny and five weeks early, exactly what your gas will cost from October - which is the one piece of certainty on offer this week, and not the pleasant kind. HMRC counted every unincorporated landlord in the country and published what they earned and what it cost them. The IPPR proposed replacing two property taxes with one, at 0.65% a year, for ever. And the planners have rewritten the rulebook and called it, with a straight face, "a clear, rules-based planning system". Certain and not arbitrary. We shall see about that.
As the last bank holiday of the summer arrives and the Budget countdown clock ticks louder - 59 days to the red box at publication, for those keeping score - our next Property Business Workshop is live and tickets are selling. This is the really juicy one for taking big steps forward in your property businesses - subject matter JVs and M&A - joint ventures done properly, and buying and selling property businesses, portfolios and companies, not just houses. Book in on the next Property Business Workshop with myself and Rod Turner - Thursday 1st October - Manchester - https://tinyurl.com/pbwoct26. The VIP dinner seats are nearly gone, so if you want that extra time in with Rod and myself, now is the time! We are proud to announce that this workshop is powered by Roma Finance, who in my experience really do think differently about bridging and development lending (they don’t just say that, like some lenders do!), and they will also be represented on the day. Super Early Bird pricing, at more than 20% off, is coming to a close.
Welcome back to Trumpwatch. If last week was the week economic warfare got its name, this week was the week the markets decided they believed the word "economic" rather more than the word "warfare" - and then a man in Wyoming reminded them that inflation is a war too. Three things to take in order: Kevin Warsh's first Jackson Hole keynote as Fed Chair, which was the set-piece we trailed last Sunday and which delivered; the oil market, which spent the week quietly pricing the end of the Hormuz premium while the diplomats went nowhere; and the assorted domestic theatre that a US President generates in an election year the way a coal fire generates soot.
Jackson Hole, then. Friday morning, the official theme was "Financial Innovation: Implications for Payments and Policy", and I said last week that precisely nobody would be listening for the payments content. Nobody was. In roughly thirty minutes Warsh managed three things at once. First, he recommitted the Fed to its 2% target measured on PCE - killing off the suggestion, floated earlier in his tenure, that he might prefer a different yardstick. That will have been a relief. Second, he named short-term interest rates as the "predominant tool", parking the balance-sheet and AI musings that had confused everybody in July. That will have been another relief, although again, these are simply basic expectations thus far. Third - and this is the bit that moved markets - he said the quiet part out loud: "While this summer's readings were better than expected, they do not tell me that underlying trends have meaningfully improved." And then the line that will be quoted until the September meeting: "We must be confident that underlying inflation is moving to our objective, clearly and at sufficient speed. Otherwise, we have work to do." This is the one where he actually sounded like a central banker. A few swallows does not a summer make.
What he did NOT do is equally important, and entirely deliberate. No forward guidance. No reaction function - no "if X then Y" for the traders to code into their models. His framing: "We should not indulge a regime in which market participants are looking primarily to the Fed for their next trade." Warsh wants the pre-2008 Fed back - fewer meetings, fewer words, more mystery - and a market raised on fifteen years of hand-holding is relearning how to read an institution that prefers not to be read. I said as much last week; this week he said it himself. Paul Krugman's verdict was that Warsh "did not sound at all like somebody who was going to do something different" - stuck with PCE, stuck with the framework, produced numbers that suggest he does not think this inflation is transitory. Standard Fare, in other words, delivered by a man who has spent three months insisting he is anything but.
The market's arithmetic moved accordingly. Futures pricing for a quarter-point hike at the 15th-16th September meeting went from roughly 39% a week ago to a coin flip, and around 56% after the speech, per CME's tracker. Treasury yields rose, the dollar firmed, gold and silver gave back 3-4% of last week's gains, and US stocks pared their advance. This all landed on top of a PCE print earlier in the week showing the Fed's preferred inflation gauge at 3.7% over the year, with July payrolls having already shown the economy shedding 23,000 jobs and a further downward revision of 79,000 to earlier months landing this week. So the Fed Chair is hawkish into a softening labour market with a war premium in the energy complex. If you sit on the FOMC, which of those do you set policy on? The three July dissenters - Hammack, Kashkari, Logan - will say inflation, and they now have a Chair who sounds like them. The doves will say you do not hike into a jobs market losing bodies. I suspect September is at least as close as the pricing suggests either way, and I would not put real money on it, which is precisely the state of affairs Warsh wants. We can say the same as we say about the Bank of England right now though - the next move is upwards, we just aren’t sure exactly when that is - and in both jurisdictions, there’s reasons to hold off (although the end of the middle east conflict gets less and less likely as things go on, just as all conflicts do).
One thread from last week that deserves following, because it is the one I flagged as "might well be time to revisit". The 30-year Treasury hit a 19-year high the week before last, and fell back briefly only after Treasury Secretary Bessent announced plans to double the maximum size of the government's long-dated debt buybacks. Buybacks are not yield curve control - the Treasury buying its own long paper with the proceeds of short paper is a maturity swap, not a printing press - but they rhyme with it closely enough that the bond market noticed, and so should you. They certainly feel like a YCC warmup, at the very least. I’ve said myself in the past this is exactly what I’d be doing in the UK, which means (I believe) this is certainly active interference beyond the current remit of the central bank (I believe we are “too independent” for this to happen above the line in the UK vis a vis the US system). I will write the yield-curve-control piece properly again in the coming weeks, because the shape of that conversation matters more for UK long-end gilts than anything the MPC says.
Elsewhere in the theatre: the President has reinstated his attempt to remove Governor Lisa Cook from the Fed Board, on grounds that CNBC's analysis politely described as weak evidence. Sigh.
Which brings us to the war, and the oil price, and a week in which the two parted company. Earlier in the week the US Treasury announced what it called the "toughest sanctions in history" on Iran, explicitly aimed at forcing the Strait of Hormuz open; Tehran called them "an inhumane and hostile act" that had "lost their effectiveness", which is roughly what you would say if they hadn't. Washington confirmed on Thursday that it is not talking directly to Tehran, whatever the Omanis and others are trying to broker, and there was reporting that the President has no interest in returning to the previous deal terms. On the water, the Iranian military says it has a revenue-sharing arrangement with Oman over the strait - a corridor, a toll, call it what you like - while stressing that this does not mean an immediate reopening; vessel-tracking still has Hormuz traffic at a fraction of pre-war levels.
And yet Brent fell more than 5% on the week, to around $88 as I go to print from the $94 it was pushing at last Sunday's print, snapping a two-week winning streak. Why? Because the market has decided, for now, that "economic warfare" is the cheaper kind. Goldman Sachs put recent Gulf oil exports at 15-16 million barrels a day - that is 7-8 million below pre-war levels, but 5-6 million above the March low when the strait effectively shut - and ING's note captured the mood: producers are adapting to the new realities and becoming increasingly comfortable navigating the strait. Translation: the tankers have worked out how to get through, the sanctions target Iran's revenue rather than anyone's supply, and the risk premium is being priced down. Add Russia to the mix - Putin said talks with Ukraine had yielded nothing and that Russia was preparing to intensify, while Ukrainian strikes keep hitting Russian refineries and ports - and you have a market that spent Thursday briefly worrying about Eastern Europe instead of the Gulf, which tells you how used to the Gulf it has become. Remember the shape of the year: $60-ish in January, $120-ish at the dated-Brent panic peak in March, low $70s on the summer's peace hopes, $94 on economic warfare, $89 as the warfare turned out to be mostly paperwork. I would not bet the mortgage on the direction from here, and I would gently remind everyone that a ceasefire that keeps collapsing has a mirror image, which is an escalation that keeps not happening. The market is currently pricing the second. It has been wrong on both before.
Why does all this matter to your portfolio in Derby or Darlington rather than a trading desk in Manhattan? Because this week the two transatlantic channels pulled in opposite directions, and the tug of war landed directly in the gilt market. Warsh's hawkishness pushed the 10-year gilt yield up 11 basis points on Friday alone to 5.15%, its highest since May - the high-beta passenger on the Treasury bus, again. But the oil sell-off pushed the other way on the Bank of England: LSEG pricing now has less than 4 basis points of tightening in for the 17th September meeting - call it a 15% chance of a hike, down sharply on the month - with around 24 basis points by December and 36 by February, and the "next move is up" expectation pushed from late 2026 into 2027. So the US long end says higher, the oil price says later, and your five-year swap sits in the middle trying to hold both thoughts at once. A really basic analysis sees inflation under 3% in the UK and well over it in the US, and a US market at a coin flip as to whether they will hike rates or not - whilst there are a million nuances beyond that, if they are at a coinflip we are comfortable odds against simply by being a lot closer to the target (let alone having an administration that spells out that they want to help with the cost of living and is actively fiscally intervening, versus an administration that says paying more at the pumps is a worthwhile price to pay to ensure Iran doesn’t have nuclear weapons). A quiet fortnight in the Gulf plus a hawkish Fed is, perhaps counter-intuitively, not the worst combination for UK mortgage pricing - the front end relaxes even as the long end frets. Whether either lasts past the next headline is anyone's guess. More in the gilts section.
Phew - stepping away from the transatlantic macro-storm, let's get back to the reassuringly measurable reality of the real-time UK property market.
As is customary, Chris Watkin has been relentlessly crunching the portal numbers and then publishing them at Property Industry Eye. His analysis for Week 33 of 2026 - the week ending 23rd August - is where it is at, as always. If you want to know how the macroeconomic gridlock translates to the local high street, and the REAL property market on the ground, look no further. Chris's headline this week is a good one and I am going to borrow it: the UK property market is not broken, but something is definitely blocking it. His diagnosis, which I share, is that the blockage is not demand and it is not stock. It is the price - or more precisely, the number of homes that have entered the market at the wrong one.
Supply first. 31.8k new listings this week, down from 32.6k last week, against a 2026 weekly average of 36.3k and a ten-year average for week 33 of around 33k. So the tap has slowed to a touch below the long-run norm for the last full week of August, which is what you would expect in the last week of the school holidays - but for a few weeks in a row, the bath has been filling ever so slightly slower than it has historically. Fractions of a percentage point though, at the moment. The year-to-date picture is where the story sits: 1.198m new listings so far in 2026, which is 0.4% BELOW 2025's pace, 2.9% ahead of 2024, and 10.2% above the 2017-19 pre-Covid average. Regular readers will remember my year-plus adage of "10% more stock than a normal market" as the ready reckoner - after drifting above 12% in the early summer it settled back to the reckoner level last week, and this week it has nudged a fraction below it. Two readings does not imply mean reversion much more than one did, but the direction has now held for the best part of two months. The bathtub is still fuller than normal; the tap is now running slower than normal; and the second of those is new.
Demand, then. 23.3k homes went sold subject to contract in Week 33, down from 24.4k the week before, against a ten-year week-33 average of 25.1k and a 2026 weekly average of 24.5k. After last week's bounce-back, this is a step back towards the wobble, and I will say what I said a fortnight ago: reserve judgement. One week does not make a trend in either direction, least of all in late August. What I would note is that the year-to-date gross figure has now landed EXACTLY on the decade average: 809k homes sold subject to contract so far in 2026, and 809k is the ten-year average for this point in the year. That is 7.2% below 2025's pace and 0.9% below 2024, but 10.9% ahead of the limp lettuce that was 2023 and 5.8% above the pre-Covid 2017-19 years. Should we then accept that 2026 is a normal year for transactions following an exceptional 2025, which borrowed a chunk of activity from this year ahead of the stamp duty reset? No. Look more closely, as we do every week, and a more sensible conclusion is that 2026 started well with rates on the way down and confidence returning - then we had political upheaval of our own with a change of PM, a world cup and a war which moved our interest rates by 1% or so - and the past few months have been tricky, looking more like an early 2024 or a late 2023 than anything else. Functional - just about - not frothy. The Five Ds - death, debt, divorce, downsizing and the diddy ones - are doing what they always do, but let’s see what keeps happening to the supply of new stock because we could do with more water escaping the bath via the overflow.
Net sales: 17.7k for the week (down from 18.1k), against a ten-year week-33 average of 19.1k and a 2026 average of 19k, taking the year-to-date to 627k. That is 5.8% below 2025, 0.5% below 2024, 13.7% ahead of 2023 and 4.1% above the pre-Covid average. Note the gap between the gross and net comparisons to 2025: gross is 7.2% down, net only 5.8% down - because fewer of this year's agreed sales are falling over. More on that in a moment.
Now the friction, which is where this market continues to earn its description as brutally price sensitive. In July, 79.5k homes exchanged and completed while 75.6k withdrew from agents' books unsold - both figures will move as late reporting comes through, and both have already moved a fraction since last week's first cut - but on the current numbers only 51.3% of homes leaving the books in July actually sold. Call it half, against a seven-year average of 57.6% (I do worry about how much 2021 and 2022 affect this figure, but then the 7 years has seen enough sluggish markets versus booming ones I suppose). Half the homes leaving estate agents' books are leaving without a sale. I keep quoting this stat weekly, because it is the single best corrective to the "market is fine" and "market is dead" headlines alike: the market is fine FOR CORRECTLY PRICED STOCK and dead for the rest. Chris's show this week digs into how dramatically a home's chances change the longer it sits unsold; the honest summary is that the first fortnight is the whole game. Price for it, or plan to own the house for another year. It looks very similar to the war analysis for both Ukraine and Iran - the longer it goes on, the longer it is likely to go on - and the more unrealistic the demands, the less likely there’s a resolution.
The pricing evidence this week is, if anything, more pointed than last week's. 23.3k price reductions this week on 767k homes for sale. 13.7% of the entire stock was reduced during July, down a touch from 14.3% in June, with the 2026 year-to-date average at 13% against a six-year long-term average of 11.2%. The gap between the average asking price of all listings (£385k) and the average asking price of the homes actually going sale agreed (£358k) has narrowed again on Chris's current measure to 7.8%, from 9.9% last week, against a longer-term average he puts at 16-17%. I said last week this was a typical August anomaly that would blow out the other way when September's first figures land, and nothing this week changes that view - if anything the anomaly has deepened, which just means the reversal will look more dramatic. Do not conclude too much from a late-August gap reading. Do conclude something from the fact that one home in seven on the market had its price cut in July.
Here is this week's back-of-the-envelope, and I will hedge it as rough working as ever. The sell-through rate - the share of homes on agents' books going sale agreed in a month - was 14.2% in July, up from 13.8% in June, against a pre-Covid average of 15.5%. Compounded over a typical twelve-week sole agency, that is still roughly a 37% chance of going under offer inside the contract, as it was last week. Now stack the flows. In a month, 14.2% of a 767k book is around 109k homes going under offer; 13.7% is around 105k homes being reduced; and around 75k are being withdrawn unsold. So in round numbers, for every home that goes under offer in a given month, another one has its price cut and roughly two-thirds of another one gives up entirely. Meanwhile the pipeline of agreed sales stood at 487k on 1st August against 508k a year earlier, on a total stock of 767k against 763k - the pipeline is 63% of the stock, against 67% twelve months ago. Same size bath, thinner outflow. If you are buying, that ratio is your negotiating context: the agent's book is fatter than their pipeline and they know it. If you are selling, the reductions column is your pricing memo, and the withdrawals column is what happens to the people who ignore it.
The quietly encouraging line, promised above: the fall-through rate came in at 23.9% against a decade average of 24.5% - so, for once, BELOW the norm, having been slightly elevated at 25.7% last week - and only 5.32% of homes sold subject to contract in July actually fell through, against a 2025 average of 5.3% and a ten-year average of 5.8%. The deals that are being agreed are, on the whole, sticking. Fewer offers, but better offers - which is, I would suggest, exactly what you would expect when the retail buyer has been squeezed out and the remaining bids come from people who have done their sums. Stock 767k on 1st August against 760k a month earlier and 763k a year ago: flat, and flat is the new normal. The machine is processing what it is fed, just slightly less of it, and pricing it slightly better.
On pricing itself, July's agreed sales averaged £345.41 per square foot - 1.2% ahead of twelve months ago and 11.9% ahead of five years ago, and off June's record £350.22. Sub-inflation nominal growth, which is to say falling real prices, which is to say improving affordability - with CPI at 2.9% the real-terms move is about minus 1.7% on the year. Chris's £-per-square-foot at sale agreed matches the Land Registry index with 98% correlation, five months in advance - when the official December number lands next spring, you read it here in August. Zoopla's index, covered in the Macroscope below, sings from the same sheet.
The rental side: £1,792 per month was the average asking rent in week 33, with August 2026 averaging £1,795 against £1,800 in August 2025 - so essentially flat year on year at the national level on Chris's portal measure, and remember this is asking rents on new listings, the most real-time cut there is. Against that, £1,394 in August 2021 tells you what the last five years did. Availability continues its quiet improvement: 323k rental properties available in August against 319k a year ago, with 135,928 new rental listings in the month - up from 128,821 a year earlier and massively up from the 111,080 of 2022. The stock-shortage era is easing at the margin, rents have stopped rising nationally on the new-listing measure, and the ONS's index of all tenancies, renewals included, still shows 3.7% growth in the year to July because it is catching up with rises already banked. Both things are true. Affordability ceiling? Or, supply constraints easing? A bit of both. What landlords are actually declaring to HMRC about all of this is the first stop in the Deep Dive, and the number is bigger than you think.
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! This week Chris is joined by Steph Vass from TAUK to go through the numbers and then run the data ruler over Bath's estate agents - market share is one thing, but which agents are actually getting their sellers moved, and how big is the gap between them? Some of the numbers, as Chris puts it, are difficult to ignore. Watch the Week 33 show here: https://youtu.be/5RebgIlD9Jc.
Dust off the Macroscope, then - a thinner week than most, as the calendar always is before the August bank holiday, and I will not pretend otherwise. What we do have is important: Ofgem's price cap for the fourth quarter, which is the war arriving at your meter for the second time. Then the housing prints - Zoopla's August index and HMRC's July transactions, which between them tell you what the sales market did in the summer and what it might do in the autumn. Then the consumer, via the CBI's distributive trades survey, which had a shocker. The PMIs, my darlings of the real time economy, return with their final August readings next week. Bringing up the rear the contract states that we have to talk about the gilts and swaps, and will do so for many years to come, I'm sure.
OK. The cap. On Wednesday Ofgem announced a 4% increase in the energy price cap for the period 1st October to 31st December - a rise of £60 a year to £1,723 for the typical dual-fuel household paying by direct debit, from £1,663 today. A word on the arithmetic before anyone compares that £1,723 to the £1,862 I quoted last week and concludes bills are falling. They are not. Ofgem changed its "typical household" consumption assumption on 1st July - down from 2,700 kWh of electricity and 11,500 kWh of gas to 2,500 and 9,500 - so the same unit rates now produce a smaller headline number. On the OLD basis, the like-for-like comparison with every quarter before July, the new cap is around £1,935 against £1,862: up £73, or 3.9%. Same rates, two labels; the village watchman strikes again. Read closely what that implies though - people are using less energy. Is that pricing? Is that a victory for the heat pump? You’d think the latter would mean less gas and MORE electricity. Is it just average household numbers falling? A question for another day, as it needs closer scrutiny and deeper research.
Underneath the headline the rise is lopsided in a way that matters for the CPI. Gas is where the damage is - the unit rate goes to around 7.97p per kWh, and gas costs for cap customers will run close to 27% above the same period last year on Uswitch's read of the tables, because the assessment window caught June's oil spike and wholesale gas has recently been at a three-year high. Electricity, by contrast, barely moves - less than 1% for homes with no gas at all - and the reason is a policy one: the government has removed VAT from electricity bills from 1st October to 31st March. That is the Burnham cost-of-living measure we discussed in July when the gilt market first reacted to it, and it now has a number attached: 5% off the electricity component of every capped bill for six months, funded from general taxation. Around 35% of households - some 11 million - are on fixed tariffs and will not see the rise; the 22 million on default tariffs will.
What does this mean for the inflation path? Two things, pulling against each other, which is rather the theme of the year. First, the direct effect: a 4% cap rise is a much smaller step than July's 13%, so the household energy contribution to the annual rate rises again in October but by less - and the Bank's August projection of a 3.2% peak in the fourth quarter almost certainly had a rise of roughly this size baked in. The question is whether it had the VAT removal baked in as well, because that is a mechanical push DOWN on the measured electricity price for six months, which flatters the headline through the winter and then unflatters it in April when the VAT goes back on (April, the month when inflation kicks in when it comes to public sector numbers and welfare payments - so look out there for April 2027’s inflation number, you heard it here first!). I said last week that Burnham's fixation on the cost of living would show up in the inflation numbers as policy rather than as economics, and here is exhibit one. Second, the forward effect: the next cap announcement is 25th November, covering January to March, and the ECIU's read of the wholesale curve is that further rises are expected in January - that assessment window is being filled right now, at $89 Brent rather than $94, which helps, but with European gas firm on LNG supply risk. So the war's third wave is being priced as we speak, and it lands in your January bill and the February CPI - the print the MPC will be looking at when it decides whether 2027 is the year the next move is up. I remain unconvinced of any orderly glide back to target while that is the shape of the winter, and equally unconvinced that the Bank hikes into a labour market losing bodies. Treacle, still. We'll see what the November window looks like.
Next up, the housing prints - and there are two of them, which is a luxury in a week like this. Zoopla's August house price index landed on Friday with UK house price inflation at 0.9% in the twelve months to July, down from 1.3% in June, and an average price of £272,800. Flat to falling across most of southern England: minus 0.3% in the South East, minus 1.0% in London, and Bournemouth propping up the city table at minus 2.2%. The other half of the country is a different market. Prices are 1.7% higher in Yorkshire and Humberside, 3.1% higher in the North West, 5.4% higher in Northern Ireland, and the city table is a roll-call of the north and the Celtic nations: Belfast plus 4.4%, Liverpool plus 4.2%, Newcastle plus 3.0%, Glasgow plus 2.6%, Manchester plus 2.4%. Nominal-flat nationally, real-terms falling nationally, and a North-South divergence that is now the structural story rather than a footnote.
The bit of Zoopla's release that deserves the most attention, though, is the demand mechanism. Richard Donnell's team reports that home searches in the four weeks to 16th August were 7% higher than a year ago - the strongest for twelve months, and up on the year in every region for the first time since last August, led by the South East (plus 8.9%) and the East of England (plus 8.5%). But sales agreed are still 6% lower than a year ago, and the stock of homes for sale is 5% higher. So buyers are looking again, in numbers, with plenty to look at, and not yet buying at last year's rate. Why? Zoopla's answer is arithmetic I rather like: average five-year fixed rates have gone from below 4% in January to around 4.8% now, and a buyer who could afford a £200,000 mortgage in January can borrow around £182,000 for the same monthly payment today - a 9% cut in buying power - or must find an extra £18,200 of deposit to buy the same home. In London that extra deposit is nearer £35,500; in the North East, £10,200. I have checked their sums on the back of an envelope and they hold, roughly. That, in one paragraph, is why the searches are up and the sales are not: the buyer is back, but nine per cent poorer, and the seller has not yet noticed. Donnell's own line: buyers are returning but have plenty of choice, so sellers will need to price carefully this autumn. Chris's blockage, described from the other end of the pipe. Demand will be building in the background for as and when rates DO come back a little though - that’s assuming they will, of course, whereas you have a world very focused on national debts for the next 5 minutes or so because of the symbolism of the uncontrollable US National Debt clocking above the big 4-0; $40 TRILLION was surpassed this week and that triggered 1000+ podcasts about it, of course. Then they will do the only thing they can do - keep calm, and carry on, of course.
The second print is HMRC's July transactions, out on Friday: 96,710 residential transactions on a seasonally adjusted basis, 1% lower than July 2025 and 2% below June's 98,390. Unadjusted it was 106,620, 5% up on the year and 3% up on the month. These are completions, two to four months behind the offer, so July's number is April and May's sale-agreed data turning into keys - and it says the spring was fine. The financial-year-to-date comparison, April to July, is the most interesting line in the release: 393,800 seasonally adjusted, against 342,900 in the same four months of 2025 and 368,100 in 2024 - the strongest April-to-July since 2022. The 2025 base was hollowed out by the April stamp duty cliff, so the comparison flatters, but this is not a weak market by any historical measure. It is a market doing decade-average volume, as Chris's 809k-on-809k number said above, at prices that are drifting down in real terms. Not broken. Blocked. Next HMRC print on 30th September. How we’ve got here is more volatile than the average year, I’d say - although perhaps just as volatile as we should be expecting in today’s world and political climate?
The consumer, then, and I am afraid the CBI's distributive trades survey for August was a shocker: the retail sales balance dropped to minus 48 from minus 26 in July, against a consensus of around minus 35, the fastest annual fall in over a year, with sales for the time of year at minus 26 from minus 18. Wholesale fell faster too (minus 23 from minus 9); only the motor traders had a decent month (plus 9 from minus 27). Two things to hold against that. First, the heatwave: an August in which the whole country was outside is not one in which it was shopping, and the CBI themselves flag the weather. Second, the forward-looking lines are less bleak than the headline: September's expected balance of minus 22 would be the strongest expectation since March, and investment intentions rose to minus 16 from minus 52 in May, their best since early 2024. Martin Sartorius, the CBI's lead economist, used the release for the usual pre-Budget pitch - business rates reform and a cut to employer NICs - which tells you where the lobbying is going. Set all of it against the GfK consumer confidence reading of minus 14 a week earlier - a two-year high - and you have the same picture the retail sales data gave us last week: a consumer who FEELS better than they are spending. Confidence surveys measure mood; the CBI measures tills. When the two disagree for more than a month or two, the tills usually win, although I would love to be wrong going into the pre-Christmas run.
Gilty, or not Gilty? The court reconvenes, and this week the jury spent Friday afternoon listening to a speech from Wyoming. The 5-year gilt closed the week at 4.638% from 4.595% last Friday, up roughly 4 basis points, with almost all of that move coming on Friday after Warsh. The 30-year closed at around 5.80% against 5.81% last week - essentially unchanged, and the one part of the curve that did not flinch. Twelve months ago the 5-year stood at a little over 4.1%, so we remain 50-odd basis points on the wrong side of where we were. The 10-year was the mover: 5.15% at Friday's close, up 11 basis points on the day and the highest since May, a fourteen-week high. So the week's shape was heavy in the belly and the ten-year, flat at the long end: the pattern you get when the market reprices the policy path rather than the long-run inflation premium. The Treasury bus, once again: we did not vote for the driver, we cannot reach the pedals, and this week the driver told us he will not be indicating before he turns.
Swaps: the 5-year SONIA at 4.3% against 4.32% last week, and the 3-year at 4.23% against 4.25%, with the twelve-month comparison still 50 basis points adrift of the 3.7s/3.8s we saw last August. The mortgage market evidence continues to point one way: Zoopla's own read, above, has average five-year fixes at around 4.8% for a 75% loan-to-value borrower at the big banks, Rightmove's tracker has the all-in average back above 5%, and the sub-5% five-year money of the early summer has not returned. Bank Rate, for the avoidance of all doubt, remains at 3.75% - the 4.00% you keep hearing is the level three MPC members voted for in July and the level the curve has at times priced a year out (currently pricing 4.25%); it is not the current rate, and nothing this week changed it.
What DID change this week, and it is the most important thing in this section, is the timing. The oil sell-off has pushed the market's expectation of the next Bank of England move - which remains a hike, not a cut - from late 2026 into 2027. LSEG's pricing on Thursday had less than 4 basis points in for the 17th September meeting, roughly a 15% probability of a hike, with about 24 basis points by December and 36 by February. The Bank's own yield-curve data has the one-year-ahead rate at 4.37% on 19th August against 4.36% a month earlier: the curve still expects two (and a half!) hikes in the next twelve months, just not before Christmas. Sterling fell against a firmer dollar to around $1.35, which is what happens when the Fed sounds like it might hike and the Bank sounds like it might not. Put the two halves of this section together and you get my current read of the 17th September meeting, unchanged from last week but with a little more conviction: hold, with a 6-3 or 5-4 split, and the QT envelope decision the same day doing more for long-end gilts than the vote itself. I would put a little real money on 6-3 if it was being offered at odds-against, but it is best kept at 50p or so - such is the strength of my conviction at this point! What this means in practice, as ever: if you have completions between now and the Budget, the case for securing rates rather than surfing the curve is what it was last week. The front end has relaxed a little; the long end has not; and Warsh has told you, in terms, that he will not warn you before the next move. Boring advice, again. It keeps ageing well.
OK. Here endeth the lesson on current rates - here comes the Deep Dive. The through-line this week, trailed at the top, is certainty: Adam Smith's second maxim, and what it costs when a system lacks it. We have HMRC counting 2.88 million landlords and what they earn, the most complete ledger of the sector that exists, quietly missing everyone who does it through a company. We have the IPPR, under the by-line of Ben Ansell, proposing to tear up stamp duty and council tax and replace them with one annual charge, and diagnosing why nobody ever does. We have the new National Planning Policy Framework, which promises rules over discretion and a "default yes" near railway stations. And we have the IFS asking, in the plainest terms, whether rent controls work - with a counter-report from Autonomy landing the same week saying they can. Four documents, one question: does anyone in this country know what they will be paying, and to whom, and for how long? Let's get into it.
First up, the ledger. HMRC published its Property Rental Income Statistics for 2026 on Friday morning - the annual count of what unincorporated landlords declared on their Self Assessment returns, this year covering the 2024-25 tax year, with five years of history behind it. The least glamorous document in this Supplement, and the most useful for anyone who wants to know what the sector actually earns rather than what people say about it. One word on method, because it matters this week: HMRC restates the previous year in each summer's release as late returns arrive, so I will say which basis I am using every time I compare two years. Four themes.
Theme 1: The Headcount
The Summary: In 2024-25, 2.88 million unincorporated landlords declared income from renting property. Of these, 2.85 million - 99% - were individuals declaring through Self Assessment, and 0.03 million were partnerships. Individuals declared £49.81 billion of property income and partnerships £9.18 billion. Last year's release put the 2023-24 headcount at 2.83 million individuals and 2.86 million in total, so on the published-to-published comparison both counts rose by around 20,000; the 2026 release itself gives no year-on-year headcount change, only the five-year rise from 2.81 million in 2020-21. The statistics cover Self Assessment returns only: incorporated landlords file Corporation Tax returns and are excluded, as are overseas entities and anyone below the reporting threshold. Geographically, 17% of landlords were based in London and accounted for 28% of income; London and the South East together housed 33% of landlords and produced 44% of the income; the North East was the smallest English region at 2%. The split is by the landlord's registered address, which HMRC notes may not be where the property is.
The Propenomix Perspective: 2.88 million is a number I want everyone to sit with, because it answers a question the debate keeps getting wrong. The narrative says the landlord is a rare and rapacious creature; the ledger says roughly one adult in twenty declares rental income, and 99% of them are individuals rather than anything you would recognise as a business. Now the headcount. On the figures HMRC published a year ago, individuals are UP 20,000, which is the opposite of the exodus every intentions survey has promised since 2016, and I have been saying since 2016 that the surveys measure mood rather than completions. Two things I see in my own conversations. Landlords are coming out of the woodwork - I still speak to people who lower their voice when they mention the one flat they have never quite got round to declaring, usually because it makes them no money - and HMRC's compliance machinery finds a few thousand more every year. And landlords with several properties are trimming to one or two rather than to zero, which keeps them in the count while shrinking the stock. The company landlord, meanwhile, is invisible here entirely, and anyone who has watched the incorporation trend since Section 24 knows that is where the professional end has been migrating, so the individual count can rise while the serious operators quietly leave it. The regional split is the market in miniature: the South has the value, the North has the returns. I have built a business on the second half of that sentence.
Theme 2: The Plateau, or the Revision
The Summary: Total property income declared by individuals and partnerships was £58.99 billion in 2024-25. Against the 2023-24 figures published a year ago - £47.62 billion from individuals, £7.90 billion from partnerships, £55.53 billion in total - that is a rise of a little over 6%. The 2026 release, however, describes total income as having "remained fairly consistent" with 2023-24, which it now puts at £59 billion, and explains that "estimates may increase between publications as additional tax returns are received and processed", with revisions "particularly noticeable for partnership income". On the five-year view total income rose £12.3 billion, or 26%, from 2020-21. Average income per landlord was £20,500, the highest in the series, against £20,300 for 2023-24 as restated in this release (£19,400 as published last year). Furnished holiday lettings: 0.13 million landlords declared £2.46 billion, 4% of the total, against £2.47 billion in 2023-24. Nearly half of all landlords - 1.3 million, or 45% - declared property income of £10,000 or less.
The Propenomix Perspective: So did declared rent rise 6% or not at all? Both, depending on which year you let HMRC finish counting. The 2023-24 total grew by around £3.5 billion between last summer's release and this one - late returns, most of them from partnerships - so the "plateau" headline you will have seen in the trade press compares a mature year with a green one, and I would expect this year's £58.99 billion to be revised upwards next summer for exactly the same reason. Smith wanted the quantity to be paid plain to the contributor; it turns out the quantity paid is not entirely plain to the collector for a year or so either. Two things I take from the composition on any basis. First, the 2024-25 tax year ran from April 2024 to April 2025, the tail end of the big rent rises; the flat year Chris's portal data has been showing us since the spring is 2025-26, and it arrives on this ledger next August. Rental increases are down from 8%+ on the ONS data set to a much more sustainable low 3%s, and that will come down further if history is any sort of guide (assuming the big waves of inflation are over, which could be a big assumption - but the next wave comes from a black swan just like the last one). Second, the average is a top-heavy number: £20,500 across the lot, but 45% of landlords under £10,000, which tells you the median sits a long way below the mean and the sector is mostly people with one or two properties and a very large one-off bill when the boiler goes. A hedge on the headline, because it invites overconfidence: £58.99 billion is gross rent declared on one tax form. It is not the size of the sector, it is not what landlords keep, and it is not the number a Chancellor should build a surcharge on. Which brings us to the expenses.
Theme 3: Expenses Eat the Rent
The Summary: Unincorporated landlords declared £34.75 billion of allowable expenses in 2024-25, and 87.7% of them declared some expense. The largest category by value was residential finance costs at £12.82 billion, 37% of all expenses; repairs and maintenance came next at £6.41 billion. Total expenses rose 11% on the year and by £12.42 billion, or 56%, over five years; average expenses per landlord reached £13,700, up 12% on the year and from £8,900 in 2020-21. For comparison, last year's release put residential finance costs for 2023-24 at £9.05 billion, or 31% of expenses, subject to the same revision caveat as the income series.
The Propenomix Perspective: Fifty-nine pence in every pound of declared rent now goes back out in declared costs before HMRC gets its turn, up from forty-eight pence in 2020-21, and £12.8 billion of it is interest on residential lending. On the published-to-published comparison that interest line has gone from £9.05 billion to £12.82 billion in a single year - a 40%-odd jump, revisions permitting - which is the refinancing wave arriving on the tax return: the 2% fixes of 2021 and 2022 rolling onto 5% money through 2024. Here is the arithmetic the headline never does. Take the £58.99 billion, subtract the £34.75 billion of costs, and the sector's operating profit is around £24 billion. But because of Section 24, the individual landlord is taxed on profit BEFORE residential finance costs - roughly £37 billion of taxable property income - and handed back 20% of the £12.82 billion as a credit. For the higher-rate taxpayer that is tax at 40% on interest paid to a bank, less 20% back: a 20-point levy on money they never saw. That is the phantom income I have written about for a decade, seen in aggregate for the first time. Rough working, and dependent on a basic-versus-higher-rate split HMRC does not publish here (and don’t forget the additional rate, or the fact that the mechanics of s24 push landlords falsely into a higher bracket because they go from gross income FIRST before the credit in terms of the calculation), but the order of magnitude is right. Now overlay April 2027: the two-point property income surcharge lands on the £37 billion base, not the £24 billion profit, which on my envelope is worth something like £700 million a year before anyone changes their behaviour. And a question I have not seen answered anywhere: does the Section 24 credit stay at 20% when the basic property rate goes to 22%? It shouldn’t of course, but how is it drafted? If it does, the phantom-income levy widens by two points for every landlord above basic rate (or an extra 10% if you compare 22% to 20%). They will change their behaviour. They always do. I would be delighted for someone to check my sums and let me know if I’m a long way out, or even a little out.
Theme 4: What the Ledger Cannot See
The Summary: The release carries an explicit scope limitation: it covers property income reported through Self Assessment only, excluding every landlord operating through a limited company, every overseas entity, and income below the reporting thresholds. It contains no information on tenants, on the number of properties let, or on capital transactions. Estimates are subject to revision as late returns arrive, with the next release due in summer 2027. Because the series is built from returns for a tax year ending in April 2025, the figures pre-date the abolition of the furnished holiday lettings regime, the Renters' Rights Act, and the 2026 turn in the rates cycle. For a more current read, the Scottish Government's landlord register showed registered landlords down 1.0% on the year in July 2026 with registered properties broadly unchanged, and UK Finance data shows the number of buy-to-let mortgages outstanding, 1.92 million at the end of the first quarter, is 6% lower than at the end of 2022.
The Propenomix Perspective: So what would the full ledger look like, if it existed? I will hazard a guess and flag it as such. The incorporation wave since 2016 means a meaningful slice of the sector - the newer, larger, more leveraged, more professional slice - sits in Corporation Tax returns, where interest is fully deductible and the 2027 surcharge does not reach. So an individual headcount rising on the published series while the surveys promise an exodus is not as puzzling as it looks: the small landlord who stays is still counted, the one who incorporates disappears from this table rather than from the market, and the one who finally declares the flat appears for the first time. Three flows, one net number, and the ledger cannot tell them apart. The second thing it cannot see is the year we are in - 2024-25, describing a world before Section 21 went, before FHL went, before the war reversed the rates cycle - and for that you go to the live gauges. Scotland's register, published monthly, is down 1% on the year. The buy-to-let mortgage count is down 6% in three years. Both point the way the anecdotes do: gently down, nothing like a rout. What this ledger CAN tell the Chancellor: the base for any property income tax rise is 2.85 million individuals, 45% of them under £10,000 a year, with £20,500 coming in and £13,700 going out on average. Not a cartel. A cottage industry, with a mortgage. Smith asked that the quantity to be paid be plain to the contributor. On this ledger the quantity is plain, the contributor is rather smaller than advertised, and the collector has quietly reserved the right to change the total next year.
Second, then, to the IPPR, who published on Wednesday a discussion paper by Ben Ansell - an Oxford professor rather than an IPPR staffer, which matters for how you read it - under the title "Taxing Times: Policies, Politics, and Principles". It is an argument about why Britain cannot reform its tax system, followed by a list of the reforms it would make if it could. One of them would abolish stamp duty. Regular readers know I have opinions about that.
Theme 1: Three Pathologies
The Summary: The paper opens from the premise that Britain's tax system is not fit for purpose: its historical evolution has produced great complexity, yet it struggles to raise the revenue the public demands of the state. Ansell argues that almost two decades of weak growth and an ageing population have moved the country from wanting a European-style welfare state on American levels of taxation to something closer to an American-style welfare state on European levels of taxation, with pressure set to grow as the population ages, immigration falls, and defence and climate demands rise. The press release puts numbers on the ageing part: ageing accounting for almost four-fifths of additional fiscal pressure by then, close to 10% of GDP. He attributes decades of reform inertia to three pathologies: a public pathology, in which voters shy away from the tax implications of their own expectations; a political pathology, in which politicians reach for frozen thresholds and stealth taxes rather than fiscal truth; and a press pathology, which amplifies the losers from any tax rise and distorts debate.
The Propenomix Perspective: I find myself nodding at the diagnosis and reaching for my wallet at the prescription. The three pathologies are real - fiscal drag is the political pathology in its purest form, and I have spent years calling it the stealth tax that never has to be announced - and the "American welfare state on European taxes" line is the best one-sentence description of the last decade I have read. Where I part company is the implicit conclusion that the cure for a public that will not pay more is a politician brave enough to tell them they must. Perhaps. Or perhaps the public has noticed that the money already raised buys less each year - the public service productivity numbers we covered earlier this month would support them - and their reluctance is a rational response rather than a pathology. The fiscal arithmetic underneath I would not dispute: ageing at four-fifths of the pressure by 2075 is the Ageing Britain question I have promised a proper Deep Dive on, and I will not spend it piecemeal here. The bill is coming either way. The paper is about who gets it.
Theme 2: Funding - Capital Gains and the Big Three
The Summary: On raising revenue over the next five to ten years, the paper argues there are limits to what can be raised while leaving the "big three" - income tax, employees' National Insurance and VAT - untouched. One option identified is reversing the Conservatives' 2022-24 National Insurance cuts, which the author suggests could command public support, especially if hypothecated. The most obvious immediate alternative, in the paper's framing, is equalising capital gains tax with marginal rates of income tax, offset by an investment allowance, which the press release describes as protecting normal returns.
The Propenomix Perspective: Equalising capital gains tax with income tax is the proposal that never dies, and for landlords it is the one to watch in 59 days. Residential property gains sit at 18% and 24% today; at income tax rates the higher-rate landlord pays 40% and the additional-rate landlord 45% on the same gain. The investment allowance is the interesting bit, and the bit that always gets lost: an allowance protecting a normal return would, depending entirely on how normal is defined and whether it is indexed, remove a good part of the taxable gain on a property bought in 2005 - cumulative CPI over the period is 77% - and might leave some long-term holders paying LESS than now. Ansell knows this, which is why it is in the paper; the Treasury knows it too, which is why the allowance would be the first thing to shrink in the drafting. My read, hedged: the headline rate goes up at some point in this parliament, the allowance is thinner than the academics want, and transactions fall because the lock-in effect is real. If you have a disposal you were going to make anyway, the date matters more than it did a month ago. That is not advice; it is arithmetic.
Theme 3: Fairness - The 0.65% Property Tax
The Summary: Under fairness, the paper argues the system has tilted too far in favour of age and wealth. Its measures include extending the 2% National Insurance surcharge, currently paid by employees under 65 on earnings above the upper earnings limit of £50,270, to pensioners; replacing both stamp duty and council tax with a single proportional property tax, at a rate of around 0.65% of property value to replace the combined revenue of the two abolished taxes; and levies on what the author terms "Spiv Britain" - a tax on the net winnings of amateur gamblers, estimated at £1-3 billion, greater auditing and monitoring of crypto asset sales, and additional taxation of commercial property transactions. A separate campaign model the Prime Minister has previously backed, Fairer Share, proposes 0.48% with a doubled rate for second homes, empty properties and foreign-owned homes.
The Propenomix Perspective: Let me run the numbers, because a proportional property tax lives or dies on them. 0.65% of Nationwide's average house - £277,542 - is about £1,800 a year, against an average Band D council tax bill of £2,392 this year: so the average home pays less than it does now, and the difference, plus the lost stamp duty, is found at the expensive end. On Halifax's £536,000 London average, about £3,500 a year; on a £1 million flat in the prime postcodes we discussed last week, £6,500, for ever, on an asset falling in real terms. So it is a transfer from South to North and from the asset-rich to the transaction-heavy, and it abolishes stamp duty - which in principle unblocks Chris's blockage at a stroke by removing the £30,000 friction on the £400,000 move. I have argued for years that stamp duty is the worst-designed tax in the British system, so I should be applauding. Two hesitations. First, who pays it on a rented property? If the owner, it is 0.65% of asset value against a 7% gross yield - a tenth of the income - and it would be in the rent within a cycle. Second, the transition: the family who paid £40,000 of stamp duty in 2024 and now pays an annual charge has been taxed twice, and nobody has yet designed a credit for that which survives contact with the Treasury. Good tax, terrible politics, and Ansell's own three pathologies explain why it stays on the shelf. Note the company the idea keeps, though: the Prime Minister has backed a version of it before, and that is not nothing 59 days from a Budget.
Theme 4: The Future, and the Budget
The Summary: The paper's third section addresses wealth concentration and potential mass unemployment arising from artificial intelligence. Proposals include a progressive consumption tax offset by a tax-free consumption allowance, a negative income tax or a universal basic income, in response to widening inequality; an "AI token tax" to capture a portion of the value created by AI models, which the author acknowledges would require consistent auditing of AI companies and would likely rely on international cooperation; and, in the extreme scenario of highly concentrated AI wealth, a tax on AI "unearned rents" to sustain consumption for the majority. Ansell frames the package as intended not only to raise revenue but to bring citizens along with it, which he argues depends on politicians willing and able to tell a convincing story about rebalancing the system towards the young and those in work. It is published as a discussion paper rather than an IPPR institutional position.
The Propenomix Perspective: Where does this leave us, 59 days out? My honest read is that nothing in this paper is in the 28th October Budget, and at least two of its ideas are in the one after. The AI taxes are a decade from being administrable. The pensioner NI surcharge is politically radioactive but arithmetically inevitable, and I would not be surprised to see it floated in a Sunday paper before the red box. CGT is the live one. What ties the four themes together for a property investor is Smith's question: is the quantity to be paid plain to the contributor? Today, for a landlord, it is not - Section 24 makes it opaque, the surcharge makes it worse, and stamp duty makes the decision to move a tax event rather than a housing one. Ansell's paper would make it plainer and, for most of my readers, higher. Whether that trade is worth having is a political question on which I will keep my counsel, and a spreadsheet question on which I will not: model 0.65% against your portfolio value this week and see whether you would take the deal. I did. I would, just.
Third, the rulebook. The Ministry of Housing published the new National Planning Policy Framework on 17th August under the banner "Creating a clear, rules-based planning system" - it slipped past us while the Blue Book hogged the nerd bandwidth, and BuiltPlace's return from its summer break this week put it back on the desk. Four themes, and the fourth is the only one that matters to anyone with a spreadsheet.
Theme 1: What Actually Changed
The Summary: The NPPF 2026 follows the December 2024 revision and the December 2025 Planning and Infrastructure Act, and responds to a consultation that ran from 16th December 2025 to 10th March 2026. Its decision-making policies, now separated from plan-making, took effect on publication. Headline changes from the consultation draft include: expanding the "default yes" for development around well-connected railway stations from the top 60 to the top 80 Travel to Work Areas by gross value added; supporting more development within the curtilage of existing homes; tailoring minimum densities around stations to avoid unviable requirements; a distinct category of strategic sites of around 1,500 units or more; limiting local quantitative standards, including on energy efficiency and internal layout, to cases where variation is justified; and reinstating a presumption against major development in Protected Landscapes. A standard Section 106 template for medium sites, expected to become the default, is to be consulted on shortly.
The Propenomix Perspective: "Default yes near stations" will get the headlines, and deserves them: eighty travel-to-work areas is most of urban England, and a presumption in favour is the most powerful phrase a planning document can contain. But read the clause about local standards, because that is the one I would have paid for. For years scheme viability has depended on which council you drew - one demands its own energy standard, the next its own space standard - and the developer prices the uncertainty into the land bid or walks. Limiting that to "where variation is justified" is Smith's maxim applied to planning: the rules plain to the applicant and to every other person. The strategic sites category is the sleeper - 1,500 units is a new town in miniature, and if the mayors get the call-in powers below, that is where they will use them. My hedge: this is the best-drafted framework since 2012 and it will take three years to know whether it changed a single start date.
Theme 2: The Consultees
The Summary: Alongside the Framework, the government published its response to the November 2025 consultation on the statutory consultee system, following more than 1,600 responses. Sport England is retained but the range of applications on which it must be consulted is narrowed to development involving the loss of playing fields or substantial sports or school development on them. The Gardens Trust and the Theatres Trust lose statutory status, replaced by a notification requirement. Four of the highest-volume national consultees - Active Travel England, National Highways, Historic England and the Mining Remediation Authority - are being reformed as consulted on, and the moratorium on new statutory consultees is maintained. Regulations "will follow in due course". For scale, the Strategic Land Group calculated in June that the median planning application took 107 days to determine in 2010 and 349 days in 2025.
The Propenomix Perspective: Anyone who has waited eleven weeks for a highways response on a forty-unit scheme will read this with something close to joy, and then remember that "regulations will follow in due course" is the Whitehall phrase that has buried more reforms than any Commons vote. The consultee system is where planning delay actually lives: not in the committee, which is theatre, but in the statutory windows that stack up behind each other while the finance clock runs, which is how a median decision went from fifteen weeks to fifty in fifteen years. Narrowing Sport England will not build a house; reforming National Highways and Historic England might, if reform means response times rather than reorganised letterheads. I am cautiously in favour and entirely unconvinced it happens before the Budget, or the one after. And the Mining Remediation Authority on that list will mean nothing to most readers and everything to anyone who buys in the coalfield towns where I do a good deal of my business - a mining report that takes three months to arrive is a deal that dies. That’s one reason why we skip or indemnify against them.
Theme 3: Mayors and Money
The Summary: Two announcements followed the Framework. First, the government will consult on giving mayors "call-in" powers over key planning applications affecting local growth: in the government's words, a mayor "will now be able to take over key planning applications which impact on local growth, and direct a council to take it forward or refuse it". The consultation is open. Second, the first wave of the Social and Affordable Homes Programme was announced, with £9.58 billion allocated to 33 strategic partners outside London; Homes England has published the partners but not the projects. Housing Delivery Test results for 2024 and 2025 were published together, with the 2025 results to be used for decision-making. MHCLG's statutory homelessness statistics for the first quarter of 2026 showed households in temporary accommodation at a record high.
The Propenomix Perspective: The mayoral call-in is the Burnham premiership in one policy - a former mayor giving mayors the power he wished he had - and I will be honest that I am torn. On one hand, a single accountable decision-maker who can override a refusal on a strategic site is precisely the certainty developers have been begging for. On the other, "direct it to be approved or refused" cuts both ways, and a mayor with an election to fight is not obviously a better guardian of a 1,500-unit scheme than a committee with no election to fight. This is the mayoral homework thread from the start of the month: "who decides" is now clearer, "what they will decide" is not. The £9.58 billion is real money and 33 partners is a real list, but note "outside London" and the absence of projects: this is allocation, not construction, and the 119-years-to-clear-the-queue arithmetic from June does not move until a spade does. I remain a supporter of properly funded social housing, and someone who counts completions rather than press releases; the temporary accommodation record in the same week is the reason why. It’s one person to influence either way….which influence will win the battle? It may depend on timing more than anything in terms of electoral cycles, and the ideology of the mayor (and their backers) will also make a huge difference.
Theme 4: Will It Build?
The Summary: The Framework arrives against a supply backdrop the week's other data describes in detail. BuiltPlace's leading indicator - the 52-week rolling total of Energy Performance Certificates on new-build homes in England - stood at 208,071 for the week commencing 17th August, the lowest since December 2025. The Home Builders Federation recorded 58,814 residential units approved in the first quarter of 2026, 9% down on the previous quarter though 6% up on a year earlier, and its SME survey in July found development viability had overtaken planning delays as the most frequently cited constraint. Bellway reported a private reservation rate of 0.55 per outlet per week including bulk sales, against 0.57 a year earlier, with cancellations at 12%. Knight Frank's land index reported development land values still declining in the second quarter as borrowing costs, build costs and weaker sales rates weighed on viability.
The Propenomix Perspective: Here is the synthesis, and it is uncomfortable for everyone who wrote a press release this week. The planning system has just been handed the clearest rulebook in a generation, the mayors are about to get the pen, the money for social homes has been allocated - and the new-build EPC count, the best real-time proxy for completions we have, is at its lowest since Christmas. Approvals are down 9% on the quarter. Land values are falling because nobody can make the numbers work, and the housebuilders themselves now name viability above planning as the thing stopping them. Like the rest of the market though - when there is a fairly sharp adjustment in one of the factors (in this case - construction inflation with regulation tied into that) - the price adjustments are not fast but instead quite slow, amortised over a number of years. I said in the construction special earlier this month that the constraint on housebuilding in 2026 is not permission but viability - and have been saying ever since Labour came into power, trumpeting how they would revolutionize the planning system, that that alone would not be enough - I was not alone in saying that, but it seemed that the ruling party were sick of experts before they started. The cost of money, materials and Section 106 - the NPPF touches the third of those, a little, and the first two not at all - is not nothing but how material is it? A default yes is worth nothing to a developer whose appraisal says no. That is not an argument against the reform, which I broadly welcome; it is an argument for patience about what it can achieve while five-year money sits near 4.8% and the war premium sits in the price of a bag of cement. Hold that thought for 2028.
Last up, the argument. On Tuesday the IFS published an explainer, by Matthew Oulton and Tom Wernham, under the plainest title of the week: "Do rent controls work?" The same week, the Autonomy Institute published "Rebalancing the Market: Designing Feasible Rent Controls", which concludes that well-designed controls are a viable option for government. Two papers, one question, opposite answers - and a Prime Minister who, as a mayor, asked for the power to freeze rents. I am a landlord reading this, so please check my working as I go, because this needs to be objective and realistic, not partisan.
Theme 1: What the Evidence Says
The Summary: The IFS notes that housing costs absorb over 11% of household income on average and 28% for private renters in 2024-25, and sets out the theoretical prediction that a cap below market rent creates excess demand, with landlords exiting, cutting quality or seeking compensation elsewhere. It then summarises the empirical literature, anchored on Kholodilin's 2024 near-exhaustive review. Tenants in controlled properties do generally pay less than they otherwise would. But every study in the review found reduced supply of rental housing, with landlords selling to owner-occupiers or converting to other uses, and some found lower construction, including recent evidence from Ireland. Exemptions produce their own distortions: Germany's 2015 controls had no effect on average rents after about a year, and Oslo's produced adverts demanding babysitting or deposits of ten to twenty times the monthly rent, which all but vanished when controls were removed. The authors see no clear reason the UK would differ.
The Propenomix Perspective: I will start with the hedge, because the temptation for someone in my position is to wave this paper around like a flag. The IFS is careful to say that rent controls DO lower rents for the tenants who have them - the redistribution works, for the incumbents - and that they reduce the uncertainty those tenants face, which is Smith's maxim from the tenant's side of the door and a point I would not wave away. The evidence question is what happens to everyone else, and there the literature is about as unanimous as economics gets: supply falls, every time, in every market studied. I have said for years that you cannot regulate your way to more of something by making it less profitable to provide, let alone loss-making. As I’ve stated very clearly in the past - rent controls DO work for this year’s crop of renters. They just ruin the market for everyone that follows . The Oslo babysitting story is my favourite footnote of the week, and it is not a joke: it is what happens when you fix the price of a thing people badly want. The price finds another form. It always does. Like water - it flows around whatever you put in its way.
Theme 2: Quality, and Who Ends Up Where
The Summary: Beyond supply, the IFS reviews two further channels. On quality, most studies find significant declines in the condition of controlled properties; a 2026 working paper by Bressler found a 36% increase in "immediately hazardous" building code violations after New York strengthened its controls, consistent with landlords facing a queue of tenants cutting maintenance without losing rent. Quality regulation could mitigate this but is costly to enforce, and renovation exemptions reopen the door to rent rises. On allocation, controlled tenants move less often and the stock is used less efficiently: Glaeser and Luttmer found New York's controls produced more large families in small homes and vice versa, while San Francisco's prompted landlords to convert rentals into exempt condominiums. Lower-income and lower-wealth tenants with the least scope to leave the sector are, the authors argue, most exposed.
The Propenomix Perspective: The quality channel is the one that should worry tenant advocates most, and the one they discuss least. A landlord with a queue at the door and a capped rent has every incentive to let the boiler limp another winter, and a 36% rise in hazardous violations in New York is what that incentive looks like in data. Britain's answer would be the Decent Homes Standard extended to the private sector - and the IFS is right that enforcing it against nearly three million small landlords, most of them declaring under £10,000 a year, is a different proposition from writing it down. The allocation point is subtler and, I think, more important for the long run. A controlled tenancy is worth keeping, so people keep it - the empty-nester in the three-bed, the couple in the flat that suits a family - and the stock stops flowing to the people who need it. Adding it to the private sector, where the whole point is that people can move, would be, I would suggest, a mistake of some size. Perhaps the Autonomy authors have a design that avoids it. Let us see.
Theme 3: The Autonomy Counter
The Summary: Autonomy's report argues that well-designed rent controls are "a viable option for the government, as part of a strategy to tackle problems in the private rental sector". It builds on the institute's May 2026 paper by Garcia and Stratford, "The Property Premium", which used English Private Landlord Survey data to estimate landlord returns and concluded that being a landlord had been the best investment available to individuals over the past decade and that returns had turned down in 2024 from a very high level. The IFS addresses the excess-profit claim in a footnote: landlords' unobserved costs, including their own time and uninsurable risks, and the fact that expected rather than realised returns drive decisions, mean risk-adjusted profits could be well below what the data first suggest. For excess profits to persist there would need to be barriers to new investors becoming landlords, and the IFS finds none in the UK - an instance, it says, of the "joint hypothesis problem".
The Propenomix Perspective: So the whole argument reduces to one question: do landlords make excess profits? If yes, controls can bite without breaking supply; if no, they cannot. Autonomy says yes, from the landlord survey; the IFS says the survey cannot tell you, because it does not see the risk, the time or the alternative uses of the money. Let me be careful here, because I have an interest and I know it. HMRC's ledger at the top of this Deep Dive says the average unincorporated landlord takes £20,500 in and pays £13,700 out before tax, before their own time; and nearly half of them are under £10,000 a year. Last week's Pegasus data said 86% make a profit and 14% do not. That is not the return profile of a cartel; it is a cottage industry adequately but not lavishly rewarded for a great deal of hassle. Excess returns in pockets? Certainly. An entry barrier stopping the next investor competing them away? None I can see beyond stamp duty and Section 24, which are the government's own doing. The IFS has the better of this exchange, I think, although I would say that. The decisive evidence is what capital does when a market pays excess returns: it enters. Scotland's landlord register is down 1% on the year and the buy-to-let mortgage count is 6% below where it stood at the end of 2022. That is not what an excess-profit industry looks like from the outside, whatever the survey says from the inside. I suspect - a bit like appraisals of property developers - that idealogues particularly look back at markets of yore with green-eyed goggles on, and don’t trouble themselves with the reality of a market with suddenly normalised interest rates and more regulations than ever to adhere to. Making today’s (potential) investors pay for profits of yesteryear is as sensible as turning bigotry on its head and being prejudiced against groups that have had a historic advantage; two wrongs don’t make a right and as usual the innocents of today pay unnecessary penalties for the mistakes of yesterday.
Theme 4: The Rest is Politics
The Summary: The IFS notes that rent controls were commonplace in England and Wales when the private sector was far smaller, and were abolished for new tenancies in 1988. The Scottish Government introduced temporary controls during the pandemic and plans permanent local powers; Plaid Cymru, now governing in Wales, pledged controls in its manifesto; the new UK government has indicated it does not plan to introduce them in England. Andy Burnham, as a mayoral candidate in 2016, pledged to seek powers to regulate rent increases in Greater Manchester, and in February 2023 signed an open letter with other Labour mayors calling on the government to introduce an immediate freeze on private rents. The IFS concludes that, unless the UK market differs substantially from those studied, controls would be a costly way to support renters; a government wishing to reduce housing costs should address supply through investment and planning reform, and one wishing to redistribute should use the tax and benefit system, citing housing support through Universal Credit as better targeted.
The Propenomix Perspective: Read that conclusion alongside last week's NRLA paper and this week's NPPF and something rather satisfying happens: the IFS has written the Propenomix position out in full. Reduce housing costs by building - the planning reform two sections up, and the viability problem underneath it. Help poorer renters through the benefit system - last Sunday's LHA argument, which the Prime Minister has said in terms he agrees with. Do not fix the price. Which makes the politics the interesting bit, because this Prime Minister asked for a rent freeze from the other side of the table three years ago, and his government's line is now that it has no plans for controls in England. Read that as growth winning the argument inside Number 10, or as a man who has read the IFS, or as "no plans" being a phrase that has aged badly before; I lean to the middle reading, without much conviction. My suspicion, offered as no more than that: in three years the IFS will have its UK evidence, it will point the way Ireland's did, and England will be debating how to help the tenants the controls hurt. Meanwhile the rules that DO now apply to the private landlord - Section 8 grounds, the once-a-year rent review, the Ombudsman and database still to come - are the certainty Smith asked for, whether or not anyone likes the quantity. I will take known rules over the alternative; most of my readers, I suspect, would too.
As we get towards the end for this week - I can't wait for our next workshop in Manchester on 1st October. JVs and M&A is the topic, and this week's Supplement has made the case for it without meaning to: HMRC counting 2.88 million landlords, nearly half of them under £10,000 a year, with partnership income the fastest-growing line on the return; the IPPR sketching a tax system that would continue to reward the well-structured and punish the accidental; and a planning framework that hands the big sites to whoever can assemble the capital to build them. The deals are moving up a level, and structuring them properly - on both sides of the table - is precisely what Rod and I will be working through, with the real-life case studies and the no-holds-barred Q+A as always. The VIP dinner returns too, in the smaller setting where the properly specific questions get answered. Super Early Bird pricing, at more than 20% off, is in its last days. Book your tickets for Thursday 1st October, Manchester at: https://tinyurl.com/pbwoct26 - powered by Roma Finance, now covering all of your needs from bridging and development to term finance.
Above all - please remember to Keep Calm, ALWAYS listen to or read the Supplement, and Carry On. It has been a week about certainty and its absence: a Fed Chair who will not tell you what he will do, an energy regulator who told you to the penny, a tax authority that counted every landlord and reserved the right to recount, a planning system that promises rules. In weeks like this the temptation is to mistake the loudest number for the important one. Resist it. The fundamentals have not changed: there remains a shortage of 2-3 bed terraces and semis, there was one fifteen years ago, and nothing in this week's data - not the 0.9% house price print, not the 208,071 new-build EPCs, not the 2.88 million landlords HMRC counted - does anything but deepen it. "Investing in property" is not a guaranteed win, and with the next rate move still pointing up, just later, the discipline around leverage matters as much this month as last. But buying below replacement cost, strong on the spreadsheet, strong on yield, in a market repricing slowly in real terms rather than crashing in nominal ones, from sellers who have finally started reading the reductions column - that remains as close to a guarantee as this game offers. The houses still aren't being built. The people still need somewhere to live. The rules, for once, are getting plainer. KCCO!