On the morning the Office for National Statistics reported that the UK economy had grown by 0.7% in the three months to August 2026, it also reported that one of the industries the government most needs to expand had gone the other way.
Private housing new work in Great Britain fell by 3.2% over those three months compared with the three months to May, according to the ONS monthly GDP estimate published on 15 October. Construction output as a whole fell 1.0% over the same period. In August alone, construction output slipped 0.2%, while private housing new work fell 1.9%.
The accompanying ONS construction output figures show that total construction activity had declined for four consecutive months, following revisions to earlier estimates. Private new housing made the largest negative contribution to the latest three-month fall.
Those figures are not a count of homes. They measure the volume of construction work carried out, and a fall can reflect slower progress on existing sites, fewer new phases getting under way, or a shift in the mix of work. Nor should the two private housing figures be stacked together: the 3.2% is a rolling three-month comparison, the 1.9% the latest single-month movement within that same series.
But read for what they are, they describe a sector contracting while the wider economy expands, and that divergence is the more interesting story.
Why the two numbers can move in opposite directions
An economy can grow on the strength of services, consumer spending or business investment elsewhere while housebuilding retreats, because housebuilding responds to a different set of signals. A developer is not deciding whether today’s economy is healthy. It is deciding whether homes started now will sell, at a price that covers the land, the build and the finance, when they are finished a year or more from now.
That makes residential development unusually sensitive to expectations. Most of the money goes out long before most of it comes back. Land, planning, infrastructure, materials, labour and interest all accrue before the first completion, and the developer carries that exposure until the homes are sold. When the expected sale price and the expected cost of building come too close together, the rational response is often not to stop entirely but to slow down: to pace a site to the rate at which homes are actually selling, or to defer the next phase until conditions are clearer.
Interest rates sit at the heart of both sides of that calculation. They affect what developers pay to finance work in progress and what buyers can afford to borrow. With the Bank of England holding Bank Rate at 3.75% in September and three of its nine policymakers voting for a rise, neither developers nor buyers have much reason to assume borrowing is about to get cheaper.
There is now evidence of that caution in the trading statements of major housebuilders. On 13 October, Bellway reported that it had completed 9,695 homes in its financial year to July 2026, an increase of 10.8% on the previous year. Yet it expects to complete only 9,000 to 9,400 homes in the year ahead, with a weaker opening order book and continuing pressure on affordability, construction costs and operating margins.
The company’s full-year results announcement also revealed that sales incentives had increased to an average of around 5%, compared with 4.1% a year earlier. Bellway intends to maintain selective land investment, control work in progress and increase the contribution from bulk sales to support delivery and cash generation.
That is not evidence of a housebuilder in immediate financial trouble. Bellway increased its profits and cash generation during the year. It is evidence of a profitable developer adjusting its expectations and protecting its financial position in a market where selling additional homes has become more difficult.
There is a further complication. Bellway believes the government’s proposed Your First Home scheme could eventually support first-time-buyer demand, but it also acknowledges that anticipation of the scheme may influence when some buyers decide to purchase. A measure intended to encourage home ownership could, in the period before its introduction, give certain prospective buyers a reason to wait for details of the support available.
That is a possibility identified by one developer, not an established explanation for the national slowdown. But it illustrates how decisions about future demand can influence construction activity well before a government policy has taken effect.
These are plausible explanations, and they are consistent with how the industry works. They are not, however, what the ONS data shows. The output figures record that activity fell; they do not say why, and they cannot apportion the decline between weaker demand, financing costs, planning delays, build costs or anything else. Anyone claiming to know precisely which factor drove August’s number is going beyond the evidence.
The figures that seem to point the other way
Three weeks before the ONS release, the Ministry of Housing, Communities and Local Government published its own indicators of new housing supply in England. They show an estimated 35,910 new-build starts in England in the second quarter of 2026, seasonally adjusted, up 6% on the previous quarter and 20% on a year earlier. Completions were estimated at 35,800, down 3% on the quarter.
At first glance, rising starts sit awkwardly beside falling construction output. They do not actually contradict each other. The MHCLG figures cover April to June; the ONS three-month comparison runs to August, so the periods only partly overlap. The MHCLG data is for England, the ONS output series for Great Britain. And the two measure different things: one estimates dwellings recorded as started or finished through building control inspections, the other the volume of construction work carried out in a given period.
A rise in the number of homes started in spring is entirely compatible with less work being carried out across all sites over the summer.
There is also an important caveat in the MHCLG release itself. The Building Safety Regulator, which became a standalone body in January 2026, reported 5,758 starts in the second quarter, against a previous quarterly maximum of 1,206. The department says the substantially higher number reported by the regulator partly explains the increase in recorded starts.
The result is that the headline 20% rise cannot safely be interpreted as an equivalent increase in underlying building activity. Some of the growth may represent additional homes genuinely entering construction, while changes in reporting also affect the comparison. The data does not establish precisely how much should be attributed to each.
The same release notes that new-dwelling Energy Performance Certificates, an imperfect proxy that includes conversions, were down 7% year on year in the 13 weeks to mid-September. Over longer periods the EPC figures can show different movements, which is another reason not to treat any single measure as a complete picture of housing delivery.
There is also evidence of caution elsewhere in construction. The latest published ONS construction new-orders figures showed an 11.8% fall in the second quarter of 2026. That decline was driven mainly by private commercial and public other work, rather than private housing, so it cannot be presented as proof of a similar collapse in new housing contracts. It does show that weakening future workloads were not confined to the latest monthly output figures.
None of this means one data source disproves the other. It means the starts figures are less buoyant than their headline suggests, and the latest output data does not stand alone.
Who feels it first
A national housebuilder facing a softer market has options. It can slow sales releases on some sites while pressing ahead on others, lean on cash generated elsewhere in the business and wait out a weak quarter. A smaller developer with two or three schemes, a development loan and a single planning consent has far fewer.
One project turning marginal can put the whole business under strain. Finance costs continue to accumulate, contractors may already be committed and land acquired under earlier assumptions may no longer deliver the expected return. A business with limited working capital cannot necessarily afford to wait indefinitely for buyers or lenders to become more confident.
The ONS output data does not break down activity by size of firm, so it cannot show whether smaller builders have been hit harder this summer. But a slower market compounds the financing and land pressures they already face. For a smaller developer deciding whether to purchase another site, the immediate question is not whether the national economy is expanding. It is whether that particular scheme remains financially viable under realistic assumptions about sales prices, interest costs and delivery times.
Landowners and promoters face a related problem. Planning permission may make a site developable in principle, but it does not guarantee that a builder will pay the price its owner expects. If developers revise their sales assumptions or increase allowances for borrowing and construction costs, their residual land values can fall. Land transactions may then stall because buyers and sellers cannot agree a commercially workable price.
The effect travels quickly down the supply chain. Groundworkers, bricklayers, electricians, builders’ merchants, architects and surveyors all depend on a steady flow of starts and phases. They may feel a slowdown in site activity well before it appears in annual completion totals, and it is in their order books, rather than in the housing statistics, that a prolonged slowdown first becomes visible.
The longer that uncertainty persists, the greater the risk that businesses hold back on recruitment, equipment and training. A developer may be able to restart a delayed phase when demand improves, but rebuilding a weakened local contractor or specialist workforce is not necessarily as straightforward. That is a potential longer-term constraint on the industry’s capacity to accelerate delivery.
Affordable housing is exposed too, if less directly. A substantial part of affordable supply in England is delivered through planning obligations attached to private schemes, so a market-led slowdown can delay the affordable homes built alongside them.
The link is not automatic. Affordable delivery also depends on grant funding, agreements with registered providers and the terms of each consent, and some schemes are not tied to private sales at all. But where affordable homes are a by-product of private development, they move at private development’s pace.
This creates an uncomfortable dependency for housing policy. The government may want more affordable homes, but where their delivery relies on market-led schemes, it also needs those private developments to remain sufficiently attractive for builders to proceed.
What it means for buyers, and for supply
The temptation is to draw a straight line from less building to higher prices. The evidence does not support anything so simple.
In the short term, slower construction may itself be a symptom of developers struggling to sell at the prices they need. For a buyer looking at new-build homes, that can mean more room to negotiate or more incentives on completed stock.
Bellway’s increased use of incentives offers a practical example, although its reported average is a company-wide figure rather than an offer available on every property. Mortgage contributions, deposit assistance and upgrades can have different financial values to a purchaser and are not necessarily equivalent to an outright reduction in the agreed purchase price. Buyers need to assess the actual terms and the price of comparable properties rather than judge an offer solely by its advertised value.
The position varies locally, too. Developers with completed homes that are proving difficult to sell may have stronger reasons to negotiate than those with limited remaining stock and healthy reservations. Existing homeowners selling nearby properties may find themselves competing against incentives that are unavailable in the second-hand market.
Lower output today also does not mean fewer completions today: homes already well advanced will still be finished, and annual new-build completions in England, at 143,770 in the year to June 2026 on MHCLG’s building control figures, were unchanged on the year before.
That figure should not be compared directly with the government’s ambition to deliver 1.5 million homes in England over this Parliament. The target concerns net additions to the housing stock, which also reflect conversions, changes of use and demolitions, rather than simply the building control record of new-build completions.
MHCLG’s latest indicative estimate puts net additional homes in England at 196,900 for 2025–26, down from the previously published 208,600 in 2024–25. The newer figure is based on EPC information and an adjustment for demolitions, and will remain provisional until the full annual statistics are published.
The distinction matters because a decline in the volume of construction work is not directly convertible into a number of homes lost from the government’s target. The indicators measure different stages of delivery and do not move in perfect step.
The longer-term concern is different. Housing supply is the product of decisions taken years before a home is marketed. Phases deferred now are homes that may arrive later, or not at all, in the places that need them.
That matters all the more against the government’s commitment to 1.5 million additional homes in England, an ambition that requires delivery to rise substantially rather than drift. Faster planning decisions and more allocated land may remove important obstacles, but neither guarantees that builders can finance and sell the resulting properties at a commercially viable pace.
There is a fundamental difference between the capacity to build more homes and the commercial incentive to build them sooner. If a developer believes that releasing too many homes into a weak local market will reduce achievable prices or leave capital tied up for longer, holding back can be rational for the business even when additional supply would benefit the wider community.
Whether reduced supply eventually feeds into prices will depend on demand, mortgage affordability, the number of existing homes for sale and local conditions, not on construction output alone.
Reading the next set of numbers
A single month’s rebound would be welcome but would prove little. Evidence of a genuine recovery would look like sustained growth in private housing new work, improving construction new orders pointing to a stronger pipeline, delayed phases being brought forward and, eventually, completions rising rather than holding flat.
The next ONS construction output figures are scheduled for 12 November 2026. MHCLG’s full annual net additional dwellings figure for 2025–26 is provisionally due in November, while the next quarterly indicators of housing supply are expected in December.
The numbers worth watching extend beyond official construction statistics. Developers’ reservation rates, incentive levels, forward order books, site openings and land purchases will indicate whether confidence is translating into actual commitments to build. An improvement in mortgage affordability would be encouraging, but the stronger evidence would be builders increasing activity rather than merely expressing optimism about future demand.
For smaller developers, a more accessible development-finance market and stronger site viability would also matter. A national recovery driven largely by the biggest housebuilders would not necessarily resolve the difficulties faced by businesses whose next project depends on securing a single funding agreement or viable plot of land.
The significance of the August data is not that Britain has stopped building homes. It has not. It is that private housebuilding has become a drag on construction at a moment when the economy around it is growing and the political commitment to more homes has rarely been louder.
If developers are pacing their sites to a market they judge too weak to justify building faster, that is a commercial decision no target can override on its own, and the consequences may outlast the months in which the slowdown was first recorded.
This article analyses provisional economic, construction and housing-supply statistics available on 15 October 2026. ONS GDP figures cover the United Kingdom, construction output data covers Great Britain and MHCLG housing-supply statistics relate to England. The datasets measure different activities and are subject to revision. Commercial consequences discussed are conditional and should not be treated as forecasts of house prices, housing delivery or individual business performance.
