Permission is not the constraint any more

Almost everything in planning policy this year has pushed in one direction. More land, in more places, with fewer reasons to refuse. The revised framework published on 17 August 2026 supports development inside settlements, opens up land around well-connected stations, and gives grey belt a clearer route through. Strategic planning is being reinstated. Authorities are being pushed to plan for more.

That is a real shift and it will change where development happens. It will not, on its own, change how much gets built, because consent stopped being the scarce input some time ago.

What the cost stack looks like now

The policy costs loaded onto a new home are not marginal and they are not one thing. The Home Builders Federation, which is the housebuilders' trade body and has an obvious interest in the number being large, puts new taxes and policy costs at more than £76,000 a home. Their component figures are worth looking at separately, because they land at different times and some are still arriving.

The squeeze underneath the cost stack

Policy costs are the visible part. The structural problem is the relationship between what it costs to build and what the finished house sells for. Construction cost inflation has been reported as running around 80% higher than house price growth between the final quarter of 2020 and the final quarter of 2025. A margin that was thin in 2020 has been compressed from both ends for five years.

That is why the market keeps being described as having a demand problem when it does not. There is no shortage of people who want houses. There is a shortage of schemes where the numbers work at the price the land was bought for.

What the output figures are actually doing

Delivery is not responding to the policy loosening, at least not yet. In the second quarter of 2026, private residential completions were reported down 21% on the previous quarter at a little over 28,000. Total construction starts fell 1.5% on the quarter and 5.7% annualised. Housing starts were up 14% year on year, but against a low base, which is a different claim from recovery.

The distribution matters more than the totals. The share of homes delivered by the smallest housebuilders fell by 16% in the year to the second quarter of 2026. A policy environment that opens up more sites, in a cost environment only large balance sheets can absorb, concentrates delivery rather than spreading it.

What this means depending on where you sit

If you own land, the risk has moved. A consent is worth what somebody can afford to build out, and the pool of buyers who can afford to build out a mid-sized site has thinned. Pricing land off a consented value that assumed 2021 build costs is the most common mistake we see.

If you are an authority, the pipeline in your plan is not the same thing as delivery, and the gap between them is now largely financial rather than regulatory. Allocating more sites will not close it. Allocating sites that work will.

If you are a parish or town council, this is the honest version of what is coming. More land will be allocated near you. Less of it will be built quickly than the numbers imply, and the sites that do come forward first will be the ones with the fewest abnormal costs, which is not always the ones the plan expected.

What we are watching

Three things will tell us whether the loosening turns into houses. Whether the Building Safety Levy lands as estimated when it commences in October. Whether the Autumn brings the further levy the industry is expecting, and at what rate. And whether permissions granted under the new framework start converting into starts within a year, or sit.

Our own view is that the policy changes of 2026 will show up in where development happens well before they show up in how much. Anyone modelling a site on the assumption that a friendlier framework means a faster build out is modelling the wrong constraint.