England has not built a new town since Milton Keynes was designated in 1967. It is about to try again, at a scale that would be the largest planned settlement programme since the post-war Acts, and the process has gone quiet at exactly the point where it was supposed to conclude.
What has actually been decided.
The New Towns Taskforce reported in September 2025 with twelve shortlisted areas. On 23 March 2026 the government cut that to seven and put them out for consultation, alongside a draft New Towns Planning Policy, a strategic environmental assessment and a note explaining how the locations were chosen.
The consultation ran from 10am on 23 March 2026 to 11:59pm on 19 May 2026. The government said it intended to publish final proposals and confirm the locations "later in summer 2026".
As at the date of this piece, it has not.
The seven locations.
| Location | Homes proposed |
|---|---|
| Tempsford, Bedfordshire | up to 40,000 |
| Brabazon and the West Innovation Arc, South Gloucestershire | up to 40,000 |
| Milton Keynes, Buckinghamshire | around 40,000 |
| Crews Hill and Chase Park, Enfield | up to 21,000 |
| Leeds South Bank, West Yorkshire | up to 20,000 |
| Manchester Victoria North | at least 15,000 |
| Thamesmead, Greenwich | up to 15,000 |
Tempsford is the one to watch, because it is the only one of the seven that is a new settlement on open land at that scale, built around a new East West Rail station connecting to Cambridge, Oxford, Milton Keynes and London. Milton Keynes appears as a renewal of the town it already is. Thamesmead, Victoria North, Leeds South Bank and Brabazon are urban extensions and regeneration at scale rather than towns on fields.
That distinction matters more than the headline number, and it is the first thing to check when the final list lands. A programme weighted towards brownfield regeneration is a different proposition, politically and environmentally, from one weighted towards Tempsford.
The number in context.
191,000 homes sounds enormous and is, against any single local plan. Against the national target it is smaller than it looks.
The government is committed to 1.5 million homes over five years, which is 300,000 a year in England. The new towns will not deliver at that pace. New settlements take decades: land assembly, infrastructure, a development corporation, then building out at whatever rate the market absorbs. Milton Keynes was designated in 1967 and reached 100,000 people around 1990.
So the new towns are not the answer to the five year target. They are an answer to the decade after it, and they should be judged on that basis rather than against a number they were never going to touch.
The 40% that is the real test.
The programme sets an affordable housing target of 40%, with at least half of that for social rent.
If delivered, that would be remarkable. Section 106 delivered roughly 27,700 affordable homes across the whole of England in 2023/24, and its share of national affordable supply fell to 36% by 2024/25, the lowest since 2014/15. A single programme committing 40% at scale, half of it social rent, would be a different order of ambition from anything the planning system has produced in decades.
This publication has followed enough schemes to know where the risk sits.
Every affordable housing percentage this publication has followed has been set in a plan and then reduced in a viability assessment. The question for the new towns is not whether 40% is a good number. It is whether anything protects it.— The Editor · Ground Level, August 2026
The mechanism that erodes an affordable housing percentage is not a policy change. It is a viability assessment, prepared by the developer's consultants, arguing that the scheme cannot carry the obligation and remain deliverable. Councils see a great many, and a planning department short of staff is not always placed to test them. The 50% requirement attached to grey belt release under the December 2024 framework was written with no viability opt-out precisely because of this.
Whether the new towns carry the same protection is the single most important detail in the final programme, and it will not be in the press release.
Why the land might behave differently here.
There is one structural reason to be less pessimistic than the record suggests, and it is worth setting out fairly.
New towns are not delivered through the ordinary planning system. They are delivered through development corporations, which assemble land, hold it, and grant themselves the framework to build. The Planning and Infrastructure Act 2025 strengthened both development corporation powers and compulsory purchase.
That matters because of the arithmetic this publication has set out before. Development land is valued residually: whatever is left after build costs and developer profit are taken out of sales value. When a private promoter has already paid for land at a price that assumed a lower affordable percentage, the obligation has nowhere to go but into a viability argument.
A development corporation acquiring land compulsorily, without paying for the planning permission the land does not yet have, starts from a different number. The uplift that would otherwise have gone to the landowner stays inside the scheme, and it is that uplift which pays for the social rent. This is the one model in English practice where a 40% target has a plausible funding mechanism behind it rather than an aspiration.
Whether the corporations are given the money and the mandate to use those powers is a separate question, and it is a Treasury question rather than a planning one.
What to watch when the list lands.
Four things, in order of how much they tell you.
Whether the affordable target has a viability opt-out. If it does, treat the 40% as an opening position. If it does not, this is something the system has not produced before.
How many of the seven survive. Twelve became seven in March. Seven becoming four in the final programme would be a signal about appetite, not about the sites.
Whether Tempsford is in. It is the only one that tests whether this government will build a new settlement on open land against local opposition, which is what the post-war programme actually did and what every programme since has flinched from.
What the development corporations are given. Powers without funding produce a masterplan and nothing else. The corporations are where this programme either works or becomes another document.
The thing nobody is saying.
There is a quieter point underneath all of this, and it connects to everything else on this site.
The standard method distributes housing need to districts by formula, and the formula loads it onto places where houses are expensive relative to local wages. That is how a National Landscape district ends up with a target it cannot physically meet, and how fields at the edge of market towns end up allocated.
New towns are the opposite approach. They put a large number of homes in one chosen place, planned, with infrastructure designed in from the start, rather than distributing the same number across a hundred settlement edges by arithmetic.
Both approaches are running at once, and they are not reconciled. Nothing in the new towns programme reduces the number any individual district is told to find. A district facing a target it cannot meet gets no relief from 40,000 homes being built at Tempsford, because the formula that set its number does not know Tempsford exists.
The consultation documents remain published. The final programme was promised for summer 2026 and has not appeared.