There is a situation that until this month was ordinary and is now interesting.
A promoter submits an outline application on a greenfield site at the edge of a town. The council does not determine it. Meanwhile the same council, in a different building with a different set of officers, is preparing a local plan and proposes to allocate that same field.
This has always been awkward. It has never been consequential in the way it is now, because the framework published on 17 August 2026 attached a specific legal effect to allocation that it did not previously have.
Why every council in England now wants a local plan adopted.
Start with the incentive, because it explains why this is about to happen at scale.
Buried in Annex A of the new framework is a provision that will shape council behaviour for the next two years. Where a local planning authority can demonstrate a five year supply of deliverable housing sites with the appropriate buffer, and its Housing Delivery Test result is above 75%, a local housing need figure higher than the housing requirement in its adopted plan does not count as evidence of unmet need. The protection runs for five years from the date the plan is adopted (Annex A, paragraph 3).
Read what that actually protects. It is not protection against running out of housing land. It is protection against the gap between the standard method figure and the number a council has planned for. A council that adopts a plan with a requirement below its standard method figure, and then delivers against its own lower number, is safe from the argument that the shortfall is unmet need.
That is the most valuable thing in the framework for a local planning authority, and the only way to get it is to adopt a plan. Every council in England now has a reason to move.
What allocation does under the new framework that it did not do before.
Allocation has always made development easier. A site allocated in an adopted plan is development in accordance with the development plan, and section 38(6) does the rest. Nobody should be surprised that allocating land for housing makes housing more likely on it. That is what a plan is for.
The change is what allocation now does to the national policy that applies.
The framework decides housing outside a settlement by category. Policy S5 sets out ten kinds of development that should be approved outside settlements, and a speculative housing scheme has to get into one of them before the balance is even reached. Inside a settlement, policy S4 applies and there is no list.
The framework then defines a settlement in its glossary, and the definition includes land which is allocated, or which has permission, and which will form part of the built-up area once that development is complete.
So allocation does not merely add support for a scheme. It moves the site across the line that decides which policy governs it.
The two conditions that disappear when a site is allocated.
This is where it stops being a technicality.
Policy S5(1)(j), the route a speculative scheme takes where a council cannot demonstrate five years of supply, carries two conditions. The development must be physically well related to an existing settlement. And it must be of a scale that existing or proposed infrastructure can accommodate.
Both are conditions of qualifying, not factors in a balance. A scheme that fails either does not get into the category at all.
Neither exists in policy S4.
Infrastructure capacity is the one worth dwelling on. Objections about roads, schools, surgeries and sewers have historically been the weakest kind, endlessly raised and rarely decisive, because they went into a planning balance that housing need generally won. As a gateway condition they are a different instrument. Allocation turns that instrument back into an objection.
The same council decides the application and writes the allocation. Until this month those were two separate jobs. Now one of them changes the test that governs the other.— The Editor · Ground Level, August 2026
How this works on a real file, at Lechlade.
Lechlade in Gloucestershire is the clearest example we have, because both halves are on the public record.
Cotswold District Council's Regulation 19 plan, approved by Full Council on 12 August 2026 and out for consultation from 24 August, allocates 165 homes at Lechlade. Of those, 150 are at land south of Ferrers Park.
Application 25/04047/OUT, for up to 150 homes at land south of Ferrers Park, was submitted last year. Its determination target date was 17 April 2026. It has not been determined, no officer report has been published, and no committee date has been set.
The allocation and the application are the same field and the same number.
The council cannot currently demonstrate five years of supply, so policy S5(1)(j) is available to the applicant today, with both of its conditions attached. Its own Topic Paper projects 4.7 years at the intended adoption date in 2028, recovering above five only by 2029/30. If the allocation is adopted, policy S4 applies and the conditions fall away.
There is no window in which neither route is open. The supply shortfall opens one now. The allocation would open the other later. And because the council's own projection is that it will not hold five years at adoption, it does not get the Annex A shield on day one either.
The objection to all this, which is a fair one.
Someone will say, fairly, that this is what plan-led development is supposed to look like. A council identifies where growth should go, allocates it, and applications on allocated land get approved. The alternative is speculative development in places nobody chose. Making allocation meaningful is a feature.
That is right, and it is worth saying before going further.
The difficulty is narrower. It is that the effect arrives through a definition in a glossary rather than through a policy anybody votes on, and that it lands on applications which are already in the system and undetermined. When a council allocates a site with a live application on it, the allocation does more than signal support. It changes the test that application will be judged against, in favour of the applicant, while the council is still holding the file.
There is also a timing problem the framework does not address. An application submitted under one framework, undetermined for a year, decided under another, on a site whose status changes again on adoption, has moved through three different policy positions without anything about the land changing.
What survives allocation, and the refusal duties that do not move.
Not everything. Both S4 and S5 provide that the benefits are likely to be substantially outweighed where a proposal fails one of the national policies that require refusal, and those duties apply inside a settlement and outside it. They are the part of an objection that allocation does not touch.
The list is longer and more explicit than the equivalent in the December 2024 framework. Policy N6 on sites of special scientific interest, which permits assessment in combination with other developments and which reaches land outside a designation where the value of the designated area is affected. Policy N2(2), where significant harm to biodiversity cannot be avoided, mitigated or compensated. Policy TR6(4), which now names cumulative impacts on the transport network in the policy itself rather than leaving them to be argued in. Flood risk, heritage and design each carry their own.
An objector whose case rests on the character of the countryside loses most of its force on allocation. An objector whose case rests on a named refusal duty does not.
What a parish council should take from this.
Two things, and the first is uncomfortable.
The consultation on a local plan is now more important than the objection to an application on the same land. It was always the more strategic of the two. It is now the one that determines which national policy the application is decided under, which is a different order of importance. A parish that spends its effort on the application and lets the allocation pass has fought the wrong document.
The second is that the settlement boundary is worth knowing before you need it. Policy S2 requires plans to identify settlements and their boundaries. A lot of adopted plans do not, or use development boundaries drawn decades ago for another purpose. Where that line sits, and whether anyone has drawn it, now decides which test applies to every field around a town.