The 1 July 2026 NPPF buffer: seven London boroughs about to tilt

For most of the past eighteen months, the conversation about the 2024 NPPF reforms has been about the new standard method for housing need. That conversation matters strategically, but for the small-site applicant sitting in front of a refused application, the more practical change is one that has had almost no press coverage: a clause that quietly retires the old Housing Delivery Test buffer regime and replaces it with one that bites in seven specific London boroughs from 1 July 2026.

The mechanic, set out in NPPF paragraph 78c, requires a council to apply a 20% buffer to its five-year housing land supply calculation where three conditions are met simultaneously: a housing requirement adopted within the past five years against the pre-December 2024 NPPF, an annual requirement that sits at or below 80% of the standard-method local housing need, and a 20% buffer not already triggered by the 2023 Housing Delivery Test result. Lichfields’ April 2025 analysis identified the London-wide cohort that lands inside all three filters: Barnet, Bexley, Brent, Merton, Waltham Forest, Wandsworth and Westminster. From 1 July, each of these seven boroughs must demonstrate not the standard five-year supply but five years plus a fifth on top — a meaningful increase that several of them will struggle to meet against current trajectories.

Why the trigger matters for refusal defensibility

Where a council cannot demonstrate the required supply, NPPF paragraph 11d engages what officers and inspectors call the tilted balance: the presumption that planning permission should be granted unless the adverse impacts of doing so significantly and demonstrably outweigh the benefits. In effect, the balance of argument on appeal shifts toward the applicant. A refusal that would have been comfortably defensible at appeal under a five-year supply position becomes more contestable under a four-and-a-half-year supply position, and the cost-of-fighting calculation on the council’s side moves with it.

For sub-10 unit schemes this matters in a specific and measurable way. Small-site refusals are disproportionately determined at officer delegation rather than committee, which means the case officer’s read of the planning balance does most of the work. The same officer who would have refused a borderline scheme on amenity-impact grounds at the end of March may, by mid-July, find the inspectorate calculus harder to ignore. The point is not that the trigger forces approvals. It is that it raises the cost of refusing in a way that is most visible at the marginal scheme.

What the data shows for the seven boroughs

The Perfect Scale dataset covers 2,800 small-site (1–9 unit) determinations across these seven boroughs in the three-year window to 31 March 2026. Refusal rates vary substantially:

Borough	Determined	Refusal rate
Westminster	389	19.0%
Bexley	365	29.0%
Wandsworth	551	30.3%
Merton	320	38.4%
Brent	622	53.2%
Barnet	277	56.3%
Waltham Forest	276	66.7%

The reading is not that the buffer will move every borough by the same amount. Westminster is already approving 81% of small-site applications and is unlikely to be the borough where the tilt makes a numerical difference. The cohort that should be reading this most carefully is the three at the higher-refusal end. Waltham Forest, Barnet and Brent sit between 53% and 67% small-site refusal and are exactly where a swing of even three or four percentage points in the borderline-decision band would represent a material shift in the achievability of a typical conversion-led or backland scheme.

The lesson for an applicant in any of the three is not that pre-1-July refusals are a wash. It is that an application validated in the second half of June, capable of being put before the case officer in late summer, sits in a different appeal posture from the same application validated in February.

The wider 26-borough context

It is worth being precise about what is and is not new on 1 July. The seven boroughs above carry adopted local plans that are still within their five-year window, which is the condition that makes the buffer trigger applicable. The other twenty-six London boroughs sit in the opposite position: they do not have a borough plan adopted since 2021, and as of March 2026 their five-year supply calculation already runs against the standard method rather than the London Plan figure, because the London Plan itself crossed the five-year line in March 2026.

The practical effect is that the tilted balance has already been quietly available across most of London since March, depending on each borough’s individual supply position. The 1 July buffer is the visible, dated, named version of a structural shift that has been moving in the same direction for the past two months. For applicants whose pipeline runs across the 33-borough estate, the conclusion is therefore not “wait for July” but “recognise that the planning balance has already moved” and apply that recognition to the cases that are already in the system.

What to do with this between now and 1 July

Three concrete moves are worth considering. The first is timing-led: any application with a tight refusal/approval edge in one of the seven buffer boroughs is better validated late June than early June if the choice is open, on the simple basis that determination then falls inside the post-trigger window. The second is appeal-led: applicants holding a refusal that pre-dates 1 July retain the right to appeal under the post-trigger framework if the appeal sits after the trigger date, and that lengthens the menu of viable appeal strategies in the cohort. The third is portfolio-led: any LPA-level housing land supply position is worth reading directly off the council’s most recent Annual Position Statement before betting on a particular planning balance, because the 5YHLS number is the single piece of information that determines whether paragraph 11d engages at all.

The Perfect Scale dashboards for each of the seven boroughs publish the dataset that sits beneath the refusal rate quoted above: the area-by-site-type matrix, the design-quality vs. amenity-impact split of the refusal-reason data, and the determination-time distribution by decision route. For applicants weighing whether to push a marginal scheme in the next seven weeks, those three numbers in combination are the empirical version of the question the buffer raises in policy terms.

Sources. Lichfields, “The buffers are back in town” (Feb 2025) and “The Presumption: a London focus” (Apr 2025) for the borough-specific list and the cohort filters. National Planning Policy Framework, December 2024, paragraph 78c, for the mechanic itself.