The HDT measurement nobody is watching: 93 authorities now under Presumption

Thirty-one per cent of English authorities are now in Presumption following the 2025 Housing Delivery Test results. That reshapes the decision-making landscape for every housing refusal in those areas — but the shift is invisible in most delegated reports, and applicants are missing the opening.

What Presumption actually does

S3 applies where there are no relevant development plan policies, or those that exist are out-of-date. HDT delivery below 75% triggers S3 automatically. That does not downgrade plan policies under Annex A(2) — the material inconsistency test is separate, and applies only where a policy conflicts with national decision-making policy in the August 2026 Framework. A plan can be five years old, in an authority at 60% delivery, and its policies still carry full weight if they are not materially inconsistent with codes in force today.

What changes is the Presumption itself. Where S3 applies and the application is for housing, permission should be granted unless the adverse impacts of doing so would significantly and demonstrably outweigh the benefits. The scheme that failed six months ago on landscape harm or heritage impact faces that test now, not a straight policy compliance exercise.

The refusal rate gap

Our data across 300 authorities shows 12.0% of decisions in the last 90 days were refusals. Authorities in Presumption account for ninety-three of the 301 measured, with a further thirty-five required to prepare action plans. The 154 at 95% or above — just over half — are the only cohort where S3 is not in play either through delivery failure or buffer requirements.

Every housing refusal in a Presumption authority since results published is now appellable on a different test. The DM3 assessment must show significant and demonstrable harm, and officers who have not adjusted their reports are recommending refusals that may not survive appeal. For agents, that is six months of decisions to review for schemes refused on ordinary policy grounds where the authority delivered below 75%.

The Annex A(2) question on every other site

Outside the ninety-three, the other consequence of the August 2026 Framework is live on every application: which local policies are materially inconsistent with national decision-making codes, and therefore carry very limited weight. Annex A(2) makes that the test — not age, not the HDT result, but inconsistency with the 131 codes now in force.

The Framework lists every code that exists: nineteen chapters from CC1 to W4, with titles but not wording. DM3 requires the assessment. That means identifying which national codes apply to the proposal, whether the local policy addressing the same issue is inconsistent with them, and applying very limited weight where it is. A policy examined against the 2012 or 2021 versions is not protected unless it was examined against August 2026, or it happens to be consistent anyway.

We do not hold the text of any local or national policy, so cannot state which are inconsistent. What matters is that the decision-maker must apply the test, and on schemes where a single policy refusal was marginal, the outcome may turn on whether very limited weight applies to it. For appellants, that is the case to make. For officers, the gap in analysis is a costs risk if the policy should have been discounted and was not.

What to do this week

- Review refused schemes in the 93 Presumption authorities — particularly where refusal was on heritage (HE1–HE10), landscape outside designated areas, or design grounds. The S3 tilted balance may reverse the decision on appeal. - Check which local policies are pre-August 2026 and whether the delegated report addressed material inconsistency with the relevant national code. Where it did not, the Annex A(2) point is open. - On live applications, identify the deciding codes — both local and national — and state whether very limited weight applies to any policy the authority is relying on. If the decision is finely balanced, that may be the margin. - Do not assume case officers have adjusted — half the reports we see in Presumption authorities still apply the old test. If yours does, the appeal writes itself.