Housing Delivery Test 2025: the presumption now applies to 93 local planning authorities, up from 87

UK Construction and Law

Housing Delivery Test 2025: the presumption now applies to 93 local planning authorities, up from 87

By Staff Writer  |  20 August 2026

A tower crane standing over a scaffolded residential block among tiled roofs in an Oxfordshire town at sunset

The 2024 and 2025 Housing Delivery Test results were published on 17 August alongside an updated Rule Book. On the 2025 measurement 93 authorities fall below 75 per cent of the homes required of them and are recorded with the presumption, 19 carry the buffer and 35 must produce an action plan. The Chief Planner has told every chief planning officer in England that the 2025 results are the set to use in decision-making.

Both years were released at once, at 14:09 on 17 August 2026, with the updated measurement Rule Book a minute later. The Chief Planner set the position out to authorities in a letter dated 18 August and published the following day.

The 2025 Housing Delivery Test results should be used for the purposes of decision-making

Joanna Averley, Chief Planner

What the 2025 table records

The published spreadsheet lists 304 rows, two of which are grouping rows for Somerset and Cumberland rather than authorities in their own right. The Isles of Scilly is listed with no measurement and no consequence attached. That leaves 301 authorities with a measurement, and each is given one consequence.

Of those 301, 93 are recorded as Presumption, 19 as Buffer, 35 as Action plan and 154 as None. The bands are visible in the table itself. Every authority marked Presumption measured below 75 per cent, the lowest of them at 4 per cent and the highest at 74. Those marked Buffer measured between 75 and 84 per cent, those marked Action plan between 85 and 94 per cent, and those with no consequence at 95 per cent or above.

The larger authorities are well represented at the bottom of the table. Birmingham was required to deliver 21,400 homes over the three years and delivered 12,101, a measurement of 57 per cent. Tower Hamlets delivered 5,501 against 11,698, at 47 per cent. Bristol delivered 4,192 against 10,134, at 41 per cent. Newham is at 61 per cent, Sheffield at 58 per cent and Greenwich at 42 per cent. Manchester sits just inside the buffer band at 75 per cent, and Leeds is one of the authorities above the line, at 115 per cent.

The year on year movement

The 2024 table listed 308 rows on the same pattern, with 304 authorities measured. Those 304 split into 87 Presumption, 22 Buffer, 25 Action plan and 170 None. So the count of authorities carrying the presumption has risen from 87 to 93, action plans from 25 to 35, and the number with no consequence at all has fallen from 170 to 154. Only the buffer band shrank, from 22 to 19, and that is what happens when authorities in it drop through the 75 per cent line rather than climb out.

Matching the two years by authority code, 304 codes appear in both. Twenty of those authorities were outside the presumption in 2024 and are inside it in 2025, and ten moved the other way. Several of the twenty did not slip across a boundary so much as fall through the floor. Coventry went from 111 per cent to 47 per cent, the City of London from 117 per cent to 4 per cent, Wolverhampton from 95 to 59, Merton from 95 to 63 and Brent from 93 to 56. Windsor and Maidenhead went from 97 to 74, which is a single point past the boundary and a different order of problem.

London carries a large share of it. Twenty one of the 33 London authorities are recorded with the presumption on the 2025 measurement, against 15 on the 2024 one.

What it means on a live application

For anyone advising on a residential scheme in one of those 93 areas, the consequence recorded against the authority is now a fact on the file rather than an argument to be run. It is published, it is dated, and the Chief Planner has told authorities in writing which set of results applies. The same letter records that the new decision-making policies in the Framework took effect on 17 August 2026, and closes by encouraging authorities to familiarise themselves with them.

The measurement is a three year rolling one, so a single bad year does not clear quickly and a single good one does not fix it. Twenty authorities discovered that this week from the other direction.