Cllr Henry Higgins is the founder of Blackridge Planning Insight and a councillor in Hillingdon
HMP Gartree tells you everything wrong with how Britain builds prisons.
Harborough District Council rejected the new 1,700-place jail in April 2022, against its own officers’ advice. The Secretary of State overruled that decision in November 2023. The final layout, reserved matters, in planning jargon, wasn’t approved until June 2024. That was a delay of just over two years, for one straightforward site, next to a prison that was already there.
That hesitation is not a construction problem. It’s a planning problem, and it’s the single biggest reason the government’s prison-building programme is now five years late and £4.2 billion over budget.
The Public Accounts Committee said: the entire 2021 programme was built on the assumption that planning permission for a new prison would take 26 weeks. Gartree took over two years just for the principle.
Full Sutton cleared, but only after nearly 3,000 objections. Sites in Buckinghamshire and Lancashire were refused outright and had to go to appeal.
Meanwhile, the actual building has got faster. The Ministry of Justice now delivers new houseblocks through standardised, factory-built modules, Design for Manufacture and Assembly, in the jargon, and it’s already running about a fifth quicker than traditional construction. Wellingborough became the proving ground for a repeatable prison design that doesn’t need reinventing every time.
The industry side of this problem is basically solved. It’s the consenting side that’s still stuck refighting the same battle, from zero, at every single site.
The fix already exists, for other kinds of infrastructure
This is the bit that should annoy every taxpayer in the country: government has already built the legal mechanism to solve this, for other sectors. Nuclear power stations over 50 megawatts don’t go anywhere near a district planning committee, they sit in the Nationally Significant Infrastructure regime, decided nationally because the country decided long ago that whether to have nuclear power isn’t a question for one council to relitigate every time a new station is proposed.
Scotland has gone further still, building the idea of a “national development” directly into its statutory development plan, certain types of infrastructure the country has already decided it needs, with the fight left to be about siting detail, not about whether the thing should exist at all.
Prisons don’t have that protection, so every scheme starts the argument from the beginning: should there be a prison here, should there be a prison anywhere near here, does the country actually need the places. It doesn’t matter that Parliament has already answered that question in the sentencing framework it legislates. Every application re-opens it anyway.
To its credit, government has already built a mechanism that could fix this. The Crown Development route, live since April 2025 under the Levelling-Up and Regeneration Act 2023, lets Crown bodies take developments of “national importance” straight to the Planning Inspectorate, and new prisons are named explicitly as a qualifying example, alongside defence and border infrastructure. The Justice Secretary is already trying to use it to speed up prison approvals. That’s the right instinct. The mistake would be treating it as a discretionary judgement call, made project by project, instead of prison capacity’s actual default route.
What should actually change
Settle the principle nationally, once. If Parliament has legislated the sentencing framework that drives prison capacity need, that need shouldn’t be re-litigated site by site by a district committee weighing traffic counts against national capacity. Put prison need on the same statutory footing as nationally significant infrastructure, the way nuclear already sits.
Leave siting, not principle, to the local process. Councils and residents are the right people to ask about access roads, landscaping, noise, and site-specific mitigation, exactly the questions Gartree’s residents actually raised. What they shouldn’t be asked to relitigate every time is whether the country needs prison capacity at all.
Give it a real clock, not a fantasy one. The 26-week assumption that broke the last programme was never credible. A statutory determination period that’s actually enforceable, with automatic referral to the Secretary of State if it’s missed, rather than the years of drift Gartree suffered, would do more for delivery than any amount of extra modular capacity.
Reward the repeatable design, don’t punish it. Every time a proven, already-tested prison layout gets built again, the planning burden on it should shrink, not reset to zero. That’s the whole logic behind Wellingborough’s prototype and it should be applied on the consenting side with the same discipline.
None of this is about ignoring local people. It’s about not asking them to fight the same national argument, over and over, in different postcodes, at a cost of years and billions, when Parliament already settled that argument the moment it decided how long sentences should be. Local voice on where; national decision on whether. That’s not paternalism. That’s just not wasting everyone’s time twice.