Green belt housing has returned to the centre of England’s planning debate after the Centre for Policy Studies published a report on 21 October 2026 claiming that targeted policy changes could create theoretical capacity for 4.6 million homes. The study focuses on land near existing railway infrastructure, rather than proposing the wholesale removal of Green Belt protection.
The report, The Future of the Green Belt, was written by Robert Colvile and Samuel Hughes, using spatial work commissioned from Create Streets. Its headline figure is a capacity estimate under proposed policy assumptions, not a Government housing target, a programme of allocated sites or a forecast of homes that will actually be delivered.
For landowners and local planning authorities, that distinction matters. The research raises serious questions about which Green Belt parcels should be assessed more closely, but it grants no permission and does not change the legal status of any site. I would treat the findings as a contribution to policy debate, not as a replacement for site-specific planning evidence.
What the green belt housing study found
The Centre for Policy Studies says Create Streets surveyed England’s 16 Green Belts, looking particularly at previously developed and lower-quality land close to railway lines. Its analysis identifies about 33,000 hectares of unprotected low-quality agricultural land and 115,000 hectares of unprotected medium-quality agricultural land near rail infrastructure.
On the density assumptions used by the researchers, those two categories could accommodate approximately 700,000 and 2.4 million homes respectively. The wider headline figure of 4.6 million homes is the authors’ broader modelling claim. Those component estimates should not be added to the 4.6 million as though they were separate extra supplies.
The organisation also suggests that its approach could identify room for approximately 1.1 million homes around London at relatively low suburban densities. Again, this is a subset of the wider potential, rather than an additional 1.1 million homes on top of the national total.
Rail corridors, not a blanket building permission
The central proposal is to look within roughly 1,200 metres, described by the authors as a 15-minute walk, of existing railway infrastructure. This would widen the reach of development-friendly planning policy beyond the most immediate surroundings of railway stations and include suitable land near railway lines.
There is an obvious transport question here. Being near a railway does not necessarily mean being near a station, having a safe walking route to it or enjoying a service frequent enough to make daily life less car-dependent. A straight-line buffer can identify land worth investigating, but it cannot establish sustainable accessibility on its own.
That is why green belt housing proposals located beside a rail corridor would still need to test actual station access, highway safety, public transport frequency, local services and infrastructure capacity. An apparently convenient map location may become much less attractive when assessed from the entrance to the proposed homes.
How the proposals differ from existing policy
The report advocates changes to the Government’s grey belt approach and argues for what it calls gentle-density development. It also suggests reducing affordable housing obligations that the authors consider excessive. These are recommendations from a think tank, not new regulations or amendments that have already come into force.
England’s August 2026 National Planning Policy Framework made changes to national policy, including the treatment of well-connected locations. But individual applications still have to be assessed against the operative Framework, the development plan and other material considerations. The report itself does not establish that all land near railway lines qualifies as grey belt.
For green belt housing, the practical question is not simply whether a parcel looks physically degraded. The relevant tests include the purposes of Green Belt designation, the status of the site under national policy and the harm arising from the particular proposal. Designated habitats, flood risk, heritage, landscape quality and infrastructure requirements may introduce separate obstacles.
What official statistics establish
The Government’s published 2024 to 2025 Green Belt statistics estimate that England had 1,633,220 hectares of designated Green Belt at 31 March 2025, equivalent to about 12.5% of the country’s land area. That official national baseline broadly matches the one used by the Centre for Policy Studies.
The Government also stresses that Green Belt designation is concerned with keeping land open, containing urban sprawl, preventing settlements merging, protecting the setting of historic towns and encouraging urban regeneration. Green Belt is not a label meaning that every designated field is environmentally pristine, nor is a poor ecological score automatically a reason to release it.
That distinction is vital when evaluating green belt housing opportunities. A railway-side parcel that appears unremarkable in a photograph may still perform a strategic separation function, while another parcel may have fewer policy constraints despite its proximity to an established town.
Affordable housing and viability remain contested
The report’s suggestion that some affordable housing requirements should be eased deserves scrutiny, particularly where new housing would serve expensive commuter markets. Lower contributions might improve residual land values and make particular schemes viable, but that does not establish that the homes delivered would meet the needs of households currently priced out of the area.
There is also a question of who captures any uplift in land value if a site is released. Developers, councils and communities would need to consider affordable tenure, transport links, healthcare, schools, public open space and the timing of infrastructure delivery, rather than assuming the headline capacity can materialise without associated costs.
For applicants considering green belt housing, viability assessments need evidence, not a general assertion that obligations prevent development. A nationally modelled capacity estimate tells us little about a particular site’s purchase price, remediation, servicing costs or deliverability.
What happens to the 4.6 million figure now?
The research will likely be cited in arguments for a more rail-oriented approach to housing growth, but its assumptions require testing against the full report, the source mapping and individual local circumstances. The publisher’s summary establishes the authors’ claimed figures; it does not independently prove that every identified hectare is developable.
Local plan reviews provide the appropriate context for considering strategic changes to Green Belt boundaries. Proposed allocations must be assessed alongside alternative locations, environmental evidence and infrastructure needs, with public consultation and examination where required. A private sector study does not itself change those boundaries.
My conclusion is that green belt housing near established transport infrastructure deserves serious, site-by-site investigation, but the 4.6 million figure is a scenario, not a promise. Its value is in prompting better evidence about land and accessibility. Planning decisions still have to be made on the actual facts of each proposal.
Original research and official sources
The Centre for Policy Studies published its original research overview and 21 October 2026 announcement. The official comparator is the Government’s 2024 to 2025 Green Belt statistical release. The policy context is the 17 August 2026 national planning policy reform statement.
This is an England-only policy and research story. It does not announce a change to the distinct planning systems of Wales, Scotland or Northern Ireland.
0 Comments