Affordable Housing: The Delivery Gap
London Plan targets 50% affordable housing on public land and 35% on private. The data shows a persistent shortfall between commitment and delivery
Every major residential scheme in London is required to provide 35% affordable housing. Yet the Planning London Datahub data consistently shows that approved schemes average 22-28% across the capital. The gap between policy aspiration and planning reality is structural — rooted in viability, tenure complexity, and the mechanics of S106 negotiation.
The Policy Architecture
London Plan Policy H5 sets the framework: 35% affordable housing on privately owned land, 50% where public land is involved, with a tenure split of 30% social/affordable rent and 10% intermediate (shared ownership, London Living Rent). These are headline figures — the policy also includes a fast-track route for schemes meeting the threshold without viability testing, and a viability-tested route where schemes fall below the threshold but demonstrate financial constraint.
Why the Gap Exists
Viability testing — the mechanism by which developers demonstrate that policy compliance would render a scheme unviable — is the primary driver of the delivery gap. The GLA's standardised viability toolkit was intended to bring consistency and transparency to this process, but the inputs remain contested: benchmark land values, profit margins, and construction cost assumptions all have significant headroom for professional disagreement. In practice, schemes frequently emerge from viability negotiations with 20-28% affordable housing, presented as the maximum the scheme can bear.
“Viability testing is a legitimate planning tool used to extract the minimum affordable housing a scheme can support. Whether that minimum is genuinely the maximum achievable is rarely independently verified.”
— Centre for London, 'Making Homes Affordable', 2024
Borough Variation
The delivery gap varies significantly by borough. Boroughs with higher land values — the City of London, Kensington & Chelsea, Westminster — tend to approve schemes with lower affordable percentages, as viability constraints are more acute. Outer London boroughs with more development land supply — Barking & Dagenham, Havering — tend to achieve higher affordable percentages, partly because land costs are lower and partly because the borough's development corporation model subsidises affordable delivery. The PLD data reveals this geography in granular detail.
Mechanisms That Close the Gap
- 1Public Land: GLA and TfL sites where 50% threshold applies have significantly higher affordable delivery
- 2Grant Funding: GLA Homes for Londoners grants enable higher affordable provision on marginal schemes
- 3Community Land Trust: Permanently affordable models removing land value from the equation
- 4Viability Review: Late-stage review mechanisms capturing uplift where values increase post-consent
- 5Estate Renewal: Replacement of existing social rent with higher overall provision but contested net gain
The Monitoring Problem
Even where schemes achieve policy-compliant affordable percentages at planning, the translation to occupied affordable homes is not guaranteed. S106 obligations are sometimes renegotiated post-consent; affordable housing may be phased to later in the development programme; and the specific tenure mix can shift between consent and occupation. The absence of systematic post-occupancy monitoring means the true delivery rate against planning commitments is uncertain. The PLD captures applications data — it cannot yet capture what was actually built.