Few words in planning carry as much confusion as affordable. It is used in policy documents, committee reports, marketing material and news coverage to mean things that differ by an order of magnitude in what they cost a household. A scheme delivering a headline percentage of affordable homes may be delivering something close to social housing, or something a household on a median income still cannot access, and the percentage does not tell you which.
Anyone working in or near the process needs to be able to read behind the number. The percentage is a container. What matters is what is in it.
The families of product
Broadly, affordable housing in London falls into three families, and they are not variations on a theme. They serve different households, are funded differently, and produce different outcomes for the people who live in them.
Social rented housing is set by a national formula rather than by the market. It is the deepest form of subsidy, it houses households in the greatest need, and it is the product that local authorities generally want most and get least. Its rents bear no relationship to local market levels, which is the point, and which is also why it requires the largest amount of grant or cross-subsidy to produce.
Intermediate rented products sit between social rents and the market. They are typically defined as a discount from local market rent, or capped against household income, or both. The critical variable is the reference point. A percentage discount from the market rent in an expensive part of London can still leave a rent that a household on an ordinary salary cannot pay, while the same percentage elsewhere produces something genuinely accessible. The label is identical in both cases.
Intermediate ownership products, of which shared ownership is the most common, allow a household to buy a share of a property and pay rent on the remainder. The household typically also carries service charges on the whole property regardless of the share owned. Access depends on deposit and mortgage availability, which means these products serve households with some savings and stable income rather than those with the least.
There are further variants, including discounted market products and rents pegged to local housing allowance levels, and their treatment varies. The important point is structural: a single affordable percentage can be composed almost entirely of the shallowest products, and frequently is.
Why tenure mix is the real negotiation
Because the products differ so much in subsidy depth, the mix within the affordable element matters at least as much as its size. A scheme with a smaller affordable proportion weighted towards social rent may deliver more housing need than a larger proportion composed of intermediate ownership.
This is well understood by authorities, which is why policy frameworks generally specify a split between low-cost rented products and intermediate products rather than a bare headline figure. It is less well understood outside the process, where the percentage is what gets reported, and it creates a persistent gap between what a scheme is described as delivering and what it delivers.
For practitioners the implication is clear: when assessing what a scheme offers, read the tenure schedule and the rent-setting basis in the agreement, not the summary. When negotiating, understand that moving tenure within a fixed percentage has a large viability effect, because the value of a social rented unit to the developer is very different from that of a shared ownership unit.
Viability, and why it is contested
The amount of affordable housing a scheme can support is determined by a financial appraisal: the value of what is built, less the cost of building it, less a return, leaves a residual that has to cover the land and any obligations. Where the residual is thin, the affordable element is what gives way, because it is the largest discretionary cost.
Three inputs drive most of the disagreement.
The benchmark land value. Everything depends on what the land is assumed to be worth. If the assumed value reflects a price already paid on the expectation of a low affordable contribution, the appraisal builds that expectation into its own answer. Policy approaches have tried to address this by anchoring to existing use value plus a premium rather than to transaction price, but it remains the most consequential and most argued input.
Build cost. Costs are estimated, and estimates for complex urban schemes carry real uncertainty. Reasonable people produce different numbers for the same building.
Values and yields. Sales values or investment yields at some future date are forecasts. In a moving market they are the input most likely to be wrong in both directions.
Because these inputs are uncertain and consequential, viability assessment has become a specialist adversarial exercise. The mechanisms that have grown up around it, including review of viability at later stages so that improved performance can generate additional contributions, exist because a one-off appraisal at consent locks in an answer based on assumptions that may not survive.
Grant changes the arithmetic
Public grant is what makes the deepest tenures possible on schemes that could not otherwise support them. Where grant is available, a scheme’s affordable capacity is materially different from the same scheme without it, and comparisons between schemes that ignore grant availability are not comparisons at all.
This matters for how outcomes are judged. A site delivering a high proportion of social rent with grant support and a site delivering a lower proportion without it may be performing similarly in terms of what the developer contributed. Attributing the difference to negotiation skill or developer willingness misreads the mechanism.
Service charges and the affordability of the affordable
A dimension frequently absent from the planning conversation is the ongoing cost of occupation. Buildings with lifts, communal plant, concierge provision, extensive landscaping or complex facades generate service charges, and those charges fall on affordable occupiers alongside everyone else. In shared ownership the charge is levied on the whole property, not the owned share, which produces a total monthly cost that can approach or exceed market rent for a comparable home.
A scheme can therefore satisfy every policy test on tenure and still produce housing that is not affordable in the ordinary sense, because the design generated a running cost the planning system did not examine. Simplifying the servicing strategy, reducing plant, avoiding unnecessary communal provision and designing for maintainability are affordability measures, even though they never appear in the affordable housing section of a report.
Reading a scheme honestly
- Ask what proportion of the affordable element is low-cost rent as against intermediate products.
- Ask what the rent is set against, and whether that reference point produces an accessible figure in this specific location.
- Ask whether grant is assumed, and what happens to the offer if it does not arrive.
- Ask what the service charge is projected to be, and how it is apportioned in shared ownership.
- Ask whether the affordable homes are in the same building, with the same entrance arrangements, the same aspect and the same access to amenity.
- Ask when the affordable element is triggered relative to the market element, because a phasing arrangement that leaves it until last leaves it exposed.
None of these questions is hostile. They are the questions that distinguish a number from an outcome, and the gap between the two is where most of the disappointment in this field comes from.
