Abre Etteh returns to propviews with the third and final part of his small sites series on the Outcome Differential
The most democratic feature of London’s planning system is also the source of its unfairness. Every refusal is made by people who answer to local voters, through a process that hears objections, weighs the evidence and applies the adopted policy. That is local democracy working exactly as it was designed. Yet the people who bear most of the cost are largely excluded from the decision.
By now the finding is familiar. Across around 12,000 small-site decisions, once differences in sites, transport accessibility, conservation status and scheme size are accounted for, boroughs still differ by 43 percentage points in how often they approve applications, and the gap follows house prices downward.
Small sites, everything from backland plots, to side returns, to the tired garage block rebuilt, are meant to deliver about 23% of London’s ten-year housing target, which is 119,250 homes. They are the natural work of small builders, whose share of what gets built has fallen from roughly 40% in the 1980s to around a tenth today. Most refused small schemes never return. They simply become homes that were never built.
The people who bear that cost are mostly not part of the decision-making process. They are the household that would have rented the ground-floor flat, the couple that would have bought the converted loft, the grown child who wanted to move back to the street they were raised on. The home was not built, so they pay more for less and further out, or they leave, or they never arrive at all. None of them voted on the decision, because none of them lives in the borough, and many never will.
It works rather like a members’ club in which the existing members vote on who may join. No single vote is unreasonable on its own terms, and everyone is entitled to worry about the parking, the noise, the view from the terrace. The result is predictable. If today’s members alone decide who joins tomorrow, the membership will tend to preserve itself.
This is no complaint against councillors or officers, who do the job the system hands them. Local discretion is a deliberate choice. Places have a character worth protecting, and nobody seriously wants Whitehall ruling on every backland plot in London. Both of those things are true. It is also true that the constituency carrying the heaviest cost of that discretion is the one local democracy does not represent. The objecting neighbour has a voice, the applicant has a lawyer, and the person who would have lived in the unbuilt home has neither.
Economists would describe this outcome as regressive, not because anyone intended it, but because its effects fall hardest on the places least able to absorb them. The boroughs that approve least relative to their caseload are, on average, the most affordable and therefore the ones where small-site homes would matter most. No bad faith is needed to produce this effect. Nor is it fate, because Bexley, Sutton and Enfield managed above-average outcomes at the affordable end, which means that the others are, in the end, choosing.
The question cuts both ways, though. There is an important qualification: approving applications is not the same as delivering homes. Richmond illustrates this. It approved small sites well above the rate its caseload predicted, at +21.8, and still built out only about 60% of schemes, in a borough where small sites are meant to carry some 84% of the target. So “what did your posture produce, and for whom?” is a fair question for the generous borough that does not deliver, and for a central government that has piled costs onto small sites without determining the effect on viability.
The system does respond to pressure. After the 2024 tightening of national planning policy, London’s small-site approvals rose from about 52% to about 61%, the spread narrowed, and the most restrictive boroughs moved the most. By early 2026 the rate had begun to drift back, which is the signature of a one-election shove rather than a structural change. The machinery was not altered, only pressured, and it sprang back once the weight lifted. The system proved responsive to political pressure, but not permanently changed by it. The reforms that outlast a political cycle are duller and more durable: plan-making where a borough’s plan has drifted from strategic housing policy, and a full tally of the compliance costs heaped on small sites without anyone measuring whom they screen out.
Which leaves one concrete thing worth doing: publish the differential every year. Not a league table and not a naming-and-shaming exercise, but an annual question put to each borough against a measured number. Your caseload predicted this; you produced that; where the gap is wide, what public good justified it? It would let a borough’s own residents see the trade between local character and city-wide supply being struck on their behalf. It would give the Mayor a measured basis for the conformity powers that already exist. And it would hand central government a mirror for the costs it keeps adding. None of it asks anyone to confess bad faith. The only indefensible position is the present one, in which the answer to every one of these questions is a shrug.
The differential settles nothing and prescribes nothing. The mechanism is unresolved without further scrutiny of the data. Bexley and Sutton show that low delivery is not destiny. What the differential does is make a buried conflict visible. Each individual refusal may be perfectly reasonable. The question is whether thousands of individually reasonable decisions produce a collectively reasonable outcome.
Abre Etteh is the founder of Perfect Scale and the author of the London Borough of Merton’s small-sites planning guidance. This essay is adapted from the white paper Given What You Received.
The analysis draws on the London Small Sites Planning Dataset (LSSPD): 11,689 decided one- to nine-unit residential applications across the 32 boroughs, 2022–2026. Delivery figures: Housing Delivery Test (MHCLG 2023 measurement) and London Plan 2021 small-sites targets. Full method and limitations are in the white paper, available from Perfect Scale.
Method note. The outcome differential is calculated by fitting a single London-wide statistical model to 11,689 small-site planning decisions from 2022 to 2026. The model predicts the probability of approval from site type, public transport accessibility (PTAL), conservation status and the number of homes proposed, but deliberately excludes the borough, which lets each borough’s difference from the London average be measured. The published score is the borough’s observed approval rate minus the expected approval rate the model gives for its own caseload.
Perfect Scale · perfect-scale.co.uk
