Inner London boroughs refuse 34.2% of small site applications. Outer London refuses 48.7%; a 14.5 percentage point gap. This is not explained by density (approved and refused schemes have virtually identical median densities). Borough selection is a more powerful lever than scheme design.
Is there a structural inner-outer planning gap in London?
Every developer has a hunch about which boroughs are friendly. But is there a structural pattern across inner and outer London, or is it just anecdote dressed up as strategy? We looked at 15,114 small site applications across all 33 London boroughs to find out.
14.5 percentage points — and density does not explain it
Inner London refuses 34.2% of small site applications (n=4,288 decided). Outer London refuses 48.7% (n=7,709 decided). That is a 14.5 percentage point gap — roughly one in seven additional applications refused in outer boroughs, all else being equal.
The gap is not marginal. It is the difference between a business model that works and one that does not.
The obvious explanation is density. Inner boroughs accept higher densities, so perhaps inner schemes are simply smaller and less contentious. The data says otherwise. Across London, approved schemes have a median density of 67 units per hectare. Refused schemes: 66 units per hectare. Virtually identical. Density does not predict refusal at a London-wide level.
| Metric | Inner London | Outer London |
|---|---|---|
| Refusal rate | 34.2% | 48.7% |
| Decided applications | 4,288 | 7,709 |
| Median approved density | 85 u/ha | 54 u/ha |
| Median refused density | ≈ 85 u/ha | ≈ 54 u/ha |
| London-wide approved density | 67 u/ha | |
| London-wide refused density | 66 u/ha | |
Within each category, the numbers diverge further. Inner approved density sits at a median of 85 u/ha. Outer approved density: 54 u/ha. Inner boroughs approve denser schemes, yes — but they also refuse denser ones. The threshold for what is acceptable simply sits higher, and the gap between what is proposed and what is refused is no wider.
What does predict refusal is which borough you are in.
Take a single site type: mid-terrace development and redevelopment (DMR), the most common small site category in London. In Islington, DMR schemes achieve a 92.6% approval rate from 54 decided applications. In Croydon, the same type achieves 31.1% from 264 applications. That is a 61.5 percentage point gap for architecturally identical categories of development.
This is not a quirk of two outlier boroughs. Havering refuses 70.5% of all small sites. Ealing refuses 31.7%. Newham refuses 63.4%. The pattern is consistent: outer boroughs, particularly in east and south-east London, are systematically more hostile to small site intensification.
The reasons are structural. Inner boroughs have planning frameworks built around densification — their local plans anticipate and encourage it. Many have design codes that give applicants a clear standard to meet. Outer boroughs more often rely on character-based assessments, where “out of keeping” is both the test and the verdict. When design quality is assessed against an articulated standard, more schemes pass. When it is assessed against a subjective sense of neighbourhood character, more fail.
There is also a committee effect. Outer boroughs tend to have lower thresholds for committee referral, and committee decisions on small sites correlate with higher refusal rates across the dataset. But this is a contributor, not the whole explanation.
Borough selection matters more than site selection
Consider a developer with capital for a single acquisition. Two sites are available: a mid-terrace in Islington at £650,000, and a comparable property in Croydon at £400,000. Both will deliver 6 units. Construction costs are broadly similar. The Croydon site is £250,000 cheaper.
At Islington’s 92.6% DMR approval rate, the expected planning outcome is close to certain. At Croydon’s 31.1%, two in three applications fail. A failed application costs £15,000 to £25,000 in professional fees, plus 6 to 9 months of holding costs on a property you cannot develop. Two failed applications on the Croydon site — not improbable at a 31% rate — consume the entire saving on the purchase price.
Risk-adjusted, the cheaper site is not cheaper.
This does not mean outer London is uninvestable. Ealing, at 31.7% refusal, is outer London — and that rate is lower than several inner boroughs. The signal is not about geography in the physical sense. It is about planning culture, which varies dramatically between neighbouring boroughs. The 14.5 percentage point inner-outer gap is an average. Individual borough rates vary by over 60 points.
For developers building a pipeline, the implication is that borough selection deserves at least as much analytical attention as site selection. A good site in a hostile borough is a worse bet than a mediocre site in a permissive one.
What the inner-outer classification misses
This analysis treats “inner” and “outer” as the standard GLA classification. It does not account for edge effects — boroughs like Greenwich or Haringey that straddle the boundary in character if not in designation. The dataset also cannot capture pre-application withdrawals, which may be higher in stricter boroughs (making their published refusal rates an understatement of true hostility). And approval rate alone does not measure what was approved — a borough with a high approval rate and aggressive unit compression may not be as friendly as it appears.
Borough-level profiles, area breakdowns, and officer-level patterns are covered in each of the all 33 borough reports.