The 13 London councils that publish the percentage but not the price

Yesterday we published what 16 London boroughs charge for caretaking, grounds and lighting — the only line-by-line service charge data published anywhere in the UK. This is the other half of that story: the boroughs that set the same charges, in the same public meetings, and publish only the headline.

This is not secrecy. Every one of these councils takes the decision in public, publishes the papers, and says by how much charges are changing. What none of them publishes is the number underneath — what caretaking costs, what grounds maintenance costs, what the communal lighting bill actually is. And that is the number you would need to tell whether the increase is justified.

What each of the 13 actually publishes

The number they do publish is not nothing

Taken together the uplifts do say something. They range from Sutton’s 2.7% cap to Greenwich’s 5.26% — a spread of nearly a factor of two in the same year, in the same city, under the same cost pressures. Croydon at 4.8% pegged service charges to the social rent cap; Tower Hamlets at 3.8% pegged them to CPI. Those are different policy choices and they are visible.

What the percentage cannot tell you is whether the base it applies to was right. A 3.8% increase on a caretaking charge that was already twice what a neighbouring borough pays is not a modest increase. Among the 16 boroughs that do publish, combined estate staffing ranges from £7.37 to £18.67 a week — a 2.5× spread. Applying a percentage to an unpublished base tells a resident the direction of travel and nothing about the level.

Two figures show the scale of what is not being broken down. Westminster budgets £24.677m of service charge income for 2026/27, plus £7.026m for heating and hot water — £31.7m a year, with no published figure for any individual service. Ealing’s service charge income rises from £19.325m to £22.600m in a single year, a 16.9% increase in the total collected, alongside an average 23% rise for leaseholders driven, it says, by a 15% jump in the building insurance premium and 7.2% on heating.

Those are large, real movements. They may be entirely justified. Nothing published lets anyone outside the council check.

A cap is a deferred increase, not a saving

Three of the 13 cap increases rather than recover cost: Sutton at 2.7% per property, Kingston at £3 a week, Ealing at £52.40 a week for general needs. A cap reads as protection, and in the year it applies it is.

But service charges are supposed to recover the cost of providing the service and no more. Charging less than cost does not make the cost go away — it leaves a shortfall, and that shortfall is met from the Housing Revenue Account, which is to say from everyone else’s rent. Sooner or later it has to be corrected.

You can watch that correction happening elsewhere. Lewisham’s report is explicit that full cost recovery had not been maintained across caretaking, grounds, heating and communal lighting, and sets 2026/27 charges to fix it — which is why its core basket rose 7.8% and its communal lighting charge rose 39.1% in one year. Barnet’s papers say the same thing about its leaseholders: where charges do not recover costs, the difference is “being subsidised by social housing rental income that could otherwise fund investment in safety measures, property condition, and new social housing”.

So when a council caps an increase without publishing the underlying cost, a resident cannot see how big the gap has become — or how large the eventual correction will be. The absence of a rate card is what makes a cap unreadable.

Two publish the method but not the numbers

The most interesting failures are the ones that get it exactly backwards.

Westminster publishes no rates at all, and yet its day-to-day service charge booklet is the most detailed methodology document any London borough produces. It discloses four different denominators inside a single bill: most services are apportioned by bedspaces (measured by bedroom floor area, with a 110 sq ft threshold splitting one from two), heating and hot water by habitable rooms, accountancy and administration at a flat rate per leaseholder “regardless of property size”, and buildings insurance by reinstatement cost, block size and bedroom count. It also states that most estimates are a 3–5 year rolling average of previous spend plus inflation.

Newham likewise publishes its formula — caretaking is the total cost for an area, divided into estate or block costs, then divided by the number of properties, adjusted for bedrooms; horticulture is a flat rate — without publishing what any of it comes to.

This is genuinely useful, and it is worth saying so. Knowing the apportionment is how you check that your share was calculated correctly, which is a different question from whether the total was reasonable. A borough that publishes the method has given you half of what you need. A borough that publishes neither has given you nothing.

City of London is on a different clock

One practical note. Every other London borough changes service charges in April, alongside rents, following a Cabinet decision in January or February. The City of London changes its tenant service charges on the first Monday of October, and the revised charge reflects the actual cost of providing services in the previous financial year. If you are comparing the City against anywhere else, you are comparing different years.

What to do if your council is one of the 13

The absence of a published rate card does not affect your statutory rights, and those rights get you to the same information.

Under Section 21 of the Landlord and Tenant Act 1985 you can request a written summary of the costs behind your service charge. Under Section 22 you can then inspect the invoices, receipts and contracts that support that summary, free of charge. Neither depends on the council having published anything. And the Section 19 reasonableness test applies regardless: you pay only to the extent the cost was reasonably incurred and the work was of a reasonable standard.

There is also the route that produced the best council service charge data anyone has published. In 2025 the Hackney Gazette asked Hackney Council, under the Freedom of Information Act, for the service charge paid by a two-bedroom leaseholder in every block in the borough. The council answered: 1,467 blocks, ranging from £206.76 at Millfields Road to £6,807.25 at Fellows Court, with a median of £1,925. The data existed the whole time. It simply had not been published.

That is the point worth holding onto. Every one of these 13 councils knows exactly what caretaking costs on each of its estates — it has to, in order to bill for it. The information is not missing. It is just not published.

How we compiled this

We reviewed the 2026/27 rent-setting or HRA budget report for all 33 London boroughs. Four have no Housing Revenue Account, having transferred their stock: Bexley, Bromley, Merton and Richmond upon Thames. Sixteen publish a per-service rate card, and those figures are analysed in our companion piece. The 13 above are the remainder.

Where a borough’s report could not be located we say so rather than assume it publishes nothing — Wandsworth’s rate table may exist in a paper we did not find, and Harrow’s headline average appears to predate the 2026/27 cycle. Corrections are welcome; every figure here is traceable to a named council document.

The dataset behind both pieces is open, and each row records the paper it came from. We refresh it each February when the new Cabinet reports are published.