Camera Density League: How UK Cities Compare on Surveillance — and What Residents Can Do About It
Ask most people how heavily surveilled their city is and they will guess from anecdote: the number of cameras on the high street, the ones that follow you round a shopping centre car park. Ask a council and you get a spreadsheet — sometimes. Over the spring and summer of 2026 we sent Freedom of Information requests to 46 local authorities and 12 police forces across the UK, asking for counts of public space CCTV cameras, automatic number plate recognition (ANPR) sites, and camera estates recorded as non-operational. Thirty-eight responses were usable.
How we counted
Our headline measure is cameras per 1,000 residents, combining council-operated public space CCTV with police ANPR camera sites falling within the local authority boundary. Housing association, retail and private estates are excluded, which means every figure here understates the true density of observation. Populations are taken from the most recent ONS mid-year estimates; Scottish and Northern Irish figures use their respective statistical agencies.
The caveats matter. Several councils count 'cameras' and 'sites' interchangeably, a distinction that can double or halve a total. Three authorities declined to release operational status, citing security. Two counted only cameras capable of panning and zooming. Any league table of this kind is a rough instrument, and we have flagged the biggest inconsistencies rather than smoothing them away.
The 2026 ranking
| Rank | City | Cameras per 1,000 residents |
|---|---|---|
| 1 | Westminster | 14.7 |
| 2 | City of London | 12.9 |
| 3 | Birmingham | 6.1 |
| 4 | Manchester | 5.8 |
| 5 | Glasgow | 5.4 |
| 6 | Leeds | 4.9 |
| 7 | Liverpool | 4.6 |
| 8 | Edinburgh | 4.2 |
| 9 | Cardiff | 3.8 |
| 10 | Bristol | 3.5 |
| 11 | Sheffield | 3.2 |
| 12 | Newcastle upon Tyne | 3.0 |
| 13 | Nottingham | 2.9 |
| 14 | Brighton and Hove | 2.7 |
| 15 | Belfast | 2.6 |
The two London entries sit in a category of their own and are included for completeness rather than fair comparison. Both are small, densely commercial districts with daytime populations far exceeding their resident counts, and both operate estates designed for crowd management and counter-terrorism rather than residential streets. If you strip them out, the picture changes considerably.
What the numbers actually show
Birmingham's scale is structural, not strategic. Europe's largest local authority operates one of the UK's biggest civic camera estates, much of it inherited from a period when central government grants paid for city centre coverage. The per-resident figure reflects geography and a dispersed population as much as intent.
Greater London is not uniform. Several outer London boroughs returned densities below Manchester's and Leeds's. The capital's reputation for blanket coverage is driven by a handful of central boroughs, Transport for London's roadside network and the Metropolitan Police's ANPR estate — the last of which sits largely outside council control and is therefore only partly visible in this data.
Glasgow's figure comes with an asterisk. The city reported a substantial estate, but a meaningful share was recorded as awaiting repair or replacement. Across all respondents, roughly one in nine cameras was logged as non-operational on the date of the request. A camera that does not work still appears in a press release.
Brighton and Hove punches above its weight relative to comparable seaside towns, unsurprising given the night-time economy and the seafront's public order history.
Cardiff's ANPR count is a motorway artefact. Much of it clusters around the M4 corridor and approach roads rather than residential streets. ANPR is not really a neighbourhood technology; it is a border technology, and it should be judged that way.
The transparency gap is the real finding
More striking than any ranking is how difficult the basic questions were to answer. Only eleven authorities supplied a current data protection impact assessment when asked as part of the same correspondence. Several could not confirm the number of cameras without manual counting. Two had no published retention schedule for footage held by their own estate.
The Surveillance Camera Code of Practice sets out twelve principles, among them that use should be proportionate, transparent and subject to regular review. Those principles are not optional decoration. Where a council cannot say how many cameras it runs or how long it keeps footage, the failure is one of governance rather than of technology.
A practical playbook for residents
This is not an argument for tearing cameras down. Well-run public space surveillance solves crimes, supports prosecutions and protects lone workers. The case here is for residents to test the system rather than assume it — and to do so with paperwork rather than placards.
Start with published documents. Every council operating public space CCTV should publish a code of practice, a retention policy and, increasingly, an annual report. If they do not, that absence is itself a finding worth raising with your ward councillor.
Use Freedom of Information properly. Ask for camera counts, locations at ward level, retention periods, sharing agreements with police, and the number of cameras out of service. Keep requests specific and reference the authority's publication scheme so you can tell whether the information already exists.
Make a subject access request. You have a right of access to footage of yourself. Requests to councils are usually answered within a month; police forces handle these under their own procedures and may withhold material where an exemption applies. If you want to know whether you appear on ANPR, ask your force about its retention policy — national standards cap ANPR data retention at two years for policing purposes.
Ask about the new stuff. Live facial recognition trials, drones, and AI analytics that flag 'anomalous behaviour' are all being procured now. Ask whether a data protection impact assessment has been completed, whether an equality assessment exists, and what the evaluation criteria are. Contracts over the procurement threshold are usually published.
Escalate proportionately. The ICO handles concerns about compliance with the surveillance camera code and data protection law. Your council's scrutiny committee handles questions about spending and strategy. Choosing the right channel, and bringing documents rather than suspicion, is what separates a useful challenge from a complaint that gets filed and forgotten.
The point of a league table is not to name villains. It is to make the invisible visible, so that the next camera installed in your town is installed after a decision rather than instead of one.
