How many homes in England and Wales fail EPC C?
According to EPCGuide's analysis of 27,612,160 domestic energy performance certificates, 49.6% of homes in England and Wales are rated below EPC C. That equates to 9,788,286 properties across all 318 local authorities, with an estimated total upgrade cost of £67.7B.
The proposed 2030 deadline applies to privately rented homes, not to every home. There are 1,848,327 privately rented homes below band C, and 1,512 days between 11 August 2026 and 1 October 2030 to deal with them, which works out at 1,222 rented homes every single day.
| Metric | Value |
|---|---|
| Total EPC certificates analysed | 27,612,160 |
| Local authorities covered | 318 |
| Homes rated below EPC C | 9,788,286 (49.6%) |
| Total estimated upgrade cost | £67.7B |
| Average upgrade cost per property | £6,911 |
| Rented homes needing upgrade per day (to meet 1 October 2030) | 1,222 |
| Worst performing local authority | Isles of Scilly (85%) |
| Best performing local authority | Tower Hamlets (24.9%) |
Source: EPCGuide analysis of MHCLG EPC Register data. Figures above count homes, taking the newest certificate for each property, from the August 2026 bulk register: 27,612,160 certificates covering 19,721,103 distinct homes, at roughly 1.4 certificates each. The per-authority, regional and construction-age tables below count certificates, from the 30 March 2026 per-authority extract that the area pages are built from: 29,207,013 certificates across all 318 authorities, 55.4% of them below band C. Compare any area figure against 55.4%, never against the 49.6% home rate. See full research hub for methodology and interactive tools.
Which UK regions have the worst EPC ratings?
Wales has the highest proportion of certificates below EPC C at 60.6%, followed by Yorkshire and The Humber at 60.3%. London performs best at 50.3%, which is still half its stock. Every region sits within 5.1 to 5.2 points of the 55.4% national rate: this is a problem spread across the whole country, not concentrated in one corner of it.
| Region | Below EPC C |
|---|---|
| Wales | 60.6% |
| Yorkshire and The Humber | 60.3% |
| West Midlands | 58.5% |
| East Midlands | 56.8% |
| North West | 56.4% |
| South West | 56.1% |
| North East | 54.7% |
| East of England | 53.4% |
| South East | 53% |
| London | 50.3% |
The regional gap is 10.3 percentage points between Wales and London, which is narrow next to the spread inside a single region: London contains both Richmond upon Thames at 61.8% and Tower Hamlets at 24.9%. Which region a property sits in tells you far less than which authority it sits in. For the area-by-area breakdown, see the interactive UK EPC map.
Regional figures are the sum of every certificate lodged in each authority, with no sampling. The worst and best authority named for a region exclude authorities with fewer than 10,000 certificates.
Which local authorities have the worst EPC compliance?
EPCGuide's analysis shows that Isles of Scilly has the worst EPC compliance in England and Wales, with 85% of certificates rated below EPC C. Tower Hamlets performs best at 24.9%. The gap between them is 60.1 percentage points, roughly 6 times the spread between the worst and best region.
| Rank | Local Authority | Below EPC C |
|---|---|---|
| 1 | Isles of Scilly | 85% |
| 2 | Gwynedd | 74.3% |
| 3 | Pendle | 74% |
| 4 | Ceredigion | 73% |
| 5 | Castle Point | 72.6% |
| 6 | Blackpool | 72% |
| 7 | Hyndburn | 71.2% |
| 8 | Staffordshire Moorlands | 70% |
| 9 | Burnley | 69.6% |
| 10 | Denbighshire | 69.5% |
| Rank | Local Authority | Below EPC C |
|---|---|---|
| 1 | Tower Hamlets | 24.9% |
| 2 | City of London | 35.4% |
| 3 | Hackney | 38.6% |
| 4 | Salford | 38.7% |
| 5 | Milton Keynes | 38.7% |
| 6 | Southwark | 40.2% |
| 7 | Basingstoke and Deane | 41.3% |
| 8 | Knowsley | 41.8% |
| 9 | Greenwich | 41.8% |
| 10 | Islington | 42.2% |
The full rankings for all 318 local authorities are available on the interactive EPC map, including sortable tables, property-type breakdowns, and CSV export.
How much does it cost to heat a home in England and Wales?
The average modelled heating cost across every home on the register is £736 per year. That average hides a lot: owner-occupied homes average £919 and social housing £567, a gap of £352 a year for the same country.
Geography splits it further. Across a sample of 20 local authorities, Bath and North East Somerset is the most expensive place to heat a home at £946 per year and Hackney the cheapest at £562, a £384 gap. Areas with more off-gas-grid properties and older housing stock cost consistently more.
| Area | Avg Heating Cost |
|---|---|
| Bath and North East Somerset | £946 |
| York | £905 |
| Craven | £895 |
| Blackburn with Darwen | £890 |
| County Durham | £875 |
| Area | Avg Heating Cost |
|---|---|
| Hackney | £562 |
| Islington | £567 |
| Manchester | £570 |
| Westminster | £617 |
| Brent | £626 |
For the full ranked list of all 20 areas analysed, including fuel-type breakdowns and Victorian-era heating penalties, see the heating costs by area research page.
Do rental properties have worse EPC ratings than owner-occupied homes?
Counterintuitively, owner-occupied homes perform worse than private rentals. 61.5% of owner-occupied homes sit below EPC C, against 53.3% of private rentals and 33.2% of social housing. The direction holds in 286 of the 318 local authorities, so it is not one region dragging a national average around. MEES has obliged landlords to upgrade since 2018 and owner-occupiers have been under no equivalent obligation.
| Tenure Type | Below EPC C |
|---|---|
| Owner-occupied | 61.5% |
| Private rental | 53.3% |
| Social rental | 33.2% |
The heating costs run the same way: £919 a year for owner-occupied homes against £687 for private rentals. Private rentals reach that despite carrying more uninsulated solid wall than owner-occupied homes, 33.6% against 21%. They are not better stock. They have been worked on.
Social housing performs best, at 33.2% below EPC C, after decades of government-funded retrofit aimed at council and housing association stock. For the full analysis, see the rental vs owner-occupied EPC gap research page.
How to read these numbers. These count homes, not certificates: where a property has been assessed more than once, only the newest certificate is used. That correction matters most here. Rented homes are re-assessed far more often than owner-occupied ones, usually straight after improvement works, so counting raw certificates over-weights improved rental stock and invents a gap the housing stock does not have.
The English Housing Survey 2024-25 inspects a representative sample of dwellings and finds 51% of private rented homes in bands A to C, so 49% sit below band C. Our 53.3% is 4.3 points worse than that. The two are measuring slightly different populations: a home that has not been sold, let or built since EPCs became mandatory has no certificate and is absent from our data entirely, while EHS inspects the standing stock regardless. Either is defensible; a claim should say which one it rests on.
How does property age affect EPC rating?
Construction date is the single strongest predictor of EPC performance on the whole register. 86% of certificates on homes built before 1900 sit below band C, against 7.1% of those built 2007 onwards. Nothing else on this page separates two groups of homes by 79 percentage points.
| Built | Below EPC C |
|---|---|
| Before 1900 | 86% |
| 1900-1929 | 81.1% |
| 1930-1949 | 76.6% |
| 1950-1966 | 64.3% |
| 1967-1975 | 62.5% |
| 1976-1982 | 48.9% |
| 1983-1990 | 48.6% |
| 1991-1995 | 49.2% |
| 1996-2002 | 34.7% |
| 2003-2006 | 11.1% |
| 2007 onwards | 7.1% |
The cliff is not gradual and it is not where people expect. Homes from the 1990s still fail at close to the national rate. The collapse happens at 2003, when Part L of the Building Regulations tightened, and again after 2007: post-2007 stock fails at 7.1%, which is the only group on this page that is broadly compliant already.
These bands cover 87.6% of the register. The remaining 3,611,289 certificates record an age the schema does not define, most often a bare year or a blank, and are excluded rather than assigned to a nearby band.
How much does it cost to upgrade an EPC rating?
EPCGuide estimates the average cost to upgrade a below-C property to EPC C at £6,911. Across all 9,788,286 non-compliant properties in England and Wales, the total estimated upgrade bill is £67.7B.
| Current Rating | Score Range | Avg Upgrade Cost |
|---|---|---|
| D | 55-68 | £5,500 |
| E | 39-54 | £8,500 |
| F | 21-38 | £12,000 |
| G | 1-20 | £18,000 |
These are indicative averages based on industry data from DESNZ, RICS, and the Energy Saving Trust. Actual costs vary significantly by property type, construction method, and region. Solid-wall properties cost substantially more than cavity-wall homes. Government grants through the Boiler Upgrade Scheme and other funding options can offset some of these costs.
Can the UK meet the 2030 EPC C deadline?
The numbers suggest not. The proposed rule covers privately rented homes, of which 1,848,327 sit below band C. Clearing them by 1 October 2030 means 1,222 upgrades every single day, about 446,030 a year, sustained without interruption. The other 7,939,959 below-C homes in England and Wales are outside the deadline entirely.
| Metric | Figure |
|---|---|
| Privately rented homes below EPC C | 1,848,327 |
| Days from 11 August 2026 to 1 October 2030 | 1,512 |
| Upgrades needed per day | 1,222 |
| Upgrades needed per year | ~446,030 |
| All below-C homes, rented and owned, for scale | 9,788,286 (£67.7B) |
There are currently around 800 accredited EPC assessors active in England and Wales, and an assessment is the cheap part. 1,222 completed retrofits a day, every day, for 1,512 days is far beyond the installation capacity the sector has now.
For a full breakdown of the 2030 EPC C deadline regulations, including the proposed cost cap and exemption framework, see our regulations guide.
How was this data collected?
All statistics on this page are derived from EPCGuide's analysis of 27,612,160 domestic Energy Performance Certificates sourced from the official EPC Register operated by the Ministry of Housing, Communities and Local Government (MHCLG) via the Open Data Communities API.
The dataset covers all 318 local authorities in England and Wales. Where multiple certificates exist for the same property, only the most recent is used. The analysis was last updated on 30 March 2026.
Upgrade cost estimates use industry-average figures from DESNZ, RICS, and the Energy Saving Trust. Heating cost data is drawn from the energy cost fields within EPC certificates themselves, which are modelled estimates (not actual meter readings).
For full methodology notes, data limitations, and source documentation, see the EPCGuide Research Hub.
Suggested citation: EPCGuide (30 March 2026). UK EPC Key Statistics: Analysis of 27,612,160 domestic energy performance certificates. Available at: epcguide.co.uk/research/key-statistics