{ "title": "State of Rental EPC Compliance 2026, Verified Findings", "generatedBy": "EPCGuide research", "generatedAt": "2026-08-11", "supersedes": "2026-06-15 edition, which stated every national figure in certificates and labelled Hook B as regional", "methodologyNote": "Every finding is national and computed over the full GOV.UK bulk domestic EPC register for England and Wales, downloaded August 2026. The unit is the HOME, not the certificate: the register holds 27,612,160 certificates covering 19,721,103 distinct properties, at about 1.4 certificates each, and only the most recent certificate for each property is counted. Deduplication matters because rented homes are re-assessed far more often than owner-occupied ones, usually straight after improvement works, so a raw certificate count over-weights recently improved rental stock. Certificate-level figures are still quoted where the source is the March 2026 API pull, and they are labelled as such. No figure in this file may be used in outreach unless it is reproduced here verbatim.", "sources": { "national": "public/data/epc-national-2026.json (27,612,160 certificates, 19,721,103 homes, 318 LAs)", "byAuthority": "public/data/epc-by-authority-2026.json (certificate level, 318 LAs after folding 28 abolished councils into their successors)", "independentCheck": "public/data/epc-full-dataset.json byTenure summed across all rows (28,625,767 certificates, March 2026 API pull)" }, "corrections": [ "The June 2026 edition said 'Of the 29.2 million homes with an Energy Performance Certificate, 16.2 million, 55.3%, sit below band C'. 29.2 million was a certificate count. There are 19.7 million homes with an EPC, and 9.8 million of them sit below band C.", "The June 2026 edition benchmarked 346 local authorities. 28 of those were councils abolished between 2020 and 2023, counted alongside the unitary authorities that replaced them. The correct figure is 318.", "Three abolished councils, Barrow-in-Furness, Eden and Copeland, appeared in the published worst-10 list. Their certificate stock stopped updating when they ceased to exist, so no post-2023 improvement was ever counted against them.", "Hook B findings were labelled REGIONAL and drawn from a 15-authority, 75,000-certificate sample. That sample came from a single unpaginated API request per authority, which returned the most recently lodged certificates, and it understated every below-C rate by 13 to 22 percentage points. It is retired. All Hook B findings below are now national.", "Average SAP score 63 was quoted on the EPC map page as a national figure. It came from the March 2026 capped pull and could not be reproduced at home level, so it has been removed rather than restated.", "The press release argued that a split-incentive problem explains the private rented sector lagging on efficiency, and cited private renters as 42% more likely to have uninsulated solid walls. The direction was backwards against the register and the number was wrong. Private rentals lead owner-occupied stock, and the solid-wall ratio is 33.6% against 21.0%, which is 60% more likely, not 42%." ], "angles": { "A": "The Postcode Lottery. Where a home sits decides whether its owner can afford to keep letting it. Injustice framing. NATIONAL scope.", "B": "The Reversal. The homes politicians and the press call 'slum rentals' are in better shape than the houses homeowners live in. Vindication framing. NATIONAL scope, confirmed twice by independent datasets." }, "findings": [ { "id": "F1", "angle": "A", "scope": "national", "unit": "home", "headline": "Almost half of England and Wales' assessed homes would fail the 2030 EPC C standard today", "value": "49.6%", "supporting": { "homesBelowC": 9788286, "homes": 19721103, "certificates": 27612160 }, "citable": "Of the 19.7 million homes in England and Wales that hold an Energy Performance Certificate, 9.8 million, 49.6%, sit below band C, the standard the government wants every rented home to hit by 2030.", "emotionalFraming": "The deadline is not a stretch goal for a few stragglers. It is half the country." }, { "id": "F2", "angle": "A", "scope": "national", "unit": "home", "headline": "Fixing them all carries a \u00a367.7 billion bill", "value": "\u00a367.7bn", "supporting": { "modelledCostPerBelowCHome": 6911, "homesBelowC": 9788286, "totalEstimatedUpgradeCost": 67650907482 }, "citable": "Bringing every sub-C home up to standard would cost an estimated \u00a367.7 billion, a bill currently landing on individual owners and landlords rather than the state.", "emotionalFraming": "Nobody has costed who actually pays this. The answer is whoever owns the worst homes.", "caveat": "Derived by applying the per-property modelled upgrade cost from the March 2026 authority dataset (\u00a36,911 per below-C property) to the deduplicated home count. The cost model itself is unchanged; only the property count was corrected." }, { "id": "F3", "angle": "A", "scope": "national", "unit": "certificate", "headline": "A postcode is a 3.4x swing in the odds of failing", "value": "3.4x", "supporting": { "worst": { "name": "Isles of Scilly", "code": "E06000053", "belowCPct": 85, "records": 1297 }, "best": { "name": "Tower Hamlets", "code": "E09000030", "belowCPct": 24.9, "records": 187213 } }, "citable": "In the Isles of Scilly, 85% of assessed homes fall below band C. In Tower Hamlets, 24.9% do. A landlord's compliance risk is 3.4 times higher purely because of where the property sits.", "emotionalFraming": "Same law, same deadline, same penalties, and the hill you have to climb is set entirely by your postcode." }, { "id": "F4", "angle": "A", "scope": "national", "unit": "certificate", "headline": "The hardest-hit areas are rural, coastal and northern, not the inner cities", "value": "9 of the worst 10 are rural, coastal or northern", "supporting": { "worst10": [ "Isles of Scilly 85%", "Gwynedd 74.3%", "Pendle 74%", "Ceredigion 73%", "Castle Point 72.6%", "Blackpool 72%", "Hyndburn 71.2%", "Staffordshire Moorlands 70%", "Burnley 69.6%", "Denbighshire 69.5%" ] }, "citable": "The ten worst-performing local authorities for EPC compliance are overwhelmingly rural, coastal and northern: Gwynedd, Pendle, Ceredigion, Blackpool, Burnley. All have older housing stock and lower rents to absorb upgrade costs.", "emotionalFraming": "The places least able to afford the upgrades face the biggest ones.", "note": "Recomputed after folding 28 abolished councils into their successors. Barrow-in-Furness, Eden and Copeland have been removed; Staffordshire Moorlands, Burnley and Denbighshire take their places." }, { "id": "F5", "angle": "A", "scope": "national", "headline": "Every region has a wildly uneven internal spread", "value": "318 local authorities benchmarked", "supporting": { "totalLocalAuthorities": 318, "england": 296, "wales": 22, "datasetFile": "public/data/epc-by-authority-2026.json" }, "citable": "EPCGuide has benchmarked all 318 local authorities in England and Wales, letting any landlord or letting agent see exactly how their patch compares. It is the only dataset of its kind built from the full register and corrected for local government reorganisation.", "emotionalFraming": "We can tell you, to the local authority, whether you are in trouble. Nobody else can." }, { "id": "F6", "angle": "B", "scope": "national", "unit": "home", "headline": "Private rentals are in BETTER shape than the homes owners live in", "value": "53.3% vs 61.5% below C", "supporting": { "privateRentalBelowCPct": 53.3, "ownerOccupiedBelowCPct": 61.5, "gapPp": -8.2, "privateRentalHomes": 3467781, "ownerOccupiedHomes": 9189856, "localAuthoritiesWhereItHolds": 286, "localAuthoritiesScored": 318, "gapPpBeforeDeduplication": -7.0 }, "citable": "Across all 19.7 million assessed homes in England and Wales, 53.3% of privately rented homes fall below band C, against 61.5% of owner-occupied homes. The rental stock the press calls substandard is, on this measure, eight percentage points ahead of the houses people own and live in, and it holds in 286 of 318 local authorities.", "emotionalFraming": "Landlords have spent a decade being told their homes are the problem. Owner-occupied homes are further behind. They just do not face a deadline.", "verification": "Confirmed independently against a separate March 2026 API pull of 28.6 million certificates, which gives 60.8% for private rental against 68.0% for owner-occupied, a gap of 7.2 percentage points. Two datasets, two units of count, two pipelines, same direction." }, { "id": "F7", "angle": "B", "scope": "national", "unit": "home", "headline": "Landlords got there the hard way, with worse starting stock", "value": "33.6% vs 21.0% solid-wall uninsulated", "supporting": { "privateRentalSolidWallUninsulatedPct": 33.6, "ownerOccupiedSolidWallUninsulatedPct": 21.0 }, "citable": "Privately rented homes carry far more of the hardest-to-treat housing stock. 33.6% have uninsulated solid walls, against 21.0% of owner-occupied homes, and they still come out ahead on EPC compliance.", "emotionalFraming": "They were dealt a worse hand and still beat the room. That is a decade of MEES rules forcing work owners never had to do." }, { "id": "F8", "angle": "B", "scope": "national", "unit": "home", "headline": "Tenants in rentals pay less to heat their homes than owner-occupiers", "value": "\u00a3687 vs \u00a3919 a year", "supporting": { "privateRentalAvgHeatingCost": 686.9, "ownerOccupiedAvgHeatingCost": 919.1, "annualGap": 232.2 }, "citable": "Modelled annual heating costs run to \u00a3687 for privately rented homes against \u00a3919 for owner-occupied homes, a \u00a3232 gap that follows directly from rentals' better efficiency.", "emotionalFraming": "The efficiency gap is not abstract. It is \u00a3232 a year out of a renter's pocket that an owner-occupier is still paying.", "caveat": "Heating cost only. The June 2026 edition quoted total modelled energy cost from the retired sample; that series is not reproduced here." }, { "id": "F9", "angle": "B", "scope": "national", "unit": "home", "headline": "Social housing leads everyone", "value": "33.2% below C", "supporting": { "socialRentalBelowCPct": 33.2, "socialRentalHomes": 3001110 }, "citable": "Social rented homes are the most compliant tenure in England and Wales at 33.2% below band C, against 53.3% for private rentals and 61.5% for owner-occupied homes.", "emotionalFraming": "The more a tenure is regulated, the more efficient it is. Ownership is the laggard, not the benchmark." }, { "id": "F10", "angle": "B", "scope": "national", "unit": "home", "headline": "1,222 rented homes must be upgraded every day to hit the deadline", "value": "1,222 a day", "supporting": { "rentedHomesBelowC": 1848327, "deadline": "2030-10-01", "countedFrom": "2026-08-11", "daysAvailable": 1512, "derivation": "1,848,327 homes over the 1,512 days from 11 August 2026 to 1 October 2030 = 1,222.4 a day." }, "citable": "1,848,327 privately rented homes in England and Wales sit below band C. There are 1,512 days between now and the 1 October 2030 deadline, so clearing them means upgrading 1,222 homes every day, weekends and Christmas included.", "emotionalFraming": "There are not enough assessors, installers or grant pounds in the country to do that. The deadline is arithmetically out of reach and nobody in government has said so." } ], "openItems": [ "Per-authority tenure at home level is not yet published. epc-by-authority-2026.json carries per-authority byTenure at CERTIFICATE level only. The home-level per-authority breakdown exists in the August 2026 aggregator output and should be committed to public/data once transferred.", "Constituency-level figures are still capped-sample only (constituency-epc-rankings.json, 25 constituencies, 22 of them at the 5,000 ceiling). The page is now labelled as a sample. Regenerating it from the full register needs a constituency lookup that the bulk download does not carry, so it stays a sample until that is built.", "heating-costs-by-area.json is a 20-authority capped sample (99,995 records). Its page is labelled. National heating cost by tenure is verified and lives in epc-national-2026.json." ] }