Scotland’s existing housing stock is wasting more than one and a half billion pounds in avoidable energy spend over the next three years, and the figure is not a model output. It is the literal sum of what Energy Performance Certificate assessors have written, dwelling by dwelling, on the 1.78 million Scottish homes that currently hold an active EPC. Run the addition across every certificate on the public register and the total comes to £1,577,681,297 — money already documented on official paperwork, money householders are paying out today, and money that could stay in their pockets if the improvements listed on each certificate were carried out. The geographic pattern of where that wasted spend sits is, in places, exactly what you would expect; in others, it is the opposite of what the headlines suggest.
Why the cities top the totals
The biggest pools of potential savings sit in Scotland’s biggest cities, for the unromantic reason that this is where most certified homes happen to be.
| Council | Certified homes | Total 3-year savings | Per home |
|---|---|---|---|
| City of Edinburgh | 141,711 | £117.4m | £829 |
| Highland | 86,685 | £116.6m | £1,345 |
| Glasgow City | 174,412 | £108.4m | £622 |
| Fife | 128,764 | £103.4m | £803 |
| Aberdeenshire | 85,374 | £98.0m | £1,147 |
Edinburgh leads the table on raw total, but Highland sits a close second despite holding barely half as many certificates — and the picture starts to diverge from that point on. A Highland home, on average, carries 60% more documented waste than a Glasgow home, which is a difference too large to be explained by any single variable. Stock age, heating fuel, insulation standards and geography each act as a multiplier in some councils and barely register in others. Sort the same dataset by average potential savings per home and the totals table flips entirely upside down.
Per home, the islands lead by a distance
Re-rank the councils by average savings per home and the cities vanish from the top. What replaces them is a list of places more familiar from holiday brochures than from energy policy briefings.
| Council | Per home | F or G rated | Homes |
|---|---|---|---|
| Na h-Eileanan an Iar | £2,102 | 27.8% | 10,879 |
| Argyll and Bute | £1,588 | 13.9% | 38,154 |
| Orkney Islands | £1,549 | 13.4% | 8,549 |
| Highland | £1,345 | 12.1% | 86,685 |
| Dumfries and Galloway | £1,297 | 11.0% | 54,758 |
The Western Isles tops the list with average documented savings per home that are more than twice Edinburgh’s and three and a half times Glasgow’s. More than one in four certified homes in Na h-Eileanan an Iar holds the lowest two energy ratings, F or G, against a Scottish average closer to 5%. The pattern is consistent across the table — rural and island councils dominate the per-home metric, while mainland cities only dominate when the question is reframed as a total. These are, in effect, two different headlines pointing towards two different policy conversations.
Deprivation is a poor predictor of energy waste
The Scottish Index of Multiple Deprivation gives every Scottish postcode a rank from 1 (most deprived) to 6,976 (least deprived), grouped into ten deciles. Layer SIMD over the EPC data and a pattern emerges that contradicts the easy narrative attached to fuel poverty for the past decade.
| SIMD decile | Per-home savings |
|---|---|
| 1 (most deprived 10%) | £524 |
| 2 | £626 |
| 3 | £717 |
| 4 | £813 |
| 5 | £1,050 |
| 6 (peak) | £1,176 |
| 7 | £1,089 |
| 8 | £922 |
| 9 | £884 |
| 10 (least deprived 10%) | £1,126 |
The distribution is not a straight line but a bimodal curve. Energy waste per home peaks in decile 6 — the mid-deprivation band, neither poor nor wealthy — and peaks again in decile 10, the most affluent ten per cent of postcodes in the country. Decile 1, the most deprived areas of Scotland, records the lowest potential savings per home anywhere on the table.
That sounds wrong until you look at what each decile actually contains. The most deprived tenth is dominated by urban deprivation — tenements in Glasgow’s east end, post-war estates in North Lanarkshire and social housing across central Scotland — and those homes are typically connected to mains gas, often with district heating, and frequently sit in buildings retrofitted under the past two decades of social housing energy programmes. They score badly on income, employment and health, but they score reasonably on EPC.
Decile 6 is middle Scotland in the most literal sense — market towns, suburban villages and mid-rural communities — where homes are statistically more likely to be off the gas grid, heated by oil or LPG, and to sit in older detached or semi-detached buildings that escaped the wave of social housing investment which worked through the urban stock. They score better than decile 1 on the SIMD measures, and they waste considerably more energy. Decile 10 is the genuine surprise. Prosperous postcodes — Aberdeen’s West End, Edinburgh’s New Town, Glasgow’s West End — are dominated by pre-1919 granite, sandstone and limestone villas, often with original or first-generation double glazing and aging gas boilers. High income, large floor area and low insulation combine to show that affluence is no defence against an inefficient building.
The Scottish fuel-poverty debate has historically assumed that deprivation is the dominant driver of energy waste. The EPC data shows that the assumption is structurally wrong. What does drive the waste is a tighter combination of geography, property age and tenure — SIMD score is, at best, a weak signal, and in places it points actively in the wrong direction.
The pre-1919 problem driving the total
Cross-cutting every council and every decile is a single, monotonic finding — the older the house, the worse its energy band.
| Built | Homes (Scotland) | % rated F or G |
|---|---|---|
| Before 1919 | 282,079 | 16.2% |
| 1930–1949 | 170,072 | 4.4% |
| 1976–1983 | 97,955 | 2.4% |
| 1999–2002 | 60,699 | 0.4% |
| 2003–2007 | 89,158 | 0.1% |
| 2008 onwards | 47,753 | 0.084% |
A pre-1919 home in Scotland is around 190 times more likely to be rated F or G than a home built since 2008. The Victorian and Edwardian housing stock that defines so much of urban and rural Scotland is, statistically, locked into poor performance, and roughly three hundred thousand of these properties are still sitting in active EPC records. They are loved, they are characterful, and they are very expensive to heat.
This is the underlying mechanic concentrating the per-home figures in places like the Western Isles and Argyll — high proportions of older stock combined with off-gas-grid fuel — and it is also what creates the surprise spike in decile 10, where affluent owners of large pre-1919 properties often sit inside conservation areas that restrict external insulation. The same old-buildings-hard-to-retrofit dynamic drives the entire £1.6bn number. A retrofit programme that addressed the pre-1919 stock effectively would shrink a meaningful share of that figure.
What this means
For policymakers, the headline message is that targeting energy efficiency funding by SIMD decile alone will miss most of the available opportunity. The biggest per-home wins sit in middle deciles and in the most affluent decile, not the most deprived, and a blanket programme keyed to deprivation would leave large chunks of the £1.6bn untouched. The case for layering property-age and rural-fuel data over the existing deprivation framework is now a numeric one rather than an argumentative one.
For local authorities in rural and island Scotland, the analysis provides a numeric anchor for something those councils have been arguing for years. Na h-Eileanan an Iar has 27.8% of its certified homes in the worst two bands, with Argyll and Bute, Orkney and Highland not far behind. These are the councils where retrofit programmes deliver the biggest per-home gain, and they are also where the cost of doing nothing compounds fastest as fuel prices move.
For homeowners, the £1.6bn is not a promise. The figures sit on each EPC certificate as a potential — what the assessor calculated would be achievable if the recommended works were carried out, at average energy prices, and with average occupant behaviour. Realised savings will depend on the works being installed, on energy prices, and on how a household actually uses heat and light. But the figures are concrete enough to be defensible, because every penny of that £1.6bn is written down on a government certificate somewhere.
For Scotland’s new-build sector, the contrast with the older stock is stark. A new home built to 2008 standards or beyond is around 200 times less likely to be rated F or G than a pre-1919 property. New-build is no panacea, but the data does not support the view that retrofit alone can deliver the warmth and efficiency that future buyers will expect — some of Scotland’s heat-leaking stock will need to be replaced, not just renovated. The £1.6bn defines the size of the problem; where it sits, who lives there, and what kind of building it lives in are the questions that will determine how, and how quickly, it gets fixed.
Frequently Asked Questions
It is the sum of the 'three-year potential energy savings' field on every active Scottish Energy Performance Certificate — what the EPC assessor wrote on each certificate. We summed 1,782,417 records, excluding 469 rows with non-numeric values.
Na h-Eileanan an Iar (Western Isles) has the highest share of F or G rated homes at 27.8% — roughly five times the Scottish average. Argyll and Bute (13.9%), Orkney (13.4%), Highland (12.1%) and Dumfries and Galloway (11.0%) follow.
Not in the way the headline assumption suggests. The most deprived SIMD decile has Scotland’s lowest per-home energy savings opportunity at £524. Energy waste peaks in middle Scotland (decile 6, £1,176) and the most affluent decile (decile 10, £1,126).
A pre-1919 Scottish home is around 190 times more likely to be rated F or G than a post-2008 build. Victorian and Edwardian stock typically has solid walls without cavity insulation, original or first-generation glazing, and is often in conservation areas where external retrofit is restricted.
No. The figure is 'potential' — what each EPC assessor calculated would be achievable if the certificate's recommended works were installed at standard prices and usage patterns. Realised savings depend on the works being done, on energy prices, and on occupant behaviour.
Methodology and sources
Data sources. Energy Performance Certificate records were drawn from the UK EPC register (published by the Ministry of Housing, Communities and Local Government (MHCLG, formerly DLUHC) for England and Wales, and by Energy Saving Trust on behalf of the Scottish Government for Scotland). The dataset used covers all certificates lodged between 2002 and the most recent quarterly release (Q4 2025). Only the most recent certificate per dwelling was counted, using the is_latest = 1 flag on each record. Deprivation data was taken from the Scottish Index of Multiple Deprivation 2020v2, published by the Scottish Government, with the September 2025 postcode and data-zone lookup refresh. SIMD ranks data zones from 1 (most deprived) to 6,976 (least deprived) using seven domains — income, employment, health, education, geographic access, crime and housing.
Scope. The 1.78 million Scottish homes counted represent approximately 68% of Scotland’s roughly 2.6 million dwellings (Census 2022). EPCs are typically issued only when a property is sold, rented, or significantly improved, so older and less recently transacted properties are underrepresented in the register. The real F/G rates for the full Scottish housing stock are likely worse than the figures shown here, not better.
The £1.6 billion figure. Each EPC carries a field labelled ‘Three-year potential energy savings’ — the assessor’s calculation of how much could be saved over three years if the certificate’s recommended improvements were installed. Summing this field across 1,782,417 Scottish dwellings (excluding 469 rows with non-numeric values) gives £1,577,681,297. This figure is described throughout as potential rather than ‘realisable’ because it assumes the works are installed, energy prices stay broadly stable, and occupants behave as the standard assessment methodology expects.
Cross-references. Postcodes were joined between EPC records and the SIMD 2020v2 postcode lookup to produce all decile-level and council-level breakdowns. Spatial polygons used for the maps were dissolved from the SIMD 2020 data-zone shapefile (gov.scot, EPSG:27700) up to council level (32 councils). Source data files used in this analysis are available from gov.scot. Analysis and visualisation by Axiom Eco Homes.