Terraced houses are the workhorse of the English housing market — roughly a third of every home that carries an energy certificate. As a group they rate only a shade below the national average for energy efficiency. But that average hides something sharper: in a handful of districts the typical terraced home rates far below the homes around it — and where it does, the reason is almost always the age of the bricks, not the price on the door.
Homecost matched every current Energy Performance Certificate (EPC) to its address and kept the latest certificate per home — 10.7 million dwellings across England and Wales. Of those, 39.9% sit in EPC band C or above (band C is the common benchmark for an efficient home). Narrow the field to terraced houses and the figure is 37.5% — a gap of just 2.4 percentage points. On that headline number, a terrace is barely colder than the average home.
The district-level picture is where it gets interesting.
The national picture
| Stock | Homes (E&W) | EPC band C or above |
|---|---|---|
| All dwellings | 10.7m | 39.9% |
| Terraced houses only | 3.4m | 37.5% |
Across 294 English districts with at least 1,500 terraced homes on the register, the median gap between all-stock and terrace efficiency is essentially zero — 0.2 points. In 143 of those districts terraces actually rate higher than the local average. So "terraces are cold" is simply not true as a general rule.
What is true is that the gap has a long tail. Fifteen districts show a terrace penalty of eight points or more. These are the places where reading a district's headline efficiency figure would badly mislead a buyer about the terraced home actually in front of them.
Where the terrace penalty is largest
These are the English districts where terraced houses rate furthest below the district's all-stock average — latest certificate per home, districts with at least 1,500 terraces:
| District | 2025 median price | All-stock C+ | Terraced C+ | Gap |
|---|---|---|---|---|
| Bolsover | £183,000 | 46.3% | 30.0% | 16.2 pts |
| Tower Hamlets | £500,000 | 58.6% | 44.4% | 14.3 pts |
| Islington | £642,000 | 41.6% | 28.4% | 13.2 pts |
| North West Leicestershire | £268,000 | 50.7% | 38.2% | 12.5 pts |
| Ribble Valley | £284,000 | 43.2% | 30.8% | 12.4 pts |
| Hartlepool | £156,000 | 44.3% | 32.9% | 11.5 pts |
| Middlesbrough | £155,000 | 40.7% | 30.1% | 10.6 pts |
| Barnsley | £180,000 | 44.7% | 34.8% | 10.0 pts |
| Sunderland | £155,000 | 42.4% | 32.5% | 9.9 pts |
| Cumberland | £175,000 | 36.2% | 26.4% | 9.9 pts |
| Darlington | £178,000 | 37.9% | 28.1% | 9.8 pts |
| Rossendale | £190,000 | 31.4% | 22.3% | 9.0 pts |
Prices are HM Land Registry medians for standard 2025 sales; EPC shares are the latest certificate per address.
Two very different kinds of place sit on this list. Most are the cheaper, ex-industrial North — Middlesbrough, Sunderland, Hartlepool, Darlington and Cumberland in the North East, Barnsley in South Yorkshire, Rossendale and Ribble Valley in Lancashire, Bolsover in the Derbyshire coalfield. Here the district's all-stock figure is lifted by newer semi-detached and estate housing, while the mill- and pit-town terraces at the core stay stubbornly in band D and below.
But the list also holds some of the most expensive postcodes in the country. Tower Hamlets (a £500,000 median) and Islington (£642,000) show terrace penalties of 14 and 13 points. The mechanism is the mirror image: their all-stock figures are inflated by a wall of new-build flats in Docklands and along the canals, all built to modern standards, while the Georgian and Victorian terraces that give those boroughs their character sit far below.
So the terrace penalty is not a "cheap equals cold" story. Across all 294 districts the correlation between price and the size of the gap is weak (−0.24). It shows up at both ends of the market.
Where terraces are the greener stock
Flip the ranking and a different kind of district appears — places where terraced homes rate higher than the local average, sometimes by a wide margin:
| District | 2025 median price | All-stock C+ | Terraced C+ | Gap |
|---|---|---|---|---|
| Bournemouth, Christchurch & Poole | £338,000 | 33.4% | 50.7% | −17.3 pts |
| Castle Point | £365,000 | 23.4% | 37.7% | −14.3 pts |
| South Holland | £235,000 | 43.6% | 57.1% | −13.5 pts |
| North Kesteven | £245,000 | 48.9% | 61.4% | −12.5 pts |
| Epsom & Ewell | £550,000 | 31.3% | 43.8% | −12.5 pts |
In these districts the terraced stock is largely modern — townhouses and terraces built on estates from the 1990s onward, to tighter building regulations than the older detached and semi-detached homes around them. A new terrace is not a period terrace, and the certificate shows it.
Build era, not price, sets the rating
The single variable that explains the whole pattern is when the terrace was built. The register-wide gradient is stark: homes built before 1945 rate C-or-above only about 19% of the time, while homes built from 2003 onward do so around 93% of the time — the cumulative effect of successive tightenings of the Building Regulations, from cavity-wall insulation to double glazing and condensing boilers.
Line the districts up by the share of their terraces built before 1930 and the terrace efficiency figure follows almost mechanically — whatever the price:
| District | 2025 median price | Terraces built pre-1930 | Terraced C+ |
|---|---|---|---|
| Islington | £642,000 | 71% | 28.4% |
| Darlington | £178,000 | 67% | 28.1% |
| Bolsover | £183,000 | 62% | 30.0% |
| Middlesbrough | £155,000 | 55% | 30.1% |
| Bournemouth, Christchurch & Poole | £338,000 | 17% | 50.7% |
| North Kesteven | £245,000 | 22% | 61.4% |
Islington and Darlington make the point on their own. One is prime inner London; the other is a former railway town in the North East that costs less than a third as much. Their terraces rate almost identically — around 28% in band C or above — because roughly seven in ten of them were built before 1930. Price tells you nothing about the certificate; the calendar tells you almost everything.
Older terraces are harder to move up the scale because they were built with solid walls rather than the cavity walls that became standard from the 1930s. Solid-wall insulation is more disruptive and more expensive than filling a cavity, and bodies such as the Energy Saving Trust and Historic England publish detailed guidance on the trade-offs for period and traditionally-built homes. None of that is captured by a district's headline efficiency figure.
What it means for a buyer
The practical point is narrow and specific: a district's average EPC tells you very little about the particular terraced house on your shortlist, especially in the districts above. Comparing running costs between a modern estate and a Victorian terrace in the same town is comparing two building eras, not two addresses.
A few caveats matter. EPC bands are a modelled assessment of a building's fabric, not a measured bill; the pound figures printed on a certificate rest on dated cost assumptions and should be read as directional, not literal. The analysis here covers the 10.7 million homes with a current, address-matched certificate — a large and representative sample, but not every home. And the ranking includes only districts with enough terraces (at least 1,500) to give a stable percentage.
This is general information, not advice. The efficiency and running cost of any individual home depend on its specific fabric, heating system and condition — read the actual certificate, and speak to a qualified adviser before acting.
See it for a real street
Homecost calculates the all-in cost of any UK home — mortgage at the current Bank of England quoted rate, council tax, and energy drawn from the property's own EPC. Try a terraced heartland like Middlesbrough (TS1), or read the regional deep-dives on Teesside and County Durham's cold terraces and East Lancashire's retrofit gap. For the wider view, see how cheap-to-buy districts map against energy efficiency, what the EPC A–G scale actually means, or browse more market analysis. All figures are drawn from HM Land Registry Price Paid data and the EPC Open Data register — see how we work.