IN Brief:
- England recorded approximately 46,000 EPC lodgements for new dwellings in Q3 2026, 4% below the same quarter of 2025.
- Six of nine regions registered year-on-year falls, led by Yorkshire and the Humber at 19%.
- EPC lodgements track new-builds, conversions, and changes of use but are not a direct count of net additional homes.
BCIS has recorded a 4% year-on-year fall in Energy Performance Certificate lodgements for new dwellings in England during the third quarter of 2026. Ministry of Housing, Communities and Local Government statistics show approximately 46,000 registrations between July and September. The 12-month total to September was around 199,000, up 1% on the preceding year. A weaker latest quarter has therefore followed a period in which the rolling annual figure remained marginally higher than a year earlier.
EPCs are lodged when properties receive an energy assessment, making the register a relatively frequent indicator of new housing supply. The new-dwelling category includes newly built homes, conversions, and changes of use. It also records an energy-efficiency rating, giving the dataset a wider function than measuring building activity alone. Registration can occur at a different point from a contractor’s practical completion, and a certificate may be generated from a design assessment before a later record is lodged for the finished property. The government applies data cleaning to reduce double counting within the published new-dwelling figures.
The official measure of net additional dwellings uses a different calculation, adjusting additions for demolitions and other changes to the housing stock. EPC registrations enter that wider statistical picture but cannot, on their own, establish the net increase in homes. Their value is the frequency of reporting and the ability to compare activity across equivalent periods and regions. For construction businesses, the data indicate where properties are reaching the assessment and registration stage, while planning decisions, project starts, and completion statistics track earlier or different points in delivery.
Regional results show that the quarterly reduction was widespread rather than confined to one market. Six of England’s nine regions recorded fewer new-dwelling certificates than a year earlier. Yorkshire and the Humber fell by 19%, the West Midlands by 15%, and the East of England by 9%. Both the North West and South West recorded an 8% decline, with the North East down 1%. These percentages describe changes in certificates lodged, not the same percentage change in construction site starts or signed contracts.
The South East recorded the strongest increase, with lodgements up 9% on the corresponding quarter of 2025. The East Midlands rose by 4%, while London’s total was unchanged. The regional differences mean the national decrease was driven by uneven local movements, with a more pronounced fall in Yorkshire and the Humber and the West Midlands than in other areas. Planning pipelines, project sequencing, and the timing of energy assessments can each affect the distribution of certificates across a quarter, so the regional figures are best compared with subsequent releases using the same definitions.
Provisional building-control counts for April to June 2026 provide a separate view of housing delivery. The figures cited by BCIS recorded 37,530 new-build completions in England, an increase of 11.2% from the preceding quarter and 1.4% from a year earlier. Housing starts reached 38,510, up 11.8% quarter on quarter and 20.7% year on year. These statistics cover an earlier period and count different activities from the new-dwelling EPC data. They indicate how many homes began or completed construction in the second quarter, while the EPC release records qualifying registrations in the third.
The July–September comparison is made with the same quarter in 2025 because the government’s quarterly EPC figures are not seasonally adjusted. The frequency of registration can fluctuate with the timing of assessments and handovers, particularly when developments complete in stages. As a result, the 4% fall establishes that fewer qualifying new-dwelling EPCs were lodged in this quarter than a year earlier, without showing how many construction projects began or how many net homes were added. The underlying rolling total offers another time frame, but it cannot replace a consistent comparison of individual quarters.
Developers must bring projects through land acquisition, planning, financing, groundworks, and construction before the resulting dwellings are normally recorded in the supply statistics. Changes in funding costs, labour availability, construction prices, and sales demand can affect the pace of that sequence. Dr David Crosthwaite, BCIS chief economist, identified financial viability and sufficient buyer demand among the constraints on turning permissions into new homes. The effect on EPC registrations may emerge only after developments have moved through earlier site stages, which helps explain why planning approvals and EPC totals need to be analysed separately.
BCIS chief economist Dr David Crosthwaite linked the weaker quarterly result to constraints on financially viable construction and demand, while cautioning against judging a sustained slowdown from one quarter alone. He noted that the next releases will help establish whether the fall is temporary. That is consistent with the annual moving total, which remains slightly higher even though the latest three-month period has declined, and with the separate starts data showing earlier pipeline activity in the second quarter.
The housing target of 1.5 million net additional homes in England over the parliament will be assessed against net housing additions, while EPC lodgements provide an earlier and more frequent signal of supply. The government is expected to publish its next quarterly EPC figures, covering October to December 2026, on 28 January 2027. That release will allow a new year-on-year comparison after the six-region decline recorded in the third quarter. Housing starts, building-control completions, and the eventual net additions statistics will provide the complementary measures of what has reached site and what has entered the housing stock.



