Chapter navigation
Overview
At the time of this report, Housing supply: indicators of new supply statistics, Housing supply: net additional dwellings, and Dwelling stock (including vacants) estimates are published by the Ministry for Housing, Communities and Local Government (MHCLG) as accredited official statistics.
The statistics measure recent housebuilding activity, annual changes in housing supply, and the size and tenure of dwelling stock in England. They support housing policy, economic analysis, local planning, public debate, and the assessment of housing delivery. OSR last reviewed these statistics in 2018.
This report presents findings and recommendations from our review of whether the three sets of statistics continue to meet the standards of Trustworthiness, Quality and Value set out in the Code of Practice for Statistics (the Code).
Why we did this review
These high-profile statistics provide complementary measures of housing supply and stock in England. The annual net additional dwellings statistics are MHCLG’s most comprehensive measure of housing supply and are used to update the annual dwelling stock estimates. They are also the key source used for measuring annual progress against the UK Government’s target to build 1.5 million homes by the end of this parliament.
The quarterly indicators provide earlier information about housebuilding activity. They have recently evolved to include more timely estimates of net additional dwellings which draw on Energy Performance Certificate (EPC) lodgements. These new estimates provide information for periods not yet covered by the annual net additional dwellings statistics and allow housing delivery to be estimated from the beginning of the current parliament. They are increasingly used across government and by the media.
OSR reviewed the underlying EPC official statistics in May 2026 and found that they are produced in line with the Code. The review did not assess the suitability of using EPC lodgements to estimate net additional dwellings. The review also highlighted forthcoming changes to the EPC policy area and related metrics.
The UK Government recently changed the way it calculates housing targets for all English local authorities as part of the National Planning Policy Framework. The framework now uses local-level dwelling stock estimates as the baseline for housing needs in each area.
Findings
The three sets of statistics answer distinct but related questions and form a coherent and valuable suite of statistics. The housing statistics guide supports users to navigate and use the different publications, though it states that it was last updated in 2019 and information about the new EPC-based net additional dwellings estimates in the guide is limited and not prominent.
New EPC-based net additional dwellings estimates are published prominently within the quarterly and annual housing supply statistics. The estimates combine the latest annual net additional dwellings statistics with more timely monthly and weekly EPC data. Their timeliness allows for an earlier assessment of housing delivery than the annual statistics, and they are increasingly used by government and the media.
The status of the new EPC-based net additional dwellings estimates and governance surrounding their development is unclear. Published information about the EPC-based estimates’ methods, limitations, and uncertainty is limited and mainly based on an earlier ad hoc analysis. There is currently no structured evaluation of the estimates to assess the reasons for, and extent of, EPC data revisions, including on regional estimates, or to inform guidance and future changes to methods.
Each publication is supported by technical notes containing substantial information about coverage, revisions, and limitations, using OSR’s Quality Assurance of Administrative Data (QAAD) Toolkit. However, some important information about uncertainty and the impact of revisions is not sufficiently prominent in the main releases. This includes uncertainty associated with the EPC-based estimates and how revisions to annual net additions and periodic Census reconciliation can affect the dwelling stock series.
The team engages regularly with government, including ONS, industry bodies, data suppliers, housing experts, and the media, but there is scope to broaden this engagement.The contribution of user engagement and user feedback to statistical development is not consistently visible in the published supporting material.
The statistics are supported by strong departmental governance and quality assurance processes which are reflected in a variety of departmental policies and standards. The use of multiple sources to reconcile and validate the statistics is a particular strength.
Our judgement
OSR concludes that these accredited official statistics meet the standards of Trustworthiness, Quality and Value as set out in the Code of Practice for Statistics. The suite shows many strengths, including triangulation across multiple housing data sources, and the transparent communication of limitations and revisions for established measures.
The new EPC-based net additional dwellings estimates are a valuable innovation that addresses a clear user need for more timely evidence. However, their growing prominence creates a corresponding need for clearer labelling of their status, a stronger published quality assurance and evaluation approach, and updated guidance on their appropriate use within the wider statistics suite.
We have identified one requirement that MHCLG must fulfil in order for its statistics to remain accredited official statistics (called National Statistics in the Statistics and Registration Service Act 2007).
Requirement 1:
To ensure users understand the status of the new EPC-based net additional dwellings estimates and can use them appropriately, MHCLG must clearly label their status. This should include an explanation of whether the estimates are official statistics, official statistics in development or another type of analysis.
In addition to the requirement, we have identified four recommendations that MHCLG should address to further enhance the statistics’ Trustworthiness, Quality and Value:
Recommendation 1:
To ensure users understand the quality of the EPC-based net additional dwellings estimates and can use them appropriately, MHCLG should strengthen the supporting methods and quality information to explain the methods, assumptions, limitations and uncertainty of the estimates more fully, in a dedicated document or clearly identifiable section.
Recommendation 2:
To help users assess the reliability of the EPC-based net additional dwellings estimates and understand plans for further development, MHCLG should publish its approach to evaluating their performance and fitness for purpose. MHCLG should publish further details of the nature, extent and direction of the revisions between initial and revised estimates, and annual outturns, including regional differences where possible, and consider the potential impact of planned changes to EPC policy and metrics.
Recommendation 3:
To help users select and interpret the different housing supply measures appropriately, MHCLG should update its housing statistics guide and include more prominent guidance on the EPC-based net additional dwellings estimates. MHCLG should ensure that the most important information about uncertainty, coverage and revisions is prominent in each statistics release andapply OSR’s QAAD Toolkit at the data source level for sources with known quality risks that could materially affect the statistics.
Recommendation 4:
To ensure that the development of the statistics reflects user needs and to demonstrate the public value of the statistics, MHCLG should engage proactively with a broader range of users and publish the main findings and resulting actions from their feedback. MHCLG should involve users in the future development of the statistics, including seeking views on preferences, understanding and use of the EPC-based net additional dwellings estimates.
Next steps
OSR expects MHCLG to address the requirement the next time the EPC-based net additional dwellings estimates are published. This is currently scheduled for November 2026. OSR expects MHCLG to publish an update by March 2027 setting out how it plans to address the additional recommendations. We will review progress following that update and continue to engage with the team as it clarifies the status of the EPC-based estimates, evaluates their performance, and updates relevant guidance for users.