Affordable housing is a topic that consistently attracts significant public, political, and media interest in Wales. Whether it’s questions in the Senedd, scrutiny from the press, or queries from local authorities and the public, there’s a clear demand for timely, reliable information on the amount of affordable housing in Wales.

Darllenwch y dudalen hon yn Gymraeg

But as anyone who’s worked with these figures knows, affordable housing statistics can be far from straightforward. Different definitions of affordable housing are used for different purposes. For the affordable housing statistics published this week, two main definitions are used.

Two definitions, two perspectives

The first is the TAN2 definition (from Technical Advice Note 2: Planning and Affordable Housing), which for many years has been the standard for official statistics. This definition is broad, covering all additional affordable housing units delivered – whether through new build, purchase, or conversion. It’s designed to provide a comprehensive, consistent picture of affordable housing provision across Wales.

The second is the target definition used to measure progress against the Welsh Government’s commitment to deliver 20,000 new low-carbon homes for rent in the social sector by May 2026. This is narrower: it focuses specifically on homes for social rent, intermediate rent, and shared ownership delivered by registered social landlords, local authorities, and some private providers. Notably, it excludes shared equity units but includes certain properties brought back into use with government funding (known as “long-term voids”).

These differences mean that, depending on the question, the “right” number can vary, and it’s important to be clear about which definition is being used.

Two data sources: Accredited official statistics and management information

As well as two definitions of affordable housing, there are also two main sources of data used in the statistics publication, each with their own strengths and weaknesses.

Accredited official statistics

Statisticians collect data annually from local authorities, registered social landlords, and National Park authorities. This data collection is used to produce accredited official statistics, meaning they have been independently assessed as meeting the highest standards of trustworthiness, quality, and value. The data undergo rigorous validation and quality assurance, but this means they come with a time lag (typically around seven months after the end of the reporting period).

Management information

This is collected throughout the year by policy teams to monitor delivery associated with government-funded housing projects. It’s drawn from a live database, updated as schemes progress, and provides a more up-to-date picture of delivery based on the information available at that time. The data goes through a different type of validation process to official statistics, but typically the overall end-of-year figures are now not too dissimilar to the official statistics at a Wales level.

Recognising the strong user demand for more timely figures, we’ve started to include management information alongside the official statistics in our annual publication on affordable housing. This approach allows us to provide both the robust, quality-assured figures that underpin long-term policy and the more current estimates that support immediate decision-making and public accountability. The Office for Statistics Regulation has supported this effort to meet user needs.

A quick guide to published data on affordable housing

Official statistics

Use:

• for robust, quality-assured figures that meet the Code of Practice for Statistics
• for monitoring long-term trends

Do not use:

• for the most up-to-date delivery figures, there’s a publication lag
• to monitor total delivery which includes “voids”, only management information captures voids

Management information

Use:

• for timely, in-year estimates and real-time monitoring of delivery
• to complement official statistics, especially when there is interest in recent progress

Do not use:

• as a substitute for accredited, quality-assured statistics
• for precise, final delivery figures; numbers can fluctuate month-to-month based on the latest available information
• for local-level data, as there are larger differences with official statistics

Understanding progress towards the 20,000 homes commitment

Owing to the strengths and limitations of each data source, measuring progress towards this target needs a mix of official statistics and management information.

For the period April 2021 to March 2025, this means using the accredited official statistics, supplemented by management information on void properties brought back into use. For a more up-to-date view on the number of homes delivered, management information covering April to September 2025 can also be used. Both sets of figures are included in the statistical report.

Data on planned delivery in 2025-26 is also included in the statistical publication.

The statistical publication does not include any data beyond the 2025-26 period. As dates get further away, there is more uncertainty in these estimates which makes it unsuitable for a statistical publication. However, management information can provide some intelligence on projects due to be delivered over a longer time period. To support transparency, data has been published separately from the statistical release on the number of homes projected to be delivered after 2025-26.

Looking ahead

The quality and reliability of management information have improved significantly in recent years. As a result, the gap between management information and official statistics has narrowed, and the two sources now provide a broadly consistent picture of delivery.

Given these improvements, we’re actively considering how we can make more use of management information in the future (similar to the statistical approaches in other parts of the UK) and whether there’s an ongoing need for two separate sets of figures. Our aim is always to provide the best possible evidence to support decision-making, scrutiny, and public understanding.

As ever, we welcome feedback from users on how we can continue to improve our statistics and the way we communicate them. Please email stats.housing@gov.wales

Stephanie Howarth
Chief Statistician

This post was updated on 25 November 2025 to remove leasing from the TAN2 definition of affordable housing. This was included in error.