House building data, UK QMI

• National Statistic: no

• Data collection: administrative data

• Data source: House building data from Department for Levelling Up, Housing and Communities, Welsh Government, Scottish Government, and the Northern Ireland Department of Finance; based on data collected by National House Building Council Inspectors and other bodies

• Frequency: quarterly (UK and country) and annual (local authority)

• How compiled: aggregation of source data

• Geographic coverage: UK, Great Britain, England, Wales, Scotland, Northern Ireland, local authority level

Related publications: House building data, UK, House building, UK: permanent dwellings started and completed by local authority

This Quality and Methodology Information report contains information on the quality characteristics of the data (including the European Statistical System five dimensions of quality) as well as the methods used to create it.

The information in this report will help you to:

• understand the strengths and limitations of the data

• learn about existing uses and users of the data

• understand the methods used to create the data

• help you to decide suitable uses for the data

• reduce the risk of misusing data

• This release provides data on newly built dwellings in the UK, presented separately for dwellings started and completed and split by sector (private enterprise, housing associations or local authorities).

• Individual country-level statistics are available separately from the producers of each country's statistics; we create the Great Britain (GB) and UK data by aggregating previously published data from the constituent three or four countries.

• Each release provides the most up-to-date data available for each geographical area and time period at the time of publishing; data for individual countries may cover more recent time periods.

• The UK and country level dataset contains three sets of tables, with estimates available for different periods (quarters, financial years, calendar years).

• These data are broadly comparable across the UK and available for flexible time periods with a small amount of lag; however, for policy purposes in England, DLUHC's annual "Housing supply: net additional dwellings" is the quoted measure.

This quarterly publication compiles house building data published by the four UK countries to allow users to more easily compare over time and between countries or local authorities, as well as publishing figures at a Great Britain (GB) and UK level.

A dwelling is counted as "started" on the date work begins on the laying of the foundation. It is counted as "completed" when it becomes ready for occupation or when a completion certificate is issued.

"Sector" refers to new dwellings constructed by private enterprises, housing associations, or registered social landlords and local authorities. Other data providers may use the term "tenure".

Uses and users

The datasets in this release give timely information relating to starts and completions of new dwellings in the UK. They are used to inform government housing policy, and for housing market analysis and research. They are also used by forecasters and decision-makers, including at the Bank of England. The compilation of the UK data in this release also allows international comparisons of housing supply.

Strengths

• Data for all four UK countries are available in a single place.

• Data are timely and available with little lag.

• Data are available for quarters, financial years and calendar years.

• Data are broadly comparable across the UK, providing a measure of activity for house building.

Limitations

• Unfortunately, the coronavirus (COVID-19) pandemic affected quarterly data collection in Wales in 2020 and 2021; financial and calendar year figures are available for this period for Wales and, therefore, the UK, but quarterly figures are not available for April to June 2020 until January to March 2022.

• This data does not cover the full extent of house building activity or housing supply in England and Wales; for example, conversions and demolitions, house building activity not covered by National House Building Council (NHBC) inspections, and house building implied by the decennial censuses are not included here, but are included in another source for England described in Section 5: Quality characteristics of the house building UK data.

• The sector of the home can differ between the time of inspection and when it is inhabited, leading to uncertainty over sector breakdowns.

• Start dates can be uncertain, and are defined differently in the Northern Irish data to other countries.

This section provides a range of information that describes the quality and characteristics of the data and identifies issues that should be noted when using the output.

Relevance

These statistics are relevant to a wide range of users who are interested in house building in the UK or its constituent countries or local authorities. 

However, the statistics presented in this release do not cover all house building in England. Additional data sources are used to produce the annual Department for Levelling Up, Housing and Communities' (DLUHC) Housing supply: net additional dwellings series, which captures additional house building based on annual returns from local authorities and the Greater London Authority, and censuses. Therefore, these include homes not inspected by The National House Building Council (NHBC) inspectors, which are typically homes built by small house-builders, or those not for private sale. 

The difference varies from an additional 4% above the estimates in this release (in 2009 and 2010) to 27% (in 2018 and 2019).

DLUHC's net additional dwellings series, and the Scottish Government's Supply of new housing series, also provide estimates of all new housing supply by including conversions, demolitions and changes of use. Therefore, this "net additional dwellings" approach is the more complete picture for looking at the provision of new homes. 

The figures in this release, however, are available earlier, for a variety of time periods, and are broadly comparable across the UK.

Accuracy and reliability

As the data are based on administrative sources, they usually present a complete picture of the house building activity covered by the main types of building inspector in the UK. However, occasionally data must be imputed. 

Housing may occasionally be allocated to the incorrect sector. 

There are some inconsistencies with historic data which we have identified and are investigating. 

Imputations

Scotland's data contain imputed estimates for a small number of quarters for estimates for new private house building starts in Highland local authority.

A small amount of data for England are missing and, therefore, imputed. 

Uncertainty over final sector

House building figures reflect the sector of the developer building the dwelling, rather than the intended final sector. This may lead to an undercount of housing association and local authority starts and completions recorded in these tables, and a corresponding overcount of private enterprise figures. This problem is more likely to affect starts than completions.

Wales's new house building data are based on the reports of local authority building inspectors and the NHBC, a private approved inspector; it does not include information from other private approved inspectors.

It is sometimes difficult for building control officers and the NHBC to identify the intended final sector of the property (the basis for the sector information). This may lead to an under-count of social sector new house building and an over-count for the private sector in Wales. Within the social sector, it may also lead to an under-count of local authority new house building and an over-count for the housing association sector. Therefore, the sector data should be treated with caution.

Historic data

We have identified some inconsistencies with historic data. As a result, small clusters of figures by sector may not add up to the correct total between 1972 and 1992. Only a small proportion of figures are affected, and differences are small. We are working with the data producers to see whether we can correct these. However, for now, we have decided to leave them unchanged to be consistent with the data sources.

Coherence and comparability

There are some differences in the coverage of house building by country. This is because the house building inspection systems work differently across the UK, specifically:

• England - Housing supply: indicators of new supply statistics (live table 213): local authority new build form returns; National House Building Council (NHBC) data; approved inspector data returns

• Wales - New house building statistics: local authority building inspectors; NHBC data

• Scotland - Housing statistics quarterly update: new housebuilding and affordable housing supply: local authority new build form returns; Scottish Government Affordable Housing Supply Programme

• Northern Ireland - New dwelling statistics: District Council Building Control

The sources for Scotland and Northern Ireland are the sole sources of building control statistics, and all homes are required to be registered with these for access to services. Therefore, they are expected to have complete coverage.

Identification of housing association homes in Scotland

In Scotland's housing statistics, it is not possible to directly identify housing association starts. Therefore, housing association new build approvals, from the Scottish Government Affordable Housing Supply Programme, are used as a proxy for housing association new build starts.

Delays in reporting and recording site starts for Housing Association homes in Scotland tend to be more significant, and as a result the data quality for new build starts is less robust. Therefore, approvals are used as a proxy for starts for housing association new build. This proxy introduces a clear spike in the quarterly time series of starts, in Quarter 1 (Jan to Mar) of the year, when a large proportion of housing association new build has traditionally been approved. This spike is not only observed for housing association starts, but it is pronounced enough to be clearly seen for total starts incorporating all tenures.

The housing association new build statistics record approvals, starts and completions, for whole developments (or phases of larger developments). This differs from the figures collected for other sectors which relate to starts and completions of individual dwellings. This means that the housing association new build statistics contain a greater amount of aggregation and the quarterly count of starts and completions may be subject to some inaccuracy.

Differences in measuring start and completion dates

Northern Ireland's new dwelling data count the number of starts and completions inspections conducted by building control officers in each council area. The date of a new dwelling start is the date that the first building control inspection takes place, rather than when the foundations are laid.

With the NHBC completions figures, a registered house is deemed completed when the NHBC building inspector considers the house is satisfactorily completed in respect of NHBC's technical requirements. This is normally prior to (but generally signifies the house is ready for) occupation. The Scottish Government completions statistics regard a dwelling as complete when it is ready for occupation, whether it is in fact occupied or not and, if occupied, whether or not a full completion certificate has been issued.

Further information is available in the methodology documents produced by each country (see Section 7: Other information).

Timeliness, punctuality, and revisions

This publication uses timely data, typically publishing about three to six months after the reference period. Between publications, more timely data can sometimes be found on the individual countries' webpages. This is one of the advantages of this publication over the more complete annual Housing supply: net additional dwellings publication for England.

Inspection reports can be processed long after the period in which the start or completion was observed. A building control inspector may not visit a site following completion until the following quarter. The completions would subsequently be allocated to the correct quarter of completion, resulting in a revision to the previously published statistics.

Therefore, this release reflects the revisions made in the source data from each country on each publication. This ensures that statistics for the UK and Great Britain (GB) are consistent with the latest statistics, at time of publication, in each constituent country.

Revised figures are marked with an [r] in the publication.

Earliest time periods available (starts and completions, and by sector, unless stated)

Data by country, quarterly

Worksheets: 1a to 1f

• Great Britain: only featured in this section.

• Wales: April to June 1974 until January to March 2020; April to June 2022 onwards.

• England, Scotland, and Northern Ireland: January to March 1978 onwards.

Data by country, by financial year

Worksheets: 2a to 2e

• England, Northern Ireland, Scotland, and Wales: year ending March 1970 onwards.

• Statistics on starts for Wales are not available split by sector from the financial year ending 2012.

Data by country, by calendar year

Worksheets: 3a to 3e

• England: 1946 for completions, 1978 for starts.

• Wales: 1946 for completions, 1974 for starts.

• Scotland: 1946 for completions, 1980 for starts. 

• Northern Ireland: 1949 for completions, 1978 for starts.

Data by local authority, by financial year (earlier time periods are available from the original sources)

• England, Scotland, and Wales: financial year ending March 2010 onwards.

• Northern Ireland: financial year ending March 2015 onwards.

Geography (including list of changes to boundaries)

For the local authority series, the latest geography boundaries, names, and codes are used for the entire time series. These can be found on the Open Geography portal. This means the entire dataset is updated in line with the latest geography changes, so previous local areas that have experienced changes, such as mergers with other areas, will not be shown on the datasets. If you require to access data for an area before it had changed, the previous year's publication will still contain the data we had published for the previous year.

Administrative data is collected from online resources, provided separately by the individual countries of Northern Ireland, Scotland, and Wales. Data for England is provided by the Department for Levelling Up, Housing and Communities (DLUHC). Further information can also be found in Section 1: Output information and Section 4: Quality summary.

Data is passed through a formulaic process that takes data provided from the source and aggregates quarterly figures to produce the required figures for the UK and Great Britain (GB). Data is tested for any revisions between releases, allowing us to validate changes and identify them by placing the following shorthand marker in the final dataset [r].

Source data will also have its own symbol and styling replaced to represent data under the count of 10 with the shorthand of [low] in the data set, while where no data was collected we use [x1] to represent this in our dataset. Where data has been aggregated for the UK or GB, we use [x2] to show where data is missing in part from one or more countries and hence a total can't be calculated.

We undertake various quality assurance testing methods on the data, including checks to evaluate that data points produced in part, aggregate correctly as a whole, while accounting for any inaccuracies in rounding.