1. Main points

  • There are multiple estimates of the number of empty homes in England and Wales, which tell us different things.

  • Council Tax numbers of empty homes in 2021 were lower than figures from census data and this difference was more pronounced in Wales compared with England. 

  • There are definitional, methodological and practical reasons for the differences in measures of empty homes, with the main differences being long-term versus short-term vacancies, coverage and addressing, caravans and coronavirus (COVID-19) impact.

  • If you are interested in the widest possible coverage of empty homes, including those that have only been temporarily vacant and dwellings such as caravans on holiday parks not paying Council Tax, we recommend using census data.

  • If you are interested in exploring trends over time, or those that are registered for Council Tax, we recommend using Council Tax or English Housing Survey data.

Back to table of contents

2. Overview of empty homes data

Housing is an important topic for the public and government. Central, local and devolved government want to have a good understanding of how many homes are empty across England and Wales, so that this can feed into decisions on local housing needs.

Some data sources refer to empty homes as vacant homes. An empty or vacant home typically refers to a residential property that has no one usually living there, although definitions vary across sources. Government analysts have been investigating differences between data sources available for empty homes.

The purpose of this explainer article is to:

  • outline data available on the number of empty homes in England and Wales

  • explain the main differences in the number of empty homes across data sources

  • outline the definitions of empty homes from each data source

  • explain which data source is suitable for different purposes

Back to table of contents

3. Available data on empty homes

The Office for National Statistics (ONS) released Census 2021 data on the number of vacant and second homes in England and Wales. These statistics estimated that in March 2021 in England, there were 1,507,100 unoccupied dwellings, of which 1,352,130 were vacant dwellings and 154,970 were second homes. It was estimated that in Wales, there were 120,450 unoccupied dwellings, of which 102,875 were vacant dwellings and 17,575 were second homes.

The Ministry of Housing, Communities and Local Government (MHCLG) release Council Taxbase statistics for England annually. In September 2021, there were 653,025 dwellings classed as vacant in the Council Taxbase and a further 253,357 dwellings classed as second homes.

MHCLG also produce a regular estimate of the number of vacant dwellings based on the English Housing Survey (EHS). Data are not available for 2021 because of the coronavirus (COVID-19) pandemic, but in 2022 there were an estimated 1,139,000 vacant dwellings in England.

The Welsh Government uses annual local authority Council Tax dwellings data to infer the number of empty homes chargeable for Council Tax in Wales. In the financial year ending March 2021, there were 24,671 chargeable empty properties and 24,423 chargeable second homes.

For more information on each of the data sources and quality, please see Section 7.

Estimates of the number of empty homes tell us different things, with Council Tax numbers of empty homes typically being lower than figures from census data. This is more pronounced in Wales compared with England.

Back to table of contents

4. Reasons for the differences in the number of empty homes

There are some methodological and practical reasons for the differences, which will be explained in more detail, but they include long-term versus short-term vacancies, coverage and addressing, caravans and coronavirus (COVID-19) impact.

Long-term versus short-term empty properties

In census and English Housing Survey (EHS) data, the property is assessed as empty on a particular day (either Census Day or the day of the interviewer's visit for EHS) for example, by checking with neighbours.

In Council Tax data, the vacant dwelling is assessed for tax discounts or premiums, which typically only applies when a property has been empty for a longer time, for example, six months.

If a dwelling is empty for a short time period, such as in-between rental periods or house sales, the dwelling would be more likely to appear as vacant in census and EHS data than in Council Tax data. This is most likely to affect estimates in areas with high numbers of rental properties such as Westminster or Kensington and Chelsea, and tourist areas such as the Isles of Scilly.

Coverage and addressing

The Office for National Statistics (ONS) made every effort to ensure Census 2021 counted everyone, so any address that may have someone living in it was sampled. It was important to limit the number of non-residential addresses that were included, as well as limit the under-coverage. Where there was uncertainty over whether an address was residential or non-residential, the decision was generally made to include these addresses. However, this meant that there was an average over-coverage of 2.1% in England and 1.7% in Wales, with examples such as dwellings that were still being constructed.

These would have been counted as unoccupied, vacant dwellings. Based on the average over-coverage across local authorities, it is estimated that around two in five vacant dwellings in England and one in four vacant dwellings in Wales were because of oversampling. With the majority (61%) of these later found to be under-construction.

To produce the number of vacant dwellings from census, it required identifying unoccupied dwellings (those with no one usually living there) and then combining intelligence from census and administrative data sources to split these into truly vacant dwellings and second homes. Address matching was required to link data sources together, which adds extra complexities. Address matching is widely used but can be less accurate for specific accommodation types including caravans, flats, maisonettes and apartments; as well as addresses in the Welsh language. This may mean that it is more difficult to allocate unoccupied addresses with poor address matching to either a second home or truly vacant dwelling.

Caravans

Council Tax statistics in England and Wales are based on dwellings that are eligible for Council Tax, including those that have discounts. Some dwellings are usually not liable for Council Tax unless it is someone's main residence, such as some caravans on holiday parks, so are not included in Council Tax statistics.

For census data, caravans will have been included as a dwelling, and many more than usual were vacant in March 2021 because of coronavirus (COVID-19) restrictions at the time. This may result in larger differences in the number of vacant dwellings between data sources in areas with high numbers of caravans or mobile homes, such as East Lindsey or Denbighshire.

Around three in five (95,500 out of 168,300) caravans or other mobile or temporary structures in Census 2021 were also in domestic Council Tax data in England, for Wales it was lower at two in five (4,800 out of 12,800).

Coronavirus impact

As Census 2021 was carried out during the coronavirus (COVID-19) pandemic, it is possible there may be more vacant dwellings recorded, although it is difficult to quantify the impact. Living situations may have changed because of restrictions; for example, overseas students or those privately renting may have moved back in with family members and caravan parks were shut leaving more unoccupied. The pandemic may have caused delays in completing building projects, inflating the number of properties still "under-construction" at the time of the census and recorded as empty.

Back to table of contents

5. Definitions of empty homes

There are varying definitions of empty or vacant homes, which are useful to know to help understand the differences between the numbers provided.

In the census release produced by the Office for National Statistics (ONS), a "truly vacant" dwelling refers to a residential dwelling that is unoccupied on Census Day (21 March 2021) with no usual residents, has no indication of being used as a second home and is not inhabited by short-term residents. A property would be "truly vacant", if the property was not where anyone spends most of their time (that is, place of usual residence) and there were no usual residents in England and Wales who stayed at the property for more than 30 days, as well as at least one of the following: 

  • a census response confirms the address is truly vacant - see guidance around Empty properties

  • a housing association provided details around the property being vacant 

  • a Council Tax exemption or discount indicates the property is vacant 

  • utilities data indicate a property is unoccupied or de-energised 

In measures of empty homes based on Council Tax data (Council Taxbase statistics for England and Council Tax dwellings for Wales), vacant dwellings are defined as empty properties as classified for Council Tax purposes. They include all empty properties liable to pay full Council Tax or with a premium and those that are empty but receive a Council Tax discount. They exclude empty properties on the valuation list that are exempt from paying Council Tax.

In the English Housing Survey (EHS), an assessment of whether or not a dwelling is vacant is made at the time of the interviewer's visit. Clarification of vacancy is sought from neighbours. Surveyors are required to gain access to vacant dwellings to undertake full inspections. Both properties in-between lets and those that are vacant for a longer period are classified as vacant in the EHS.

Back to table of contents

6. Which data to use, and when

These are all official statistics, which use different definitions and methodologies that result in different measures of empty homes. When deciding which figures to use we would recommend that you take into consideration what you are wanting to measure and the different strengths and weaknesses of each.

If you are interested in the widest possible coverage of empty homes, including those that have only been temporarily vacant and dwellings such as caravans on holiday parks not paying Council Tax, we recommend using census data.

For exploring trends over time, Council Tax data or the English Housing Survey (EHS) provide a longer time series of data. If you are interested specifically in those that are registered for Council Tax and how many homes have been empty for a long time, we recommend using Council Tax data.

Back to table of contents

7. Data sources and quality

Number of vacant and second homes, England and Wales: Census 2021

The Office for National Statistics (ONS) released Census 2021 data on the number of vacant and second homes in England and Wales. This was the first time the ONS sought to allocate unoccupied dwellings into vacant or second homes. Estimates were produced by combined intelligence from census data and administrative data sources, as the split of unoccupied dwellings is difficult to determine from census data alone.

To produce estimates of vacant dwellings from the census, address matching was required to link data sources together, which is widely used but can be less accurate for specific accommodation types including caravans, flats, maisonettes and apartments; as well as addresses in the Welsh language. This may mean that it is more difficult to allocate unoccupied addresses with poor address-matching to either a second home or truly vacant dwelling from available data sources. If an unoccupied address is not allocated from census and administrative data sources, the allocation is provided through a statistical model. See Figure 1 in our bulletin Number of vacant and second homes, England and Wales: Census 2021.

The number of dwellings in Census 2021 data is derived from the sampling address frame. More information on quality can be found in our article Evaluation of addressing quality: Census 2021. There were many changes to the addressing of the census between 2011 and 2021, including data sources and methodological changes. In 2011, only unoccupied dwellings where responses identified the household space was unoccupied or was solely occupied by short-term residents were included. In 2021, additional dwellings where no census response was received but were determined as vacant were counted. Therefore, the counts of unoccupied dwellings in 2011 and 2021 are not directly comparable.

Council Taxbase statistics for England

The Ministry of Housing, Communities and Local Government (MHCLG) release Council Taxbase statistics for England annually derived from a snapshot of dwellings liable for Council Tax. A snapshot of the number of dwellings is taken on the second Monday in September by the Valuation Office Agency (VOA), then local authorities report the number of dwellings in different types of discounts, premiums and exemptions as at the first Monday in October. From this the number of dwellings classed as vacant are derived.

A vacant property (that is not exempt) is classified as a property that is empty and unfurnished.

Local authorities have discretion over what discounts or premiums will be applied, but empty properties receiving no discount are also reported. Local authorities provide information on the different exemption categories, which include some where the property is empty. Where local authorities award zero discounts for empty properties, there is less incentive for owners to report their property as empty, which could mean an undercount in Council Tax data.

Council Tax dwellings for Wales

The Welsh Government release annual local authority Council Tax dwellings data for Wales. The Valuation Office Agency (VOA) provide a snapshot of the number of dwellings in each Welsh local authority as at 31 October. The Welsh Government collect taxbase information from local authorities on the Council Tax dwellings return (CT1), where local authorities produce estimates for the upcoming financial year, using the valuation list and accounting for any changes likely to occur in that year. This provides information on the number of dwellings eligible for Council Tax, which includes the number of properties eligible for Council Tax discounts and exemptions. This information is used to infer the number of empty homes chargeable for Council Tax in Wales.

The Welsh Government also publish separate data on empty homes exempt from paying Council Tax, for example, if the dwelling is new and undergoing structural changes or if the occupier is in hospital.

English Housing Survey

The English Housing Survey (EHS) is a continual national survey commissioned by the Ministry of Housing, Communities and Local Government (MHCLG). It collects information about people's housing circumstances and the condition and energy efficiency of housing in England. Each year a sample of addresses is drawn at random from a list of private addresses held by the Royal Mail.

If an address is vacant this is identified at the time of the interviewer's visit, and further clarification is sought from neighbours and full inspections of the property. As surveyors are required to gain access to vacant dwellings and undertake full inspections, data on vacant dwellings were not available in 2021 where this was not possible during the coronavirus (COVID-19) pandemic. Estimates of vacant dwellings are available prior to the pandemic and for 2022 onwards.

Collaboration

This article has been produced in collaboration with the Ministry of Housing, Communities and Local Government (MHCLG) and the Welsh Government; we would like to thank them for their support.

Back to table of contents

9. Cite this article

Office for National Statistics (ONS), released 9 August 2024, ONS website, article, Comparing empty homes statistics in England and Wales

Back to table of contents

Contact details for this Article

Housing Analysis team
better.info@ons.gov.uk