Homelessness accommodation provision and rough sleeping: quality report

This monthly publication covers rough sleepers and the provision of temporary and long-term accommodation for homeless people who present to local authorities for housing support.

The data in this statistical release form the basis of evidence on temporary accommodation provision in Wales. These data are used by the Welsh Government, local authorities, homelessness agencies and other housing organisations to monitor the use of temporary accommodation across Wales.

At the start of the pandemic, the Welsh Government adopted a ‘No One left out’ approach to ensure everyone had access to a safe place to live. This stance has been maintained through the transitional legal arrangements which came into force on 24 October 2022. This ensures that anyone who is street homeless or at risk of being street homeless is now recognised in legislation as being in ‘priority need’ for housing support and emergency accommodation.

On 19 May 2025 The Homelessness and Social Housing Allocation (Wales) Bill and the associated Explanatory Memorandum was laid before the Senedd. This sets out a range of amendments to existing homelessness legislation, including the abolition of the ‘priority need’ test. The Bill has been referred to the Local Government and Housing Committee for Stage 1 scrutiny of the general principles. Further information on this process and the accompanying consultation can be found on the Senedd Cymru website.

Users and uses

The monthly information is used to inform policy; by providing data to support understanding of homelessness in Wales. This includes providing a clear rationale to secure additional funding to support local authorities in the ongoing ‘no-one left out’ approach adopted at the start of the pandemic.

Local housing authorities are both providers and users of the statistics and use the data to plan services, allocate resources, monitor performance and benchmark against other authorities. The data are also used for answering Senedd Questions, Ministerial correspondence, Freedom of Information Act 2000 cases and queries from the public. There are a variety of other users of these statistics, including researchers, academics and students. The voluntary sector also uses the statistics to monitor and evaluate housing policy and for campaigning and fundraising purposes.

A key recommendation in the Ending Homelessness Action Plan 2021-2026 was to develop a new Ending Homelessness Outcomes Framework. This data is used to measure several data indicators within the framework, enabling Welsh Government to monitor progress made towards the overarching goal of preventing and ending homelessness in Wales.

Data collection

This monthly data collection, initiated in August 2020, replaced the weekly gathering of intelligence conducted during the coronavirus (COVID-19) pandemic.

Data is provided by all 22 local authorities in Wales and is collated using both electronic and manual methods. The homelessness data returns are based on information held by the local authorities, which are mainly used to monitor and provide provision to those experiencing homelessness. Most of the data is extracted into reports from electronic systems, but not all information is fully captured by electronic systems, and manual processes are needed to verify details such as household composition, reasons for homelessness, and temporary accommodation placements.

Some local authorities use systems and processes set up to report statutory homelessness data, which tracks households rather than individuals, and undertake manual adjustments to populate the monthly data return.

Local authorities run out of hours services to support individuals who present over public holidays and weekends. These individuals are recorded into data systems when normal working hours are resumed.

At the start of each month, Welsh Government sends the data collection forms to homelessness data leads in local authorities, to gather information on the previous month. Local authorities have a minimum of 10 calendar days to complete and return these forms.

The data collection forms are provided in Excel spreadsheet format, which includes built-in guidance to assist users in accurately completing the forms. The Welsh Government has worked with local authorities to enhance the guidance, aiming to improve the accuracy and consistency of data items across different areas. More information is available in the quality section. Once completed, local authorities upload the data collection forms to Welsh Government via a secure system - Objective Connect.

The spreadsheets incorporate validation checks that allow data leads to verify certain data before submission. Additionally, data leads have the opportunity to provide contextual information to explain substantial or irregular changes in the data.

Copies of the current homelessness accommodation provision and rough sleeping data collection forms are available online.

Rough sleeping

Local authorities are asked to base their rough sleeper estimates on local intelligence. This is different to the methodology used for the national rough sleeper count, undertaken between November 2015 and November 2019.

The way in which rough sleeper information is collected varies by local authority, which affects the comparability of the data across local authorities.

Some local authorities operate targeted outreach services (run by the local authority itself or a third sector party) who regularly interact with people sleeping rough. In other local authorities, members of the public, charities and support agencies report instances of rough sleeping, either directly to the local authority or raising alerts through platforms such as StreetLink.

StreetLink is a platform that connects people rough sleeping in England and Wales to support provided by local authorities and charities. To make these connections, the platform relies on alerts submitted by members of the public and people sleeping rough.

The collected data is forwarded to the appropriate team within the local authority and stored in databases that differ across each authority. Local authorities use this data to engage with those sleeping rough, to ensure appropriate support is offered and to monitor rough sleeping within their area.

At the end of the month local authorities take a snapshot of this data to provide Welsh Government with an estimate of rough sleepers within their area. As the figures are a snapshot, individuals who slept rough during the month but were no longer sleeping rough on the last day of the month are not counted.

Validation 

Individual local authorities are responsible for providing high quality data to Welsh Government each month. The collection process is dependent on the accurate entry of information into local authority systems. The validation of data varies across local authorities with some requiring sign-off from senior management before submission to the Welsh Government, while others are collated and submitted by senior management without needing sign-off. All local authorities compare figures to previous months and review any returns that appear incorrect before sending them to the Welsh Government.

The data collection and housing statistics team are responsible for validating and undertaking reasonable checks on the data to ensure that they meet the requirements for ‘Official Statistics’.

Once the data has been securely submitted to the data collection team, it goes through further validation and verification checks, for example:

• common sense check for any missing/incorrect data without any explanation
• arithmetic consistency checks
• cross checks against the data for the previous month
• sense checks against the annual and 6 monthly statutory collections
• thorough tolerance checks

If an error is found during the validation process, the data collection team will contact the local authority directly to seek resolution. In most cases, the local authority will correct the figures within their return form and securely resubmit the data to Welsh Government. However, in certain cases where errors are small, the local authority will provide Wesh Government with the updated figure/s and the data will be updated manually by the data collection team.

If a local authority is unable to submit their collection form on time, fail to respond to data collection communication, or if a data issue cannot be resolved before the publication, a decision will be made by the data collection team on what to include in the release. This usually involves imputing the previous month’s figure, though the method may vary depending on the circumstances.

The decision will be shared with the local authority ahead of publication to allow time for a response and any imputed data will be clearly noted in the release. If the authority later identifies minor data errors, they can submit corrected figures to the Welsh Government, which will be treated as revisions in the next publication.

This ensures completeness of the data and reduces the potential effect on Wales’ trends. If the local authority provides the data retrospectively, their figures will be updated in the next release. The method used and the affected data will be highlighted within the statistical release footnotes. Further details can be found in the revisions section below.

Quality

Welsh housing statistics adhere to the Welsh Government’s Statistical Quality Management Strategy, and this is in line with the Quality pillar and principles in the Code of Practice for Statistics (UK Statistics Authority).

This monthly data collection was introduced during the coronavirus (COVID-19) pandemic initially as Management Information, and when data for August 2020 was first published in November 2020, they were published at a Wales level only. The data had not undergone the same level of quality assurance as Official Statistics and not all local authorities were able to provide comprehensive data.

In the early stages of data collection, several issues were identified that impacted the accuracy and completeness of the data. These issues included local authorities reporting household figures instead of individuals, omitting data on homeless at home individuals or those in private sector leases, and instances of underreporting that were only revised retrospectively to a certain point in time, rather than for the entire time series. These challenges highlighted the need for more detailed guidance and improved data collection processes to ensure the reliability and consistency of the information gathered.

Welsh Government analysts and policy officials have since worked closely with local authorities through a series of workshops and feedback forms to enhance the guidance provided in the data collection forms. This collaborative effort has led to the development of more detailed and comprehensive instructions, which have greatly improved the comparability, consistency, and accuracy of the data submitted by local authorities. Through this communication and input from local authorities, Welsh Government has ensured that the data collection process is both efficient and effective, ultimately leading to more reliable and robust data for informed decision-making.

In January 2024, homelessness accommodation provision and rough sleeping data was published on StatsWales for the first time. For homeless individuals temporarily accommodated at the end of the period, data for Wales is available from April 2023, however, local authority breakdowns are only available for October 2023 onwards as this is when data quality is seen as sufficient. Similarly, for occurrences of people placed into temporary accommodation and homeless individuals successfully moved into suitable long-term accommodation during the period, data for Wales is available from April 2023, however, local authority breakdowns are only available from January 2024. Data on the number of rough sleepers per local authority is published for April 2023 onwards.

Wales only level data for the period prior to April 2023 is available to download from the previous release pages. It is important to consider the data quality issues when handling the data and comparisons to April 2023 onwards are not advised.

Following a comprehensive review by the Office for Statistics Regulations (OSR), the data collection processes have now undergone the appropriate quality assurance protocols for Official Statistics. These statistics were first published as Official Statistics in March 2024.

Publication

Once the data has been finalised, the publication and charts are drafted using a reproducible analytical pipeline created in RStudio. RStudio is an integrated development environment for R, a programming language used for statistical computing and graphics. This process ensures accuracy by checking for errors such as missing data and identifying any revisions from previous months.

The statistical release includes commentary and charts at a Wales level only. A more detailed breakdown of the data by local authority is available on the homelessness accommodation provision and rough sleeping StatsWales website.

Once the release has been drafted, it undergoes an independent review and quality assurance by a fellow statistician within the housing statistics team. A final sense check by the relevant statistician is then carried out before it is published on the website.

Disclosure control and confidentiality

Disclosure control has been applied to the occurrences of people placed into temporary accommodation in this release and to accompanying data on StatsWales. All figures less than 3 have been suppressed and shown as a ‘*’. All other figures are rounded independently to the nearest 3. As a result, there may be a difference between the sum of the constituent items and the total. All percentages quoted in the report are calculated using unrounded data.

Timeliness and punctuality

All outputs adhere to the Code of Practice for Statistics by pre-announcing the date of publication through the upcoming pages on the Statistics for Wales website. Furthermore, should the need arise to postpone an output this would follow the Welsh Governments revisions, errors and postponements arrangements.

Releases are usually published on the last Thursday of each month, with a two month lag for data. For example, data for April 2024 would be published in June 2024.

Revisions

Revisions can arise from events such as late returns from a local authority or when a data supplier notifies the Welsh Government that they have submitted incorrect information and resubmits this. Occasionally, revisions can occur due to errors in our statistical processes. When a revision occurs, a judgement is made as to whether the change is significant enough to publish a revised statistical release. Where changes are not deemed to be significant i.e. minor changes, these will be updated in the following months’ statistical release. Revised data is marked with an (r) within the statistical release. 

We also follow the Welsh Government’s statistical revisions policy, details of which are available online.

Strengths

• The data is updated monthly, providing timely information that can be used to monitor trends and make informed decisions.
• The publication provides detailed breakdowns by local authority, which helps in understanding regional variations and targeting interventions more effectively.
• The data collection is comprehensive and covers various aspects of homelessness in Wales.
• Outputs have a clear focus on Wales and have been developed to meet the internal and external user need in Wales.
• Most of the data provided to Welsh Government is used within local authorities in their own reports, for monitoring and informing service delivery and case management.

Limitations

• The data captures information about individuals who engage with local authorities. It is likely to underrepresent the true scale of homelessness in Wales, as not everyone who is homeless will engage with their local authority.
• Current electronic systems vary across local authorities and are limited in terms of what information can be easily accessed, resulting in manual data retrieval processes being relied upon.
• There will be a lag in recording data in some local authorities where information from out of hours services provided at weekends and bank holidays are not registered until the next working day. This is likely to have a bigger impact where bank holidays fall at the end of the month, or the month ends on a weekend.
• Rough sleeper estimates aren’t comparable across local authorities due to the different approaches they use to record the information.
• The short time between the end of the reporting period and the data being published potentially means less accurate data as the local authority may not be able to obtain accurate information from all partners ahead of reporting deadlines. For example, local authorities can only record an individual no longer being accommodated when the accommodation provider has notified them. A greater lag between the end of the reporting period and the data deadline will result in more accurate date.
• Data quality has improved over time. We saw an improvement in data quality from April 2023 and have therefore started the Wales’ data series from this time point. This limits the ability to analyse long term trends, and not all data is available at local authority level from this date.
• In some months, the data from the previous month may have to be imputed due to missing local authority data. This imputation process can introduce potential inaccuracies, which may affect the overall quality.
• We do not recommend making month-to-month comparisons, as there may be seasonal impacts on homelessness or other reasons for fluctuations.
• The rough sleeper estimates represent a snapshot of the number individuals sleeping rough on the last day of each month. Consequently, individuals who slept rough during the month but were no longer sleeping rough on the last day of the month are not counted.

There are a number of known data quality issues for data prior to April 2023. These figures should be treated with caution when compared to the homelessness accommodation provision and rough sleeping figures published on StatsWales.

Temporary accommodation

Temporary accommodation is defined as suitable accommodation that is likely to last for under 6 months, including short-term supported accommodation. This includes accommodation ordinarily considered temporary accommodation by this definition, but that may last more than 6 months due to high demand on housing.

Long term accommodation

Long term accommodation is defined as suitable accommodation that is likely to last 6 months or more.

Rough sleeping

Rough sleepers are defined as persons who are sleeping overnight in the open air (such as shop doorways, bus shelters or parks) or in buildings, vehicles or other places not designed for habitation (such as stairwells, barns, sheds, car parks, tents, cars/vans).

Statutory homelessness

Statutory homelessness refers to the legal criteria used to determine whether an individual or family is eligible for assistance due to homelessness. According to Section 55 of the Housing (Wales) Act 2014 (UK legislation), a person is considered homeless if they, along with anyone who normally resides with them, have no accommodation in the United Kingdom or elsewhere which they have a legal right to occupy. This definition is crucial for assessing eligibility for housing support and services provided by local authorities.

It is important to note the following differences between this monthly data collection and our publications on statutory homelessness:

• The monthly information relates to the number of individuals experiencing homelessness and being supported by local authorities into emergency accommodation or suitable long-term accommodation.
• Our collections on statutory homelessness capture data on the number of households, not individuals. That data relates to homelessness as defined by the Housing (Wales) Act 2014 (UK legislation).
• Statutory homelessness figures do not include persons sleeping rough but cover information on the number of households applying to local authorities for housing assistance under the Housing Act (Wales) 2014 because they are homeless or threatened with homelessness. Some people who sleep rough may have been in contact with a local authority for assistance with housing either recently or in the past and some may not have.

Rough sleeping

Rough sleeper data is included in the monthly collection, where local authorities are asked to base their estimates on local intelligence, rather than the methodology employed for the previous annual rough sleeper counts.

In August 2024, Welsh Government made the decision to cease the annual national rough sleeper count with immediate effect. Further details can be found in the Proposals to stop the annual national rough sleeper count: summary of responses.

Due to the different methodologies, it is not appropriate to compare rough sleeping figures from this monthly collection with those from previous annual rough sleeper counts.

Deaths of homeless people in England and Wales

On 23 November 2022, the Office for National Statistics (ONS) published a statistical release detailing experimental statistics of the numbers of deaths of homeless people in England and Wales (ONS). Figures were given for deaths registered in the years 2013 to 2021.

Homelessness statistics for the other UK countries

During 2018, in order to address the problem of inconsistency and support users of homelessness statistics, the Government Statistical Service (GSS) Harmonisation Team explored the feasibility of harmonising the definitions of homelessness for official statistics across the UK. In February 2019 it published the Harmonisation of Definitions of Homelessness for UK Official Statistics: A Feasibility Report (GSS). This report concluded that it is not possible in the short term to develop a consistent definition of homelessness for UK official statistics, due to substantive differences in the administrative data systems and legal definitions of homelessness. The devolved nature of housing and homelessness legislation across the UK means that, definitions of homelessness vary. Data collection techniques for homelessness data have been designed to individually reflect the legislation in each country. Operational differences in collecting homelessness information results in differences in homelessness statistics across the four nations and currently there is insufficient guidance on how to compare the related statistics.

It also recommended creating a separate, more detailed stand-alone report on UK comparability of homelessness (ONS) statistics which includes a conceptual framework for homelessness. The conceptual framework allows users to visualise the different definitions of homelessness currently being used for official statistics and how these fit together. This was published in September 2019.

In June 2025, the Homelessness in the UK (ONS) release was published, showing trends in homelessness statistics across the four UK nations and the cross-UK comparability of the data.

England

The Ministry of Housing, Communities and Local Government (MHCLG) produce a number of statistics on homelessness in England, including a quarterly and annual statutory collection, a rough sleeping snapshot and a rough sleeping data framework.

The data used in the statutory releases are from the Homelessness Case Level Information Collection (H-CLIC) data system. This system is used as a reporting requirement of local authorities to provide data on statutory homelessness for those approaching local authorities for help with homelessness in line with the Homelessness Reduction Act (HRA).

The latest homelessness statistics available for England can be found here: Homelessness statistics (Ministry of Housing, Communities and Local Government).

Scotland

Scottish Government produce an annual and biannual statistical bulletin on homelessness. The statistics bulletin presents information based on administrative data collected by Scottish local authorities in the course of carrying out their homelessness activities. The annual bulletin provides statistical information on all stages of the homelessness process, the circumstances from which households became homeless, use of temporary accommodation and the final outcomes of their application. The biannual is a slightly reduced data set and is used to monitor trends for key aspects of homelessness.

The most recent bulletin is available here: Homelessness statistics (Scottish Government).

Northern Ireland

In Northern Ireland statistics on homelessness are sourced from the Northern Ireland Housing Executive (NIHE). The Northern Ireland Homelessness bulletin is a biannual publication which contains information on a range of areas relating to homelessness. The report is currently divided into three sections: Homeless Presenters; Homeless Acceptances; and Temporary Accommodation.

The most recent bulletin is available here: Northern Ireland Homelessness Bulletin (Department for Communities).

Longer-term, Welsh Government are developing an individual-level homelessness data collection to replace our current aggregate statutory and monthly homelessness returns. The new collection will record detailed information on individuals and families who are homeless or threatened with homelessness, their passage through the Housing (Wales) Act 2014, and details of their temporary accommodation placements. Data on rough sleeping is currently outside the scope of this project and will not be included in the initial data collection phase. Collecting more granular, linkable data will move us in-line with the approach taken in England and Scotland and facilitate more in-depth research to be conducted into homelessness, especially as a cross-cutting issue. This will greatly increase the quality and detail of information from which homelessness policies are made in Wales.

The Programme for Government makes a commitment to reform housing law and fundamentally reform homelessness services to focus on prevention and rapid rehousing. This is underpinned by Welsh Government’s Ending homelessness in Wales: a high level action plan 2021 to 2026. In line with this plan, the Welsh Government launched a consultation of a White Paper on ending homelessness in Wales on 10 October 2023. Our longer-term strategy for individual-level data will enable us to obtain more granular data regarding a service user’s journey through the homelessness system.

Engagement with local authorities and data users is key in moving towards this individual-level data collection and hence officials in Welsh Government’s Knowledge and Analytical Services (KAS) have established an External Data Reference Group for homelessness statistics. The group brings together local authorities and external stakeholders, such as charities, academics and members of the private sector, with the aim of helping guide the design of this new collection and inform us of the needs of data users. If you are interested in becoming a part of this group or would like to find out more information, please contact stats.housing@gov.wales.

We always welcome feedback on any of our statistics. Please contact us via email: stats.housing@gov.wales

Media: 0300 025 8099