Homelessness accommodation provision and rough sleeping: October 2022

This monthly collection replaces the weekly gathering of intelligence during the coronavirus (COVID-19) pandemic.

It covers temporary accommodation and provision of long-term accommodation for people who present to local authorities for housing support as they are at risk of homelessness.

This data has not undergone the same level of quality assurance as official statistics and the data may be revised in future.

The figures for the latest month should be treated as provisional. Not all local authorities have been able to provide comprehensive data and this is reflected, where relevant, in the footnotes.

For the estimates of rough sleeping, we publish a breakdown by local authority (from November 2020). For data on use of accommodation, we are initially publishing data at the Wales-level only.

Ability to compare with statutory homelessness statistics and the rough sleeper count

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

• This monthly data relates to the number of individuals experiencing homelessness and being supported by local authorities into temporary accommodation or suitable long-term accommodation.
• Our quarterly and annual collections on statutory homelessness capture data on number of households, not individuals. That data relates to homelessness as defined by the Housing (Wales) Act 2014.

In addition, we do not recommend comparisons between the rough sleeping estimates from this monthly collection and the annual rough sleeper count (up to November 2019). In this monthly collection, local authorities are asked to base their estimates on local intelligence. The annual rough sleeper count has a different methodology: a two-week information gathering exercise, followed by a one-night snapshot count.

Main points

Comparisons over time

At this stage, it is important not to put too much emphasis on data for an individual month or comparisons with previous months. This monthly collection of management information and the guidance provided is continuing to be refined and improved.

Use of temporary accommodation

• Throughout Wales,1,567 people presenting as homeless were placed into temporary accommodation during the month, 53 more than in September 2022. Of these, 475 were dependent children aged under 16, an increase of 3 from September 2022. ^{[footnote 1]}
• At 31 October 2022, 8,906 individuals were in temporary accommodation, an increase of 254 from 30 September 2022. 2,672 of these were dependent children aged under 16, an increase of 43 from 30 September 2022. ^{[footnote 2]}
• Between the beginning of the COVID-19 pandemic and the end of October 2022, over 28,800 people who were previously homeless have been supported into emergency temporary accommodation. ^{[footnote 1]}                                                         

Homeless individuals moved into suitable long-term accommodation

• 793 homeless individuals were moved into suitable long-term accommodation, 184 more than in September 2022. Of the individuals moved into suitable long-term accommodation, 288 were dependent children aged under 16, an increase of 76 from September 2022.

Rough sleeping

• As at 31 October 2022, there were an estimated 127 individuals sleeping rough throughout Wales. This is a decrease of 33 from the 160 individuals sleeping rough at 30 September 2022. ^{[footnote 3]}
• As at 31 October 2022, Newport (33), Pembrokeshire (19), Gwynedd (12), Cardiff (11) and Caerphilly (11) were the local authorities reporting the highest numbers of individuals sleeping rough. All other local authorities reported 10, or fewer, individuals sleeping rough, with eight local authorities reporting zero. ^{[footnote 3]}

Footnotes

[1] One local authority did not include ‘homeless at home’.
[2] Two local authorities were unable to obtain data from women’s refuges in their local authority.
[3] Local authorities are asked to base these estimates on local intelligence, not a one-night count.