Dear Mr Healey,
Thank you for your letter of 6 February in which you expressed concern about the quality of the Ministry of Housing, Communities and Local Government’s (MHCLG) Rough Sleeping Statistics for England. This was a key area of concern identified in our 2015 assessment of those statistics.
In 2016 the then Secretary of State recognised the importance of high quality figures on this topic and asked departmental statisticians to improve them. Rough sleeping statistics have never been designated National Statistics, and the Secretary of State acknowledged that they may never satisfy the requirements to achieve that status. Nevertheless, MHCLG did commit to improvements.
The department’s Rough Sleeping Strategy includes welcome plans for that. However the only timetable in the plan is to make the improvements over the next nine years. This seems to us so lengthy a period that it could be taken as not really being a commitment. We have been in contact with the department to urge greater speed and clarity about the plans.
You asked the Authority’s view on whether the CHAIN (Combined Homelessness and Information Network) model should be rolled out across England.
Our 2015 assessment reflected a range of users’ feedback, including whether the CHAIN model could be extended beyond London. At that time, we concluded that to roll out CHAIN across England may not be proportionate but could bring benefits for large cities. We expect MHCLG to publish its judgements about the statistical viability of expanding a CHAIN-like approach, its considerations of the cost-benefit of doing so, and how this information might be used to enhance the current official statistics.
You also asked about comparability, given changes to the way local authorities measure rough sleeping. The latest official statistics show a significant change between 2017 and 2018 in the way many local authorities are recording rough sleeping, with many more authorities taking a street count or spotlight count approach than in recent years. This is particularly apparent for areas receiving additional funding under the Rough Sleeping Initiative (RSI). More could have been done to acknowledge and explore this apparent methodological change this year and its impact in particular on the RSI areas. We welcome the substantially expanded statistical narrative and quality information in the latest statistics but expect greater clarity about what is driving this change in methods, the relationship to RSI funding, and the extent to which it may have affected comparability of the series.
In summary, we would expect the department to plan for better statistics on rough sleeping in a period shorter than nine years, to publish those plans, and to give greater clarity about the impact that the apparent change in rough sleeping methods used by some Local Authorities may have had on the comparability of the statistical series.
I am copying this letter to the Secretary of State and to Sandra Tudor, Head of Profession for Statistics.
Sir David Norgrove
 Technical notes, Table 6, https://www.gov.uk/government/statistics/rough-sleeping-in-england-autumn-2018