Is homelessness increasing or decreasing?

Homelessness in England has increased a lot in recent years.

The increase in rough sleeping has been especially marked. Data from the Ministry of Communities Housing and Local Government shows that for every 10 people sleeping rough in 2010 there are now 26. In London as a whole, rough sleeping has increased even faster than the national trend.[1] In Westminster it has more than doubled. The greater rate of increase in London as a whole may in part be due to displacement to other boroughs, reflecting the already high level of rough sleeping in central London.

It is not clear that the slight fall in the reported national total in 2018 (74 people, or less than 2% of the total) corresponds to an actual fall in rough sleeping. The fall is more than accounted for by a change in one local authority (Brighton and Hove), where the figure for the number of people sleeping rough fell by 114 (64%). If this fall had not happened the national total would have risen slightly. Similarly large falls were found in some other local authorities, for example Southend saw a fall of 61 (85%). This appears likely to reflect changes in the data collection approach, as local authorities have moved from estimates to street counts. Where there is a change from one method to the other, as was the case this year in Brighton and Hove and Southend on Sea (among others), the estimate can change quite radically even where the actual number of people sleeping rough is little changed.

Rough sleeping nationally may therefore have increased somewhat even though the total has shown a small decrease. The chair of the UK statistics Authority (UKSA), Sir David Norgrove has stated that the official figures for 2018 should not be used to make claims that rough sleeping in England has declined until the government has addressed concerns about methodology.[2]

It is any case not possible to determine trends on the basis of a single year’s data. What is clear is that rough sleeping has increased enormously since 2010.

Further evidence on the trend comes from the separate Combined Homelessness and Information Network (CHAIN) database for London, which tracks individual rough sleepers. It shows a similar trend to the survey data, giving further confidence that the overall picture is accurate (dotted green line on chart). The CHAIN data for London in 2017 shows a different trend from the street count data, with a small drop. However this does not appear likely to continue in the 2018 data. Data for Q3 2018 shows a sharp increase over Q3 2017, although full year data is not yet available.

Data on number of people with no fixed address receiving treatment by the NHS has also shown a strong upward trend (see below). The consistency across different sets of data gives robust evidence that the trend since 2010 has been strongly increasing, even if there are uncertainties in individual data sets.

Trends in rough sleeping in England (2010=100)

Trends in rough sleeping in England

Note: The chart shows street count data gathered by local authorities and assembled by the Ministry of Housing, Communities and Local Government (MHCLG, formerly DCLG) for England, London and Westminster, together with the separate Combined Homelessness and Information Network (CHAIN) data for London. CHAIN data is for corresponding financial years, so for example 2010 data is for 2010/11.

This increase in rough sleeping at the national level and across London has happened despite efforts by local authorities. In fact, the number of preventative cases has risen over the same period, but this response has failed to stem the increase in rough sleeping, or wider homelessness.

During the 2000s, rough sleeping remained roughly constant or falling (see chart below). It then increased sharply from 2010. The basis for collecting data on rough sleeping changed in 2010, so direct comparison of current data with absolute totals before 2010 may be misleading. Nevertheless, the trends within the two distinct periods – before 2010 and after 2010 – can be assessed, since the methodology within each period was the same. Changes in methodology cannot account for the stark contrast between the largely stable or falling levels in the 2000s and the rapid increase since 2010.

Contrast in trends in rough sleeping pre-2010 and 2010 onwards

Contrast in trends in rough sleeping pre-2010 and 2010 onwards

Sources: 2005 to 2008 collated from Audit Commission data. Summer 2010 onwards DCLG/MHCLG data consistent with previous charts. Chart based on the one here:

Trends in healthcare demand from rough sleepers

Healthcare data gives additional evidence on the increase in rough sleeping and its consequences. The need for healthcare for those with no fixed address (a group that closely overlaps with rough sleepers) has risen sharply in recent years, increasing by a factor of about 3.5 since 2010/11. This is consistent with the observed rise in rough sleeping. However healthcare needs show an even greater increase than that for rough sleeping, which may indicate deteriorating health among rough sleepers. In addition to the harm to rough sleepers themselves this represents an increasing burden on an already overstretched NHS.

Reproduced from the Guardian
Reproduced from The Guardian, 20.02.2019


[1] Source: MHCLG data. The CHAIN data base shows a similar increase for London. However this is measuring something different – the number of separately identified people sleeping rough over the course of a year. It is also in practice more focussed on inner London Boroughs. There two data sets are thus showing different things but the trend of increasing rough sleeping in London is clear from both, with the exception of CHAIN data from 2017.