UK's official rough sleeping numbers 'far lower than reality'
The government has been accused of dramatically under-reporting the scale of rough sleeping following council data showing numbers almost five times higher than Whitehall estimates.
On the eve of the housing ministry’s annual snapshot of rough sleeping, which last year said that 4,677 people slept outside, the council data showed almost 25,000 people slept rough in 2019.
The figures were obtained directly from councils using the Freedom of Information Act. They relate to people sleeping rough at least once during the year. The government uses a different method, taking a snapshot count on one night.
On Wednesday the Labour party called for the UK Statistics Authority to launch an investigation into the accuracy of government data, which it said were “seriously misleading”.
The government’s snapshot for 2018 shows that there were 45 rough sleepers in Oxford. But over the whole of 2019 the local council said 430 people were recorded as sleeping rough at least once, according to the data gathered by the BBC. In Manchester the government’s figure was 123, while the council’s total was 679.
Related: Rough sleeping figures cannot be trusted, says statistics chief
Boris Johnson, the prime minister, appeared to refer to the government’s own figures last month when he claimed the number of rough sleepers was “lower than any time in the last eight years”.
John Healey, the shadow housing secretary, said: “These figures expose the shameful scale of rough sleeping on our country’s streets. They also confirm that the government’s own published statistics are seriously misleading and an unreliable undercount of the number of people sleeping rough.”
But the government claimed it was completely untrue to say its figures were wrong. “Our figures are independently verified by Homeless Link – a leading homeless charity – and the UK Statistics Authority has said our method is the most comprehensive available,” said a spokesperson for the housing ministry. “The two figures aren’t comparable and it is factually inaccurate to suggest otherwise.”
The UKSA said this was “a very generous interpretation” of its view, which was that “the snap-shot methodology was limited” and was the most comprehensive approach “currently available in England”.
The BBC data showed places with the highest rates of rough sleeping, recorded as a proportion of the population; these were Westminster, in London, Hastings, in Sussex, Oxford, Nottingham, and Camden, in London.
Jon Sparkes, chief executive at the homelessness charity Crisis, said: “We still do not have a clear picture of how many people are forced to sleep on our streets throughout the year. The government’s own statistics watchdog has long called for a better way of recording rough sleeping in England. We agree … but ultimately we need to address the issues forcing people on to our streets in the first place, such as the chronic shortage in social housing and the fact that housing benefit doesn’t cover the cost of people’s rents.”
In April, Sir David Norgrove,chairman of the UK Statistics Authority, wrote to the housing ministry telling it to stop using its rough sleeping statistics to support claims about the success of its rough sleeping initiative. He cited a change in the method of gathering the data and demanded greater clarity about how the figures were gathered.
Healey told Norgrove in a letter on Wednesday: “The statistics are an unreliable undercount and are an unsound basis for public policymaking or debate. I would be grateful if you would investigate the flaws in these figures and how the government’s statistics could be improved so they better capture the level of rough sleeping in our country.”
Ed Humpherson, director general for regulation at the Office for Statistics Regulation, the regulatory arm of the Statistics Authority, said he would be carrying out compliance checks of the official statistics over the coming weeks.
“Due to the inherent difficulties in measuring the rough sleeping population accurately we will expect to see these statistics include clear guidance around their relative strengths and weaknesses, and how they should and shouldn’t be used,” he said.
