Online Briefing: immediate release
EHRC report on social housing allocation to immigrants
relied on invalid statistical reasoning
The claim of a report published 7th July by the Equality & Human Rights Commission (EHRC) - to have demonstrated that there is 'no bias in allocation of social housing to immigrants' - has been shown to be baseless by independent academic analysis.
According to a leading statistical analyst, Professor Mervyn Stone of University College London, the figures that EHRC has disseminated as if they were evidence for the claim are of zero inferential value.
The report therefore constitutes a serious betrayal of the public interest that whatever is the truth of the matter should be established scientifically. In consequence, Civitas has made a formal complaint to the UK Statistics Authority asking it to appraise the reliability of the statistical methods used by the report and the statistical reasoning that underlies its claims.
Failure to compare like with like
In support of its claim, EHRC misrepresents the meaning of two factual assertions:
1. That in 2007 'less than two per cent' (1.8%) of social housing was occupied by migrants who arrived after 2002.
2. That 'nine out of ten' (87.8%) such homes were occupied by people born in the UK.
To make any sense at all, a comparison has to be like-with-like, but this contrast is no such thing.
In 2007, the social housing stock was four million of which 72,000 (1.8%) were occupied by migrants and 3,500,000 (87.8%) by UK born. To estimate the chance of a new-migrant applicant getting a home, you would have to divide the 72,000 by the total number of migrant applicants entitled to housing. To estimate the comparable chance for the UK-born, you would first have to establish the number allocated between 2002 and 2007, before dividing it by the number of UK-born applicants for the same period.
No calculation of that sort was done for the EHRC study. In fact, the extra data that would be needed to do it are nowhere to be found in the EHRC report. If it were done, the correction would almost certainly reduce the gap between the 1.8% and the 87.8%. Could it even be reversed and accepted as evidence against the EHRC claim? That is a possibility because, as the EHRC report concedes, 'most new migrants have no entitlement to housing' and because most of the 3,500,000 homes occupied by the UK-born will have been allocated before 2002.
Statistics and Democracy
A democracy relies on the honesty of official statistics so that our differences can be settled by peaceful debate, but the EHRC report on social housing fails the most fundamental tests. According to David Green, Director of Civitas:
"Government agencies have a duty to use public funds to commission objective research but the EHRC has failed the meet even the minimal standards of statistical rigour that the public is entitled to expect."
For more information ring:
Mervyn Stone 01895 674715
David Green 01344 773453
Robert Whelan 020 7799 6677
Notes for Editors
i. The online briefing can be found at this link.
ii. Civitas is an independent social policy think-tank. It receives no state funding either directly or indirectly and has no links to any political party.