I've invited housing minister Grant Shapps, via Twitter, to correct an inaccurate statistic which he promoted in several print and broadcast outlets over the Bank Holiday weekend.

The claim appeared in these words, from the Press Association on Saturday http://uk.news.yahoo.com/nearly-900-000-drop-benefit-claims-225950176.ht... 'More than a third of people who were on incapacity benefit dropped their claims / rather than/ complete a medical assessment, according to government figures. A massive 878,300 chose not to be checked for their fitness to work under tests brought in when the benefit was replaced by Employment Support Allowance in 2008.' The Guardian states 'Shapps issued a news release over the weekend claiming that 878,300 people claiming incapacity benefit – more than a third of the total – had chosen to "drop their benefit claim entirely /rather than/ face a medical assessment".

I've emphasised the phrase 'rather than' because that is where the inaccuracy lies, and as it occurs in both the PA and Guardian stories, it is presumably included in the press briefing document issued by Mr Shapps (which I haven't seen). What is being asserted is not just that a certain number of people dropped their claims, but that they all did so for a specific reason - in order to avoid assessment. This assertion is clearly inaccurate.

Every month [correction! every three months], about 130,000 people leave Employment Support Allowance (ESA). Of these ESA leavers, about 20,000 [a month! 60,000 a quarter] have not yet undergone a Work Capability Assessment (WCA). There is no mystery about this: claims for ESA are counted from when people are issued with an ESA 50 form to fill in, and there is an inevitable gap between the form being issued and the Work Capability Assessment. Indeed according to Nomis there are currently 488,000 claims in the 'assessment phase' of ESA, meaning that there has not yet been a decision on entitlement.

Obviously, during the gap between the initial claim and assessment, many people will see an improvement in their condition or will return to work (with or without an improvement in their condition), and DWP research has shown that these - and not a wish to avoid assessment - are the overwhelming reasons for people dropping their claims before the assessment http://statistics.dwp.gov.uk/asd/asd5/rports2011-2012/rrep762.pdf . For those familiar with the sheer scale of monthly on- and off-flows for ESA and its predecessor, Incapacity Benefit, these findings are unsurprising.

The same research found that in 'only a handful' of cases had people who had dropped their claims failed to attend a Work Capability Assessment. For 'a small number of claimants', receipt of an invite to a Work Capability Assessment triggered them to contact DWP to withdraw their claim. Note that even in this small number of cases, there is no reason to believe that the WCA invite contributed to the decision not to pursue the claim. People who drop their claims may or may not notify DWP of their decision- which means they may continue to be counted on the ESA caseload for some time after their decision.

Thus to claim, as Mr Shapps seems to have done, that everyone who leaves ESA before assessment is leaving in order to avoid the assessment is patently inaccurate.It is not just that he has no evidence to back up his claim. There is evidence, and it shows he is wrong. He should issue a correction and apology.

[Update: Grant Shapps is of course the Conservative party chairman, not the housing minister. That means that his inaccurate claims were made in a party political role, so they are not subject to the rules governing ministerial use of statistics, although this would presumably be the case were his claims to be repeated by a minister. There's been some confusion about the source of the 878,300 figure: it comes from table 1 in this release http://research.dwp.gov.uk/asd/workingage/index.php?page=esa_wca and, as I should have said originally, represents all ESA claims which ended before assessment between October 2008 (when ESA was introduced) and May 2012. Finally, among those who should have known better who uncritically repeated the claim, Tim Worstall deserves special mention http://timworstall.com/2013/04/01/interesting-number-eh/?utm_source=twit... simply because he accused another blogger, and indeed all 'lefties', of being 'lying scumbags' http://timworstall.com/2013/04/01/yes-of-course-lefties-are-lying-scumba... while failing to locate, let alone understand, the original data.]