On 25 November 2014, Minister for Immigration and Security, James Brokenshire, said:

The new Immigration Act enables us to take tougher action to crack down on those who try to cheat our immigration system by abusing marriage laws. In 2013-14, we intervened in more than 1,300 sham marriages – more than double that of the previous year.

I queried with the UK Statistics Authority whether that number was accurate. I was concerned that in fact there are not 1,300 proven sham marriages and that the Minister was quoting the number of suspected or alleged sham marriages. I was right:

Thank you for your enquiry of 26 November relating to the Minister for Security and Immigration’s recent statement about sham marriages. We have consulted with the statistical Head of Profession at the Home Office about this matter, and confirm that the figure quoted relates to suspected, not actual, sham marriages. You may be interested to know that these figures have been the subject of a recent Parliamentary Question: http://www.parliament.uk/business/publications/written-questions-answers-statements/written-question/Commons/2014-11-24/215733/

These figures are taken from Home Office management systems, and are not routinely published as official statistics. They are therefore not subject to the provisions of the Code of Practice for Official Statistics. However, in discussing this matter with the statistical Head of Profession at the Home Office, we heard that the Home Office is looking into how descriptions and references relating to these data can be made clearer in future.

I also made a complaint to the Home Office but have heard nothing back.

That number of 1,300 was picked up in a number of media outlets. This kind of exaggeration and misinformation fuels mistrust in migrants and immigration enforcement and plays right into the hands of UKIP and other xenophobes.

It also begs the question of how many of those 1,300 ruined weddings were between completely innocent couples who had wrongly been reported by oversuspicious registrars. Quite a few, I would hazard to guess.

It has also been claimed, by the Prime Minister no less, that there are as many as 4,000 sham marriages per year. Some time ago a friend sent in a debunking of that figure that I think is worth sharing. The note was written as the Immigration Bill was passing through Parliament.

In the explanatory notes, the Home Office claims that ‘sham’ marriages/civil partnerships “pose a significant threat to the UK immigration control”. They then cite their own estimate that there are between 4,000 to 10,000 applications to stay in the UK a year made on the basis of a sham marriage or civil partnership. They do not give any indication upon what this estimate is based or how it was come by. However, this information is buried in a footnote in the Impact Assessment. Basically, it’s plucked out of thin air.

How to pluck a figure out of thin air in order to get your Bill accepted

First, choose a neighbouring country which keeps some statistics (the HO have no figures for non-EEA nationals marrying/CPing in England and Wales, so they borrowed Scotland’s figures).

Second, adjust slightly to show you are not just exaggerating for effect, and there is some smart maths involved in this (Scotland keeps records on the number of people born outside the EEA getting married in Scotland which includes EEA nationals born outside the EU, so they rounded down, a bit).

Third, include some jargonistic, technical sounding language with which to make your claim: 7% of all marriages/CPs in Scotland in 2011 had “an immigration advantage”. Mention in the small print (“a caveat”) that “some of these would include non-EEA nationals with ILR”. (Presumably it would also include those on other exempt visas, but ah well). Scale up from here to 12% coz England’s full of foreigners, init, (although, again, oh yes, this includes those with ILR, whoops!). Then multiply 8% (no idea, perhaps someone with Maths GCSE arrived in the office at this moment) by the number of suspect marriages reported by registrars in, oh, no particular year, and hey presto you arrive at the bottom end figure of 4,000 sham marriages a year. If this figure is not sufficient to justify spending 1.5 million over a five-year-period (see Impact Assessment) or you think your Bill might not get through, then you can always add a higher end figure which relies purely on individual estimates of people working in a related field (Senior Caseworkers working on EEA applications whose estimation of 20% was actually based on marriages between an EEA and a non-EEA “high-risk nationality”; shhhhh!).

So, to reach the figure of 4,000 sham marriages a year, the figures on non EEA nationals marrying were borrowed from Scotland because there are no equivalent figures for England and Wales. A wild and provably inaccurate  assumption on “immigration advantage” was applied. The figures were then scaled up on the simplistic basis that there are more migrants in England and Wales than in Scotland. An additional 8% was factored in for unclear reasons but which already seems likely to be incorporated into the assumption about “immigration advantage” in any event.

The higher figure of 10,000 a year was simply a guess by some case working staff at the Home Office but which was specifically about “high risk” nationalities and cannot therefore be used to extrapolate to other marriages, of which there are presumably a lot.

On this foundation of statistical sand, the rules on all marriages have been changed and mandatory reporting of marriages to non EEA nationals has been introduced.