The ONS is seriously considering issuing figures for all types of ‘marriage’ together, without distinction: if it does, how will we know if gay marriage is working?

'The Office for National Statistics is reviewing how it publishes its figures on marriage, civil partnership and divorce' (PA)

“Will traditional marriage be written out of official statistics?” That’s a question the Family Education Trust (FET) is asking in this month’s edition of its indispensable news letter.

The first same-sex marriages are due to take place by summer 2014: and the Office for National Statistics (ONS) is reviewing how it publishes its figures on marriage, civil partnership and divorce. These figures should be of vital importance in determining how the different forms of what will according to the law be called marriage – whether such relationships are between those of different sexes or the same sex — actually work out in practice. How stable will same sex marriages be, for instance, compared with traditional marriages between one man and one woman? We surely need to know.

Disconcertingly, however, the ONS is considering adopting an approach to the collection of statistics on marriage which will necessarily make it impossible for the statistics it collects to be used to answer any such question: for after the Marriage (Same Sex Couples) Act comes into force during the summer of 2014, the ONS, which has the status of a non-ministerial government department, is seriously considering whether or not to merge marriages between a man and a woman with same-sex marriages in its statistical releases. FET tells us, for instance, that it is contemplating, issuing in future figures for divorce that draw no distinction between the type of marriage being ended; it is even thinking of including civil partnerships within the marriage statistics to provide figures simply on “legally recognised partnerships”.

A consultation on these plans was launched virtually unnoticed (it certainly wasn’t noticed by me; without the FET I wouldn’t have heard about it: would you? did you?) on October 8, on the very same day that the ONS published a report showing that female couples were almost twice as likely to end a civil partnership as male couples. If, however, if the UK’s only recognised national statistical institute ceases to draw any distinction between different types of “legally recognised partnerships”, it will no longer be possible to compare and contrast the relative stability or otherwise of different types of “registered union”.

The judgement of the director of the FET, Norman Wells, is surely unanswerable: “It is vital”, he says “that the ONS is completely open and transparent about the statistics it publishes on marriage, civil partnership and divorce. If we are going to be able to assess the impact of same-sex marriage on traditional marriage, the figures will need to be published separately and not merged into a genderless mush.

“Decades of research have demonstrated that a marriage between a man and a woman is considerably more stable than other types of relationship and produces better outcomes for children. The Prime Minister and some other supporters of the recent redefinition of marriage are assuming that same-sex unions will produce identical results, but without separate figures the argument cannot be settled one way or the other. To adopt a gender-blind approach to marriage and divorce would severely limit the ability of researchers to assess the relative benefits of different types of registered relationships and stifle healthy debate in a key area of public policy.

“If the government is serious about pursuing family policy based on sound evidence, it is of the utmost importance that all the relevant statistics should be readily available and not hidden from view.”

The fact is, and it is a fact highly highly inconvenient to the government and all other proponents of gay marriage, that the evidence so far appears to be that Norman Wells is right, and that marriage between a man and a woman is considerably more stable than other types of relationship and produces better outcomes for children: is that why the ONS proposes to muddy the waters by withdrawing the evidence from public scrutiny? In other words, is the supposedly independent ONS actually neutral on this issue: or is it parti pris? Is its independence as mythical as that of the BBC? I really hope my suspicions are misplaced: but we live in paranoid times.

Consider the evidence from various studies surveyed recently by Glen T Stanton in First Things. According to one writer, studies have found “higher dissolution rates among [legally registered] same-sex couples” in Scandinavia than among married heterosexual couples. This study, published in Demography, found that even though same-sex couples enter their legal unions at older ages, male same-sex marriages break up at twice the rate of heterosexual marriages.

For lesbians, the rate is 77 percent higher than that of same-sex male unions: this confirms the ONS findings quoted above (but will there be any more such dispassionate ONS reports?) Other research confirms these findings. A study of two generations of British couples (one born in 1958, the other 1970) in same-sex cohabiting, opposite-sex cohabiting, and heterosexual marriage relationships found the same-sex relationships are very much more likely to break up than the opposite-sex cohabiting and married relationships.

The figures on the probabilities of the various relationships surviving are striking. After eight years, 82 percent of married opposite sex couples are still together, 60 percent of opposite-sex cohabiting couples, and only 25 percent of same-sex couples. We need to KNOW if gay marriages in England will be any more stable than these figures suggest they might be: we cannot take it for granted that they will be (and my own feeling — though it’s no more than that, I’ve no evidence, how could I have — is that they won’t be). But if the ONS decides to muddy the figures, it will be very much more difficult to find out. It’s vital that they don’t: this is something we HAVE to know about.