Since 1999, politics has not entered marriage – we have had a decade where the structure of the family has barely been an area for discussion. Ten years ago David Blunkett felt obliged to close off the debate by describing marriage as “the gold standard”, which, of course, it is if you are fortunate enough to have a good one, and don’t want to remain single, and you don’t object to it as a matter of principle as strongly as those who believe it to be a holy sacrament.
But in the interim report on the family launched by Iain Duncan Smith’s Centre for Social Justice at the end of last year – a report of otherwise much good sense – marriage made a return as the answer to the ills of the family.
The CSJ has fallen for cod statistics and adopted journalists’ standards. “[This report] says that only changes that will reinforce marriage should be adopted. It restates the mounting body of evidence showing that marriage produces better outcomes for both adults and children. This review is working from an underlying assumption that marriage should be supported both in government policy and in the law and that, fatherlessness (or motherlessness), far more likely when relationships are informal, should be avoided…”
Alas its researchers evidently never heard the cautionary tale of my statistics course at Manchester Business School in 1972. In the US, a positive correlation was found between the number of university professors and the consumption of alcohol. But correlation does not prove cause. Thus reducing the number of professors would not reduce alcohol consumption. A common factor might be at play.
The simple act of getting more of the unmarried to marry will not cause better parenting en masse and will not cause the outcomes for their children to improve. Life is just a little bit more complicated than that. Marriage signals stability, it does not create it. Thus the positive effects of marriage are more due to the distinctive characteristics of the individuals who marry and stay married (the ‘selection effect’). For example, it would seem that such individuals have better mental health, are better off, have a more positive attitude to family, and have other attributes which make it easier to find and sustain a successful relationship.
The CSJ statistics also suffer from a repeated failing found in much policy analysis – the tyranny of the average. Thus much analysis concludes that average X produces better outcomes than Y and thus X should be adopted universally. This neglects the full distribution of outcomes and ignores the fact that the outcomes for some unmarried families are better than for some married. Thus, unmarried people can and do produce as good or better outcomes and we know that children from high conflict homes do better after divorce. But the average is used to tell them they are wrong.
However, more sophisticated and useful analysis is very slowly creeping into the civil service and academic world, called consumer segmentation. This is based on the startling notion that people need, want and use public services in different ways. The ‘can’t pays’ before court for their debts need a different approach to the ‘won’t pays’. In Iain Duncan Smith and Graham Allen MP’s excellent report, Early Intervention: Good Parents, Great Kids, Better Citizens, they write that “There are no quick fixes, no ‘one size fits all’, we need an integrated approach and a resolve that is shared by people across the political divide”. Is this the same Iain Duncan Smith?
Just how much would trying to promote marriage really achieve? Changing social behaviour is notoriously hard. Governments fail by trying to do too much –carefully selecting a few policies for consistent implementation is essential.
But assuming that a political focus on marriage and on structure is the answer is an easy trap to fall for. The English are prone to this black and white thinking. Schools have suffered for 40 years from the tug between the comprehensive and selective systems. Either system can work well – the trick is to commit to it, pull in the same direction, and all do the many things needed to make the system work. The criminal justice system suffers a similar diminution as punishment and rehabilitation are juxtaposed rather than cohered. Perennial structural arguments absorb energy and motivation, and take attention away for the real issues.
Instead we should be honest about the limits of our knowledge. A good marriage is great to experience, but quite what makes one is not known with enough clarity to give a government justification for pulling certain levers. Some people are lucky and find the right partner at the right time, one whose idiosyncrasies amuse in the long run and do not grate. Some people are not.
Society benefits from variety in its parenting output. The CSJ report implicitly measures family success by the educational attainment, addictions and employability of the children. These are all important symptomatic measures but the end-goals of our society are surely more about contentment and fulfilment.
Diversity produces a vibrant society which is resilient to unanticipated change – as it contains the social variety to respond.The social totalitarians would allow the minimum divergence from their norm. We would all be the worse off for it.
The Fabian Review, Spring 2009
