Wedding bells and dog whistles

David Cameron’s Tories are turning right.

This is not about preaching to people about how they should lead their lives.
…
government has a responsibility to help people make the right choices.
…
We propose that government send a clear and unambiguous signal about marriage
– Conservative Social Justice Policy Group report

Why give taxpayers’ money to married couples (but not cohabiting couples) regardless of whether they have children? One might understand wanting to incentivise and/or financially help supposedly stable families for the benefit of the children involved, but what’s the sense in spending money on childless couples and couples that are already well off?

Here’s the Tory policy commission’s argument. First their report [PDF, p64] quotes an American researcher called Nancy Burstein:

Policymakers are prone to focus on marriage penalties and bonuses for low income families, because that is the most vulnerable segment of the population. I would argue, however, that these families do not make their decisions in a vacuum. They are affected by societal norms, reflecting marriage patterns of higher-income couples with and without children. Policymakers should therefore be concerned about these less vulnerable populations as well. While they may never need government assistance, their choices to marry or cohabit have cumulative effects. Marriage needs to be encouraged, honored, and rewarded for all.

The report agrees, saying that this:

indicates the need to avoid the trap of only focusing on the most vulnerable families in this area. Support for marriage cannot be simply dismissed as giving money to those who are already comfortable.
…
Marriage is concentrated in these more educated… middle class… sections of the population. To acknowledge the enduring value of this institution is to shore up marriage in its heartland. It also sends the signal to the lower deciles, characterised by greater informality and therefore instability, that marriage, to which many aspire… is a social good.

Let me try to summarise this. It would be good to get poor people to marry, and offering them some money will encourage them to do so. However, as poor people look up to rich people for inspiration in their relationship decisions, it would be good to encourage the rich people to get married as well. So let’s offer them some money too.

A few things occur to me here. First, how paternalistically patronising can you be? Second, given that higher earners already have much higher marriage rates, where is the need to spend my hard-earned money on what they’re doing anyway? Third, if the lower orders – sorry, the “lower deciles” – already “aspire” to marriage, then they hardly need state-financed inducement, either directly or given to their better-off role models.

Fourth, will this have any effect? The report says (p71) that the proposed transferable tax allowance

would provide only modest financial support for marriage - £20 a week to those making use of it – encouraging rather than incentivising it. The main rationale for the allowance would be to provide symbolic recognition of the institution of marriage. It would indicate that marriage is valued because of its benefits to children and the wider society.

I’m not completely sure what the difference between “encouraging” and “incentivising” is, but I presume this means that the money itself won’t make any difference: it’s about sending a message. But if so, then why not, well, just send a message? That David Cameron’s got a lovely speaking voice and he comes across as ever so encouragingly earnest.

The report suggests that the policy would cost £3.2 billion a year. And this calculation doesn’t seem to anticipate any projected increase in the marriage rate resulting from this “encouragement”. As such, the idea seems to be an even more expensive and ineffectual piece of public relations than the new Olympic logo.

A fifth point is that it’s still unexplained why marriage is a social good even in the absence of children. There’s a fraction of a hint of an answer in the comment (p71) that there are “proven advantages to children and the wider society (for example, married couples play a key role in caring for elderly relatives)”.

Well wrap me in a red flag and call me a clunking top-down statist, but why not just target the financial help on people who are in fact caring for their children or for their elderly parents, or doing whatever else that might be beneficial to society?

What makes families stable?
There’s another key strand to the Tory case, which is that marriage is not just associated with but positively causes increased family stability:

Statistics indicate that cohabitation is inherently less stable, so there is not the same justification for recognising it in the tax system.

In this respect, the report relies heavily on the argument made in the policy group’s previous report [PDF] that married couples are less likely to split up than cohabiting ones. A study included as appendix 3 to that report by the Bristol researcher Harry Benson carries much of the intellectual weight.

It’s an analysis, carried out recently in the UK, of differences in family stability in the first three years of a child’s life. Benson finds that 6% of married couples split up during this period, whereas 20% of cohabiting couples do – thus a topline result that cohabitees are about 3.3 times as likely to separate over this period.

However, there are demographic differences between married couples and cohabiting couples: the latter tend to be poorer, younger and less well educated, all of which account for some of the difference. After accounting for these factors, and a couple of others, Benson calculates that a cohabiting couple remains about 2.2 times as likely as a married couple to split up in the first three years.

This seems suggestive. However, I’d venture two observations. First, in looking at income and age differences, Benson groups couples by income quintile and by decade of age (a 20-year-old mother is grouped with a 29-year-old and a household earning £5,000 a year is grouped with one on £15,000). These are fairly wide categories, and given that both factors have been established as causally relevant, their contributory role will almost certainly have been underestimated.

Second, there is no mention of relationship duration. If it’s the case that new relationships are more likely to end in any given period than more established ones, and that married couples that have new children have been together longer than cohabiting couples that do so, then that would be another contributing factor unaccounted for here. There may well be other relevant demographic factors.

Furthermore, Benson himself accepts that there are “selection effects” – pre-existing differences between the kinds of people that choose to marry and those that cohabit – which will affect the different separation rates. Most notably: “married couples have a higher level of commitment to one another compared to unmarried couples in the first place” (p127).

So the evidence for a specific causal link from getting married to staying together, independent of other factors, is uncertain at best.

One interesting question is whether the Tories are concerned to promote marriage instrumentally as a powerful engine of family stability or to promote it moralistically for its own sake. If the latter, we would expect a few flaky arguments for why marriage is good regardless of whether there are children. There are certainly those.

But if the former, we would expect that some non-marriage tools for promoting family stability might be promoted as well. If cohabitation really is less stable, then why not try to make it more stable? Ah, but remember, cohabitation is “inherently less stable” than marriage. So all hope may be lost.

There is a fascinating part of Benson’s paper that unwittingly packs a killer punch here. Table 3 (p124). compares the post-birth separation risk for married and cohabiting parents in different income quintiles.

Married couples in the bottom quintile have an 8% risk; in the second quintile a 6% risk. Cohabiting couples in the bottom quintile have a 23% risk; in the second quintile a 12% risk (disregarding other factors). Benson comments, correctly: “Across every income group, cohabiting couples are at least twice as likely to split up compared to married couples.”

But what he misses is this: the drop in separation risk for cohabiting parents between the bottom and the second income quintile is far greater than that for married parents (23% to 12% vs 8% to 6%). This means that prioritising financial help and economic opportunities for low-income cohabitees is a better way to increase family stability than prioritising low-income married couples.

This is not proposed. So, it seems that the aim in throwing £20 notes at married couples is to blow the ‘dog whistle’, playing to Tory prejudices and moralising at the rest of us. Hence the following two reactions.

The Child Poverty Action Group’s Chief Executive, Kate Green, said:

A marriage certificate does not end addiction, it does not cure a mental health condition, it does not cancel debt, it does not increase skills and qualifications and it does not provide employment. Addressing these problems will do more to support relationships and lift children out of poverty than using the tax and benefit system to penalise children for their family background.
… David Cameron must resist the temptation to squander the resources needed on tax breaks that will do nothing to help the majority of children living in poverty.

And the Daily Mail said:

Finally! After months of wittering about wind-power and the work-life balance - not to mention the spectacular own-goal over grammar schools - David Cameron has come up with something recognisable as a true Tory policy.
With his support for the family as 'the most important institution in our society', he's given great cheer to the mass of traditional Conservative voters who were beginning to despair of him.

Recall that the report said it wanted to “shore up marriage in its heartland”. No. The real aim is to shore up Tory support in its heartland. Where Peter Lilley had his “little list”, Cameron and Iain Duncan Smith have a huge report – but they’re creeping back to Tory basics all the same.