Thursday, September 28, 2017

Getting married is a surprisingly rational thing to do

So who'd get married? Typically, for the first marriage, it's a man aged 30 and a woman aged 28 to 29. That's a big advance on the early days of the contraceptive pill in the 1970s when, according to the Bureau of Statistics, the typical ages were 21 and 23. But even in recent decades, we've been doing it later. Twenty years ago we married at 27 and 25.

And we are staying together longer. The figures I am quoting are "medians", meaning half of all marriages last longer and half are shorter. Those that get divorced typically do it after 12 years of marriage, up from 8 years in the 1970s. And we're less likely to do it. Twenty years ago six of every 1000 married Australians got divorced each year. Today it's 5.3.

We're also less likely to get married in the first place. Forty years ago eight Australians per 1000 got maried each year. Now it's just 4.8.

All up, we're taking marriage more seriously. And we are declaring our commitment before the Australian nation rather than God. Forty years ago, 82 per cent of marriages were conducted by religious ministers. Today, it's just 25 per cent. Civil ceremonies overtook religious ones in 1999. Only 10,000 of the 113,600 couples who got married in 2015 did so before a Catholic priest; only 4700 did it before an Anglican one.

Australia is unusual in allowing churches the role they have. In most of the countries to which we normally compare ourselves, the marriage forms have to be signed in an office. They can't be signed in a church, no matter how grand the ceremony.

Who are we marrying? It's increasingly people like us. Forty years ago the mainstream thinking was that highly educated men married less educated (but better-looking) women. In a seminal paper, A Theory of Marriage, economist Gary Becker argued that those sort of matches would enable women to specialise in domestic duties while their men specialised in earning money. Whether or not it was true at the time, it's much less true since. Dishwashers and hired help have reduced the importance of homemaking, while the increasing value placed on educated children has increased the payoff for highly educated men who marry highly educated women.

In a paper just published in the American Economic Review, economists Pierre-Andre Chiappori, Bernard Salanie and Yoram Weiss find that "assortative mating" has become much more likely. High-income, highly educated men are more likely than they were to marry similar women, making it even harder than it was for poorly educated women to get ahead and increasing the financial returns from education to women beyond those that are apparent.

People who are married are, on average, happier than those who aren't. Until recently it was thought this might be because happy people got married rather than the other way around. The good news is that a detailed examination of British happiness surveys by two Canadian economists shows pretty clearly that, whether or not happy people get married, they do indeed become more happy after marriage. It had been thought that happiness blast didn't last – that married couples lost the sparkle after two to five years. Married couples do indeed become less happy over time, the researchers find, but that happens to everyone of marriageable age. The important finding is that at every age, married people are on average happier than ones who aren't married.

So which marriages last? Melbourne University demographer Rebecca Kippen along with Australian National University economist Bruce Chapman and researchers Pen Yu and Kiatanantha Lounkaew have attempted to track success by tracing the history of 2500 married couples who signed up for the Household, Incomes and Labour Dynamics survey, which records personal data and life changes over time.

They find that what helps most is being similar. Couples who are close in age have less than half the risk of separation as couples where the man is nine or more years older. Couples with different views about whether or not to have children are twice as likely to split. Couples where the man is much better educated than the woman are 70 per cent more likely to split. If one partner smokes and the other doesn't, separation is 75 to 95 per cent more likely. If the woman drinks more than the man, separation is two-thirds more likely.

What each partner brings with them matters too. If they bring low incomes, they are twice as likely to split. If the husband is or becomes unemployed, they are three times as likely to split. If one or both of the partners have divorced parents, they are 60 to 85 per cent more likely to split. If one or both brings with them children from earlier relationships, they are two thirds more likely. Differences in race and religion turn out not to matter at all.

Getting married is a leap into the unknown. There are no guarantees. But it works for most who try it, which might be why so many more want to.

In The Age and Sydney Morning Herald