Divorce may be a Good Thing (quel horreur!)

By skepticlawyer

Obviously, that headline can be taken any one of a number of ways, and if you’re recently divorced, it may even inspire you to find the homo economicus who wrote it (me) and burn my house down. So – with that in mind – I’ll just ask you to hear me out, and see where the argument leads.

This post had its origins in Kim’s argument over at LP (derived from Stephanie Coontz) that feminism is good for the family. I’m actually going to go a step further, and argue that divorce – that creature of panicked headlines and (alleged) moral decay is not only underrated (Tim Harford‘s argument), but is sometimes a genuine good. Yes, you read right. I think divorce has had some social effects that are so positive it’s simply impossible to ignore them. However, in order to appreciate those effects, you’ll need to remember that an economist knows that – for the most part – people behave rationally. That is, they respond to incentives and weigh up costs and benefits (which, I might add, are often non-monetary). I argue that divorce has provided incentives to men (in particular) to change their behaviour; these behavioural changes are entirely rational. They’re also positive.

First up, though, a bit of background. Coontz’s piece interested me because – unlike much feminist writing – it actually gave fair play to capitalism:

Wherever women enter the labour force in large numbers, certain processes unfold. Women begin to marry later and have fewer children, especially as they make inroads into higher education or more remunerative careers. They are also more likely to challenge laws and customs that relegate them to second-class status in the public sphere or mandate their subordination within the family. Often, governments and employers then find that it is in their interest to begin to remove barriers to women’s full participation.

The dramatic decrease in laws and customs perpetuating female subordination over the past 40 years has been closely connected to women’s expanded participation in paid employment. Societies where women remain substantially under-represented in the labour market, such as in the Middle East, remain especially resistant to women’s rights.

About the only thing she’s left out is capitalism’s talent for sucking women away from the countryside and into the cities, where they are less likely to be under familial control. That’s an important part of the mix, but (arguably), labour-market participation is even more important. Indeed, it was Coontz’s own research (in Marriage: A History) that showed just what the developed world looked like in 1965, before women entered the labour market in massive numbers: the average married woman worked fewer than 15 hours a week in paid employment. That average was only dragged up from zero by ‘maiden aunts’ and the very poor. By contrast, the average married man worked over 50 hours a week. In terms of household work, married women did 40 hours per week, while married men did fewer than ten.

What then, does this say about the family? And what does it have to do with divorce? First, Coontz’s research illustrates a fundamental economic truth (one that Adam Smith first spotted in 1776). That is, the family is the economist’s division of labour writ large. Next, it illustrates something that Gary Becker first noted in 1981: the division of labour in marriage allows for economies of scale because it is an expression of comparative advantage. Becker discovered that any difference in ability between the genders at either homemaking or labour outside the home would have a disproportionate effect on how people spent their time. He suspected that any differences in expertise between the genders in terms of child-rearing skills were small, and likely brought about by a combination of biological differences, socialisation and discrimination in the workplace. His main point, however, was that – regardless of how it came about – there didn’t have to be much difference to produce a very large effect.

Tim Harford illustrates Becker’s point by asking his readers to engage in a thought experiment: imagine a hypothetical married couple where the woman is both a more productive worker than her husband and a better parent. The man is a bad worker but a worse dad. In circumstances like these, it becomes rational for the woman to do the parenting and for the man to eke out whatever living he can. Compounding the effect of comparative advantage is the fact that the capable woman cannot leave. Before the reforms to divorce laws enacted in various countries since 1960, she was trapped by law.

It’s fair to say that there are a large number of people who yearn for a world before feminism, with low divorce rates and (apparently) stable families. They’re often called ‘conservatives’, and when it comes to marriage, they’re barking up the wrong tree very badly. What they really want is a return to a world where the division of labour is plain for all to see. Indeed, sometimes the same people argue for tariff protection and other, less savoury forms of economic autarky. Either that, or they want free trade and free markets, but not for the family – that has to be shielded from economic reality.

However – as Coontz’s argument above illustrates – capitalism and its creative destruction is no respecter of divisions of labour when those divisions become inefficient. Economist Betsey Stevenson – using California as her test-tube, where Governor Ronald Reagan signed a bill into law making divorce ‘no-fault’ in 1969 – figured that there was more to the effects of divorce than met the eye. Recall, of course, that one of the effects of easier divorce – as Harford documents – was more women going to college, having fewer children and getting better jobs. However, this cuts no ice with serious conservatives, who will still argue for the old division of labour based on research indicating better outcomes for children raised by both biological parents. Libertarian feminist Kerry Howley summarises that research neatly:

Children do better on average, along a variety of dimensions and across all income groups, when raised by both of their biological parents. Poor children are more likely to be born out of wedlock, and those that are born to married parents are more likely to see their parents divorce later.

However, Stevenson noticed something when she was compiling her no-fault divorce data. As each state passed no-fault divorce laws, women also acquired the ability to leave the marriage. Her statistics suggest that many chose not to exercise this option – the threat was enough. Even when not exercised, however, the credible threat gave men stronger incentives to behave well inside marriage. Domestic violence fell by almost a third, and the number of women murdered by their husbands fell by 10%. There was also a statistically significant drop in rates of female suicide. These figures hold even when controlling for the increased reporting of domestic violence facilitated by feminism during the 1970s.

These findings may come as a shock – in part because they suggest that criminals and would-be criminals are rational and respond to incentives, too. I think it’s fair to say that while there are some irrational criminals out there (Heath Ledger’s Joker in The Dark Knight is probably representative of the type), most criminals (even wife-beaters) are rational. They respond to incentives, and are prepared to weigh up the relative costs and risks attached to obtaining the things they want. Indeed, what makes the Joker so scary is that he wants nothing – ascribing a motive to his acts is thus nigh on impossible.

Over at LP, Adrien made this comment:

The assertions that feminism wrecked the family are, in my view, grounded in the high divorce rates apparent after the 60s. Remembering just one of the stories about a female ancestor of mine who was trapped in a loveless, hostile, unfair marriage to an absolute arsehole courtesy of the law, society and Holy Bloody Church (of course) my view is that the family can go fuck itself if it needs to ride on the back of a woman’s misery.

In light of Stevenson’s findings, he may be right, although Coontz’s recent work showing a correlation between education (for both partners) and marital stability indicates that familial relationships may well be finding a new, rational level.

Homo Economicus will still want to do the inevitable CBA, and ask serious questions about the relative weight one should attach to children having better outcomes when there is less divorce (while women get beaten and murdered) or children having worse outcomes when there is more divorce (while domestic violence rates drop through the floor). To my mind, however – one schooled in law and criminology, and one that argues if there are any crimes a society should strive to prevent as much as possible, they are murder and serious assault – the choice is a no-brainer. Divorce – a fair bit of the time – is a good thing.


  1. conrad
    Posted August 28, 2008 at 9:14 am | Permalink

    Sorry I can’t state the literature here, but I was under the impression that there were now many studies that find that divorce, when measured from a similar baseline of married dysfunction, has, on average, an overall positive effect in many areas, both for the children in terms of long term adjustment and the females involved in terms of mental health (at least in Western countries where these studies are always done). Too bad about being male :). Thus I don’t think you even needed to qualify the idea that divorce is sometimes good.

  2. Posted August 28, 2008 at 9:29 am | Permalink

    The general rule (stated very crudely, and with all the usual provisos) is that divorce is bad for children, but good for women. There is a lot of evidence around now that children raised by two biological parents enjoy better outcomes, and that the outcomes are traceable to family arrangements.

    What I’m trying to do here is put the conservative (pro-family) argument at its highest, then put the libertarian (pro-individual woman) argument at its highest, and set them against each other (something Harford and Stevenson don’t bother to do).

    Maybe this is because I’m a lawyer and interested in conflict (tee hee) but it’s also because I think that many people don’t appreciate how difficult and complex public policy decisions (especially in the form of legislative changes) actually are.

  3. pedro
    Posted August 28, 2008 at 11:15 am | Permalink

    Well, divorce does not need to be ultra bad for children. My second wife and I have my son from my first marriage half the time and we and the 1stW work together pretty well to make it work. His 2 little sisters are never called half-siblings and we never do anything significant without the whole family. But that’s just us and lots of others have different and tragic experiences.

    I hate the idea that parents should be kept in unhappy marriages for the sake of the children. I also hate the idea that a parent should bail out or be squeezed out because of a divorce. I don’t think anyone can solve that dilemma with a law. The instrument is just too crude and I suspect that our view of the history of family life is wrong anyway.

    Was the family really so great in the pre-divorce past? Do we have a good old days problem here?

  4. Posted August 28, 2008 at 11:41 am | Permalink

    Was the family really so great in the pre-divorce past? Do we have a good old days problem here?

    It would depend on how much data we’ve got over time, and then how it’s analysed.

    Studies like Stevenson’s only become possible when you can use legal changes as a type of ‘flagfall’ and then use other areas (states) that don’t enact the same laws as controls. In terms of available modern data, it seems that divorce is a prophylactic against violence against women, and that children of two biological parents have generally better outcomes (measured against various indicia, as Howley notes). That said, preventing violent crime is probably orders of magnitude more important than ensuring that every little Johnnie & Julie goes to college – well, that would be my public policy view at any rate.

    Of course, these are massive data sets, and don’t account for individual outliers – eg the violent (biologically related) family where divorce would undoubtedly be better for all concerned – mum, dad, the kids – probably even the neighbours.

  5. conrad
    Posted August 28, 2008 at 4:30 pm | Permalink

    “There is a lot of evidence around now that children raised by two biological parents enjoy better outcomes”

    Yes, I agree — I was not being clear. My point, which is basically what everyone else is saying is that if you look at the left end of the distribution, then you see positive effects of divorce. That’s obviously the case for DV style situations, but they represent the worst of the worst. Even if you look further up the functionality distribution, you see benefits (Adrien’s quote is a good one on that). It’s how far up the distribution you can go before the benefits turn into losses and to whom the losses get attributed to which I think is where a lot of the interest lies (and how to define/measure it).

    The other thing to consider is that more well controlled studies show little overall effect on overall wellbeing due to marriage (make sure you only look at studies that controlled for the people going in — many early studies didn’t and found positive effects — however, it turns out that this was an artifact of the type of personalities that do/don’t get married). However, within that null effect you see males benefit and females lose. So the conservative idea of marriage and having kids is a loser for women on average. It would be interesting to look at some groups further to see if win-win groups are identifiable, such as when both parties are well educated and get married later in life (I’m sure people must have done this in fact).

  6. Posted August 28, 2008 at 5:34 pm | Permalink

    Speaking as the eldest child of people who should’ve been divorced years before they actually got divorced I’ll say divorce is a good thing. Staying together when it’s over is bollocks. And doing it for the children is a lame excuse. It makes things worse.

    In terms of the division of labour argument one thing that’s become apparent to me is that when children are very young most women in my circle assume the homemaker role not reluctantly but imperiously. They turn into drill instructors.

    Motherhood obviously creates problems for women who wish to pursue a career. I imagine th solution is a relay-rave type division of labour where the father assumes more responsibility in the second half of the kids’ first ten years of life while she gets back to being a hard-nosed career bitch.

    All depends on what’s possible however.

  7. Posted August 28, 2008 at 5:47 pm | Permalink

    There is a lot of evidence around now that children raised by two biological parents enjoy better outcomes, and that the outcomes are traceable to family arrangements.

    Yeah I wonder how much of this is due simple to the parents’ splitting and how much comes from the fall-out afterwards. The conventions of parentage call for a two parent nuclear family living at home blah blah. There have been other models involving extended families.

    Thanks to the industrial revolution this village life is no longer possible for most of us. And the free association of adults in modern society may mean that kids grow with a variety of adults from whom they can expect very little by way of a relationship. Mum’s new boyfriend is not required to be Dad in any way. And in fact Mum might have trouble getting a new boyfriend is that was so.

    However the ‘village life’ model does have advantages. Children can, for example, seek out role models if their parents don’t cut it some way.

    My youngest brother had the worst trouble with my parents’ split. He didn’t see it coming, it upset him terribly and he started to get into trouble. He also developed a pattern whereby he could escape one city when things got too hot and move to another where the other parent lived. There was no post-divorce structure. My parents did not sufficiently deal with the possibility that their marriage would end.

    These days things are different and I suspect ways of dealing with higher divorce rates might evolve. From experience however I reiterate that it’s no fun living with a married couple who are effectively not married anymore.

    My brother who is now a husband/father for example would, in the unfortunate circumstance of divorce, not make the same mistakes our parents made.

  8. conrad
    Posted August 28, 2008 at 7:53 pm | Permalink

    Adrien, I think your situation suggests that people in bad situations should get divorced. Here’s a more extreme version of this, which should be a good moral dilemma (and policy question if you must): If marriage doesn’t, on average, increase life satisfaction, then it suggests that 50% of people might be better off if they got divorced, which is higher than the current divorce rate. So a trickier question is whether people should be encouraged to get divorced even if they are not especially unhappy being married (assuming here the average couple isn’t especially unhappy), since it might lead to better life outcomes.

  9. Sinclair Davidson
    Posted August 28, 2008 at 7:58 pm | Permalink

    Getting divorced is a wonderfully liberating experience. A bit emotionally draining before its going to happen but given that I had gotten married, the best decision I had made in a long, long time.

  10. Posted August 28, 2008 at 8:04 pm | Permalink

    However, within that null effect you see males benefit and females lose. So the conservative idea of marriage and having kids is a loser for women on average. It would be interesting to look at some groups further to see if win-win groups are identifiable, such as when both parties are well educated and get married later in life (I’m sure people must have done this in fact).

    I’m pretty sure Stevenson has done studies like this; I didn’t mention it in the post, but I think it’s pretty clear that (biological) marriage is beneficial for children, neutral or slightly positive for men, and slightly negative to very negative for women. I don’t set much store by happiness research, but it’s possible to find this by looking at outcomes.

    What does perplex me is the possibility that marriage is a zero-sum game. This would mean that the extent to which men are now ‘losing’ in the marriage race is inevitable, as modern law changes have simply redistributed the gains more evenly (men used to ‘gain’ everything; now the – finite – gains have to be shared).

  11. TerjeP
    Posted August 28, 2008 at 10:06 pm | Permalink

    Yes, you read right. I think divorce has had some social effects that are so positive it’s simply impossible to ignore them.

    I think you are conflating two separate issues, perhaps deliberately in order to better grab our attention but maybe not.

    “Divorce” and the “Freedom to Divorce” are not the same thing. The crux of your argument does supports the notion that the latter might be a good thing. However that doesn’t mean the former is also a good thing. Just as illicit drug taking and drug prohibition might both be bad things, divorce and divorce prohibition may both be bad things also. Doing something is not the opposite of being prohibited from doing something.

  12. Posted August 29, 2008 at 12:59 am | Permalink

    Oh well, as a last resort a bloke can always bring home a Filipino bride. Or in this day an age, a Russian one.

  13. Posted August 29, 2008 at 3:33 am | Permalink

    I’m claiming that both may be a good thing, Terje – the legal ‘permission’ and the actual act. When it comes to the actual act, there’s Oswald and Gardner’s big study showing that divorcees, unlike widows and widowers, are happier one year after the marriage ends than while they were still married.

    However, if you find ‘happiness research’ to be rather dodgy (I know I do), then it’s possible to fall back on more conventional economic arguments – finding a marriage partner is rather like finding a job, and we all know that a job market where nobody could quit or be fired wouldn’t work very well. People – lots of people – would finish up in jobs they hated or were crap at.

    I’m not going to hazard a guess at what an ‘optimal’ divorce rate would look like, but I’m positive it isn’t zero.

    Mel: an interesting but high-risk strategy, as quite a few blokes have discovered.

  14. conrad
    Posted August 29, 2008 at 6:33 am | Permalink

    “What does perplex me is the possibility that marriage is a zero-sum game.”

    Whilst on average it’s a zero sum game (or close to it), there is of course a huge distribution of people where both lose, one loses and the other wins etc. You could certainly set up laws or social campaigns that would cause it to be a winner. We already do that — it’s why people frown on early marriage now. That’s why I wonder if we should be encouraging divorce more, rather than sponsering marriage counselling.

  15. Posted August 29, 2008 at 3:39 pm | Permalink

    I believe Edward de Bono suggested that marriage should be reformed as a short-term contract of 5 years with an option to renew. I thought the idea quite brilliant.

    A lot of people think marriage is something at which you fail or succeed. A refusal to admit failure can see people dogging it out when there’s no point. However if marriage is short term it means there’s no failure as such. It also means you don’t take each other for granted as the tender’ll be up soon enough.

  16. John Greenfield
    Posted August 30, 2008 at 9:27 am | Permalink

    Ah, Edward de Bono, that name really takes me back. Big hair, shoulder pads, Duran Duran and Edward de Bono.

  17. John Greenfield
    Posted September 1, 2008 at 7:07 am | Permalink

    Mem Fox is going to be receiving a lot of hate mail this week,

    Leading children’s author Mem Fox has drawn criticism from working mothers but qualified support from family groups for likening childcare for very young babies to child abuse.


One Trackback

  1. […] the community quickly when it first became clear that — while unilateral no-fault divorce lessened the incidence of domestic violence dramatically — there was still a fair bit going on, and the culprits weren’t being prosecuted. This […]

Post a Comment

Your email is never published nor shared. Required fields are marked *