« Should Marriage Be Subsidized?--Posner's Comment | Main | How to Use Pensions to Improve Judicial Tenure-Becker »



Feed You can follow this conversation by subscribing to the comment feed for this post.

Paul N

Why do children brought up in intact families do better at school?

You appear to have neglected to consider the possibility that people who do better in school are themselves less likely to get divorced, independent of whether their parents were divorced. For instance, performance in school and duration of marriage both correlate with intelligence.

Matt C

What Paul N is trying to say I think is that the majority of these studies that find benefits to marriage, do not take into account the selection effect into marriage. Becker touches upon this but not quite enough:

Consider a simple bivariate analysis where the two variables of interest are state of parent's relationship and the child outcome variable. To go to the extreme, say there are three types of parents: very good, decent, and bad, and the parent type correlates heavily with child outcomes (very good parents produce high quality children, etc).

Say both very good and decent parents always decide to enter into marriage, and bad parents always stay out. However, only very good parents stay married, with decent parents always divorcing after having children. Any study without the proper controls for parental quality will deduce that marriage has a positive effect on child outcomes, when in fact marriage is acting as an unwilling proxy for some unobserved parental characteristic.

Now, move from to the extreme example to one where there are a range of parent types, and the "better" they are, the more likely they are to become/stay married. The result is the same.

You learn that sort of thing in Basic Econometrics

Gu Si Fang

I agree that no good can come from subsidizing marriage - or divorce for that matter. In France, both are subsidized through tax cuts and welfare schemes, and the resulting effect is anything but clear.

As the number of divorces continues to grow with respect to marriages, people often consider this to be a social fact with economical consequences. Why not consider things the other way around? Divorce is in some cases a need - to be single, to change the course of one's life, etc. This need is not easy to satisfy, as it is expensive, and creates emotional stress for the parents and the children.

But living in a rich country makes it more and more affordable. Therefore, one could say that the growing rate of divorce is an economical fact with social consequences. Implementing a subsidy to encourage people to get married would be the same as encouraging them to use their ressources for "better goals" than divorce. But better for whom? As Becker argues, there is little evidence that divorce has any long term detrimental effect on children. Besides, who is in a better place to judge what is good and bad for a child than his own parents? Who can tell if the relief of getting divorced will outweigh the stress of the process? I doubt that a Minister of Marriage could.

Shaoping Zhou

When analyzing the tax impacts over incentives, we need to be clear of the market elements, including "product", producers and the consumers. Can I infer that the product in this case is the marriage, producer being parents, and consumer being children?


Regarding "Arguments based on the external effects of parental divorce and other separation decisions on children are often not applicable because the vast majority of parents do love their children. These parents take account of their children's interests in deciding whether to separate, and in their other decisions."

Is professor Becker suggesting that such positive "externalities" is already largely internalized by the parents, thus leaving the argument of subsidizing even weaker from a pure economics point of view?


Does anyone remember Emperor Augustus and his desire to curb extravagence, adultery, and increase the birth rates of native Romans? Sounds like similiar problems confronting modern industrial states. Agustus's program resulted in the "lex Julia de maritandis ordinibus" which regulated taxation, marriage, non-marriage and the like. Turned out to be really, really unpopular with the population and was later modified by the "lex Papia Poppaea" named after two consuls who were bachelors. So much for promoting marriage and the virtues of family life. The same thing will happen today if legislation is passed. We're not that far removed from the ancients.

As for 40% of all British marriages ending in divorce, better that, than ending up in the Criminal Courts on homicide charges. Just remember, "familiarity breeds contempt." ;)


Is their any correlation between remarries and children success? Or is the intial divorce so painful, and the social mess that may accompany it so bad, that it permanently screws up the child?

If we look at gay marriage, should civil unions be given the same tax treatment as traditonal married people?

Thomas Hardy

Hasn't Cameron read Jude the Obscure? Puts this issue to rest.



After the high calibre comments I feel embarassed to ask this but I will anyway.

Can anyone explain the following sentence to me please?

"Different outcomes between boys and girls of growing up in families without fathers suggest that this rather than which parents continue to live together is what harms children."

I have read it several times and still don't understand it. Who are the parents which continue to live together that are being compared to fatherless families in terms of harm on children?

I am completely confused. Besides this I found the post a very interesting read.


Aaah yes, the old "I'm am persuaded" despite "evidence far too limited" conundrum! It's indeed alive well and very popular these days!

"I am persuaded that children raised by two gays or lesbians do worse than children raised by heterosexual parents, although the evidence is far too limited to be certain about this."

Indeed this would be a tough measurement to make as we'd have to adjust for high rates of adoptions, with many being adopted after spending time as a foster child etc. Then! there would be the sticky definition of "doing worse". Would the "standard" be that of high earners who don't create much disruption?

There are other dimensions to the "doing worse" aspect of raising kids such as living in poverty in this wealthy country, and that schools in the low income areas, both urban and rural, are routinely under-funded. But since those are both tough problems (ie the solutions involve money!) moralists and others seem to have a lot more fun engaging, uselessly, in hand-wringing over marital status and the like.

As for the marriage tax subsidy/penalty issue itself, though the socialistic Earned Income subsidy that is becoming a substitute for our corporations paying a living wage, was mentioned, the truth is that any "marriage tax" penalty is barely worth discussion until household income is WELL above our median household income.

If we were really interested in family income and the welfare of our children instead of tripping over nickels on this issue we should be actively pursuing a health care system that is not twice as costly as that of other leading industrial nations, but still leaves many out and makes medical costs our number one cause of bankruptcy.


Two things.
First, the key point here is selection. If you've read Freakonomics, you know that children with more books in their household do better academically. However, it is obviously not the books that are causing the improved performance, but the fact that parents and households with lots of books are also likely to care about and encourage academic achievement. Similarly, it is not THAT the parents are married that is the decisive factor [although I think the greater stability used to help - no fault divorce has made the expectation of permanence a thing of the past], but WHO the parents are that are choosing to get married. Simply marrying two people won't make them good parents, much like stocking a child's house with books will make them smarter. By encouraging marriage, we'd just be spending taxpayer money on people who probably shouldn't be getting married in the first place.
Second thing: I'm not holding this against Becker, but isn't it kind of unfair of some people to say that children of gay couples underperform by some measures, when the reason for such underperformance is, at least partially, discrimination? It's like pro athletes saying a gay teammate would be a distraction, when it's only a distraction because you make it so. I don't think the children of gay couples can be fairly judged by any objective measure until such discrimination, and the effects of such discrimination, are fairly accounted for.


When it comes to the issue of breeding and child rearing, just remember, rats and other vermin do it quite successfully without intervention. The real qustion is quality vs. quantity. To achieve the first, perhaps the State needs to remove all childen from the dysfunctional family environment and raise them properly. Much like the Spartans or the children of the Brave New World. That way society gets what its wants and needs. And while we're at it, let's implement a truly scientific eugenics program. ;)

Dostoevsky's Poodle

Since most of the good (or long-lasting) marriages I have observed seem to be between men and women who did not make careful cost-benefit analyses of yoking themselves, I entertain much doubt that tax/monetary incentives/disincentives are worth a “family values” advocate’s consideration. (As for David Cameron, it’s my public duty to speak a little ad hominem: his public statements are even less worth our attention than those of most politicians. Yes, that is a rebuke, of Becker, obviously, not of Cameron.) One very difficult issue, away from which Becker is most wise to shy, is whether there are some actions that are motivated by a (spiteful) desire to do the opposite of whatever an economist would predict. Much of so-called “romantic” behavior, for instance, might be considered in this light. What one might call the logic or psychology of such anti-scientific and anti-economist spite is rather well-explored by Dostoevsky in Notes from Underground. I applaud economists (and ethologists, sociobiologists, et al.) for attempting to understand such (“irrational”, “imprudent”, etc.) behavior, but often such attempts fail to take due account of the level of spite. (Obviously, I am being rather unscientific (but I hope amusing) in using such a word as spite.)
I find it difficult, however, to see my comments as a response to what you, Becker, have written. As for the comments of my fellow respondents, I hope you, Becker, will find them as dreadfully boring as I have.


"Subsidies to marriage could be easily implemented by allowing larger per person standard deductions to married couples than to single persons, "

This may improve the income inequality problem too! It creates an incentive to establish nominally legal marriages between very wealthy people and lower income people to bring the averaged income into lower brackets.


To the Poodle, I certainly hope you're house broken. As for your comment on Cameron, the guy's a Tory and should explain it all. Half of them still think of Britian as the Empire. ;)


"Romantic" behavior is actually quite rational, and fits very well within economic analysis, especially when you analyze it in an evolutionary light. "Romantic" behavior, in a sense, is similar to the feathers on a peacock. The large, colorful feathers on a peacock are a huge evolutionary disadvantage: they attract predators and slow escape. But that's how the peacock signals its genetic superiority: a male that survives DESPITE these hindrances must have fantastic genes. Similarly, a man who can spend on fancy dinners, diamond rings, and limousines is demonstrating that he has great means, and, by implication, good genes and a promise of a safe future for the women's genes as well. In a sense, the man spends money in a very inefficient manner [it'd be simpler to just write a woman a check] to show that he can afford to spend money inefficiently and that he will succeed despite such inefficient spending. Thus, such romantic behavior is actually quite calculated: the cost of being inefficient in such behaviors does not exceed the signaling benefits. Economics wins again.
My apologies for the digression. You may return to the discussion of marital subsidies, and whether they're appropriate [no].

Dostoevsky's Poodle

To Hatfield: Cheers, I love a dog-fight. "Tory" explains nothing. For Tony Blair is classified (correctly) by most followers of Linnaeus as an invertebrate, "homo slimus", and most political taxonomists have (punnily) labelled Cameron "Tory Blair". Cameron might get elected because he has aped Blair so well. Apparently Brown has not been doing such a good job of this.
To Haris: Assuming for the sake of argument that this is way you prefer to be addressed, the premise of your entire post ("actually...") is irrelevant, since we are in deep agreement. Thus my quotation marks around all those words that belong to the category flatus voci: romantic, imprudent, irrational. My point was only that the behavior that fetches such labels is often more difficult to predict than many economists would care to realize or admit--but such behavior is no less difficult for others to predict, and many of these, too, don't like to realize or admit it. Incidentally, Posner's book Sex and Reason is worth perhaps a whole library of psychological, moral, and philosophical literature on such subjects as "romance", "love", and "sexuality". To exhibit my economic reductionist credentials further I could offer you the observation that a slut is a woman who sleeps with a man after 2 hours' acquaintance, whereas a girlfriend is a woman who sleeps with a man after 22 hours' acquaintance. But you will justly complain that I have omitted a key point about the species slut. A slut is a woman who sleeps with another man instead of you, provided she is not your sister or mother.
But, Haris, I beg your pardon, for it was difficult not to have read my ENTIRE post as sarcastic, and "anti-economic-reductionist". Only some of it was. You only made the efficient mistake of jumping to conclusions.
I must cease here--and not reply to any rejoinders--because I have already hijacked this blog too far to be pardoned.


I understand that subsidizing married couples means taxing them less than unmarried individuals (this is true even if subsidies are direct cash handouts rather than tax-brakes). This can mean two things: either 1) the total national tax funds available for everything besides marriage subsidies decreases, or 2) if those tax funds must remain constant then, tax burden is shifted to the unmarried population.
If it’s the first scenario, then it simply means tax cuts for at least one part of the population without increases in taxes for the rest. This sounds too good to be true: governments seldom give up tax revenues so easily. Therefore, my guess is that we are dealing with the second scenario.

In the second case, if tax-brakes actually affect couples’ behavior, as more couples decide to marry and stay married, the more taxes increase for the (shrinking in size) unmarried part of the population, since we are assuming that the government must maintain a constant tax revenue. Here, consequent tax evasions and other problems associated with tax hikes, though obvious, but are only one part of the problem.

Most important problem is that marriage-subsidizing policies, if effective, end up helping couples to raise their kids partly at the expense of everyone who is either not married, or married but with grown up children, or married but not planning to have kids. The argument, if I understand correctly, is that the beneficiary couples create positive externalities by raising children in two-parented homes. Hence, such policies are justified.
Aha! The old externality issue.

Suppose that raising children in married two-parented homes creates positive externalities. Should we subsidize it? If yes, then we should subsidize education, healthcare, physical fitness, church affiliation (it’s quite possible that religious families get divorced less frequently and have more children), farming (one could argue that well fed people create positive externalities) , and an infinite number of other endeavors. Following this logic, pretty soon we would find ourselves taxing everyone to subsidize nearly everything—anything that may appear (not even be proven) to create positive externalities.
Also, I think getting divorced is painful—and hence costly—enough to play a much larger role in couples’ decisions than any politically feasible subsidies could ever do. So the case for social engineering by wizards like David Cameron is rather weak.

Anton Kunckle

Becker says, "The tax code of the United States require joint filing by married couples"

This is a misstatement. Married couples can file jointly or separately.

scott cunningham

Becker wrote, "A little evidence also indicates that being brought up only by mothers is harder on boys than girls, probably because boys benefit more when their fathers live with them. Since single mother families are so common among blacks, this finding has been used to help explain why young African-American males do a lot worse than young African-American females in school performance, delinquency, and on many other measures. Different outcomes between boys and girls of growing up in families without fathers suggest that this rather than which parents continue to live together is what harms children."

Can someone who knows this literature point me to a survey article, or a seminal article, that establishes this result? I've never seen it before, but it sounds like there's a literature upon which Becker is drawing in making this point.

garth brazelton

I ditto Matt Howard's sentiments. I had hoped Prof. Becker was above this sort of "moral" judgement based on a whole lot of nothing. Apparently I was wrong. Statements like "I am persuaded that children raised by two gays or lesbians do worse than children raised by heterosexual parents, although the evidence is far too limited to be certain about this."
are shameful and bare little need to be discussed in the topic seeing as how gays can't marry anyway (federally).

It seems the only reason this was even added to the post was so Prof. Becker could signal his own dislike of homosexual behavior and homosexual families - especially since the 'limited evidence' tends to show that gays and lesbians do an equal if not better job (due to higher incomes) at raising children than do typical straight families.

I respect you as an economist sir, but that kind of statement is beneath you.


Interesting. All comments seem to be coming from the male side of equation and not the female. I wonder what their comments on the subject would be? For they are the ones who make the final decisions and are the final arbiters in the breeding game. Perhaps, they are far to intelligent to stoop to such nonsense. ;)


The real question is whether marriage should be subsidized even more than it is already. See the link below for a listing of the legal benefits of marriage.

And ask yourself how many married people would give up these benefits in exchange for an elimination of the monstrous marriage penalty. The so-called marriage penalty is of course a marriage benefit (and an unfair penalty on the unmarried) in the vast majority of cases where it doesn't apply.



While I have scant evidence to present, I am persuaded that you are an old-fashioned bigot. But who needs evidence anyway?

>I am persuaded that children raised by two gays or lesbians do worse than children raised by heterosexual parents, although the evidence is far too limited to be certain about this.


I'm a financial analyst, living with my physician girlfriend. We would marry but for the $15,000 annual penalty imposed by the Federal income tax code on married couples like us.

You can't just do static analysis... There is adverse selection in marriage now. The number of couples living in sin has exploded from about 60,000 in 1965 to about 4 million today. I expect that the 4 million includes highly education high earners. I believe more people live in sin because of changes in family law and taxation in that intervening years. Marriage is now a fool's fame.

The comments to this entry are closed.

Become a Fan

May 2014

Sun Mon Tue Wed Thu Fri Sat
        1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30 31