Interesting article by Steve Landsburg in Slate about how to calculate the opportunity costs for future income of becoming a mother. He’s reporting a study by Amalia Miller (pdf), who claims that delaying childbirth for a year in your twenties increases your prospective income by 10%. I was most interested in the method, and am even more interested in hearing what economists have to say about the method and the findings (open invitation). Landsburg on the method:
How does Miller know her findings are reliable? It would never do for her to simply compare the wages of women who gave birth at different ages. A woman who gives birth at 24 might be a different sort of person from a woman who gives birth at 25 and those differences might impact future earnings. Maybe the 24-year-old is less ambitious. Or worse yet (worse from the point of view of sorting out what’s causing what), maybe the 24-year-old started her family sooner precisely because she already saw that her career was going badly.
So, Professor Miller did something very clever. Instead of comparing random 24-year-old mothers with random 25-year-old mothers, she compared 24-year-old mothers with 25-year-old mothers who had miscarried at 24. So, she had two groups of women, all of whom made the same choices regarding pregnancy, but some of whom had their first children delayed by an act of chance.
{ 42 comments }
Commenterlein 12.09.05 at 4:35 pm
It is an interesting and clever choice of instrument (as in “instrumental variables”), but I am concerned that the instrument might violate what’s called the “exclusion restriction”, which is the requirement that the instrument does not through some other channel affect the dependent variable. To cut down on the jargon: A 25-year old mother who miscarried at age 24 might for some other reason be pre-disposed to have higher income later than a randomly selected 24-year old mother.
One such (completely made up) reason might be that women who become pregnant so quickly again after a miscarriage are women who really want to have a child, i.e. a group that excludes women who accidentally became pregnant at age 24. If the incidence of accidental pregnancies is correlated with characteristics associated with low future income, then selecting a sample with no (or fewer) accidental pregnancies could produce the observed result.
More generally speaking, the authors have to (and probably did) worry about whether there are systematic differences between the sample of women who gave birth at 24, and the sample of women who gave birth at 25 after a miscarriage at 24.
John Quiggin 12.09.05 at 4:57 pm
It’s certainly a neat application of instrumental variables.
On the results, it’s worth noting that at least half the effect comes from the fact that women with children work less hours. The survey is a panel who were teenagers in 1979 (apparently born 1965) and ends in 1999, so it doesn’t cover the possibility that women who have children early will return to full-time work when their children finish school, offsetting the initial reduction in hours.
This is a common pattern in Australia, but seems to be less so in the US.
nik 12.09.05 at 5:11 pm
Is it really credible to say that (a) a woman who gives birth at 24 is not going to be a different sort of person to (b) a woman who a woman who miscarries at 24 and then gives birth a year later?
The obvious difference is that (b) is likely to be deliberately trying to have a baby, whereas (a) may well have been a victim of contraception failure. This seems a pretty big difference in terms of choices regarding pregnancy. And I’m not sure that miscarriage is an “act of chance”. She may have just swapped one inappropriate comparison as for another.
Slocum 12.09.05 at 5:17 pm
Very similar to the method used to determine the added value of an Ivy League degree over one from, say, Penn State–namely, comparing students who attended Ivy League schools to those who were accepted but, for whatever reason, chose to attend a state university instead. In that case there was no significant difference in earnings.
So it seems not to matter where you go to school and it does matter when you have your first kid. But people go nuts trying to get into the ‘right’ schools but don’t worry so much about the right time to have a child. Interesting.
LizardBreath 12.09.05 at 5:36 pm
Is it really credible to say that (a) a woman who gives birth at 24 is not going to be a different sort of person to (b) a woman who a woman who miscarries at 24 and then gives birth a year later?
Click through — she did a series of comparisons, none of which is perfect, but which do answer some of your concerns. (Comparison of mothers who got pregnant accidentally while using contraception at 24 with those who got pregnant accidentally while using contraception at 25, etc.)
shaughnessy o'brien 12.09.05 at 7:01 pm
However, per Dale and Krueger, there are income premiums associated with attending a more selective institution if you are African American, or if you measure selectivity in terms of cost of tuition.
vivian 12.09.05 at 8:06 pm
re 3: “But people go nuts trying to get into the ‘right’ schools but don’t worry so much about the right time to have a child.
Umm, you think women (of childbearing age) don’t worry about the timing of children? Some worry it’s too early, others too late, others worry about balancing the timing of career/kids/aging parents, and so on.
Isaac 12.09.05 at 8:41 pm
Wouldn’t you worry that higher propensity to miscarry is correlated with other things that aren’t being controled for (for miscarriage is a bad IV because its correlated with the error term). Like health status, or…
Rivka 12.09.05 at 9:05 pm
But people go nuts trying to get into the ‘right’ schools but don’t worry so much about the right time to have a child.
In my experience, people – especially professional women – worry a very great deal about the right time to have a child.
phil 12.09.05 at 10:25 pm
Miscarriages are unlikely to be positively correlated with good health, so if the miscarrying women are outperforming the nonmiscarrying women, it’s probably not related to health status.
cm 12.09.05 at 10:39 pm
john quiggin (comment #2): In my neck of the woods (SF Bay Area, formerly known as “Silicon Valley”) mothers in the engineering profession are often back at their full-time jobs just a few weeks up to a few months after giving birth. The children are often taken care of by visiting relatives, e.g. inlaws taking turns every few months, until they are old enough to go to daycare.
John Quiggin 12.09.05 at 11:36 pm
CM, that happens in Australia, but it’s not the socially approved norm. Although with the gigantic housing-related debts that couples are now taking on, it’s becoming more common.
cm 12.10.05 at 3:46 am
john, you are not telling me that Australian women customarily do the stay-at-home-mom thing until the kids leave school?
Otherwise re “socially approved norm”, I’m sure it’s not that they want to go back immediately, but the bills/mortgage have to be paid, and taking voluntary leaves may also have consequences for your healthcare coverage (don’t know exactly which, but I figure the employer will not want to extend full coverage for extended leaves).
abb1 12.10.05 at 5:14 am
…delaying childbirth for a year in your twenties increases your prospective income by 10%…
What does it mean, how is it possible?
A women who gives birth at 25 will work longer hours and and earn higher wages compare to a woman who gives birth at 24 – that’s it? Come on, folks.
Which phase is Saturn in?
bad Jim 12.10.05 at 6:20 am
This is a little tangential.
A young woman who worked in marketing in my company (okay, she was the brains behind the V.P., whom we eventually dumped) stayed on the job after her first child and all the way through her second pregnancy. I got to stand around while she, in her seventh (?) month, broad-bellied but unbowed, directed her union crew in the striking of our trade show booth. Capable? She had the crew, and me, wrapped around her little finger.
After the second child was born, though, that was it. She phoned in: I’m not coming back. A kid at home, another in the oven, she could cope. Two kids? She’s a stay-at-home mom.
Another friend, though, would bring her second infant into work and park her under the drafting table, but that was in an earlier time and a different company.
John Emerson 12.10.05 at 7:28 am
What does a woman gain by choosing not to have children at all? Has that been quantified?
Especially for the woman but also for the man, raising children is economically irrational. Either parents are mostly idiots — except for those who are wealthy enough to take on extremely expensive pets which are also serious legal encumbrances — or else economics is an incomplete, inaccurate, and destructive model of human behavior.
Slocum 12.10.05 at 8:37 am
Umm, you think women (of childbearing age) don’t worry about the timing of children? Some worry it’s too early, others too late, others worry about balancing the timing of career/kids/aging parents, and so on.
Oh, I know such women (and their husbands) do worry about when is the right time (my wife and I did — 1st kid at 27 BTW), it’s just that, in my experience, people aren’t as wildly obsessive about it as many are about the ‘right’ school.
My conjecture, BTW, would be that both the higher pay and greater hours arise from the same underlying factor–namely, whether or not the woman remains dedicated to her career and works parenthood around it or whether whe is a mother first and the job gets less attention. The older a woman is at first birth, the more established in her career, and the more likely she is to continue to advance in it (as opposed to seeking part-time work where flexibility is given greater emphasis by the woman and pay and opportunities for advancement less).
Daniel 12.10.05 at 9:40 am
women with children work less hours
pleeeeeassse, “fewer”. I have given up on “different than”. I have departed the field on “data” being plural. I no longer care about “hopefully” and people can say “inflammable” and “disinterested” as much as they like for all I care. But please let us hold the line on mass nouns and count nouns. Yes I was given six copies of that Lynne Truss book for Christmas.
nik 12.10.05 at 10:44 am
John;
I think the standard response of the economist would be that if people choose to have children, and in doing so forego income, then it can be rational to do so because the intangible benefits of having children are to them worth the income they forego. This is rational in the same way thatgoing to the pub instead of spending the evening at work is rational. It just means you place a higher value on what you choose to do than the value of the money you could otherwise have earned by going something else.
Brett Bellmore 12.10.05 at 2:01 pm
And having children earlier is a rational response to the fact that female fertility drops off so rapidly; Deciding to delay having children can be a decision to NOT have children, if you’re unlucky.
John Emerson 12.10.05 at 2:13 pm
By the economists’ logic the Doukhobors are rational when they burn down their houses and destroy everything they own. You just invent some fictitious house-burning-down mega-util, and voila!
Describing children as hyper-expensive consumption objects (with lots of opportunity cost) does not strike me as an intelligent way to treat childraising.
John Emerson 12.10.05 at 2:20 pm
So let’s pose the question differently. Calculate the average total income or average ultimate net worth of two couples, one of which raises children and one of which doesn’t. The difference divided by the number of children would be the utility-value per child.
Net worth is a better measure, since by calculating “total income”, you wouldn’t allow for the fact that children actually consume things. That is, if I buy food and a child or pet eats it, I don’t get to eat the food, but since the child / pet is “mine”, money spent on the child/pet’s food counts as if it were spent on me, or as if I’d eaten the food. But children/pets do detract from net worth.
Imagine a couple looking at their kids and having a number: “Has Susie really been worth $500,000 to us?” (A friend of mine started calculating by the pound: per pound, kids are remarkably more expensive to fatten up than cattle are.)
cm 12.10.05 at 3:18 pm
john emerson: I like your characterization of economics as a model of human behavior fallen short. Unfortunately, our technocratic society has an obsession with dogmatically explaining (perhaps actually rationalizing) everything with simplistic and moralistically laden models, while all the time claiming objectivity. But what else are the alternatives other than religious or ideological dogma?
abb1 12.10.05 at 3:35 pm
But does it really claim to be a comprehensive model? Isn’t it usually quite clear that it’s just a two-dimensional slice of a complex multi-dimensional object?
John Emerson 12.10.05 at 4:44 pm
But does it really claim to be a comprehensive model?
Economists seem to play a double game, speaking very modestly when cornered, but forgetting their modesty when making policy recommendations or talking about political issues.
But what else are the alternatives other than religious or ideological dogma?
Economics, as it is, already includes ideological dogma, and a better version of economics would be nice. As it is, economics ignores or misrepresents essential areas of human life.
John Emerson 12.10.05 at 4:47 pm
Brett — my poor mother had her first child when she was 28, so she was only able to have 7. Women can do quite a lot of waiting.
cm 12.10.05 at 5:43 pm
john emerson: Yes of course. I meant “ideological” as imputed purpose on society & human life outside of the realm of economics, if this concept has any merit (perhaps not, as by most appearances every ideology comes down to defining who gets the goodies, directly or through the backdoor).
Perhaps the underappreciated Sociology and Psychology should be merged and subsume Economics.
John Emerson 12.10.05 at 6:20 pm
Another way of looking at it is to ask whether the economic analysis of the family is normative or descriptive. If it’s intended as descriptive, a lot of sensible people seem to be damaging their net worth spending money for no return (like someone who spends a third or more of their income at a pub, as someone said). From a normative point of view, based on what economists say about other kinds of careless spenders whose net worth doesn’t rise, the message of economics would be that there should be no parents, or many fewer parents.
Brett Bellmore 12.10.05 at 7:13 pm
Indeed they can, John. I think that’s why it’s called “luck”.
John Quiggin 12.11.05 at 1:11 am
cm, it’s common in Australia (and widely regarded as preferable) for one parent, usually the mother, to hold a part-time job, and the other to hold a full-time job, particularly when children are of pre-school age, but also often during school.
The proportion of part-time jobs is, I think, higher in Australia than in the US and their relative pay and status is also higher than in the US (though not as good as might be hoped).
elliottg 12.11.05 at 1:26 pm
“From a normative point of view, based on what economists say about other kinds of careless spenders whose net worth doesn’t rise, the message of economics would be that there should be no parents, or many fewer parents.”
That is exactly what happens when income rises and/or reproductive choice becomes more available.
derrida derider 12.11.05 at 10:43 pm
“Economists seem to play a double game, speaking very modestly when cornered, but forgetting their modesty when making policy recommendations or talking about political issues.”
It’s not a bad rule of thumb, john emerson, to assume that any economist well-known for his or her public dogmatism ain’t much of an economist. There are exceptions – Krugman, Friedman, etc – but they’re just that, exceptions. Economists attached to financial institutions and business lobby groups should be certainly be ignored in public debate.
But if you actually have some policy responsibility, then you have to recommend *something* – absent a better model you use the one you know, warts and all. For absent an explicit model you’ll be using an unacknowledged implicit model, whose assumptions and logic are therefore untested.
bad Jim 12.12.05 at 5:32 am
There are times when investing in your children pays off. My brother and I started a company, and got our mother to invest in it. We did rather well. So the diapering and the Encyclopedia Americana and every other sort of parental effort paid off; mom’s got all sorts of disposable income as a result.
Truth to tell, nowadays, a great deal of her wealth is due to the original choice to buy a house in coastal California, for which most of the credit is due my late father, who made the mortgage payments for thirty years and died in timely fashion.
Joshua W. Burton 12.12.05 at 9:07 am
Daniel, reply 18:
pleeeeeassse, “fewerâ€.
The OED finds contrary citations from King Alfred (Swa mid laes worda swa mid ma) to Caxton (Be cause he had so grete plente of men of hys owne countre, he called the fewer and lasse to counseyll) to Lyly (I thinke there are few Vniuersities that haue lesse faultes than Oxford) and on down the centuries to recent numbers of Nature. It’s very elegant, but like most pop-grammarian pedantry the alleged “fewer”/”less” distinction is a mare’s nest, unsupported by the language’s actual history.
Native speakers only: please fill in the blanks.
Write “Grammar is a descriptive science” on the blackboard 100 times, and not one time _____.
Complete this sentence in 25 words or _____.
My dictionary cost sixty quid, much _____ than I expected.
The baby is _____ than six months old.
America has _____ than 200,000 troops in Iraq.
Our building is eight stories high, one _____ than the campus average.
Yesterday I left a fifty-cent tip; today I’m leaving five cents _____.
Six is one _____ than seven.
a rational economist (aka nik) 12.12.05 at 1:02 pm
I’m sympathetic to what John’s trying to do, but the economists way of looking at things does have some power. All you have to do is ask which of a group of options people would prefer (which choice maximises utility). John’s argument is just one from incredulity – i.e. “children can’t be worth *that* much”. He assumes he’s rational so other people’s choices must be irrational.
But is having children really “economically irrational”? Empirically, you can make the case that people do seem to place a very high value on them. If you ask people whether they’d rather lose their house or their child, they’ll probably say their house. So the child has to be worth more. And that’s even when children are quite old – and have depreciated in terms of the utility you get from playing with them when they’re small.
You could phrase it another way, why is giving children up for adoption unusual? People have a clear choice: keep them or give them away. Presumably the reason people choose to keep them is that knowing their child is worth more that the loss of time and income that the other choice brings. I know there are sunk costs in terms of the pregnancy, but even if you offered to compensate people for this (i.e. I’ll buy your first born son for $100,000), I’ve a feeling most of them would say no.
I think by “economically rational” there’s an assumption this just means maximise the amount of cash you earn. I’m not sure that’s the case.
steve kyle 12.12.05 at 9:18 pm
This is a good case of misapplication of economic logic. To imagine that having children is only or even primarily an economic decision is something only a middle aged economist with tenure could believe. Being a middle aged economist myself, I think it is ridiculous to imagine that monetary considerations drive most people to have or not have children and I know for a fact they didnt enter into my own decision to have kids.
When your spouse or partner first caught your eye was it visions of twenty dollar bills that danced through your head?
One of the things that makes people good analysts is knowing the limitations of their models. I know quite a few economists who suffer from deficiencies in this department but there is no reason for us all to follow them into pixie land.
John Emerson 12.12.05 at 9:34 pm
But if you actually have some policy responsibility, then you have to recommend something – absent a better model you use the one you know, warts and all. For absent an explicit model you’ll be using an unacknowledged implicit model, whose assumptions and logic are therefore untested.
If you are involved in policymaking you use the implicit model if the explicit model leaves important factors out. I don’t see the problem with this. It sounds like ass-covering — “Sure, I was wrong, but I was using a scientific model”.
John Emerson 12.12.05 at 9:45 pm
Obviously people do have children, but if that proves that it’s economically rational we’re back at the Doukhobor case. We’re in the zone of “revealed preference” and “effective demand”, where a sketchy spot in economics is waved away with a slogan.
Are alcoholics rational? Their revealed preference is to blow all their money on booze and die young.
In the vast majority of cases, childraising reduces your net worth, often greatly. How do economists treat other behavior patterns which lead to a reduction of net worth?
Is childraising a recreational activity, comparable to raising expensive pets.
If potential parents realized the problem and quit having children, as happens in some places, is that a good thing?
nik 12.13.05 at 3:42 am
John;
What’s the alternative to “revealed preferenceâ€?
If someone has kids (deliberately), hates it and would rather have got a new pair of golf clubs, I’m guessing that’s irrational due to insufficient information. If someone decides they want kids, tries very hard to have kids, has them, loves it, and wouldn’t change it for the world, I’m guessing that’s rational. Other than to to ask their opinion? (And I’m guessing an alcoholics expressed opinion would be that alcholism is not their choice).
steve kyle 12.13.05 at 1:40 pm
Economic analysis likes to pretend that people do things on the basis of all available information. The problem is that they dont or at least cant see how it applies to them when the issue at hand is a big one-time-only or irreversible decision whose implications are only clear later on. Children are certainly one of these cases. Nobody truly understands what it is like to have kids until they do. Buying a house is another – Most people do this once or twice in their lives – No learning from experience there either. Note that these are the two biggest “financial” decisions that most people ever make.
The bottom line is that while economic considerations play a role, so do lots of other things and it is just silly to pretend that economics is what really drives all of it.
bitchphd 12.13.05 at 1:53 pm
Not an economist, but my lay argument would be that the flaw in the economic argument about the value of choosing to have kids is that it is backwards. People don’t choose to have kids. They choose NOT to have kids. Having children is a basic biological function; you might as well ask why people spend X hours of otherwise potentially productive time breathing, eating, or pooping.
Hence, when incomes go up, people choose to have fewer children, because children are, strictly speaking, a poor investment. The “value” that parents place on kids–I’d give up my house before my kid, I’d give up my life before that of my kid–isn’t an economic calculation. It’s love.
Or is there some rational economic explanation for love, too, that I’m unaware of?
nik 12.13.05 at 3:44 pm
I think what I’m trying to get at is that I can’t agree with John when he says that if something reduces your net worth, then it’s economically irrational. I think that’s a misrepresentation of economic thought: if this were the case buying a tasty sandwich, as opposed to an otherwise identical bland sandwich, would be irrational.
How do I get out of the “revealed preference†dilemma with the Doukhobor and alcholics? I’m not really sure. I suppose the best attempt I can make is that “preference” matters, rather that “revealed preferenceâ€. So doing something that goes against the preferences of your “true rational self” – because of a compulsion or mania – would be irrational.
Comments on this entry are closed.