## Do women want more children than they end up having?

In his column, Ross Douthat states that “most women want more children than they have”, linking to a medium article about the gap between actual and intended fertility.

The big argument/finding of the linked-to article is this: the scale of the gap between actual and intended fertility in the United States is between .3 and .6 kids, and this has been relatively stable for a few decades. The analysis seems plausible. But it doesn’t get you to “most women want more children than they have”, and I’d argue it’s only barely consistent with it. If you assume we’re at the top of the range (.5-.6) and that all of the women who want more kids only want one more kid, Ross’s line would be accurate. But, for instance, if you instead assume the midpoint of that range (which is also the most recent point estimate), and that half the women with below-intended fertility wanted two more kids and half wanted one, you’d instead be looking at 30% of women who wanted more children than they have. (To get a sense of the distribution, you’d want the person-level data.)

My guess would be that Douthat’s quote is derived from the article’s line that “women have fewer children than they say they would have liked to have.” But that’s not the same thing.

You could be right—but it must be possible to break down the survey data for individual respondents, so Douthat’s question, even if it can’t be answered from averages alone, probably can be answered from the raw data. My guess is that if we were to label:
X = the percentage of women who have had fewer children than they wanted
Y = the percentage of women who have the exact number they wanted
Z = the percentage of women who have had more children than they wanted,
then the data would show that X > Z. But not necessarily that X > 50%. Douthat literally claimed X > 50% but I think that’s just because he’s being casual with his math, and I expect that X > Z would still make his point.

In the real world, all this is complicated by the fact that people’s intentions change. A couple can fully intend to have 2 kids, but then baby #2 is so adorable that they decide to try and have a third kid . . . and, after a few years, they succeed. So the number of children is 3, but what’s the desired number? It’s ultimately 3, but it was originally 2. Or, maybe a couple isn’t planning to have a fourth child, but they have one by accident. Mistakes happen! But then they’re very happy to have had kid #4. So, they had more children than they originally wanted, but they’re happy with the number of children they had.

I’m sure some sociologists have looked carefully at such questions, as many of them are directly answerable from the data, if you have a series of surveys asking people of different ages how many children they have, how many they plan to have, and how many they would like to have.

1. oncodoc says:

Since the advent of agriculture eight to nine thousand years ago, population grew very slowly till technological and economic development. Then there was a rapid growth phase which has now returned to the same rate as previously in most places including Asian and African nations that were late to achieve economic growth and the return to the historic reproductive patterns that Europe achieved a generation ago. People had more children in the past, but childhood mortality was high; for example only one of Abraham Lincoln’s four children survived to adulthood despite being a prominent family. People seem to limit family size to replacement rates more or less throughout history. Pre-technology this came about due to childhood death from infections; we are blessed to use contraception to do this.

2. Z says:

Also, very few parents would admit to a pollster that they have more children than they want.

• Alison Gemmill says:

You’d be surprised to hear that a fair amount of older women (40-44) state that their ideal number of children is smaller than their actual number of children. (I obtained this result analyzing the National Survey of Fertility Barriers for a related project.)

• Martha (Smith) says:

“very few parents would admit to a pollster that they have more children than they want.”

Especially if one of their children (or anyone else in the family or who knows the family) is in earshot.

3. This has been extensively studied by demographers/sociologists. There’s a nice paper from Morgan and Rackin (“The Correspondence Between Fertility Intentions and Behavior in the United States”) that uses a longitudinal dataset to assess how intentions measured at age 24 correspond to achieved fertility.

From their article:
“We฀ﬁnd฀that฀the฀mean฀or฀net฀error—the฀mean฀difference฀ between฀ intended฀ parity฀ at฀ age฀ 24฀ and฀ completed฀ fertility฀ in฀2006—was฀relatively฀modest:฀–.25฀and฀–.40฀fewer฀children฀than฀intended,฀for฀women฀and฀men฀respectively.”
Also from their article:
“we฀ examined฀ whether฀ individual฀ women฀ and฀ men฀ were฀ achieving฀ their฀intended฀family฀size.฀We฀found฀that,฀for฀NLSY79฀respondents,฀the฀answer฀is฀largely฀“no”—by฀2006฀(at฀age฀45),฀only฀43฀percent฀of฀women฀had฀realized฀their฀intended฀parity฀at฀age฀24.฀Instead,฀frequent฀errors฀in฀under-฀and฀overachieving฀fertility฀intentions฀are฀partly฀compensating฀to฀produce฀similar฀overall฀levels฀of฀intent฀and฀behavior.฀But,฀as฀indicated฀by฀the฀negative฀net฀error,฀underachieving฀is฀somewhat฀more฀common฀than฀overachieving.”

And yes, fertility intentions are fluid across the life course. (See, for example, Hayford (2009) “The Evolution of Fertility Expectations over the Life Course” I also have a working paper (with a conditional acceptance) showing that among those who remain permanently childless, ~56% intended no children from a young age. The rest intended children well into their 30s/40s.

Unfortunately, there are few longitudinal surveys for more recent cohorts that allow for this kind of investigation. Instead, we have to rely on the synthetic cohort approach used by the Medium author.

4. Carlos Ungil says:

> A couple can fully intend to have 2 kids, but then baby #2 is so adorable that they decide to try and have a third kid

Or maybe baby #2 is so awful that they decide to try again.

• Martha (Smith) says:

I recall a couple who decided to have more kids after their first batch was grown, because they felt they didn’t do a good job raising the first set and wanted to try to do better with the second set.

5. Motherhood is the most fulfilling endeavor. I would have liked 4 children. Had 2. However, staying home is pretty unfashionable in several circles I’ve come across.

• Corey says:

Do you mean motherhood specifically or parenthood in general? Do you mean the most fulfilling endeavor unconditionally?

• It turns out in my opinion, that in the US a major “cause” for people to stay home after having kids, is the tax code.

Suppose you have a family of 2 adults, and they have 2 kids. Suppose they’re both reasonably well educated and pre-kids had reasonably well paying jobs. Let’s say between the median income of about 55k and an upper-middle class income of 100-150k. Suppose one parent has somewhat higher income than the other pre-kids (say 10%)

Now, the family needs to provide child-care for their children. This means: pre-school age all day long, and school age, getting them to school, picking up from school, dealing with school related minor issues (potty accidents at school, jammed fingers, skinned knees, etc) and pick up after school (around here at 2:15pm) or paying for an after-school program (around a few hundred dollars a week total real cost, but sometimes subsidized so that local govt is paying a portion)

On top of this, kids produce a lot of laundry, and kids between ages of say 5 and 10 consume as much food as a full grown adult woman, between 10 and 18 potentially as much as a full grown adult male. So, shopping and cooking are a major endeavor.

If you want to invest time and effort into your kids after-school activities (sports, music lessons, etc) you need even further time investment.

Now, for example in round numbers, perhaps the “first” earner earns \$75k, and the second earner, before kids earned say \$65k.

To procure the required child care, shopping, food prep, laundry, after-care, etc costs, well let’s go with \$15/hr and 20hrs a week minimum, and 30 or 40 hrs a week during pre-school ages: \$15-30k in labor alone for the most part post-tax dollars. Practically speaking there is additional overhead, let’s round to \$20k/yr for illustration.

Evidently the higher earner shouldn’t quit their job, so the person earning \$65k needs to make the following tradeoff:

For every dollar this person earns, 11% goes to ss&medicare, 25% goes to federal taxes, and lets call it 8% state taxes, and another 8% sales taxes on items like pre-made food and laundry services etc that you buy instead of produce…

Basically, 11+25+8 = 44%, so the person takes home 65k * 0.56 = \$36k at best, with no retirement savings, subtract the \$20k cost of taking care of the children, and they work an 8+hr day with commute (cost perhaps \$5k/yr in commute expenses) to earn something like 36-20-5 = 11k/yr take home, and \$29k to taxes, and \$20k to “nanny services” or whatever. Assuming 2000 hr a year at their job that’s \$5.5/hr take home. Or less than half of minimum wage here in CA. This person is working above median wage middle class job with a college education, and taking home less than a high school student serving ice cream.

Now, this might still favor going ahead and doing the work, but let the primary earner increase in wage from say \$75k to \$100k … and state taxes go up a little due to a bracket change (say from 8% to 11%), longer hours for the primary worker due to higher job requirements, etc and pretty soon you can’t justify having the second person working for less than *more than the primary earner earns*

Now, if you want the second earner to be treated symmetrically with the first earner, it’s easy to prove that the taxes *need to be linear* aka FLAT TAX. For all IncA and IncB values for the two spouses we would have to have:

Tax(IncA + IncB) = Tax(IncA)+Tax(IncB)

is the symmetry requirement, which is in fact the definition of linear.

We can also add a government basic income so that your take home income is:

Inc_takehome = UBI + (1-TaxRate) * (IncA + IncB)

Any other system treats *second earners* extremely badly and destroys about 50% of the productive capacity of a modern society by forcing the bulk of second earners to stay home and take care of kids when they could instead have a reasonable choice to be paying a nanny and doing more productive work (worth more to society in dollar terms at least)

Fortunately, there’s a mathematically unique simple family of solutions (the UBI + flat tax, with a tax rate parameter and a UBI parameter)

Unfortunately, the existence of a simple single family of solutions which could provide massive improvements in real growth is not sufficient to get people to adopt it.

/end rant

• Let’s not forget that for 3 months of the year there’s no public school… so we’re talking summer camp or full time staying home… boosting the annual cost even more. Realistically, you’re probably talking \$30k/yr in after tax dollars you’re putting. Add an additional child, it’s even more, basically middle class families can not really afford to have a second earner actually working if they have 2 or more kids and less than \$100k earnings potential for the second earner, and this is virtually entirely due to the high marginal tax rate of close to 50% on the second earner’s income.

If instead you had a flat tax of say 30% on all income, and a UBI of say \$500 per adult and \$250/child then literally TENS OF MILLIONS of educated women would most likely return to the labor market, TENS OF MILLIONS of unemployed people would have opportunities in nanny type labor, the overall labor market would allocate MUCH more efficiently, and in the medium term you might expect real economic growth to go from a current average of a percent or so, to maybe 5 or 6% with far more broadly spread income, and reduced inequality.

• Terry says:

It’s very odd (and sad) that preventing extinction is seen by many as an unworthy endeavor.

6. Parenthood in general. Obviously it can be difficult endeavor. In my own case, I guess what I mean that it was important to watch my kids grow. That doesn’t mean it is what fulfills others. Each person has goals and needs. My mother hated being a mother I think.

7. Bryan Sayer says:

But the statement “on average, women want more children than they have” might be accurate, although I am still confused as to whether desired fertility is “right now” versus completed.

Not to mention the fact that half of pregnancies are unintended, so it is hard to reconcile the unintended rate with a lack of achieving desired fertility.

• Abigail says:

Coincidentally, I’m in the middle of Jonathan Last’s “What to Expect When No One’s Expecting”, which is what you would expect from the title. It’s a fun read. The biggest surprise to me so far is that U.S. fertility has likely been declining nearly-monotonically (the baby boom being the big exception) for at least a couple of hundred years. There’s an NBER paper showing some data here: http://www.nber.org/chapters/c9686.pdf (I don’t recall if that’s what Last cites.)

8. Terry says:

It is helpful to know the actual question that the survey asked.

Clicking around a bit, you find that “the main measure used in this indicator is the mean personal ideal number of children, which reflects the number of children that people consider as ideal for themselves personally as averaged across respondents (Chart SF2.2A). (See https://www.oecd.org/els/family/SF_2_2-Ideal-actual-number-children.pdf.)

So responses are not wishful thinking. Respondents are not saying “sure I’d like to have more children” in the sense that they would say “sure I’d like to have more money or a bigger house”. Rather, the survey asks for the ideal number of children for them, and presumably that includes the costs and benefits of having children, so the responses are realistic assessments and not just pie-in-the-sky wishful thinking.

Hence, the survey will usually produce a shortfall of actual children on average if women have a substantial say in how many children they have, so finding shortfall is not surprising. Presumably, if a woman has an ideal number of children in mind, she will tend to stop when she reaches that number. But, some women will stop before the ideal number for other reasons (such as infertility), so the average actual number of children will usually be less than the ideal.

Yes, some women will have more children than their ideal. For instance they may be forced to in some sense or it may just happens because they are not trying very hard to limit the number of children. But, in a society where women have substantial control over the number of children they have, a shortfall is not at all surprising. The question is therefore a more difficult one: is the shortfall too large?

• Andrew says:

Terry:

You write, “Yes, some women will have more children than their ideal.” My point is that “their ideal” can change over time, either from circumstances (you find you have more or fewer resources than you thought you have) or from new information (you find children to me more adorable or more difficult than you expected) or even just from general trends (it used to be that 2 children was the norm, now it’s 3).

Don’t get me wrong—I still think it’s useful to measure these things and ask these survey questions. We just need to be careful in how we interpret the responses.

• Terry says:

I agree you have identified another ambiguity which needs to be considered.

• JFA says:

The sentence following the one you quoted: “This information is available
for European countries from certain waves of the Eurobarometer survey, based on survey responses to the question: “And for you personally, what would be the ideal number if children you would like to have or would have liked to have?” “

I’m going to assume that the question uses ‘of’ rather than ‘if’. To your point about the answers not being wishful thinking, I’m not sure we can make any judgment on that. If a question starts out “For you personally, what would the ideal…”, the emphasis doesn’t seem to be on constraints. So if a woman who has a job that she likes and has 2 kid but her ideal would be 3, she could be thinking that “I would love to have 3 kids, but daycare is so expensive or we’d have to buy a bigger house or etc., but my ideal is 3.”

There’s no way to tell whether the respondents are thinking about current constraints or not. If surveyors wanted to know that, they could just ask “Given your income and life goals (career aspirations, desire to travel, early retirement, etc.) what number of kids would you would like to have or would have liked to have?”

• Martha (Smith) says:

I agree that the wording including “ideal” is likely to be interpreted as “if there were no constraints”

9. JFA says:

Just to put an economist’s take on it: maybe we should look at the revealed preference rather than stated preference. There are lots of things that I would like more of, but due to budget constraints and greater marginal net benefits of other things, I have less of those things.

• JFA says:

E.g. I don’t have three kids, not because I don’t want three kids, but because I would rather use the money I would spend on a third kid and buy a house for my spouse, my two kids, and I to live in. Do I have fewer kids than I would like? Yes. Does that mean I am worse off or failing at life? No. I just want other things more than a third kid.

It kind of reminds me of surveys asking people how much they would donate (hypothetically) to various environmental/conservation causes and those responses adding up to some absurd amount which they would not (collectively) be able to afford.

• Alison Gemmill says:

Although, how much of childbearing is truly rational? Nearly half of all pregnancies in the US are unintended, contraceptive failure is a reality, pregnancy intentions are nuanced, etc. Fertility is beautifully complex. :)

• JFA says:

Whether or not the pregnancies are unintentional, the stated preference is still larger than the revealed preference. Bringing in the issue of unintended pregnancies (some of which are unwanted, others just ill-timed) just means that the average stated preference for the number of children is even more inflated and that we probably shouldn’t put too much confidence in that number or infer anything meaningful about women’s true desires.

My guess is that most unintended pregnancies are just ill-timed rather than unwanted (in so far as it relates to the number of kids a woman would want), and in the end, the number of children a woman has is (on average) the number she wants given her desires (career goals, hobbies, desired free time) and budget constraints.

• Abigail says:

>An unintended pregnancy is one that was either mistimed or unwanted (45% of all pregnancies). If a woman did not want to become pregnant at the time the pregnancy occurred, but did want to become pregnant at some point in the future, the pregnancy is considered mistimed (27% of pregnancies). If a woman did not want to become pregnant then or at any time in the future, the pregnancy is considered unwanted (18% of pregnancies).

• Beyond contraceptive failure… there is also the people trying hard, maybe going to fertility doctors, and still not being successful. There are also people who it takes long enough that now they’re afraid of the very highly increasing rate of genetic disorders as the mother increases in age over 40… so it isn’t just all about revealed preferences for resource usage.

• In fact, given the high costs of fertility docs. you might say that many of these people are revealing a very very high economic preference for something that they just can’t biologically accomplish.

• JFA says:

Infertility is certainly an issue. I found this article (https://www.reuters.com/article/us-couples-infertility-idUSBRE90A13Y20130111) that suggests about 15 percent of couples couldn’t get pregnant after a year of trying (though it was a small sample but that does line up with other estimates I’ve seen). Not insignificant, but it is a small percentage of the population. Success rates of various fertility treatments vary (see the first table here https://blog.episona.com/the-cost-and-success-rate-of-treating-infertility). The successful treatments have a higher likelihood of ending up with multiple births (about a quarter of the pregnancies ending in live birth. The CDC puts the percent of all women who have used fertility treatments at about 12%, percent of women with impaired fecundity was 12%, and infertile was 6% (https://www.cdc.gov/nchs/fastats/infertility.htm), they didn’t give the fast stats on men). So there are a lot of women who experience infertility and get fertility treatment. You’d have to look at the age breakdown and marry that with data on success rates to get a sense about the percent of women with infertility issues who end up having children.

People are pushing back family formation these, getting married later and having kids even later. The risk of infertility increases with age. These things are choices. But overall, given the stability of the gap between stated and actual fertility, it seems that women are revealing they don’t want as many kids as they say.

• Lots of women don’t really know the stats on fertility etc, particularly the shape of the risk curve wrt age (it climbs very rapidly in the vicinity of age 40). I don’t think it’s fair to say that pushing back your conception date counts as “revealed preference” unless the people involved are actively considering the tradeoff and have real data.

I don’t entirely disagree with you, I mean, I do think you bring up a good issue, it’s just that I’m kind of pushing back against an inappropriate circularity in many Economic arguments: “you can’t ask people what they want, you need to go with their revealed preferences, and therefore when you look at revealed preferences… it turns out everyone is getting exactly what they want! there are no problems to be solved in the world!”

I think revealed preference is a more useful concept when people make some concerted effort to consider the tradeoff they’re faced with, and when their model of the world is roughly correct so that their predictions for the outcomes correspond to the reality relatively well.

For example, suppose manufacturer A makes pen Q which works well, and manufacturer B makes fake pen Q’ which appears to do virtually the same thing but rapidly breaks before you even finish writing your first page of notes… on the other hand, it costs just a little less.

Now, consumers go out and buy large quantities of pen Q’…

Economist says: See consumers clearly prefer product Q’ with short lifespan and slightly lower cost. They may “say” they like reliable products, that write for hundreds of pages, but they really only care about the first 3 sentences…

well, that’d be true *if there was a clear label on the product THIS PEN TYPICALLY WRITES LESS THAN 35 WORDS BEFORE BREAKING* but if the consumers don’t know what they’re getting… you can hardly use the fact that they buy the inferior product as evidence that they prefer inferiority.

• In particular, one might argue that when faced with good information, 15% of women go to fertility doctors and spend tens to hundreds of thousands of dollars to have more children, thereby indicating that once informed, their preferences are strongly to try to “correct” what they perceive as an error in planning etc.

Furthermore, this attempt at correction frequently leads to multiple births (twins, triplets etc) and that probably increases the number of people who have the “wrong” amount for their preference. I certainly know people who tried to have a third child and wound up with a set of twins. “Revealed preference” would say, hey they really must have wanted 4 kids anyway. But that’s clearly an insane Pollyanna way to think about the world: everything is right in the world, and everyone gets just exactly what they “really” wanted!

• JFA says:

I appreciate the push back. It’s good to temper my type of reasoning. Regarding your pen example, if people keep buying the pen, then I would say they reveal their preference for the “inferior” pen. You have to make a lot of assumptions (no consumer reviews, no way of producer A to differentiate her pen from producer B, etc.) to have the situation end up with consumers being hoodwinked consistently overtime. I think there are situations in which realized choices don’t provide evidence for revealed preference, but the pen example might just be a poor example.

Regarding how much women know: my impression is that it is mostly educated women who wait until they are older to try to have kids. Even as an adolescent in a relatively uneducated household, I heard conversations about infertility (or at least difficulty having a child) increasing with age. Maybe people don’t know the exact probabilities, but I think most women have a good idea that having kids will be harder when they’re older. And pregnancies of those over 35 have been treated as high-risk for a while, and I can’t imagine that type of information not trickling down as well. But this is all speculation Maybe some other commenter can point us to some data that would adjudicate the issue.

Regarding whether having that extra kid after IVF is revealed preference, it seems that the no one going through these fertility treatments would be unaware of the risk of having non-singelton births. So yeah… given the expected number of kids, it seems that those parents reveal there preference for having 4 kids. Ex post, is that more than their ideal? Maybe. But at least ex ante, they preferred taking the risk of having 4 kids over the certainty of just having 2.

10. Terry says:

Was anyone else struck by this Douthat sentence?

“But there is also strong resistance to seeing a failure to unite the sexes and continue the species as a problem.”

Why isn’t a “failure … to continue the species” seen as a big problem?

What could be a bigger problem than extinction?

• Carlos Ungil says:

Some great apes are considered endangered but the Homo genus is not among them. The human race may dissapear some day in some catastrophic extinction due to nuclear war, or avian flu, or misuse of p-values, but it’s hard to imagine that it will be caused by a completely generalized unwillingness to reproduce.

• Anoneuoid says:

The human race may dissapear some day in some catastrophic extinction due to.. .misuse of p-values

I’m sure you’ve seen this quote before and the meaning hasnt sunk yet, or you havent had that life changing realization yet, etc. So you are probably being sarcastic, but this is really may happen (even according to the guy who popularized the whole thing):

“We are quite in danger of sending highly trained and highly intelligent young men out into the world with tables of erroneous numbers under their arms, and with a dense fog in the place where their brains ought to be. In this century, of course, they will be working on guided missiles and advising the medical profession on the control of disease, and there is no limit to the extent to which they could impede every sort of national effort.”

Fisher, R N (1958). “The Nature of Probability”. Centennial Review. 2: 261–274. http://www.york.ac.uk/depts/maths/histstat/fisher272.pdf

• elin says:

It’s not seen as a problem because it is not even close to happening.