Do Baby Girls Cause Divorce? A New Freakonomics Radio Podcast

(Photo: Marie Smith)

Our latest Freakonomics Radio podcast is called “Do Baby Girls Cause Divorce?” (You can subscribe at iTunes, get the RSS feed, or listen via the media player above. You can also read the transcript below; it includes credits for the music you’ll hear in the episode.)  

This episode was inspired by a question from a reader named John Dolan-Heitlinger, who wrote the following: 

My wife has observed that in marriages where there is a son there is less chance of the husband leaving the marriage.  

I wonder if that is true.  

Thanks for your consideration. 

Mr. Dolan-Heitlinger asks, and we deliver. And his wife, as it turns out, is right. In a paper called “The Demand for Sons,” the economists Enrico Moretti and Gordon B. Dahl examined differences in marital rates based on whether a first-born child is a son or daughter. Here are some of their findings:

  • Couples who conceive a child out of wedlock and find out that it will be a boy are more likely to marry before the birth of their baby.
  • Parents who have first-born girls are significantly more likely to be divorced.
  • Fathers are significantly less likely to be living with their children if they have daughters versus sons.
  • In any given year, roughly 52,000 first-born daughters younger than 12 years (and all their siblings) would have had a resident father if they had been boys.
  • Divorced fathers are much more likely to obtain custody of sons compared to daughters. 

“Son preference” is not new, of course (and we’ve dealt with a different version before on this show). Gallup has been polling on this question since 1941, and the results have barely budged. In 2011, 40 percent preferred sons and 28 percent daughters; the rest stated no preference or opinion. (In 1941 the margin was 38 percent to 24 percent.)

In this podcast, Stephen Dubner talks with Enrico Moretti about the research itself and the broader economic implications of so many girls living without their dads. 

MORETTI: For children and families with absentee fathers due to a first-born daughter, family income is reduced by about 50 percent and poverty rates are increased by about 30 percent. So these are economically important effects. 

Thanks again to John Dolan-Heitlinger for the question that sparked this discussion. Please keep your good questions coming! 

Audio Transcript

[MUSIC: The Jaguars, “By By Mai Thai” (from The Jaguars)]


Enrico MORETTI: I’m Enrico Moretti and I’m a professor of economics at Berkeley.


Stephen J. DUBNER: Okay, very good. So Enrico, a listener wrote to us with a very, very straightforward questions which is this, “In marriages where a baby boy is born, is there less chance of the husband leaving the marriage?” So we can get into the details later, but can you…You know, economists are famous for never giving a yes or no answer to anything. I’m wondering if you can give us a yes or no answer to that question?


MORETTI: Yes, it’s an easy answer. And the answer is yes. Parents who have first-born girls are significantly more likely to be divorced. And so parents who have first-born boys are significantly more likely to stay together.




[MUSIC: Pearl Django, “Samois Swing” (from Swing 48)]


ANNOUNCER: From WNYC: This is FREAKONOMICS RADIO, the podcast that explores the hidden side of everything. Here’s your host, Stephen Dubner.


DUBNER: You just heard Enrico Moretti, an economist at the University of California, Berkeley, tell us a striking fact: that parents whose first child is a daughter are significantly more likely to be divorced than if they’d had a son. How does Moretti know this? He wrote a research paper, along with the economist Gordon Dahl, called “The Demand for Sons.” They analyzed U.S. census data from 1960 to 2000, along with other data, to measure the effect of a firstborn child’s gender on marital stability.


MORETTI: We find that fathers are significantly less likely to be living with their children if they have daughters versus sons. This overall effect is fairly large; it’s about 3.1 percent lower probability of a father for families with a girl.


DUBNER: And can you put that into numbers of families, or daughters for me then?


MORETTI: Yes, we estimate that over a 10-year period, that accounts for about 50,000 first-born daughters who are living without their father.


DUBNER: Wow. So Enrico, you’re saying that there is a significant, not huge, but significant effect on marriage that a firstborn daughter will have, that a firstborn daughter will decrease, you know, marriage rates or increase the families that are splitting up. Why? What are the channels, you know, by which this firstborn daughter leads to fewer fathers living with the family? Why do male offspring seem to be better at holding marriages together?


MORETTI: There are three main channels. First of all, women who have daughters are more likely to have never been married than women who have boys. Second, parents who have firstborn girls are significantly more likely to be divorced. And third, divorced fathers are much more likely to obtain custody of their sons compared to their daughters.


DUBNER: Okay, so three channels you say. One is when there’s a couple who conceives a child out of wedlock, if that child is a girl, you’re saying the couple then is less likely then to marry, yes?


MORETTI: That’s correct.


DUBNER: Talk to me for a minute about that. A) I’m just a little curious about what your response was when you saw that in the data. And B) I’m curious if you can offer any why on that one for us?


MORETTI: We were very surprised when we found the evidence on marriage in the data. We were even more surprised when we found evidence that the gender of the kid affects the probability of shotgun marriages. In particular we find that for parents who are not married at conception of the kid, among those who learn that their future child will be a boy, they’re more likely to marry by the time of delivery compared to parents who learn that their future child will be a girl.


DUBNER: Now I just want to step back. So there’s a Gallup poll that you describe that’s been taken ever since 1941 with a survey question that asks, “If you could have only one child would you prefer that it be a boy or a girl?” And in 2011, the most recent survey year, boys led girls 40 percent to 28 percent, with 26 percent saying it doesn’t matter. So first of all, why such a large gap? Why do you think, or what do you know about why people have such a strong son preference?


MORETTI: Well, one thing we do know is that it’s mostly driven by men. When women are asked whether they would like to have a boy or a girl, they are evenly split. But when men are asked when they’d rather have a boy versus a girl, they tend to favor boys. And this is consistent with the evidence in our study that comes from actual choices that people make rather than their self-reported preferences.


DUBNER: And do we know, can you unpack the number and tell us why that is, why do fathers at least say they want and as your data show actually want sons more than daughters?


MORETTI: Our study doesn’t really address the causes. We take that they are cultural. And economists take preferences as given, they don’t discuss preferences, they don’t question preferences; the best they can do they can measure preferences. And that’s what we seek to do.


[MUSIC: Euforquestra, “Chango” (from Explorations in Afrobeat)]


DUBNER: “That’s what we seek to do,” Moretti says. Measure people’s preferences, not dictate them. That is, often, a very fine line in modern society. The world is a bed of nails, and we are all hammers. Enrico Moretti just wants to know how the nails got there, and what they’re made of. When we come back, we’ll drag Professor Moretti back onto some more-familiar ground -- the economic impact of the daughter effect:


MORETTI: For children and families with an absentee fathers due to a firstborn daughter family income is reduced by about 50 percent.


DUBNER: And, me being a hammer, I ask: what should we do about this?


MORETTI: You’re talking about a tax on divorce?


DUBNER: Yes I am.


MORETTI: Good luck with that.



ANNOUNCER: From WNYC: This is FREAKONOMICS RADIO. Here’s your host, Stephen Dubner.


[MUSIC: Wolfram Gruss, “Petit Gennevilliers”]


DUBNER: Let’s say there’s a couple – for the purposes of this conversation, a woman and a man. And they are expecting a baby. Congratulations! And then they have an ultrasound and they find out it’s going to be a girl baby – fantastic! But… uh-oh. The Berkeley economist Enrico Moretti tells us that if this couple isn’t married, having a baby daughter instead of a son makes it less likely that they will get married. And if the couple is married – well, that couple is more than 2 percent more likely to get a divorce. Why?


MORETTI: Well there are three fundamental possible explanations. The first one is that parents have a preference for boys over girls. And so when they learn they’re going to have a boy that makes them happier and keeps the family intact at least more than when they learn they have a daughter. Explanation one is therefore parents are gender biased. But it’s also possible that parents are not gender biased. It’s also possible that they like boys and girls equally but they realize that having a father might be relatively more important for boys than for girls, either because fathers play an important role model for the boys, or because fathers might have some other type of comparative advantage in raising boys versus girls. And there’s also a third possible explanation. Again, in this third explanation there is no gender preference but parents might realize that girls are more costly to raise than boys in terms of time and or monetary expenses.


DUBNER: Why are girls more costly to raise than boys?


MORETTI: Well there’s actually, believe it or not, there’s actually literature on this question and it’s not perfectly conclusive, but a growing number of studies suggest that girls tend to be more expensive because they require more time, and they tend to be more, girls incur more monetary expenses, especially in the teenage years.


DUBNER: Are talking, we’re talking expenses, are we talking about what? Are we talking about educational expenses, are we talking about personal grooming stuff, clothing and makeup?


MORETTI: Personal expenses.




MORETTI: I want to make clear this is not something that we found, but this is something that other studies have found.


DUBNER: I understand. O.K. so you identify three possible explanations for why this is happening. One is preferences, just parents prefer boys for whatever reason. Two might be what you call a kind of compensatory behavior where a father would be more likely to stay with a family if there’s a boy because he thinks it’s more harmful for a boy to grow up without a father as a role model. And the third is that it’s harder and or costlier to raise girls than boys. Which one or ones of these explanations do you find most plausible or would carry the most weight?


MORETTI: Well in order to make progress on that question we looked at fertility decisions. So we looked at the decision of adding an additional baby after you already have one child. And we compare families who already have one boy with families who already have a daughter. And we look at the probability of going for a second kid. And what we find is that in families with a firstborn daughter the total number of subsequent children rises significantly. By our estimate, a firstborn daughter causes approximately 5,000 additional births per year compared with families with firstborn boys.


DUBNER: And you looked at this over about 40 years. You’re talking about 200,000 extra children you’re saying because of what looks to be a son preference.


MORETTI: That’s right. Families whose first born is a boy seems to feel less of a need of adding a second kid relative to families whose firstborn is a daughter. And this in our opinion is very solid evidence that even today U.S. parents have strong preferences for boys.


DUBNER: Now, these fathers who skip out on their families if a daughter is born first whether the couple is married or not, this has serious negative consequences, yes? Can you talk for a moment about that?


MORETTI: The effects are large. For children and families with an absentee father due to a firstborn daughter, family income is reduced by about 50 percent and poverty rates are increased by about 30 percent. So these are economically important effects.


DUBNER: So I know the following question is always a dangerous one because economists especially generally don’t like to be prescriptive, but knowing what you know about this issue if I were the president or a senator or governor and I were to come to you and say Professor Moretti I understand you’ve done some research on this topic, can you recommend a policy idea based on this that would help more families, more children, especial more daughters, would you do anything about it?


MORETTI: It’s really hard to see what the president or congress could do about it. These are deep-seated cultural norms and values. They don’t seem to be going away over time. It’s pretty clear that being born in a broken family has long-lasting economic consequences, and I think maybe that’s where we should focus our energies. I’m not sure it’s feasible, or even desirable for the government to try to change in any ways people’s preferences.


DUBNER: What about however…I mean, given the size of that impact, how much a family suffers a result of that divorce, you do believe in incentives, incentives can even help overcome preferences. So should there be an anti-divorce preference that goes across the board that isn’t meant to directly address the son preference but manages to catch it along the way?


MORETTI: You’re talking about a tax on divorce?


DUBNER: Yes I am.


MORETTI: Good luck with that.


DUBNER: Or a marriage bonus, however you want to put it. Or I don’t know, maybe it’s, maybe it is a firstborn daughter bonus.


MORETTI: You know, one thing that I want to make clear is that the gender of the children is not the only factor in determining divorce or marital stability. In fact, it’s not even the main factor in determining divorce or marital stability. There are many, many factors that go into a couple’s decision of splitting and gender is just one of the many factors. And I’m all in favor of thinking of economic policies that can foster family stability. I’m not sure we should specifically target gender of the children as one source of economic policy.


[MUSIC: Crytzer’s Blue Rhythm Band, “Dickey’s Blues” (from Chasin’ the Blues)]


DUBNER: Enrico, are you married?


MORETTI: Yes I am.


DUBNER: And do you have any children?


MORETTI: I have a wonderful four-year-old boy.


DUBNER: And what is his name?




DUBNER: Mateo. Did you have a son preference?


MORETTI: I actually did. Before having Mateo, I was really, really hoping to have a daughter. But ever since he was born, I find myself delighted at having a boy. And I can’t believe I wanted a girl. This is the best thing that’s ever happened to me.


DUBNER: Why did you want a daughter?


MORETTI: I have, I was hoping to have a daughter that looked exactly like my wife, but much younger.


DUBNER: Does Mateo know yet that you were hoping for a daughter?


MORETTI: He doesn’t, but I don’t think he has any concern about that. He’s four, he’s happy, he’s just a bundle of joy.


DUBNER: He doesn’t listen to podcasts does he?


MORETTI: He doesn’t listen to Freakonomics Radio just yet.


DUBNER: So Enrico Moretti’s boy Mateo doesn’t listen to this program – but John Dolan-Heitlinger does. He’s a consultant in Key West, Florida, – and he’s the guy who wrote in with the original question. He said: “My wife has observed that in marriages where there is a son there is less chance of the husband leaving the marriage. I wonder if that is true.” So after we interviewed Enrico Moretti, and found out that it is true, we let John know. Here’s what he wrote back: “I will tell my wife Eileen she was right, as she typically is.”


[MUSIC: Geb Zurburg, “Sometimes When I’m Home” (from Rosewood)]


DUBNER: So John, thank you for writing. And thanks for giving us ideas maybe for future podcasts. Such as: how much more likely is a marriage to last when the wife, like Eileen in this case is usually “right”? What about when the husband is usually “right”? Or when the husband and wife are right an equal number of times? How much more likely is a marriage to last if… (fade).



Leave A Comment

Comments are moderated and generally will be posted if they are on-topic and not abusive.



View All Comments »
  1. Robert says:

    One of my first thoughts is that this study should have looked at who initiated the divorce (perhaps it did, but I’ve only read the abstract at this point, and there’s no mention of this there).

    Women have just as much right to file for divorce as do men, and so blaming the higher divorce rates for couples with daughters on male preferences for sons seems unfounded. It seems just as plausible to me that women could feel more comfortable filing for divorce when they have only daughters; girls are generally seen as easier to raise and the importance of the father is more widely recognized in a young boy’s life than in a girl’s.

    Well-loved. Like or Dislike: Thumb up 53 Thumb down 13
    • Jason says:

      While I agree with your premise, that it is incorrect to just assume the higher divorce rate is due only to a father’s preference for a son, I would challenge your statement that girls are generally seen as easier to raise. While I’m not a parent, I’ve consistently heard the opposite from parents that I know.

      Also, while it is only anecdotal evidence, the other finding of the study, that men are more likely to obtain custody of a son than a daughter, does point to the disparity stemming from a father’s preference for a son; the implication being that men fight harder for custody of sons, or are generally seen as better parents in the eyes of the court if they have a son. Of course, this could also be due to some bias in the court system.

      Well-loved. Like or Dislike: Thumb up 16 Thumb down 7
      • Brit says:

        > “I would challenge your statement that girls are generally seen as easier to raise. While I’m not a parent, I’ve consistently heard the opposite from parents that I know.”

        I have no particular opinion on whether boys or girls are easier to raise, though, if I were to make a guess, I’d say that boys are harder to raise in childhood and girls are harder to raise in their teenage years. I’d be curious if the people who say that girls are harder to raise are the ones with teenage daughters.

        Well-loved. Like or Dislike: Thumb up 13 Thumb down 2
      • JBP says:


        You’re right if my extended family’s experiences is similar to society’s. I had three teenage girls and one boy. My mom had three boys and one girl. My mom says that the teenage girl was harder than all the boys combined. My experience was similar.

        Our family does seem to have very assertive women, so we may not be representative.

        Thumb up 4 Thumb down 2
      • James says:

        But what matters here is not whether girls actually are easier to raise than boys (or vice versa), but whether a woman contemplating divorce thinks they will be.

        Thumb up 6 Thumb down 2
      • TJ says:

        Add “women are the ones who usually file for divorce” to “men are more likely to obtain custody of a son than a daughter” to “women’s divorce filings are fueled by the expectation of getting custody of children,” and it makes a great deal of sense; women file for divorce less often with a first-born son because they aren’t as sure of “winning” in divorce court.

        See below link for support for “women’s divorce filings are fueled by the expectation of getting custody of children.”

        Thumb up 3 Thumb down 0
      • RGJ says:

        I’d be interested in the timeframe when these “baby girl” divorces occur versus norm. I’ve raised both and they went through their ups and downs, can’t generalize and it would be anecdotal anyway. There is no doubt that (feminist get ready to hate) there is STILL the vestiges of sexual role typing in America….dad’s get involved in boy’s traditional masculine activites, girls cleave more to mom’s home/shopping activities.

        If you believe that is blood boiling and horrible please “like” this comment.

        Could one parent or the other feel alienated at that point? Of course. And perhaps men are a bit more likely to be psychologically impacted by missing those dad and lad activities.

        This wouldn’t happen at the infant stage, perhaps but at maybe the “sport age”, 6-7 years in?


        Thumb up 4 Thumb down 0
    • Maxwell says:

      Couldn’t the opposite claim be made? That (assuming that boys are easier to raise than girls) the father is more likely to initiate a divorce when the first born child is a girl because they aren’t willing to raise a girl?

      Thumb up 0 Thumb down 1
  2. Mimi says:

    Is this only true for families with fathers? I’m wondering if it rings true for same sex couples (speaking as part of a same sex couple with a first born daughter… still together though! woohoo!)

    Well-loved. Like or Dislike: Thumb up 21 Thumb down 7
  3. R S says:

    I recall reading recently how there is growing evidence that monogamy evolved in primates to prevent infanticide. The evidence posted here makes me wonder if there is a connection between the two. That is, there is higher rates of male on male infanticide, so there is an evolutionary link to keep men present in their male-born offspring’s lives to prevent this.

    Thumb up 7 Thumb down 3
    • James says:

      I have seen articles about that, and they leave me wondering where the researchers ever got the idea that humans are naturally monogamous. Where it exists, it seems to be externally imposed by religious/cultural authorities, while the people controlling those authorities frequently ignore their own strictures.

      Well-loved. Like or Dislike: Thumb up 12 Thumb down 2
      • R S says:

        I don’t believe that the study was looking at humans, but primates. I believe they went through already existing primatology research and performed their methodology through there. I recall reading that there is fairly strong consensus that humans aren’t monogamous (something like 60% of cultures permit men to have multiple wives, and I believe it was also acceptable in the west until around 1000 A.D. ).

        Not all primates are monogamous either.

        Thumb up 3 Thumb down 1
  4. Isabel says:

    Surely this should be “Do sexist fathers cause divorce?” not “Do baby girls cause divorce?”

    Well-loved. Like or Dislike: Thumb up 65 Thumb down 19
    • katz says:

      My thought, too.

      Well-loved. Like or Dislike: Thumb up 13 Thumb down 2
    • Ken says:

      It must be nice to be able to magically pinpoint the cause of the correlation like that. Poor mortals like me are forced to imagine all sorts of possible alternative reasons:
      - Your assumption: sexist fathers and their preference for sons over daughters
      - Even worse than sexist fathers: abusive fathers. I’d assume the statistics are far worse for daughters than sons. Mother’s could also be more likely to leave a bad situation if they have a daughter than a son.
      - Courts favoring custody for daughters going to the mother and wives filing for divorce with higher frequency. Child preferences for living with Mom or Dad would also figure in here.
      - Different effects that sons and daughters have on family dynamics – possibly exacerbating existing problems. (For example, I had a friend with an overly-attached wife and she resented any attention their infant daughter received from her husband)
      - Health issues: for example boy’s are significantly more likely to be autistic or have behavior issues, which could also play a role in outcomes

      And I’m sure there are many other possible explainations that just didn’t occur to me after thinking about it for a couple minutes. I’d expect the cause for the difference in outcomes would not be any one of these but a combination of all of them.

      Thumb up 1 Thumb down 1
  5. Adam says:

    This is clearly disturbing and sexist. I’m disgraced to know that this happens. How can we disrespect and castaway the future sisters, mothers, and daughters of our nation? It takes two– a husband AND wife (or any combination of the two)– to raise a family. If the father cannot overcome his own personal bias, that should be labelled the cause. This article is inciting undeserving guilt in the daughters of our world. You can argue all you want, but treating the bearers of our children and our future progeny as inferior is inhumane. Men and women– it takes two to make a civilization.

    Hot debate. What do you think? Thumb up 29 Thumb down 25
    • Brit says:

      Here’s the thing – at one point in the podcast, they acts like this was a big effect. But, they also said it made a small difference (of about 2%). To put it another way: “In any given year, roughly 52,000 first-born daughters younger than 12 years (and all their siblings) would have had a resident father if they had been boys.” but there are over 4 million children born each year in the US, meaning that the effect is about 2.5% (since 52,000 is about 2.5% of 2 million children). We aren’t “casting away our daughters” and this isn’t even a widespread cultural thing. For all I know, this effect is limited to a backwards opinion in some small segment of the population. It could be common among the poor, or it might predominate among recent immigrants to the US. My point is that you can’t conclude much of anything about the larger society and cultural norms based on this small of an effect. (I’ve also heard that parents who get invitro fertilization, when they decide to pick the gender of the child, tend to prefer girls over boys, which seems to suggest that people tend to prefer girls. Of course, that’s a small segment of the population, and in-vitro fertilization is a path pursued by the middle-class and rich so it’s not even a random sampling of the larger population.)

      My point is: the effect is small and you shouldn’t get too worked up over it. If you need to feel a little better about the place of girls in society, just think about the fact that girls make up 70% of high school valedictorians, are less likely to drop out of high school, and between 55% and 60% of all undergrads in college are female. For whatever reason (whether distractions or disinterest on the part of boys, or society’s focus on girl problems over boy problems), girls are actually doing quite well – on average, they’re doing better than boys.

      Well-loved. Like or Dislike: Thumb up 41 Thumb down 3
  6. Lars says:

    Hidden due to low comment rating. Click here to see.

    Disliked! Like or Dislike: Thumb up 12 Thumb down 28
  7. Brit says:

    Listening to the podcast, I couldn’t help but wonder if something else was going on. For example, I had read some time ago that men with certain professions tended to have girls rather than boys. I believe the common thread was that, if men had jobs that took them away from home for long periods of time, that they were more likely to have girls. It made me wonder: if males with jobs that take them away from home for long periods of time are more likely to have girls, and if males with jobs that take them away from home for long periods of time are more likely to divorce, then couldn’t it be that we’re seeing two effects (increased chance of girls, increased chance of divorce) are actually caused by a third factor (being away from home for long periods of time)?

    Well-loved. Like or Dislike: Thumb up 29 Thumb down 1
    • JL says:

      Thank you for seeking alternative explanations.

      I skimmed the article and found that they jumped at conclusions based on “significantly” (should read STATISTICALLY significant) correlations found in birth records. I am troubled by the variety of samples they used, and the lack of consistent data found in these samples. It appears to me that the conclusions were cherry picked from the samples that said what they wanted to say, biased towards finding that “shotgun weddings” were more successful when the father knew the child was male. This was less than 5% of the population in there sample with a SE of about 2%. Yes if that truly extrapolates to the greater population then that is interesting and a lot of babies without baby-daddys, but there is no evidence given that it REALLY does extrapolate beyond the limited samples.

      They did not play with any theory or evidence that addressed WHY biology hands us boys or girls. They did not touch on whether the mothers or fathers hormone levels might be a part of the divorce. They assume divorce is the man’s choice. They use the term “shotgun wedding” which is sexist and antiquated.

      Well-loved. Like or Dislike: Thumb up 11 Thumb down 0
  8. landless says:

    The article should also note that girls in households with unrelated males have a high rate of sexual abuse. Fathers of daughters might want to remain involved with their daughters in order to protect them.

    Well-loved. Like or Dislike: Thumb up 14 Thumb down 3