The Brother Earnings Penalty

This paper examines the impact of sibling gender on adolescent experiences and adult labor market outcomes for a recent cohort of U.S. women. We document an earnings penalty from the presence of a younger brother (relative to a younger sister), finding that a next-youngest brother reduces adult earnings by about 7 percent. Using rich data on parent-child interactions, parents’ expectations, disruptive behaviors, and adult outcomes, we provide a first step at examining the mechanisms behind this result. We find that brothers reduce parents’ expectations and school monitoring of female children while also increasing females’ propensity to engage in more traditionally feminine tasks. These factors help explain a portion of the labor market penalty from brothers.

That is by Angela Cooks and Eleonora Patacchin in Labour Economics.  Once again, family niche effects seem to matter.  Via the excellent Kevin Lewis.


Probably they controlled for the fact that larger families are going to be more traditional families?

This thing called Google:

"Consistent with the literature, we find that sibling gender diversity is associated with smaller families: having a next-youngest brother reduces the number of siblings by 0.1. However, as discussed in Section 2, the literature suggests that, if anything, smaller family size should be beneficial for long-run outcomes."

The whole story is that families like to have boys and girls. When the first child is a girl and the second a boy, most of times the parents decide to stop there. A small family can devote a lot of resources to the children. However, even in this ideal case of resource allocation, women earn less in the future.

The article treats the issue as "number of siblings".

NFL quarterback Philip Rivers and his wife have 7 kids (5 boys, 2 girls) with an 8th on the way. My wholly uninformed guess is that Mrs. Rivers is happy currently having 5 daughters but Mr. Rivers would like to have a 3rd son, so they are going to try and see what happens.

I see you didn't understand my comment. Families that want to have more children because they don't have a boy are more likely to be tradition minded families. So a set of families that have girl then boy then stops will contain more traditional families than a set that contains girl-girl and then stops.

sociology dept. mission statement?
make bold claim
whack the probability
pimp the causality
confabulate the narrative
alert narrative public radio

Lol! +1

This seems to be a discovery in search of some value to it or a reason to care about it.

Down with Little Brother!

Can we boost female earnings by drowning small boys?

More children means a source other than government to care for the parents (and each other) in old age. That's a reason Ross Douthat promotes policies that encourage child bearing. In the past, many families had a designated sibling who would care for the parent in old age. Not designated explicitly but implicitly. That was the case in my grandmother's family: her oldest daughter was the designated care giver. All of the other children had very successful careers, including my grandmother (a physician), who was second oldest (she had four younger brothers). And it was reciprocal: my great grandfather and the designated care giver ended up residing with my grandmother and mother. I was the last of six children, but there was a large age difference from oldest to youngest. Looking back, I'm not so sure much was expected of me, being the last child. Indeed, I was the last child long after I was no longer a child. My observation is that order of birth does make a difference. And it doesn't have to be a large family for it to matter. One of my best friends has three boys, now young adults who couldn't be any different: the oldest is the achiever, the youngest is the intellectual, and the middle is a love (and my favorite).

Often the designated care-giver is the one who inherits. Those who don't inherit lack the wherewithal to care-give hence aren't expected to.
Traditionally at least. When the system is in transition, some non-inheritors get stuck with the job. Usually ladies. Even when an eldest daughter gets the dough (say, if Mom signs it off to her while still breathing, hoping daughter will care-give), the husband may successfully interfere with her efforts to do so (favoring his own side of the family). Such is life in Abe's Japan.

In families with male offspring, the female children don't take on male roles?


I was at a luncheon in December where statistician Stephen Stigler and his daughter, Margaret Stigler, presented work on luck and skill in sports. It was fascinating work, overall. Truly. There seems to be a constant ratio between luck and skill for every endeavor humans find interesting. I'll leave it to readers to follow up. There was a piece in CHANCE about it, but the power point they gave was way more detailed. Again, fascinating work.

But there was this bit of demonstrated fealty to orthodoxy du jour shoehorned in about gender discrimination. I asked a question: had they looked at data that might support (or not) their assumption that fewer women compete in professional golf because the purses are smaller-specifically whether the pool of women who play golf competitively but not professionally (such as for charity tournaments, club tournaments, etc.) was comparable to the pool of men who do so.

The whole room turned in unison. Like and army of white blood cells on alert. Another attendee turned around and yelled at me. Stigler himself doubled down on the gender bias bit without even considering the question.

I found myself wondering why a statistician would ignore a serious technical point with such aversion. (if you say X because Y and remove Y and still get X, you got a bad model<-this is not controversial)

I found myself wondering if he had any sons. (I still don't know.) If not, he can use this perverse egalitarianism we are all caught up in at present to put the hopes and dreams a man has for a son onto his daughter and blame 'society' for their failure to come to fruition. And if son-less fathers of girls don't do this in general.

(There is a luck constant! How cool is that?)

My guess would be that 95+% of the women golfers on the LPGA tour were legitimate daughters of men who like golf. Similarly, women military officers frequently are the daughters of male military officers.

On the other hand, in the case of golf, I could imagine that having brothers could help a daughter become a professional golfer. A big factor in becoming a golf pro is belonging to a country club as a child because that allows kids to play huge amounts of golf over summers (10,000 hours and all that, the weak form of which is very true in golf: nobody thrives on a professional golf tour without 10,000 hours of practice, ideally before about age 21.

So, say, that having two sons makes a man more likely to join a country club, and then his daughter gets obsessed with golf. That could happen too. So I could see it going either way.

Well, Stiger and Stigler share the same profession (though it isn't golf or war), and they collected the sports statistics together over decades--more than ten thousand hours! They illustrate your point neatly.

My point was something else. I've no idea if they illustrate it, they were just involved in an anecdote which became my point of departure.

I'll take "Studies that won't replicate" for $500, please.

My bullsh*t detector is ringing. At least these two professors aren't at a publicly funded University. We really should tax those endowments.

How do we benefit from this study?

We can drown little boys to boost female earnings and close the gap.

Egad, I have two older sisters, and they will sue me.

For lost revenue.

Shorter paper: "Males are so awful that the awfulness even rubs off on their sisters. Beware!!!"

I smell Pulitzer, honestly.

Specialization and sibling differentiation might conceivably happen. There's probably a good evo psych argument out there that can give a some hypotheses to test.

If pathway is "brothers reduce parents’ expectations and school monitoring of female children" then effect should be more apparent the more that families have high expectations (crosscheck against "Tiger Mother" factor?).

Adoption study would ideally be nice as well to cross check - perhaps families who have more sons just have different daughters? Maybe their daughters tend to marry earlier and form families earlier, etc?

I don't really particularly believe it though. Birth order effects almost never work out over the long run.

Maybe there's something here, but the focus on "younger" brothers makes me suspicious. I assume that the authors didn't get this result with "all" brothers or "older" brothers and therefore kept slicing the data until they ended up with the finding that they wanted.

Luciana Rabello is an excellent guitarist who had the misfortune to be the older sister of an extraordinary guitarist named Raphael Rabello. She still plays guitar but she's primarily known for playing the cavaquinho. I've always wondered if this difference in outcomes was due to differences in talent, or differences in family expectations and investments. The family did invest a lot in her musical education, but she switched to the cavaquinho when she and Raphael formed a choro group and Raphael was the obvious choice as guitarist. As a composer, she's written some wonderful choros. Would she have become a more famous guitarist without a younger brother? (Sample size of 1 and no firm conclusions.)

Comments for this post are closed