Billionaires have more sons

January 17, 2009 1:21 PM

Billionaires have more grandchildren through their sons than through their daughters, because the status advantage is more reproductively valuable to the sons. Therefore, it would be adaptive for the mothers of their children to bear more sons than daughters. But surely that can't be; mothers can't control the sex of their children. Oh but so it is: billionaires have 60% male children.

This exciting research finding partly validates the work of LSE's seeming crackpot, Satoshi Kanazawa (also previously). It's getting harder to deny that simple evolutionary hypotheses have a lot of power to describe human life. The worldview of this pithy essay (previously) is looking more and more convincing.

posted by grobstein (69 comments total)

billionaires have 60% male children.

That's because they pay their sperm to do their bidding.
posted by jonmc at 1:32 PM on January 17

correllation ?= causation
posted by duende at 1:34 PM on January 17

I see a major loophole in these statistics, as they do not take into account the thousands of female children of billionaires who are shamefully hidden away in basements, attics, and castle turrets every year.

Yes, I totally made that fact up. But it sounded good.
posted by miss lynnster at 1:40 PM on January 17 [9 favorites]

Are you sure they aren't secretly selling female fetuses to krispy kreme to bake into their abortion donuts? Correlation may not equal causation, but does that mean the alternative hypothesis is that billionaires with daughters are more likely to lose their billionaire status? The explanation for that could be either endearing or creepy depending.
posted by BrotherCaine at 1:40 PM on January 17

I love this, and it dovetails nicely with my own reductionist understanding of human behavior.
posted by killdevil at 1:45 PM on January 17 [2 favorites]

*climbs castle wall to turret*

I'll save you!!
posted by jonmc at 1:46 PM on January 17

This exciting research...

Perhaps so for the eighty-year-old billionaires. Perhaps less so for their 24-year-old wives.
posted by terranova at 1:48 PM on January 17

If you're a billionaire, why do you have kids at all? You're not going to need to sponge off someone in your old age.
posted by Mr. President Dr. Steve Elvis America at 1:52 PM on January 17 [3 favorites]

If you're a billionaire, why do you have kids at all?

What an amoral statement. Once you die, who will

a) Look after the corgis
b) Tend the grounds
c) Look after the alligator pool
d) Pay the hitmen
e) Finish project Zephyr?

Wait a minute, there's someone at the door, why hello mr nice man with sunglasses. Yes, I mentioned project Zephyr, is a that a hfjfjkdshjdfhjfhjfdhjdfshjdfhjdfjkdfjkfdhjfdjdkfszzzzzzzzzzzzzzzzzzzzzzzzzz
posted by lalochezia at 2:01 PM on January 17 [5 favorites]

It's just because billionaires are more likely to be dicks than the average person.
posted by jimmythefish at 2:01 PM on January 17 [2 favorites]

Does this mean they will have big penises too?
posted by alteredcarbon at 2:01 PM on January 17

It's because they can buy the best fedoras.
posted by LordSludge at 2:01 PM on January 17

*enters turret*

Here I come to saaave the day....

*finds note: We're sorry jonmc, but miss lynnster is in another castle*
posted by jonmc at 2:06 PM on January 17 [3 favorites]

Once I become a billionnaire I will balance out any grandchild gender imbalance through legions and legions of Junior Achievers.
posted by Shepherd at 2:08 PM on January 17 [1 favorite]

What about the sex of their children via their mistresses?
posted by Maias at 2:18 PM on January 17 [1 favorite]

What about the sex of their children via their mistresses?

In my experience most complete bastards tend to be male.
posted by thivaia at 2:20 PM on January 17 [4 favorites]

Let's play Evolutionary Psychology Bingo!

http://punkassblog.com/2007/10/25/evolutionary-psychology-bingo/

That will have to do, because the article I wanted to link to, a study which ponders the evolutionary forces that give rise to a tendency by certain people to attribute behavior to evolutionary forces, I can't find right now....
posted by edheil at 2:29 PM on January 17 [3 favorites]

Maybe rich guys just end up blowing money they would otherwise invest on buying their daughters more ponies/mansions/pink mountains, while spending only enough on their sons to dump them at a remote boarding school.

All of the evidence I've seen* suggests this is the case.









*Direct-to-Dvd family films.
posted by Benjy at 2:30 PM on January 17

This is interesting, infuriating, and pretty much over my head. I can't rid myself of the notion that this effect might also be explained by the story that the exotic species of the male billionaire, because of his relative freedom and leisure time, also has kids outside of marriage that, because of his resources and high profile, he doesn't then have or want to be parents to, at least not in a public-records kind of way. In essence, the male billionaire can buy his way out of parenting or at least claiming the total number of his kids. Now we'll just have to find out if those offspring otherwise taken care and/or not appearing in published sources are more female than male...but hey, maybe the people of the Mammal Research Institute of Pretoria, South America, are saying just this, as they say that "in polygynous species, mothers in good condition would be advantaged by giving birth to more sons."
posted by theefixedstars at 2:36 PM on January 17

I'm convinced families that are super wealthy get progressively better looking over time. Generations of trophy wives will yield better looking kids.
posted by rageagainsttherobots at 2:40 PM on January 17 [1 favorite]

Has anyone else noticed that wallstreet traders and talking heads seem to be way more likely to be bald then other guys. Neel Kashkari is like 30 years old and a complete chromedome already. Jim Cramer, Paulson, etc. All bald.

These guys must be OD'ing on testosterone.
posted by delmoi at 2:41 PM on January 17 [1 favorite]

Let's play Evolutionary Psychology Bingo!

I glanced at the score card, and it honestly doesn't look like you're going to win with this post. I didn't link to a bunch of evo-psych cliches, I linked to a study with a very simple and persuasive design. It's not immune to alternative explanations, but (as BrotherCaine points out) the authors' theory is easily the simplest.

The P of the results being random chance is also pretty small. (This is a Stats 101 exercise: find the population statistics for sex at birth, assume normality, and estimate the probability that a sample of about a thousand would be at least 60% boys. The number's fairly low.)

When you're dealing with an argument as straightforward as the one in the linked paper, you should be able to express your skepticism by addressing it directly, rather than relying on stereotypes about how evo-psych generally is supposed to be unscientific. When it is easy to do better than a rough stereotype, it is only honest to attempt it.

does this work even count as evolutionary psychology?
posted by grobstein at 2:42 PM on January 17 [2 favorites]

Background: Trivers-Willard Hypothesis.
posted by grobstein at 2:46 PM on January 17

theefixedstars, I think that's a really interesting point. I think the authors could go a long way towards testing / eliminating that hypothesis by studying only in-wedlock kids, but from the paper it doesn't look like they tried that. However, I still think your explanation is edged out by the authors' proffered one.
posted by grobstein at 2:50 PM on January 17

I can't rid myself of the notion that this effect might also be explained by the story that the exotic species of the male billionaire, because of his relative freedom and leisure time, also has kids outside of marriage that, because of his resources and high profile, he doesn't then have or want to be parents to, at least not in a public-records kind of way.

No.

I'm not saying "No, billionaires don't have extramarital kids."

I'm saying "If the gender of any particular child is a coin toss, whether any particular parent has extramarital kids or not would not affect the percentage of his in-marital children who are a particular gender."
posted by Flunkie at 2:50 PM on January 17

It would if you added the assumption that female bastard children are more likely to be hidden from view. (theefixedstars makes this point.) I have no idea whether that is plausible.
posted by grobstein at 2:52 PM on January 17

No, sorry flunkie, you're right. But from my reading of the paper, the children studied are not limited to "in-marital children."
posted by grobstein at 2:53 PM on January 17

(which means that theefixedstars' point, about which out-of-wedlock children are likely to be publicly known, potentially stands.)
posted by grobstein at 2:59 PM on January 17

No, sorry flunkie, you're right. But from my reading of the paper, the children studied are not limited to "in-marital children."

Ah, I see. Never mind, sorry.
posted by Flunkie at 3:09 PM on January 17 [1 favorite]

But if the gender of any particular child is a coin toss...what would affect the percentage of someone's in-marital children who are of a particular gender? I'm genuinely confused.

Could not the fact that different mothers are involved in the production of extramarital children than in-marital children play some role in this scenario?

And by the way, how are we deciding that billionaires are "in very good condition"? I know that seems obvious because they have access to the most external resources, but...I guess I'm not entirely getting why they looked at billionaires.
posted by theefixedstars at 3:13 PM on January 17

This enabled us to distinguish between biases resulting from the costs [2] or the advantages [1] of raising sons.

Wait, I see. Sort of.
posted by theefixedstars at 3:15 PM on January 17

Sample size too small for causation to be proved.
posted by Ironmouth at 3:15 PM on January 17

Interesting post. Deviations from Fisher's principle of 1:1 sex rations in mammals have been widely studied cross-culturally, in Anthropology and other disciplines:

W.D. Hamilton gave the following basic explanation in his 1967 paper on "Extraordinary sex ratios"[4], given the condition that males and females cost equal amounts to produce:

1. Suppose male births are less common than female.
2. A newborn male then has better mating prospects than a newborn female, and therefore can expect to have more offspring.
3. Therefore parents genetically disposed to produce males tend to have more than average numbers of grandchildren born to them.
4. Therefore the genes for male-producing tendencies spread, and male births become commoner.
5. As the 1:1 sex ratio is approached, the advantage associated with producing males dies away.
6. The same reasoning holds if females are substituted for males through-out. Therefore 1:1 is the equilibrium ratio.

In modern language, the 1:1 ratio is the evolutionarily stable strategy (ESS

Interestingly then, the actual human birth ratio is about 105 males for every 100 female babies born, reflecting the survivorship of males to reproductive age rather than an in utero sex determination. I seem to recall that the majority of miscarriages are male as well, reflecting the less robust male genetic inheritence (the non-redundancy of the XY chromosome). But the point is, the human organism can regulate its sex ratios at conception according to ultimate reproductive success, so the only real surprise for me in this study is that Billionaires apparently regulate this in such short order...

Also, the basic methodology of this paper seems to be, "google for the children's names and genders", which I find to be not totally convincing. For example, what if emergent billionaires give their daughters names like "McKinley" which are at best ambiguous and thus removed from the study.

I'd like to see this study reproduced with, say, highly successful professional athletes who may be as effectively rich as billionaires (I mean, 10 million, 50 million, whatever) and the children of whim may be better documented and whose inherent testosterone levels may be more comparable given their high physical achievements, etc.

Interestingly too, the study notes that self-made female billionaires have more daughters, which is highly counter-intuitive?
posted by Rumple at 3:16 PM on January 17

Some of the in utero and social determinants of birth sex ratios are outlined on this wiki page. For example, certain pesticides and pollutants are thought to influence sex ratios at birth.

So, quite likely, this study is pointing to a previously undetected contaminant found in caviar, 25 year old single malt, fine cigars or silver spoons.
posted by Rumple at 3:25 PM on January 17 [2 favorites]

Materials and Methods
We then searched the billionaire's name online using Google, and used the resulting pages to determine the sex of children. Resulting pages included Wikipedia, bibliographic sites, company websites, and newspapers (particularly marriage, birth and death announcements).

Well, either that or the male children of billionaires are more likely to be mentioned on the Internet.
posted by TheophileEscargot at 3:25 PM on January 17 [9 favorites]

It makes sense to me. Rich men have more sons: good-looking men have more daughters.
posted by Phanx at 3:27 PM on January 17

Meaningless data blip or...meaningless data blip?
posted by kittens for breakfast at 3:27 PM on January 17

I wish I could find the research, but if I remember correctly there is a relationship between the relative ages of the parents and the general sex of the offspring.

In rough terms, if the father is older than the mother, the mother is more likely to birth a male child. If the father is younger, on the whole, the mother will birth a daughter.

If male billionaires have children with younger women, on average, and if this research is accurate, one would expect that they have more sons. I'd be curious about the average age of the mother, relative to the average age of the father, at the time of conception.
posted by Blazecock Pileon at 3:31 PM on January 17 [2 favorites]

Sample size too small for causation to be proved.

Meaningless data blip or...meaningless data blip?

These are among the worst objections. The sample is (roughly) all the children of the more than 400 people on Forbes US billionaires list. Furthermore, these people make up a good 40% of all billionaires.

There are problems with the methodology; some of them may be serious. Sample size isn't really one of them.
posted by grobstein at 3:38 PM on January 17

There are problems with the methodology; some of them may be serious.

I think batshitinsane is closer to the mark on their search-the-internet-with-Google methodology. If there is any tendency for males to be more prominent than females on the internet, then males are going to be over-represented in their results.

Amusingly, it's totally inconsistent with the evolutionary psychology crowd's belief that the mighty power of testosterone compels men to strive to excel, making men naturally pre-eminent in the major fields of human endeavour. That itself would massively skew the results of any Google search. (The existence of sexism in human societies would skew the results too, or course).
posted by TheophileEscargot at 3:54 PM on January 17

I'm convinced families that are super wealthy get progressively better looking over time. Generations of trophy wives will yield better looking kids.

I was all set to prove you wrong with photos of Malcolm Forbes, various Kennedy clan offspring, and Osama bin Laden. But then I remembered how dreamy Anderson Cooper is so I gave up. Ahhh, but then I remembered... Anderson's probably not going to be breeding with a trophy wife so his adorable billionaire bloodline probably stops with him. So I'm back to disagreeing with you.

Just because a woman is a "trophy wife" (jesus, I hate that obnoxious phrase) who pays a great deal of attention to her own physical appearance doesn't mean her offspring -- especially her male sons -- won't be fugly. Genes often skip generations, y'know.
posted by miss lynnster at 3:59 PM on January 17

There are problems with the methodology; some of them may be serious. Sample size isn't really one of them.

That's what my nineteen-year-old wife said -- ha, oh me. My thing is, even if the methodology were sound, is 60% really that amazing? If it were 85%, okay. But this sounds like slightly more males than average, which to my non-statistician self sounds kind of...unremarkable?
posted by kittens for breakfast at 4:28 PM on January 17

What is the likelihood that billionaires are more likely to abort daughters?
posted by ardgedee at 4:30 PM on January 17 [1 favorite]

As far as male-female birth ratios go, its already very slightly skewed towards the male side due to the fact that male sperm (Y) is lighter than female sperm (X). In fact, sorting sperm by weight is considered a fairly reliable method on gender selection, at least in livestock. So, when viewed apart from all other environmental influences, yeah, more males are born than females. I think there's some other stuff going on here, though, than any sort of evolved ability to produce males in situations where its socially beneficial.
posted by internet!Hannah at 4:34 PM on January 17

billionaires have 60% male children.

How female are the girls?
posted by swift at 4:42 PM on January 17 [1 favorite]

Theophilescargot totally nailed this. Science is all about the methodology and this one is as flawed as they come.
posted by bluesky43 at 4:49 PM on January 17

It makes sense to me. Rich men have more sons: good-looking women have more daughters.

But in either case, the fish are jumping and the cotton is high.
posted by BrotherCaine at 5:30 PM on January 17 [2 favorites]

billionaires have 60% male children.

you mean they're all hermaphrodites?!
posted by doobiedoo at 5:37 PM on January 17

The Google research is the problem. Totally inconsistent means of gathering initial data.
posted by Miko at 6:17 PM on January 17

Except there are marginally more boy babies born than girls, and if I recall right, more male conceptions, which fail at a higher rate than a XX. Female fetuses and children are naturally hardier, but it follows that a billionaire may merely have access to better maternity care. Thus the issue may be that the infant mortality rate is lower for the very wealthy, which is generally known.

On time of conception in relation to gender.
On death by gender.
posted by Phalene at 6:50 PM on January 17

"death by gender"

Sounds painful.
posted by darkstar at 7:59 PM on January 17

Well, either that or the male children of billionaires are more likely to be mentioned on the Internet.

Yes, this is definitely a major methodological flaw. The Marquis Who's Who typically includes comprehensive information about names, ages, and gender of the offspring of famous people (presumably including billionaires), but the authors don't even seem to have bothered consulting it.
posted by jonp72 at 8:30 PM on January 17

But surely that can't be; mothers can't control the sex of their children.

Mothers make decisions about whether to have another baby, based on the sex of their previous children all the time. For example, young boys are socialized to be rambunctious "holy terrors." If a woman has two or three boys in a row and her husband doesn't help with the childrearing, she may decide not to have any more children, because she already has all the childrearing she can handle. In other cases, families seek gender balance. They continue having children until they have at least one girl and one boy, and then they stop. Even children express their own preferences about whether they want a little baby brother or sister, which may also influence parental decisions about whether to bear another child.

I don't know about you, but this seems more like basic demography than evolutionary psychology to me.
posted by jonp72 at 8:38 PM on January 17 [3 favorites]

If they did not ask about:

-abortions of female fetuses in favor of males
-use of fertility treatments (thus leading to multiple fetuses which might get selectively aborted, or treatments that increase likelihood of male children)
-any actual data on who the billionaire-in-question's children were, and what gender

then this is useless.

Not actually bothering to go past "googling" should just knock this out of the "science" realm altogether. Even good reporters bother to call their sources; these researchers couldn't do the most basic footwork? Pfft.

And yeah, sloppy science that always seems to back up stereotypical associations about men and women is kind of the hallmark of evo/psych.

Plus this statement: It's getting harder to deny that simple evolutionary hypotheses have a lot of power to describe human life.

No, it isn't. In fact, "simple evolutionary hypotheses" have, as far as any study yet presented, been damn useless in "describing human life." What they mostly describe is the researcher's fantasies about some sort of nonexistant "pure" early human behavior on which all current human behavior is based and from which it can never fully deviate, despite our absolute lack of data about this very early human behavior.
posted by emjaybee at 9:28 PM on January 17 [3 favorites]

Jonp72 points out the major flaw in this study. Parents have significant control over the sex ratio of their offspring because they can decide whether or not to have more children after they see the gender of their first one. For example, one would be laughed at for coming up with an evolutionary theory for the current gender imbalance of Chinese children.

It is quite likely that millionaires are more influenced by the general cultural bias because of the tradition of passing fortunes down to male heirs and they probably have more reproductive autonomy than the general population.

I'm not saying that for sure that the Trivers-Willard hypothesis does not hold for humans, just that this study doesn't come close to explaining it. If they only looked at first born children and took into account abortions it would be more convincing.
posted by afu at 11:48 PM on January 17

Sample size is a problem with only 400 people, with heredity and geneology up for consideration.

Flip a coin ten times and there is a reasonable chance you'll get an outcome more disparate than 50/50. Gender is considerably more complex but over large groups of people you're as likely to have a son or a daughter so coins model gender and heredity pretty well.

Flip a coin 400 times or write a computer program or however you want to model this; with such a small sample one of two incomes occurring 60% of the time is not improbable at all.
posted by christhelongtimelurker at 12:54 AM on January 18 [1 favorite]

Sons are significantly more likely to live with their fathers than are daughters. If a woman is pregnant with a daughter, the father is significantly less likely to marry her than if she is pregnant with a son. This phenomenon could have quite an effect on their Google methodology.
posted by transona5 at 7:03 AM on January 18

Bad sample size? Really?

It's been 15 years since my last statistics class, so hopefully someone will correct me if I'm wrong.

Assume 2 children per billionaire (probably an underestimate) for 800 children total. At 105/100 male/female births in the general population (per Rumple's stat), you'd expect 410 males and 390 females. If they got a 60/40 split, that would mean about 480 males and 320 females. Calculating for chi-squared:

((480-410)^2)/800 + ((320-390)^2)/800 = 12.25

Consulting a handy chart gives a p-value of less than 0.0005, or a 0.05% probability of getting this result by chance.

(Ah, and I see the abstract has done all this: "X2 = 37.57, DF = 1, P<0.0001")

Come on, MetaFilter -- I read you so I don't have to wade through bogus arguments like this.
posted by bjrubble at 10:40 AM on January 18

This is something I learned in my anthropology class, I thought everyone knew it. It comes down to power. When the mother feels that she has power, socially or however, she tends to have male babies. If you're at the bottom of the heap, you'll tend to have girls. It's not a huge difference, but it is there.
posted by stoneegg21 at 10:57 AM on January 18

Wealthy Men Give Women More Orgasms. But are they real or are they Memorex?
posted by terranova at 10:58 AM on January 18 [1 favorite]

"Jonp72 points out the major flaw in this study. Parents have significant control over the sex ratio of their offspring because they can decide whether or not to have more children after they see the gender of their first one. For example, one would be laughed at for coming up with an evolutionary theory for the current gender imbalance of Chinese children."

Uh...this is not correct. No matter what the gender of your previous children is, you still have a only 50% chance of having a boy next time. You can't somehow inflate the percentage of boys by having extra children. The gender imbalance in China and India is only due to selective abortion and infanticide, practices that are unlikely among billionaires.

My guess that the effect is real, and related to testosterone. The distinction between self-made billionaires and inherited billionaires as a control is weak, since the inherited billionaires, inherited their billions, and their testosterone producing genes, from self-made billionaires.
posted by roofus at 11:14 AM on January 18

roofus: suppose you want a boy, and have access to contraceptives. If your first child's a boy, you can stop there. If your first child's a girl, you can try for another one.

If a group of people do that, it will inflate the percentage of boys in the children of that group.
posted by TheophileEscargot at 12:00 PM on January 18

> If your first child's a boy, you can stop there. If your first child's a girl, you can try for another one. If a group of people do that, it will inflate the percentage of boys...

Halting production after the first male only increases the probability that any family includes a male. It can possibly diminish the ratio of males to the total population.

If the first child is a boy, stop. Result: 1 child, population 100% male.
If the first child is a girl, try again. If the second child is a boy, stop. Result: 2 children, population 50% male.
If the first child is a girl, try again. If the second child is a girl, try again. If the third child is a boy, stop. Result: 3 children, population 33% male.
posted by ardgedee at 12:15 PM on January 18 [1 favorite]

Yes ardgedee, I realised my comment was crap just after I posted. Please ignore it ;-)

Actually I don't think it changes the ratio at all.

Consider the first three children a couple has. There are 8 combinations.
Add them all up and you get 12 boys and 12 girls:

BBB
BBG
BGB
BGG
GBB
GBG
GGB
GGG

Now suppose they're following the "stop when you get to a boy" strategy. Now the combinations change like this:

B
B
B
B
GB
GB
GGB
GGG

7 boys and 7 girls. So the ratios stay the same. D'oh!
posted by TheophileEscargot at 12:20 PM on January 18

...selective abortion and infanticide, practices that are unlikely among billionaires.

Infanticide, sure, but why not selective abortion? If we find, for example, that male billionaires actively prefer male heirs, why wouldn't they pressure their wives to abort a fetus that was the wrong gender? The answer is, we don't know, because the authors of this "study" did not bother to do any real research by asking the study subjects the relevant questions.

It's possible that there is a real connection, but their methods have failed to establish that convincingly.
posted by emjaybee at 1:36 PM on January 18

theefixedstars writes "But if the gender of any particular child is a coin toss...what would affect the percentage of someone's in-marital children who are of a particular gender? I'm genuinely confused. "

Toxoplasmosis is one common ailment that radically changes the sex ratio (probability of a male birth as high as .72 IE: 260 males for every 100 females). Rumple is probably right that there is an environmental factor at work here. Even if only 50% of the billionaires are exposed a strong factor could skew the results like this. So possibly look for pre-conception and anti-natal effects from trans Arctic flights, Rolls Royce leather, Agarwood scent, Brazilian Rosewood flooring, etc skewing live births either by differences in implantation, viability or miscarriages.
posted by Mitheral at 5:13 PM on January 18 [1 favorite]

Uh...this is not correct. No matter what the gender of your previous children is, you still have a only 50% chance of having a boy next time. You can't somehow inflate the percentage of boys by having extra children. The gender imbalance in China and India is only due to selective abortion and infanticide, practices that are unlikely among billionaires.

The overall ratio of boys to girls is what's important, and ff you have the choice of where to stop you do have a large amount of control of the final ratio of boys to girls in your family. It seems that

Say you get to flip a coin and your goal is to have a higher ratio of heads to tails and you were able to decide when you were able to stop, you would be able to succeed over half the time and the times you failed the ratio would be close to 50%. So if you took the overall percentages of the coin flips, heads would probably be more than tales. Plus things like family size preferences come into play which make it even more hard to model on a simple coin flip model.
posted by afu at 8:15 PM on January 18

"Say you get to flip a coin and your goal is to have a higher ratio of heads to tails and you were able to decide when you were able to stop, you would be able to succeed over half the time and the times you failed the ratio would be close to 50%."

If you believe this is true, I suggest you head down to a casino and try and make your fortune at the roulette table.
posted by roofus at 12:01 AM on January 19

Yeah, I thought about it and I'm wrong. mea culpa
posted by afu at 12:44 AM on January 19 [1 favorite]