I think you should apply the same skepticism to the MRKH paper that you do to the IVF paper! Three issues off the top of my head. First: per the abstract, MRKH women marry older men, maybe there is a mentoring effect where the older spouses help their younger wives move up professionally. Second, there could be pleiotropic effects from the MRKH syndrome-causing mutations, e.g. altered WNT signaling can drive both MRKH and hyperandrogenism. Third, the paper compares women who know from adolescence they can't have kids to all men. Maybe men who knew from adolescence that they couldn't father children would have the highest income of all, significantly higher than MRKH. You could tell a story like "when you know you won't have kids, you focus on building your legacy in other ways, such as starting a business/becoming an executive/whater". I'm not endorsing those hypotheses, I'm just saying that if one weren't already inclined to agree with the paper, it's easy enough to poke holes in it.
Anyway this is all silly because as you say, it's wildly obvious that children are a big driver of the gender wage gap! I'm not even a grandmother and it's obvious to me! The burden of proof is so incredibly strong on people who disagree with that statement, and even the best social science provides such weak evidence, that there's basically no point studying it!
Thanks for bringing this up. Androgenic effects are a real concern here and the first thing I checked when I read this. It does seem like MRKH women have perfectly normal sexual development otherwise, including normal levels of female hormones: https://pmc.ncbi.nlm.nih.gov/articles/PMC1832178/
I think the other points are less relevant though. The key question here is how much of the gender wage gap is due to motherhood or some other intrinsic female characteristics. If women who cannot have children choose older partners and that increases their wages, it suggests "femaleness" without children leads to choices that increase wages.
As for men with a possible analogous condition: I would expect the results to go the other way, in the direction of them earning less. Men with children earn the most and many men feel a responsibility to earn in the expectation of a family.
That last point is debatable: “men with children” selects for positive traits that cause high salaries so idk if there is clean evidence of causation. Certainly while I wouldn’t trade my kids for anything I would publish more papers without them. Nevertheless I agree my last point is probably, just not certainly, incorrect.
As for my first point, I guess it’s a question of what point you are trying to make. Accepting arguendo that wage gap reduction is mediated by older mate selection, there is an interpretation by which females are worse at work but receive a leg up due to e.g. better networking or resume help, in which case “femaleness without children” and the unfair benefits cancel out the inferiority. To be clear this is very stupid! It’s just one of very many ways that this study could be wrangled into a sexist interpretation if one had a sexist prior.
Maybe the best way to put this is, if there was still a gender wage gap for MRKH women it wouldn’t, and shouldn’t, change your beliefs! You would just say that MRKH women have their own unique psychological challenges due to societal norms claiming women are meant to have children, so therefore they are “defective”, and this causes lower wages! That is a way more plausible argument than any that I gave above, right? So if the complete opposite result from the MRKH paper wouldn’t move you off your prior then it’s just not that relevant.
I think the data shows they have higher wages post children, due to pressure to support family but selection effects certainly play a role. That being said, I think that's a different question from what would men do if they knew they'd never have kids. My bet is they'd be less ambitious, not more. With exceptions for monk like autists in intellectual fields.
I understand your other points, but we can endlessly debate abt social underpinnings. I still think "condition that doesn't allow one to have kids" is a clean ish experiment. Ofc, I also take ur point that in the end no empirical evidence is enough and you need to rely on common sense.
Very interesting, thank you! I would just add that the MRKH results tell us about the state of the wage gap in Sweden, a country which is quite an outlier in terms of gender attitudes. So while the internal validity is great, I wouldn’t be quick to conclude that the same is true in the US or Japan or Italy
Yes that's true, thanks. I should probably add that Sweden has more wage compression than US and maybe at the right end rail in US you'd observe more of a difference.
Although I entirely agree with you, George Eliot did not bear children but she did serve as a mother to her partner’s children and saw herself as a mother.
Thank you for putting in the work to defend rigorously the 'obvious' answer - someone needs to!
I admit to being absolutely baffled by the people who dispute the idea that having childcare responsibilities makes it harder to progress in your career. Obviously! Have these people never *met* children? I'm actually a father here, albeit for the purposes of these studies I suspect I would be more accurately labelled as a 'mother' given that it's shorthand for 'childcare and housework effort' (I took almost two years off when they were very small and continue to do much of the day-to-day as my wife reboots her career).
Yes, my two children have without a debt not only impacted my career progression but my choice of career direction, because they just *take a lot of time*. It's as simple as that. I have no regrets on that front, they're the light of my life, but one can be realistic that there are simply only so many hours in a day.
Finance; I got to a pretty privileged position in my career where I could trade experience either for more money or more time (whilst still making objectively good money) and picked the latter to be able to be flexible for child care.
Probably also worth mentioning that people tend to vastly overestimate how fast moving most fields actually are (unless you work in AI, I guess); I took almost two years off with my kids and it was not difficult to catch up.
It would be interesting to see “men” in that graph split into men with children and men without children. Do you know if it’s still the case that men with children earn more, controlling for education, et cetera?
I was also surprised when I first noticed that many great philosophers were bachelors. I wonder how much gain fully productivity gain outsourcing childbirth and childcare would bring, if we aren't replaced by AI first.
Probably a trend which will get stronger as dating has turned into more of a negative-sum tournament for men. The guys with the deepest career ambitions will delete the apps and hire prostitutes.
I remember when I was a young software engineer in San Francisco, with a big oversupply of single men relative to women, I felt the tradeoff between career ambitions and dating quite strongly.
Wasn’t mating more demanding before? I guess there was always a tradeoff between work and human affairs.
In any case, conceptualizing and predicting the future of people's sexual preferences will need more than a psychologist psychoanalyzing "the rich" and "the beautiful." The replication crisis should've taught us that intriguing narratives cannot replace meticulous research.
Addendum: And regardless, only a very tiny portion of people will be workaholics enough to forgo any love. And the AI revolution making whole job sectors obsolete, and perhaps designer baby tech selecting against extreme obsessiveness will reduce their numbers even further.
I think dating apps have caused an acceleration, a sort of bidding war among men for who can sink the most hours into dating (in terms of optimizing their rizz, going on more dates, etc. etc.). Women have more choice, which means a man's probability of wifing up any given date is lower.
>In any case, conceptualizing and predicting the future of people's sexual preferences will need more than a psychologist psychoanalyzing "the rich" and "the beautiful." The replication crisis should've taught us that intriguing narratives cannot replace meticulous research.
I am tempted to argue the opposite. The replication crisis taught us that grandma's intuition is worth more than "meticulous" (p-hacked) research, as Rux argued in her OP. Putting a science veneer on fundamentally corrupt research practices isn't worth much. Actual honest observations are worth more. Data quality matters more than sample size. (Little use in gathering more samples from a data source that's fundamentally engineered to be unrepresentative! As many small studies are, due to file drawer effects etc. To be fair, Youtubers face incentives which are arguably just as bad as researcher incentives.)
> I am tempted to argue the opposite. The replication crisis taught us that grandma's intuition is worth more than "meticulous" (p-hacked) research, as Rux argued in her OP. Putting a science veneer on fundamentally corrupt research practices isn't worth much. Actual honest observations are worth more.
What makes grandma's advice reliable is less her qualities and more her conveyance of accumulated wisdom of the people. The random YouTuber offering vague and baseless psychoanalysis, whatever he might be doing, is not passing down time-defying ideas and dispositions.
Also, don't be a scientific nihilist. You can make good as well as bad science. But psychoanalytic "conclusions" are all arbitrary.
I added a third paragraph to my reply concerning a) men who are obsessive enough to forgo all love are a rather small psychological minority b) their number will reduce even further as AI makes more job sectors obsolete and designer baby tech selects against extreme obsessiveness.
If I didn't got it wrong, OCD, autism, ADHD and epilepsy are all related to each other. Trying to have less of any of those would reduce obsessiveness, and even modest drops in average obsessiveness would reduce extreme (I'll be all about work and will forgo any love) obsessiveness.
It's not that they "forgo all love", it's that they conclude the juice isn't worth the squeeze. The more squeeze required for the same amount of juice, the more guys opt out. This will be especially true for men who have a high self-perceived opportunity cost of their time, e.g. high earners.
Perhaps in the past it was a tiny minority of great philosophers, but nowadays their numbers are growing since dating has become a misery tournament. Plenty of anecdotal evidence for this online.
Even reputable think tanks like Pew publish attention-grabing misleading studies fueling the gender war narrative. Treat all "intriguing gender factoids" as false until proven otherwise.
Ok, so I'm going to pick on a small side comment (forgive my autism): "For while we may safely concede that history has offered no female Einstein or von Neumann ..."
I have a theory for that. (I am deeply suspicious of GMV, so it's nice to have another explanation for the sex imbalance in genius.)
You probably need both aptitude and tutoring to get an Einstein, but - it passes the smell test for me that this kind of intense academic tutoring has rarely been tried on women. Women have been tutored to be "womanly", of course - I'm not a blank slatist either but we wouldn't have invented "finishing schools" if the skills there somehow naturally developed with female puberty.
As you say "It appears that the mere anticipation, the steady, lifelong knowledge, that one might become a mother shapes a person in ways that not trivial. Expectations laid upon us in youth can cast longer shadows than we imagine."; replace "mother" with "genius" and the principle could still work?
The one example I can think of where this has been tried and worked spectacularly on women is the three daughters of László Polgár, all of whom became world-class chess players. Judith was apparently the only woman in the (non-gendered) top 10 world ranked chess players at the time. (Sophia, the "least successful" of the three, later settled down to become a mother and housewife; she only made it to women's world #6 compared to her sisters' #1 and #2.)
László certainly was of the "geniuses are born not made" opinion, one wonders if he had taken an interest in physics rather than chess, if we'd have our female Einstein there. To quote him from Wikipedia, again returning to the theme of motherhood:
> "Men must be clever and hard," Polgár said. "Women must be beautiful and look after the family. Only then, if they have time, can they be clever."
Dumb question: what does autism have to do with the gender wage gap? Is it a belief that autistic men are more likely to contribute the tail effects (and thus the GMV hypothesis) of work performance?
Best I can find (UK specific) is https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/disability/articles/disabilitypaygapsintheuk/2014to2023 which says the disability pay gap is wider for men than for women, although I don't have to raw data to compare by both sex and specific kinds of disability. From a glance at the graphs and text though, being autistic seems to have a negative effect size that trumps almost everything else, with the closest contenders being correlated factors like "severe learning disability".
So back to the OP's question, once we accept that "autism" doesn't mean "sort of like Alan Turing" it's not so much a case of GMV's "same mean, different SD" claim as evidence for autism = much lower mean/median pay in the first place, if you're one of the lucky ones who can get a job at all.
A study looking at the people with the "good at math and science autism" (I hate that meme) would be interesting indeed, but then we're conditioning on high-performers compared to the general population in the first place.
So, going off UK statistics because I have them to hand,
The gender pay gap (non-autistic) is around 7% per hour (https://researchbriefings.files.parliament.uk/documents/SN07068/SN07068.pdf) or 12.8% in total (since women are more likely to work fewer hours per week); I'm with Teslo that this should be called a motherhood wage gap. Even the UK Government, in Sec. 3.1, agrees.
Basically, Autism overall is a huge NET NEGATIVE to predicted earnings, because it includes more of the "needs 24-hour assisted living" or "just about qualified to work in a warehouse" people than it does "needs 50% extra time and a small room, but graduates with a 1st Class MSc in Computer Science" people.
The Musks of this world are no more representative of autistic people than they are of humans as a whole.
"The MRKH natural experiment is particularly powerful because it identifies a group of women who are biologically typical in every respect except one: the congenital absence of a uterus. As a result, they are far less likely to have children unless they pursue surrogacy." I am in support of the argument for the pay gap... but these papers are all making a gross assumption that overlooks a somewhat common practice when women cannot conceive: Adoption. There would be no medical assessment possible for a woman that would reveal that she has an adopted child.
(Clears throat and steps up on stump for those in the back)
As an adopted person, it's discouraging to see how often the medical community ignores those of us who are adopted - especially in a closed adoption where we lack access to our genetic history. I just filled out a very lengthy genetics questionnaire having to select "Unknown" for each question instead of one option at the top which would allow me to submit one answer of "adopted, unknown".
This widescale ignorance also reinforces the old question that many of us get from people when we reveal we are adopted, "But do you know your REAL parents?" Our REAL parents are the ones who raised us. Our REAL moms are the ones who took the hit to their pay (or exited the workforce altogether). And I promise you that our REAL parents love and care for us the same as a biological child even though others may doubt it.
Thank you for coming to my TedTalk. Now please continue doing the great work you are doing... but keep us adopted/adoptee people in mind.
(Steps down from stump, sees someone I know and wave, exit to the bar)
Hello, great article! I would like to ask about the current consensus regarding the theory that male competitiveness (for status or mates) drives young men toward money-focused careers. Specifically, is it true that this drive pushes them to make trade-offs for higher earnings compared to, for example, hospitality professions? You mentioned this briefly in your piece.
LLMs tell me this might only have a minor effect, which isn't surprising given that Claudia Goldin’s work suggests motherhood explains 60-80% of the gap. I’d love to hear your thoughts. Thank you.
From what I've seen most of the gender wage gap is actually *within* occupation.
I think the theory makes sense and men tend to rate salary as more important than women do in surveys.
That being said, it seems like in practice women often end up choosing well remunerated careers like healthcare (doctors, nurses). That might be for reasons other than salary, including "meaning", but the end result is such that, as I said before, most of the gender wage gap ends up being within occupation.
When saying that "motherhood would explain a great deal, perhaps the lion’s share, of the disparities we observe in wages between men and women", you mean the wage gap that is already controlled for objective factors (just to confirm, because you said in the comment gender wage gap is largely within professions, which kinda confuses me, because different professions earn different wages obviously)? The objective factors are like professional choices, level of seniority, hours of working (not personality because you used to think that men's appetite for competition might play a role in explaining the wage gap).
I think we should not overestimate how just the possibility of a woman to carry a child can (subconsciously) influence her's career choices, as you seem to start inclining to. To me it doesn't pass the common sense test. Unless woman explicitly plans to have kids (so not just "still thinking whether to have kids or not", there's little reason to think that she's factoring her child-bearing into the career choices she makes.
Thanks for helping me understand why I'm not a famous writer yet: "Perhaps the raw potential for literary brilliance is equally distributed between the sexes; yet the freedom to cultivate it, uninterrupted, unexhausted, untouched by the practical and emotional claims of motherhood, is not." 😆
I read your article all the way through, but my 2 cents is that this topic barely needs to be studied, because...DUH. Of course motherhood decreases wages. I cut my work hours 33% when I had my first child and now-16.5 years later, I am finally going back to full time.
I think the anticipation of motherhood is less an internal adjustment problem but more an external one. Employer assumption of child raising demands puts you on the slow track. Same when they know you have children. It is not always opting out of assignments but not being asked in the first place. There is an excellent study you should look into in the UK about women aging. Employer discrimination increases in age even when many of these women have more time due to having an empty nest. The explanation is that the employers now anticipate menopause related health issues. The thread here is discrimination rather than employee ambition.
Thanks. If you look at the MRKh women, they earn less than men in their 20s and I think that's anticipatory discrimination from employers. But in the long run the earn the same.
That being said, I know a fair amount of women who eg choose a less time intensive medical specialty because they think of motherhood requirements in advance.
Technically speaking you could have two normal distributions, one for men one for women, with equal mean and different variance.
Maybe a log-normal distribution would be more appropriate than a normal distribution if you're in the US.
It's also possible that the median wage for men is lower, e.g. because GMV causes men to crash out of the labor force at a higher rate. But positive variation on the high end brings the male mean back up to equalize with women.
Point being if you want to study GMV I think you really need variance data.
High incomes go much higher than low incomes go low. If you earn nothing, that's 60k away from median wage. Top earners earn millions. So unless there's a lot of wage compression GMV should show as higher mean.
In principle you could have rare male top earners who get counterbalanced by extra male zero earners who torpedo their careers somehow. 60-70% of homeless are men in the US.
You'd need the ratio of employed men earning very little (this data doesn't include unemployed ppl) versus employed women who earn very little to be 20 times greater than the ratio of men earning 1M over women earning 1M
Excellent article. One quibble: “did not have children—and, more curiously still, were often celibate” — is this really more curious, given the availability & effectiveness of contraception when most of those writers worked?
For a non-celibate but still childless contemporary example, consider Edna St. Vincent Millay. Her love sonnets are the equal of Shakespeare's if not their superior (spicy take, I know), she had an amazing, tumultuous life in some ways not so different from that of a "bad boy" male author, and her very gender-conscious perspective is nonetheless both passionate and unsentimental in a way that recalls Joan Didion (consider "I being born a woman, and distressed...")
I love this!! Thanks for writing. After reading your last section I got curious whether the same applied to science so I started looking up female Nobel prize winners. Surprisingly enough most of them (70%) are married with kids! Obviously not representative but very interesting to see.
I think you should apply the same skepticism to the MRKH paper that you do to the IVF paper! Three issues off the top of my head. First: per the abstract, MRKH women marry older men, maybe there is a mentoring effect where the older spouses help their younger wives move up professionally. Second, there could be pleiotropic effects from the MRKH syndrome-causing mutations, e.g. altered WNT signaling can drive both MRKH and hyperandrogenism. Third, the paper compares women who know from adolescence they can't have kids to all men. Maybe men who knew from adolescence that they couldn't father children would have the highest income of all, significantly higher than MRKH. You could tell a story like "when you know you won't have kids, you focus on building your legacy in other ways, such as starting a business/becoming an executive/whater". I'm not endorsing those hypotheses, I'm just saying that if one weren't already inclined to agree with the paper, it's easy enough to poke holes in it.
Anyway this is all silly because as you say, it's wildly obvious that children are a big driver of the gender wage gap! I'm not even a grandmother and it's obvious to me! The burden of proof is so incredibly strong on people who disagree with that statement, and even the best social science provides such weak evidence, that there's basically no point studying it!
Thanks for bringing this up. Androgenic effects are a real concern here and the first thing I checked when I read this. It does seem like MRKH women have perfectly normal sexual development otherwise, including normal levels of female hormones: https://pmc.ncbi.nlm.nih.gov/articles/PMC1832178/
I think the other points are less relevant though. The key question here is how much of the gender wage gap is due to motherhood or some other intrinsic female characteristics. If women who cannot have children choose older partners and that increases their wages, it suggests "femaleness" without children leads to choices that increase wages.
As for men with a possible analogous condition: I would expect the results to go the other way, in the direction of them earning less. Men with children earn the most and many men feel a responsibility to earn in the expectation of a family.
That last point is debatable: “men with children” selects for positive traits that cause high salaries so idk if there is clean evidence of causation. Certainly while I wouldn’t trade my kids for anything I would publish more papers without them. Nevertheless I agree my last point is probably, just not certainly, incorrect.
As for my first point, I guess it’s a question of what point you are trying to make. Accepting arguendo that wage gap reduction is mediated by older mate selection, there is an interpretation by which females are worse at work but receive a leg up due to e.g. better networking or resume help, in which case “femaleness without children” and the unfair benefits cancel out the inferiority. To be clear this is very stupid! It’s just one of very many ways that this study could be wrangled into a sexist interpretation if one had a sexist prior.
Maybe the best way to put this is, if there was still a gender wage gap for MRKH women it wouldn’t, and shouldn’t, change your beliefs! You would just say that MRKH women have their own unique psychological challenges due to societal norms claiming women are meant to have children, so therefore they are “defective”, and this causes lower wages! That is a way more plausible argument than any that I gave above, right? So if the complete opposite result from the MRKH paper wouldn’t move you off your prior then it’s just not that relevant.
I think the data shows they have higher wages post children, due to pressure to support family but selection effects certainly play a role. That being said, I think that's a different question from what would men do if they knew they'd never have kids. My bet is they'd be less ambitious, not more. With exceptions for monk like autists in intellectual fields.
I understand your other points, but we can endlessly debate abt social underpinnings. I still think "condition that doesn't allow one to have kids" is a clean ish experiment. Ofc, I also take ur point that in the end no empirical evidence is enough and you need to rely on common sense.
How can a man know he can't have kids at adolescence?
No idea, maybe there’s some sort of genetic thing, it’s more of a hypothetical though if such a population existed it would be interesting to study
Very interesting, thank you! I would just add that the MRKH results tell us about the state of the wage gap in Sweden, a country which is quite an outlier in terms of gender attitudes. So while the internal validity is great, I wouldn’t be quick to conclude that the same is true in the US or Japan or Italy
Yes that's true, thanks. I should probably add that Sweden has more wage compression than US and maybe at the right end rail in US you'd observe more of a difference.
This is fantastic, thank you! Your articles have helped me refine my critical thinking skills, and I appreciate it.
Thank you!
Although I entirely agree with you, George Eliot did not bear children but she did serve as a mother to her partner’s children and saw herself as a mother.
Oh interesting I didn't know abt that guess i need to update
Good biography: https://amzn.to/4oEYa5S
Thank you for putting in the work to defend rigorously the 'obvious' answer - someone needs to!
I admit to being absolutely baffled by the people who dispute the idea that having childcare responsibilities makes it harder to progress in your career. Obviously! Have these people never *met* children? I'm actually a father here, albeit for the purposes of these studies I suspect I would be more accurately labelled as a 'mother' given that it's shorthand for 'childcare and housework effort' (I took almost two years off when they were very small and continue to do much of the day-to-day as my wife reboots her career).
Yes, my two children have without a debt not only impacted my career progression but my choice of career direction, because they just *take a lot of time*. It's as simple as that. I have no regrets on that front, they're the light of my life, but one can be realistic that there are simply only so many hours in a day.
Yeah, thanks for saying that. If you don't mind: what's your profession?
Finance; I got to a pretty privileged position in my career where I could trade experience either for more money or more time (whilst still making objectively good money) and picked the latter to be able to be flexible for child care.
Probably also worth mentioning that people tend to vastly overestimate how fast moving most fields actually are (unless you work in AI, I guess); I took almost two years off with my kids and it was not difficult to catch up.
It would be interesting to see “men” in that graph split into men with children and men without children. Do you know if it’s still the case that men with children earn more, controlling for education, et cetera?
Men with children earn more. Also known as the fatherhood wage premium.
I was also surprised when I first noticed that many great philosophers were bachelors. I wonder how much gain fully productivity gain outsourcing childbirth and childcare would bring, if we aren't replaced by AI first.
you mean male or female philosophers?
If I remember correctly, I first realized it when I noticed only one "great British empricist" (George Berkeley) had children.
Probably a trend which will get stronger as dating has turned into more of a negative-sum tournament for men. The guys with the deepest career ambitions will delete the apps and hire prostitutes.
https://www.youtube.com/watch?v=-Z62YOHP1uc
I remember when I was a young software engineer in San Francisco, with a big oversupply of single men relative to women, I felt the tradeoff between career ambitions and dating quite strongly.
Wasn’t mating more demanding before? I guess there was always a tradeoff between work and human affairs.
In any case, conceptualizing and predicting the future of people's sexual preferences will need more than a psychologist psychoanalyzing "the rich" and "the beautiful." The replication crisis should've taught us that intriguing narratives cannot replace meticulous research.
Addendum: And regardless, only a very tiny portion of people will be workaholics enough to forgo any love. And the AI revolution making whole job sectors obsolete, and perhaps designer baby tech selecting against extreme obsessiveness will reduce their numbers even further.
>Wasn’t mating more demanding before?
I think dating apps have caused an acceleration, a sort of bidding war among men for who can sink the most hours into dating (in terms of optimizing their rizz, going on more dates, etc. etc.). Women have more choice, which means a man's probability of wifing up any given date is lower.
https://books.google.com/ngrams/graph?content=ghosting&year_start=1800&year_end=2022&corpus=en&smoothing=3
>In any case, conceptualizing and predicting the future of people's sexual preferences will need more than a psychologist psychoanalyzing "the rich" and "the beautiful." The replication crisis should've taught us that intriguing narratives cannot replace meticulous research.
I am tempted to argue the opposite. The replication crisis taught us that grandma's intuition is worth more than "meticulous" (p-hacked) research, as Rux argued in her OP. Putting a science veneer on fundamentally corrupt research practices isn't worth much. Actual honest observations are worth more. Data quality matters more than sample size. (Little use in gathering more samples from a data source that's fundamentally engineered to be unrepresentative! As many small studies are, due to file drawer effects etc. To be fair, Youtubers face incentives which are arguably just as bad as researcher incentives.)
> I am tempted to argue the opposite. The replication crisis taught us that grandma's intuition is worth more than "meticulous" (p-hacked) research, as Rux argued in her OP. Putting a science veneer on fundamentally corrupt research practices isn't worth much. Actual honest observations are worth more.
What makes grandma's advice reliable is less her qualities and more her conveyance of accumulated wisdom of the people. The random YouTuber offering vague and baseless psychoanalysis, whatever he might be doing, is not passing down time-defying ideas and dispositions.
Also, don't be a scientific nihilist. You can make good as well as bad science. But psychoanalytic "conclusions" are all arbitrary.
I'm pretty sure the video I linked is based on observations and economic analysis, not "psychoanalysis"?
I added a third paragraph to my reply concerning a) men who are obsessive enough to forgo all love are a rather small psychological minority b) their number will reduce even further as AI makes more job sectors obsolete and designer baby tech selects against extreme obsessiveness.
why would designer baby tech select against obsessiveness
If I didn't got it wrong, OCD, autism, ADHD and epilepsy are all related to each other. Trying to have less of any of those would reduce obsessiveness, and even modest drops in average obsessiveness would reduce extreme (I'll be all about work and will forgo any love) obsessiveness.
It's not that they "forgo all love", it's that they conclude the juice isn't worth the squeeze. The more squeeze required for the same amount of juice, the more guys opt out. This will be especially true for men who have a high self-perceived opportunity cost of their time, e.g. high earners.
Perhaps in the past it was a tiny minority of great philosophers, but nowadays their numbers are growing since dating has become a misery tournament. Plenty of anecdotal evidence for this online.
You should follow Nuancepill: nuancepill.substack.com
Even reputable think tanks like Pew publish attention-grabing misleading studies fueling the gender war narrative. Treat all "intriguing gender factoids" as false until proven otherwise.
Here's a good debuking tour of many popular meme statistics, with sensible speculations on their persistance: nuancepill.substack.com/p/why-is-the-incel-narrative-so-popular
Ok, so I'm going to pick on a small side comment (forgive my autism): "For while we may safely concede that history has offered no female Einstein or von Neumann ..."
I have a theory for that. (I am deeply suspicious of GMV, so it's nice to have another explanation for the sex imbalance in genius.)
Einstein, Mozart, Newton, and most other names that come to mind for "genius" seem to have enjoyed a kind of education that some call "aristocratic tutoring": (https://www.theintrinsicperspective.com/p/why-we-stopped-making-einsteins).
You probably need both aptitude and tutoring to get an Einstein, but - it passes the smell test for me that this kind of intense academic tutoring has rarely been tried on women. Women have been tutored to be "womanly", of course - I'm not a blank slatist either but we wouldn't have invented "finishing schools" if the skills there somehow naturally developed with female puberty.
As you say "It appears that the mere anticipation, the steady, lifelong knowledge, that one might become a mother shapes a person in ways that not trivial. Expectations laid upon us in youth can cast longer shadows than we imagine."; replace "mother" with "genius" and the principle could still work?
The one example I can think of where this has been tried and worked spectacularly on women is the three daughters of László Polgár, all of whom became world-class chess players. Judith was apparently the only woman in the (non-gendered) top 10 world ranked chess players at the time. (Sophia, the "least successful" of the three, later settled down to become a mother and housewife; she only made it to women's world #6 compared to her sisters' #1 and #2.)
László certainly was of the "geniuses are born not made" opinion, one wonders if he had taken an interest in physics rather than chess, if we'd have our female Einstein there. To quote him from Wikipedia, again returning to the theme of motherhood:
> "Men must be clever and hard," Polgár said. "Women must be beautiful and look after the family. Only then, if they have time, can they be clever."
***
EDIT: found the counter-argument to this theory that I meant to post for fairness' sake: https://www.astralcodexten.com/p/contra-hoel-on-aristocratic-tutoring
That makes sense but I’d say it’s still interesting you do have exceptional female writers
Dumb question: what does autism have to do with the gender wage gap? Is it a belief that autistic men are more likely to contribute the tail effects (and thus the GMV hypothesis) of work performance?
I don't know that anyone has discussed autism and the gender wage gap.
Depends on what kind of autism, too. Asperger's maybe more likely to make men study computer science?
Best I can find (UK specific) is https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/disability/articles/disabilitypaygapsintheuk/2014to2023 which says the disability pay gap is wider for men than for women, although I don't have to raw data to compare by both sex and specific kinds of disability. From a glance at the graphs and text though, being autistic seems to have a negative effect size that trumps almost everything else, with the closest contenders being correlated factors like "severe learning disability".
So back to the OP's question, once we accept that "autism" doesn't mean "sort of like Alan Turing" it's not so much a case of GMV's "same mean, different SD" claim as evidence for autism = much lower mean/median pay in the first place, if you're one of the lucky ones who can get a job at all.
A study looking at the people with the "good at math and science autism" (I hate that meme) would be interesting indeed, but then we're conditioning on high-performers compared to the general population in the first place.
So, going off UK statistics because I have them to hand,
The gender pay gap (non-autistic) is around 7% per hour (https://researchbriefings.files.parliament.uk/documents/SN07068/SN07068.pdf) or 12.8% in total (since women are more likely to work fewer hours per week); I'm with Teslo that this should be called a motherhood wage gap. Even the UK Government, in Sec. 3.1, agrees.
The (previous) Government's Buckland Review (https://www.gov.uk/government/publications/the-buckland-review-of-autism-employment-report-and-recommendations/the-buckland-review-of-autism-employment-report-and-recommendations) notes that only 3 in 10 autistic people are employed at all (compared to 5 in 10 people with a registered disability more generally), and that's before digging into part or full time hours. "Autistic people face the largest pay gap of all disability groups", Buckland says, citing 2021 figures (https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/disability/articles/disabilitypaygapsintheuk/2021/pdf) which gives a general disability pay gap of 14% (pre-covid) whereas for autistic people, the gap is 33% if I'm reading the findings correctly, but around 10% "after controlling for pay determining characteristics". I don't have the brainpower to read up on the exact details of the regrassion analysis right now, but splitting by sex should definitely appear in there somewhere.
Basically, Autism overall is a huge NET NEGATIVE to predicted earnings, because it includes more of the "needs 24-hour assisted living" or "just about qualified to work in a warehouse" people than it does "needs 50% extra time and a small room, but graduates with a 1st Class MSc in Computer Science" people.
The Musks of this world are no more representative of autistic people than they are of humans as a whole.
"The MRKH natural experiment is particularly powerful because it identifies a group of women who are biologically typical in every respect except one: the congenital absence of a uterus. As a result, they are far less likely to have children unless they pursue surrogacy." I am in support of the argument for the pay gap... but these papers are all making a gross assumption that overlooks a somewhat common practice when women cannot conceive: Adoption. There would be no medical assessment possible for a woman that would reveal that she has an adopted child.
(Clears throat and steps up on stump for those in the back)
As an adopted person, it's discouraging to see how often the medical community ignores those of us who are adopted - especially in a closed adoption where we lack access to our genetic history. I just filled out a very lengthy genetics questionnaire having to select "Unknown" for each question instead of one option at the top which would allow me to submit one answer of "adopted, unknown".
This widescale ignorance also reinforces the old question that many of us get from people when we reveal we are adopted, "But do you know your REAL parents?" Our REAL parents are the ones who raised us. Our REAL moms are the ones who took the hit to their pay (or exited the workforce altogether). And I promise you that our REAL parents love and care for us the same as a biological child even though others may doubt it.
Thank you for coming to my TedTalk. Now please continue doing the great work you are doing... but keep us adopted/adoptee people in mind.
(Steps down from stump, sees someone I know and wave, exit to the bar)
Hello, great article! I would like to ask about the current consensus regarding the theory that male competitiveness (for status or mates) drives young men toward money-focused careers. Specifically, is it true that this drive pushes them to make trade-offs for higher earnings compared to, for example, hospitality professions? You mentioned this briefly in your piece.
LLMs tell me this might only have a minor effect, which isn't surprising given that Claudia Goldin’s work suggests motherhood explains 60-80% of the gap. I’d love to hear your thoughts. Thank you.
From what I've seen most of the gender wage gap is actually *within* occupation.
I think the theory makes sense and men tend to rate salary as more important than women do in surveys.
That being said, it seems like in practice women often end up choosing well remunerated careers like healthcare (doctors, nurses). That might be for reasons other than salary, including "meaning", but the end result is such that, as I said before, most of the gender wage gap ends up being within occupation.
Some of it it between occupations, just not the main element
When saying that "motherhood would explain a great deal, perhaps the lion’s share, of the disparities we observe in wages between men and women", you mean the wage gap that is already controlled for objective factors (just to confirm, because you said in the comment gender wage gap is largely within professions, which kinda confuses me, because different professions earn different wages obviously)? The objective factors are like professional choices, level of seniority, hours of working (not personality because you used to think that men's appetite for competition might play a role in explaining the wage gap).
I think we should not overestimate how just the possibility of a woman to carry a child can (subconsciously) influence her's career choices, as you seem to start inclining to. To me it doesn't pass the common sense test. Unless woman explicitly plans to have kids (so not just "still thinking whether to have kids or not", there's little reason to think that she's factoring her child-bearing into the career choices she makes.
Thanks for helping me understand why I'm not a famous writer yet: "Perhaps the raw potential for literary brilliance is equally distributed between the sexes; yet the freedom to cultivate it, uninterrupted, unexhausted, untouched by the practical and emotional claims of motherhood, is not." 😆
I read your article all the way through, but my 2 cents is that this topic barely needs to be studied, because...DUH. Of course motherhood decreases wages. I cut my work hours 33% when I had my first child and now-16.5 years later, I am finally going back to full time.
yeah, that's kinda my point. But since ppl study it, it's worth pointing out some studies are rly bad.
I think the anticipation of motherhood is less an internal adjustment problem but more an external one. Employer assumption of child raising demands puts you on the slow track. Same when they know you have children. It is not always opting out of assignments but not being asked in the first place. There is an excellent study you should look into in the UK about women aging. Employer discrimination increases in age even when many of these women have more time due to having an empty nest. The explanation is that the employers now anticipate menopause related health issues. The thread here is discrimination rather than employee ambition.
Thanks. If you look at the MRKh women, they earn less than men in their 20s and I think that's anticipatory discrimination from employers. But in the long run the earn the same.
That being said, I know a fair amount of women who eg choose a less time intensive medical specialty because they think of motherhood requirements in advance.
Good post. I don't think this data refutes GMV since we're only looking at means, not variances?
Well, very high incomes from men should skew the mean.
But it could be that in Sweden you have wage compression that diminishes those effects.
Technically speaking you could have two normal distributions, one for men one for women, with equal mean and different variance.
Maybe a log-normal distribution would be more appropriate than a normal distribution if you're in the US.
It's also possible that the median wage for men is lower, e.g. because GMV causes men to crash out of the labor force at a higher rate. But positive variation on the high end brings the male mean back up to equalize with women.
Point being if you want to study GMV I think you really need variance data.
High incomes go much higher than low incomes go low. If you earn nothing, that's 60k away from median wage. Top earners earn millions. So unless there's a lot of wage compression GMV should show as higher mean.
In principle you could have rare male top earners who get counterbalanced by extra male zero earners who torpedo their careers somehow. 60-70% of homeless are men in the US.
You'd need the ratio of employed men earning very little (this data doesn't include unemployed ppl) versus employed women who earn very little to be 20 times greater than the ratio of men earning 1M over women earning 1M
Excellent article. One quibble: “did not have children—and, more curiously still, were often celibate” — is this really more curious, given the availability & effectiveness of contraception when most of those writers worked?
Yeah you're right although I think what's interesting for example is that V Woolf was celibate within a marriage most likely
For a non-celibate but still childless contemporary example, consider Edna St. Vincent Millay. Her love sonnets are the equal of Shakespeare's if not their superior (spicy take, I know), she had an amazing, tumultuous life in some ways not so different from that of a "bad boy" male author, and her very gender-conscious perspective is nonetheless both passionate and unsentimental in a way that recalls Joan Didion (consider "I being born a woman, and distressed...")
Interesting hadn’t heard of her
OMG I got to introduce someone new to Millay!
Her most perfect sonnets IMO are:
https://www.poemtree.com/poems/And-You-As-Well-Must-Die.htm
https://www.poetryfoundation.org/poems/148564/i-being-born-a-woman-and-distressed
https://digital.library.upenn.edu/women/millay/figs/faithful.html
https://allpoetry.com/Love-Is-Not-All
https://allpoetry.com/Sonnet-XXVIII:-From-Fatal-Interview
https://www.poetryfoundation.org/poems/46557/what-lips-my-lips-have-kissed-and-where-and-why
I love this!! Thanks for writing. After reading your last section I got curious whether the same applied to science so I started looking up female Nobel prize winners. Surprisingly enough most of them (70%) are married with kids! Obviously not representative but very interesting to see.
I use Claude to create a chart for it: https://claude.ai/public/artifacts/3e952457-709e-43cf-b706-7f81ed1957ba (All mistakes Claude's, not mine :D)