What not to wear, economist edition

What should you wear for an academic job interview? Ariel Rubinstein suggests casual attire:

I would argue that wearing jeans and a t-shirt is your dominant strategy:

If you are a good student, then a department that will not give you a job because of your "sloppy" appearance does not deserve to have you.

If you are mediocre, then there are many other candidates like you and dressing casually is the only way for you to get noticed.

Shakespeare, speaking through the character of Polonius, suggests buying the most expensive clothes you can afford: 

Costly thy habit as thy purse can buy,
But not express’d in fancy; rich, not gaudy:
For the apparel oft proclaims the man;

These two pieces of fashion advice are not as diametrically opposed as they might seem. Both recognize that clothes act as signals. The question "what not to wear?" is asking what should, or shouldn't, a person be signalling.

The first thing to signal is health, energy and vitality.*  Whatever else: it has to look good. For men, looking good is about fit, cut, tailoring - cheap suits are my pet peeve. My least favourite look for female candidates is dark suit, white shirt, and no make-up. It's a combination that looks terrible on just about everyone. But replace white with a more flattering colour, and buy some mascara, and the odds of getting a second interview increase significantly.

This might sound sexist, but it's just an acknowlegement of reality: undergraduate programs are largely meritocratic, grad school mostly so, but on the job market, gender matters. A woman who tries to follow the Rubinstein strategy faces a dilemma: ill-fitting jeans are unflattering, but well-fitting jeans, ones that signal health and fitness, risk drawing the hiring committee's attention to the wrong assets. This is not just a paranoid-feminist worry - my favourite moment so far this hiring season was the letter-writer who accidentally penned "she is a great person to talk about".

A female candidate must walk a knife's edge. Too many feminine frills and she risks not being taken seriously; too few and she is seen as unattractive. Dressing like a man sends entirely the wrong signal, as the photo illustrates. Unlike male candidates, they face an added challenge: not looking like one of the administrative staff.

But whether you are male or female, it is never easy to produce precisely the right signal of ability. Last year jeans and a t-shirt would have been a way to signal originality and  creativity, but now Ariel Rubinstein's let the secret out, all it signals is that you waste time reading internet job market advice. You could try signalling  I-am-so-busy-I-don't-have-time-to-shave by simply not shaving, but unless you are already established and successful, that could be misinterpreted as signalling lack of attention to details.

You could wear something like, say, a silk scarf or tie from the Palm Springs Art Museum as a way of signalling erudition and taste. Yet this assumes, first, that the candidate has the time and money to visit art museums. Even more problematically, it assumes that the members of the hiring committee could distinguish a Palm Springs Art Museum tie from one bought at Target. Indeed, this is an issue with signalling strategies in general - they assume the existence of someone with the ability to read signals.

Vintage clothing from the thrift store might signal resourcefulness and entrepreneurial spirit - or poverty. That's a risky strategy, because signalling desperation is to be avoided at all costs. Remember the lessons of Marxist economics: "I don't want to belong to any club that will accept me as a member." Every employer wants to hire a candidate that is just a little bit too good for them. Rubinstein and Polonius's strategies are very different ways of sending out the same signal: I don't need to impress you. I don't need your money. I'm doing just fine.

Sometimes the best solution is a mixed strategy equilibrium - jeans with $300 dress shoes, or a beautifully tailored suit with outrageous mismatched socks. Here is where accessories come in. For a development economist, a "$100 laptop" from one.laptop.org neatly signals "I know Linux, and I know how to get my hands on one of these machines."  

Ultimately, the economics job market is a matching game - each candidate and each employer is trying to find the candidate that's right for them. Hence a candidate's signal should be tailored to the type of employer he or she wishes to attract. Jeans and a t-shirt say "I'm not willing to put on a suit just to please students or clients or hiring committees". Since that would be a bad signal to give to a teaching oriented university, jeans and a t-shirt is a way of signalling interest in research. Matching also makes things tricky for female and minority candidates - a candidate who is demographically similar to the hiring committee can aim to fit in, to dress like one of the guys. Others can try, but they are less likely to succeed.

In other words, it's a complete total and utter minefield - everything from your hair to your shoes is a signal, and you could get it wrong. Looking on the bright side, as Amartya Sen once observed, economic man is close to being a social moron.** You might have been the least fashion-conscious and most clued out person in high school, but in a group of economists, you too can be a fashion star. And even if you aren't, the average economist is unlikely to notice.

* The desire to signal health also explains the outdoorsy profile pictures  here, here, here, and here

**Yes, I know that wasn't what he meant. But it's funnier that way.

Comments

[Kevin Milligan]

Thanks Frances. I think Prof. Rubinstein's Q1 and Q2 answers are *horrible* for anyone but the most brilliant student. I wonder if he understands what research is like for the rest of us not as brilliant as he.

His Q4 answer is typical baby boomer. Sloppy jeans. Younger generations are different. Our clothing rebellion is against the sloppiness of our parents.

As Russell Smith writes, "“It’s these young guys rebelling against their boomer dads,” said Russell Smith, 45, the author of “Men’s Style: The Thinking Man’s Guide to Dress” and an advice columnist for The Globe and Mail in Toronto. “But it’s very amusing and paradoxical that the new anti-parental paradigm involves a pinstripe suit and a pocket square.”"

[Kevin Milligan]

Whoops--re clothing I meant Q3 not Q4.

[Frances Woolley]

Kevin, delighted to give you an excuse to vent!

Actually this post was inspired by one of my students, who was ranting about the unprofessionalism of profs who wear jeans to class. But I couldn't figure out an easy way of working that into the post.

[Determinant]

The age-old dress debate. IME you have three choices:

1) Casual. T-shirt and jeans. Varies to dress shirt and jeans with stylish shoes, but the jeans dominate and always will. Relaxed companies go for this. GM has a reputation of not hiring engineers who show up in suits and picking people who dress this way; they want creativity and informality.

2) Casual-Formal. Tie, button-down shirt, sweater if you want, dress pants. I wear a tie chain but that's my bow to individualism. This is what I wear to church every Sunday because I sing in the choir and my suit is saved for interviews.

3) A suit. Always a safe bet. As I said my bow to individuality is a tie chain and a reverse knot that dimples very well. Elegant presentation.

When I went for an interview at Health Canada in June I wore (3). The interviewing board were in casual dress, it was one of the most humid days of the year in Ottawa and there was hail that day. I refused to risk a faux-pas with anything less than a suit. There was a female candidate in line with me who was in a power suit.

Likewise I was in Toronto for another Health Canada test, it amounted to a written interview, my answers were in writing to be reviewed by a board in Montreal. The HC staff and the other candidate were in chic jeans, good shoes and open-necked shirts. Stripped of course, some with sweaters. I can't pull this style off, I look horrible in thin-legged jeans. I was in a suit, possibly I looked like I was trying too hard but at least I didn't shoot myself in the foot sartorially.

At any interview it is wise to size-up the target and pick your clothing strategy. At some places a suit is a bit much but you have to identify them beforehand. A suit is always the default.

Never doubt the power of a tie chain, well-knotted tie and shined shoes (shoes always suffer, the overlooked detail that sets off a suit, IMO). A pocket square is optional but nice. I have my grandfather's Trilby to match my trench-coat. I have to have some headwear to go with my suit in the cold months.

If you are going to pay me $50K+ and give me a pension, ask me to represent the Government of Canada and are frankly offering a career-making position, I will wear a suit.

A good interview candidate will peg the target and dress according to the vibe of the target. A little scouting always helps.

[Kevin Milligan]

More seriously, the attire of female professors is held to a much different standards. I have heard from female colleagues that a male can get away with sloppy and still command respect from students. For females, this is much more challenging.

(To be clear, I don't think this is the way it ought to be. I am describing how female colleagues have talked about the way students treat them.)

[Min]

It was not for an academic job interview, but some years ago I moved in with my new girlfriend, who was a college professor. I was moving from the US Southwest, where jeans were normal attire. In preparation for college social events I followed Polonius's advice and bought, get this, a raw silk sports jacket. Talk about rich but not gaudy! ;) OC, as my girlfriend pointed out, I would have stuck out like a sore thumb. It was jeans and T-shirts for me. I did get some mileage out of a leather vest, though. ;)

On the general question, though, nothing beats research. Whatever club you want to join, you should look like you already belong.

[John]

min wrotes: 'On the general question, though, nothing beats research. Whatever club you want to join, you should look like you already belong.'

Isnt what matter the most? department will always prefer the candidate with the best research if no candidates make big mistakes during interviews and flyouts... and they are reasonably nice...

[Determinant]

Spam Filter? I had a nice categorical post on sartorial choices.

[Chris Auld]

Any suit for a man is fine. A suit is necessary unless one is attempting an extreme counter-signalling strategy. Much like for weddings and funerals, a suit signals not wealth or health or a keen fashion eye, but rather that you're taking the situation seriously. A cheap, ill-fitting suit might even be better than an expensive tailored suit, because who wants to work with someone who actually likes wearing suits?

[Frances Woolley]

Determinant, it's now rescued from the filter.

Kevin, agreed, female profs are held to a different standard. The everyone-wears-jeans democracy of grad school disappears once you hit the job market. But at the same time, female profs have a wider range of choices available - I could present a paper wearing a black, brown, blue, beige, or pink suit - and have. Would you wear pink without fear?

Min - really? Have to say I'm a huge fan of raw silk sports jackets - perhaps it was something about the jacket in question.

Chris "Any suit for a man is fine."

No. No. No. Perhaps 20 years ago before every wage slave was an "account manager" wearing a suit and polyester tie. But not now.

Especially one that doesn't fit because your parents bought it for your high school graduation and you've bulked up since then.

A well-fitting sports jacket and dress pants beats a badly-fitting, badly cut suit every time.

The only exception is the jacket-and-pants-that-are-almost-but-not-quite-the-same-colour sports jacket and pants ersatz suit combos.

[Jeremy Fox]

I'm with Min: do your homework, find out how people in your field usually dress for interviews, and dress like that. If you like, you can think of this as sending the signal "I am a professional in my field, and am familiar with how professionals in my field behave." Trying to signal anything more clever or nuanced with one's clothing runs a serious risk of having the signal misunderstood. Yes, wearing jeans and a t-shirt might signal to some that you are so smart you don't need to dress the part. But it might signal to others that you're crazy, or that you don't really want the job, or that you're eccentric to the point of being unprofessional, all of which are good reasons not to hire you even if you are really smart.

As for dressing unconventionally in order to stand out, I'd have thought that, if you've landed an on-campus interview, you already stood out. Once you get to that stage in the process, I'd think you'd want your conduct during the interview to do the "signalling", not your clothes.

[Frances Woolley]

Jeremy - "As for dressing unconventionally in order to stand out, I'd have thought that, if you've landed an on-campus interview, you already stood out."

In economics we have this huge job market, held at the American Economics Association meetings on the first or second weekend in January, and each employer will conduct say 25, 30, 40, 50 preliminary interviews. Since the navy, grey and black suits tend to fade into each other after a while, there's an advantage if the committee can remember "oh, yeah, that was the candidate with the red shoes."

[Stephen Gordon]

Now I'm all self-conscious about what I'll be wearing when we're interviewing candidates in Chicago.

[Determinant]

Thank you Frances.

Min has it right, look like you want to join the club. Established members can be a degree less formal than you, you should be a degree more formal as you want to join and wish to be made.

Seriously, with the ever-present threat of unemployment and the legions of consultants, career counsellors and websites to help you, making a bad signal by dressing wrong is the least of anyone's worries. People my age have had signalling-by-dressing drilled into us since high school. We can do it and do it well. This is not rocket science.

Wearing jeans for an interview that calls for a suit is just so passe. Seriously, dress is all about knowing the rules of the game and demonstrating that you can play by those rules. You can break them later, or tell a nice story about how you broke the rules but it was all for the best and it turned out really well during the interview itself. But dress like you know what you are doing.

[Shangwen]

I have hired many academics (despite not being one--really grateful after reading all these job discussions), and I have to add the obvious point that your ability to carry the signal in other forms after the first 60 seconds really matters, in terms of non-academic behavioral issues, either in the interview or other events (esp dinner) related thereto:

1. Do not suck the vital energy from your hosts by discussing your unhappy marriage, regrets, etc.
2. Do not try to convey enthusiasm for research by describing how emotionally affected you were by it.
3. Do not try to signal sophistication at dinner by badgering the restaurant staff about unlisted wine inventories.
4. Do not make strong personal statements about likes and dislikes, for the same reason as (3); understand what the median image of "jerk" is.

Have seen it all, and worse. But my heart goes out to new grads. To maintain a perfect balance between seriousness, collegiality, and social inoffensiveness in the span of two hours is not easy.

[futur jmc]

this is a bit silly, no?

I prefer to be remember as the candidate with good publication x rather than the candidate with red shoes...

I thought department made provisional list of who the prefered before interviews...

[Greg Tkacz]

I think all Economics profs should dress like that handsome devil on page 139 of the most recent Maclean's University ranking issue: dress pants, shirt, tie and the ever-present cardigan. A cardigan signals that you can dole-out sage advice while staying warm in a chilly classroom.

[Stephen Gordon]

Actually, Laval candidates have nothing to worry about. As anyone can tell you, I am completely clueless in such matters.

[Neil]

Well, at least the xkcd tie in the picture is a good start at fitting in with pretty much any numbers-oriented hiring committee.

[Determinant]

Speaking of Laval hiring, when you said the ability to lecture in French was key, Stephen, it reminded me that there is an entire industry in Montreal and Ottawa devoted to teaching functional and somewhat technical French to Anglophones, especially aspiring Public Servants.

A spell in the French Fix-it Shop for CCC-level training and a budding newly-bilingual economist would seem to have a decent advantage.

Or am I completely off-target?

[Frances Woolley]

Neil " the xkcd tie" - I wondered if anyone would notice!

Shangwen: "2. Do not try to convey enthusiasm for research by describing how emotionally affected you were by it."

What's interesting to me is the variety of ways people interpret the question "what's the most important thing you learned when doing your doctoral research?" Some people will read this as a personal question and answer, say, I learned how to manage my time. Some people will read this in terms of research technique and answer, say, I learned how to solve XYZ problems - which is also a personal response. Only a minority will answer something like "I learned that the border between Canada and the US is thicker than the border between Canadian provinces" - i.e., some fact about the world discovered in the process of doing their research.

Future jmc "I prefer to be remember as the candidate with good publication x rather than the candidate with red shoes..."

The odds of a given candidate in the tail of the distribution are low. Most candidates, at least most candidates Carleton interviews, are in a fairly dense part of the distribution, where everyone on the interview list has a good-ish publication record and/or good letters. In that context, appearance does matter. Either that or every single person who spends money on clothes, hair or make-up is irrational - which is hardly the kind of assumption that an economist would recommend making.

Stephen "I am completely clueless in such matters."

Which doesn't mean you are immune to the influence of appearance. You are probably aware when a person looks good, or when a person doesn't look so good. Being clueless only means you have no idea why - whether it's the clothes/hair/shoes/make-up/whatever...

Greg - is that handsome devil anyone we know?

Determinant "Min has it right, look like you want to join the club."

But if you're a different ethnic background from the rest of the club members, or a different gender, can you ever look as if you want to join? Can someone wear a hijab, for example, and look like they want to join the club?

[Frances Woolley]

What I didn't say, but should have, is that Rubinstein's jeans-and-t-shirts advice is an example of "mirror image hiring", that is, the tendency of employers to try to hire someone just like themselves.

[Determinant]

Ethnic Background, Gender???

Maybe I am dating myself here, but gender has never, ever been a factor in jobs searches or coworker relations. I have always been under the impression that ethnicity or gender-based discrimination on the part of anybody is a quick way to get oneself disciplined or fired. Nobody tolerates it.

Hijabs aren't that common IME and in Canada those that wear them tend to favour the style that covers the hair but not the face. In which case I look past it. I've never been in a position of power but likewise if someone makes ignorant comments about the fact my mother is an ordained United Church of Canada minister then I start to feel the lack of diversity. The street works both ways.

I had a great summer job in 2004 at an avionics company. It had a very diverse staff. I had a choice, be friendly and make small talk about Romania or Zimbabwe based on knowledge of the newspapers and recent history (a good way to make friends is to appreciate their background) or make ignorant comments and lose the job as quickly as I had got it. Quicker, actually, I had been trying since February to get in.

I chose the former and made a lot of friends.

On the flip side, I was at an interview in Toronto at an engineering consulting firm where the interviewer (Asian background) asked me (British background, family resident in Canada for generations) where I was born, where my parents were born and where my grandparents lived. She was probing for the immigrant in my family. It was rude, uncalled for and unprofessional.

Most people my age view racism and sexism at work as a good way to get fired. I was at a call centre and one operator at another centre made appalling sexist comments to one our our own (female) operators. The offending operator was dismissed for cause.

[Mike Moffatt]

Instead of dressing down to get noticed, why not wear a tuxedo?

[Determinant]

Before dinner and while still conducting business? How gauche! ;)

[Min]

Determinant "Min has it right, look like you want to join the club."

Frances Woolley: "But if you're a different ethnic background from the rest of the club members, or a different gender, can you ever look as if you want to join? Can someone wear a hijab, for example, and look like they want to join the club?"

What came to mind were some T. S. Elliott lines about children peering through a fence. What I said was to look like you are already a member. If you are in an out-group (prejudice is not dead), looking like you want to join does not help get you in, I am afraid.

The great violinist Isaac Stern, who had many youngsters who wanted to study with him, said that he did not particularly look for technical proficiency, although of course all of his prospective students were skilled, the main thing he looked for was the demand to be listened to. :)

[futur jmc]

Frances,

I suppose on the junior market none of the candidates you interview or almost have a goodish publication...
no?

[Frances Woolley]

Futur jmc: "I suppose on the junior market none of the candidates you interview or almost have a goodish publication...
no?"

Goodish to me is top 2 or 3 labour journals (we're hiring in labour or behavioural). In the junior market I'd consider a revise and resubmit on one of the thesis chapters plus a co-authored forthcoming publication at, say, J. Hum. Res or similar a goodish publication record, and quite a few of the people we're looking at are in that category. I look a lot at rates of publication. If a person looks like they can publish one solid article a year that's generally a good sign.

One thing that junior people sometimes don't realize is that in many journals R&Rs generally do get published, so an R&R is a very good signal.

[futur jmc]

Thank you Frances!

you gave me incentives to work harder !

Looking at JMC at different universities, I didnt realise that many had r&r and publications with advisor.

Thanks once again for the information....

[Greg Tkacz]

Frances: "is that handsome devil anyone we know?"

Let's just say that after posing for Maclean's, I decided not to quit my day job.

[Andrew F]

Perhaps it depends on the faculty, but Mathematics at Waterloo, I don't think dress had much to do with the respect garnered by professors. If anything, I thought overly formal dress was a signal of insecurity.

[Frances Woolley]

Andrew F - "If anything, I thought overly formal dress was a signal of insecurity"

That's the logic behind Rubinstein's advice - to be taken seriously as an economic theorist, look like a mathematician.

Did the students at U Waterloo apply the same dress code to male and female profs?

[Neil]

@ Determinanat: "Maybe I am dating myself here, but gender has never, ever been a factor in jobs searches or coworker relations."

I think you probably just lack perspective. My background has come up on the winning side of history, and I grew up thinking we'd moved beyond gender and racial discrimination. Until I moved in with my wife and started to see the world through her eyes, it wasn't apparent to me the challenges of being a woman in the workplace.

And when I stop and think about it, while my office has lots of different ethnic backgrounds, nobody wears any cultural dress. No turbans, no headscarves, no kippas. Yet I see plenty of people in public spaces wearing these things, so I'm left thinking that if I were to experience the world belonging to one of these cultural groups, I wouldn't probably find plenty of discrimination. Even if it's not overt enough to sue over.

[Determinant]

Now turban-wearing Sikhs are a category people I have worked with.

[Andrew F]

Frances, I knew many frumpy profs. I don't think there was a different standard at all. Most of the profs that seemed to dress more formally were those that had to fulfill administrative functions, and those were mostly men for the time I was there. It did run the gamut though from jeans and a fleece pullover to suit and tie (though these were sessionals that came from industry). Most were in the khaki and oxford shirt range.

Waterloo's mathematics faculty seemed more research focused, for what it's worth. I don't think the faculty leadership was too fussed about what professors wore, and most of the students are pretty frumpy, too. I do miss it. It was the least shallow/concerned with appearance group culture I've ever been a part of.

[Andrew F]

I don't think I answered your question, Frances. Female profs usually seemed better put together on average, though not necessarily more formal. Not any outright sloppy dressers come to mind, whereas I had a male prof who wore the same shirt for every lecture (3x per week) and had chalk dust all over his pants. I also think the female profs I had were more competent on average. I only had one instructor, and that was a sessional for a bogus box-ticking course.

[Patrick]

Tangentially related from the American Mathematical Society: Debunking Myths about Gender and Mathematics Performance

I'm contemplating a future where the women are better dressed than men AND good at math. I seem to remember my wife having a science fiction novel along those lines. The men mostly lived as savages. Occasionally the women would take one of the more promising specimens for breeding and perhaps hose it down and train it for ... ummm ... entertainment purposes.

[Frances Woolley]

I've just received a clarification from some folks at the Palm Springs Art Museum.

Wal-mart would be the sartorial opposite of a Palm Springs Art Museum tie, not Target.

[Andrew F]

Patrick: I wonder if PISA scores are good predictors for brilliant mathematics theoreticians. I doubt gender differences exist for PISA-level math (especially with a school system that is increasingly failing boys--I'd expect boys to score lower than girls). My experience with math is that it seems it is less likely for brilliant mathematicians to be women than men. Perhaps all the brilliant women are drawn to careers other than mathematics for cultural or other reasons. I just have a hard time believing that the massive gender gaps in mathematics, engineering, and physics is due to sexism or purely cultural factors.

[Frances Woolley]

Andrew F - So there's what's sometimes called the "Larry Summers" hypothesis - that the means of the male and female math ability distributions are more or less the same, but the variance of the male distribution is greater, so there are more men at the very top and very bottom of the distribution (the guys at the bottom of the distribution end up e.g. in prison, on the streets, living with disabilities, etc). It looks as if the paper Patrick links to attempts to address the variance issue (though I haven't looked at it closely), but I agree with you that PISA tests are somewhat limited measures.

Often gender differences in biological traits e.g. height are much less than gender differences in socially constructed traits e.g. hair length. And the biggest gender gaps emerge when culture reinforces biology, biology reinforces culture, and the cycle continues...

[Patrick]

Andrew F: It's not the right tail that I worry about, it's the mere mortals in the middle. And I don't have a hard time believing the difference is mostly socially constructed. In any case, benching half the species in the pursuits that have big impacts on our health and wealth is an awful big opportunity cost to just wave away with a fatalistic sigh. We ought to at least try to do better.

[Frances Woolley]

What Patrick said.

[Jacques René Giguère]

There is something about learning paterns. When I was in classical college,we were taught Latin ad Classical Greek as if they were a deductive systems. We were even told that learning dead languages would improve our logical abilities. Nonsense as languages are not deductive systems. Yet it somhow seemed to suit our minds.
A few years ago, I read a paper ( can't recall neither author nor title) about asking male and female mathematicians to describe their thought processes. Male responded in the usually expected hypothethico-deductive framework. But a lot of female saw their problems as a narrative : logarithm meet square root, exponent interfere, cosinus weeps and then true meeting of hearts between the now happily reunited logarithm and root. ( I condense somewhat). But the theorems were proved by a different thought proces but as decisively and as fast.
So the problems maybe the one two of my sisters-in-law ( both endocrinology Ph.Ds)diagnosed: what womwn in scince need is good daycare during their studies...

[Frances Woolley]

Jacques Rene - "what womwn in scince need is good daycare during their studies... "

I wonder if there are differences in the % of women in PhD programs, or with children in PhD programs, in Quebec v. ROC due to Quebec's $7 child care program. I know it has been found to have an effect on labour force participation, but I don't know if anyone has looked at doctoral studies...

[Jacques René Giguère]

I don't know. It is maybe too early to gauge the effects. And given that doctoral students come from a social segment ( upper-middle technical class) that already has a high participation rates, the marginal influence may be low. OTOH, given the small numbers to begin with, and the high productivity of these graduates, a small improvement in number would still be worth pursuing.

[Chris Auld]

Good lord, the paper Patrick cites is just awful.

First, it's another in a long, long string of papers which VERIFY rather than debunk the so-called "Summers hypothesis." It clearly shows that the variance of boys' math scores is greater than the variance of girls' math scores in the vast majority of countries. Look at Table 2: is that evidence "debunking" the variance hypothesis?

Of course, Summers didn't the result up, he was just reiterating well-known facts. He got in trouble with ideologues for suggesting that *part* of the empirical regularity *might* be due to innate factors.

Second, the paper limits attention the strawman claim that the *primary* reason boys exhibit higher variance is genetic. It then makes claims about how we can identify genetic versus culturally- induced variation which are simply wrong and suggest neither author (both of whom are writing far, far outside their fields) has read anything on the subject.

The fact that the variance ratio is unstable across countries and across time does NOT imply that there is not a strong genetic component to math abilities.

As a counterexample, obesity rates have varied more than the variance ratio across time and across countries, but it is very well-established that body weight has a strong genetic component.

[Frances Woolley]

Chris - sounds like this calls for a serious debunking on chrisauld.com!

[Patrick]

Chris : Interesting. I confess I really only skimmed the paper. I'd have expected better from the AMS!

[Andrew F]

"In any case, benching half the species in the pursuits that have big impacts on our health and wealth is an awful big opportunity cost to just wave away with a fatalistic sigh. We ought to at least try to do better."

I don't think I advocated benching a gender. What is to be done? Quotas? That might end up benching many talented people because of their gender, too. I'm also unsure whether the same concern is expressed for gender imbalances in other areas, such as bricklaying, nursing, medical doctors (some medical schools are topping 70% female). When I was at Waterloo, encouraging and mentoring girls/women to pursue a career in mathematics was a pretty significant (and worthwhile endeavor, to the large extent there are cultural factors the discourage that choice). Yet apparently the gender ratio was actually worsening in many programs. One also wonders whether it is social welfare maximizing to push women into careers they might not find fulfilling, just to round out gender ratios. It's a tough nut to crack. A good place to start is to not crucify the Larry Summers of the world.

[Patrick]

Andrew F: Do you have a point? I didn't actually say anything about Summers.

[Frances Woolley]

Patrick, Andrew F - Larry Summers was my contribution to the debate, actually. I think both you guys have good points - Patrick, you're rightly pointing out that we don't just have to accept the status quo, Andrew, you're rightly pointing out that it can be hard to change it.