« The Economics of the Revolving Door--Posner | Main | Is the Growing Gender Gap in College Enrollments a Cause for Concern? BECKER »



Feed You can follow this conversation by subscribing to the comment feed for this post.

Gary Bourgeault (managersrealm.com)

This is a huge problem for those of us who hire in the workplace. If the university is set up to favor certain demographic groups for politically correct reasons, it is difficult to trust even the records presented at interviews as legitimate.

Everyone should push for meritocracy as the best way to truly judge who has it and who doens't.

Richard Mason

Posner: There is no reason why the difference in variance should result in higher average male earnings

Since the distribution of income is skewed to the right (mean income higher than median income), I believe that one would expect a subpopulation with higher variance in ability to have a higher average income. Conversely a subpopulation with very low variance in ability would have average income close to the median income of the whole population, below the mean income of the whole population.

I have no idea if this is the actual reason (or a significant reason) for higher male incomes, but in principle it seems that it could be.

Lawrence Indyk, University of Kansas School of Law

I am in agreement with Richard Mason. The calculation of average income is the multiplication of the performance distribution by the average income for that level of performance. One typically breaks it down more accurately by using subgroup income-vs-performance because of the tendency for males and females (and different races) of similar talent to nevertheless concentrate in different careers with different levels of compensation. But even if this is ignored, variance alone could still have an effect.

Depending on the shape of the income vs performance graph, it is certainly mathematically possible for differences in variance (even if we assume an equal mean and momentarily ignore factors like pregnancy, and child care) to create significantly different average incomes.

If it turns out that the numbers of people at the extremes remains small and that the slope of the income-vs-performance curve is small and stable, then Judge Posner is undoubtedly correct. However, there is good reason to believe neither is the case.

Variance in mental capacity is so large, especially is males, that the total population more than 20 IQ points away from average (and over 40 points away from each other) remains still quite significant. There is a floor to income at the lower extreme with the minimum wage but the percent of hourly workers collecting that wage remains very small according to the BLS (though the percent of workers earning within just a few dollars an hour of the minimum wage is large).

Yet there is no real ceiling at the top end of the income chart. Most income distribution studies find a near exponential (Or Lorentzian) increase at the top end. So the question becomes, "Does the income of these individuals (not just superstars and CEOs, but lets say the top quarter) rise more quickly than their population diminishes. If the answer is yes, than variance alone, ceteris paribus, can explain significant differences in average compensation.

And looking again at the income distribution graphs in most developed coutries (and especially the United States), the huge portion of total income that is earned by the top few percent gives reason to believe that the compensation rise for talent is indeed this large, allowing the top decile or quarter of performers to skew the average compensation of their entire subgroup out of proportion to their numbers.

If this is true, the key to high "average incomes" is for a population to have a high enough variance to permit a substantial population to exist where incomes start to grow exponentially.

Furthermore, given the nature of bell-curves and the distribution of IQ, two groups with similar variances (such as males of different races) can have wildly different average incomes if their means differ even by a few IQ points. This is because the shift of these few points will have radical effects in the percent of individuals produced in the range of talent associated with rapidly increasing incomes.

Arun Khanna

Posner: Hitler may have had something to do with this as well, as it is plausible that the most intelligent European Jews saw the handwriting on the wall earliest and left Europe in the 1930s before it was too late.
Arun: Jews that stayed behind in Nazi Germany simply believed in the rule of law in a western country. It is not clear why people who place trust in the rule of law and Judges who enforce the rule of law should have lower I.Q.
On a personal note, the example of Jews not anticipating concentration camps of 1940s has left a bitter taste in my mouth. However, it is a legitimate example given the topic of debate.


The admission of women students into universities is in my view not a meaningful indicator of society's attitutde towards women. The reason is simply that students are in a university for a relatively short time and interact with the university's administrators only in a peripheral manner. The hiring of professors is a whole different matter since professors usually stay with the university for a long time and the administrators feel the presence of professors much more strongly than students. A woman student becoming pregnant hardly affects the university's bottomline, whereas the same cannot be said about a pregnant professor. University administrators may be more inclined to admid women students than male students in order to project an image that they do not discriminate - an overzealous effort may produce an overshoot which may explain the temporary greater rate of admission of women students. The real test is whether universities hire and promote women professors (and for that matter other minorities) at the same rate as they admit women students. On that measure the universities still fail miserably.


Jews that stayed behind in Nazi Germany simply believed in the rule of law in a western country.Or they hadn't won enough awards for being smart for countries like the USA to accept them as immigrants. As anyone in a developing country can tell you, being allowed to move to another country is not as simple as deciding that it's in your best interest to leave the country you're in.


From Prof. Mankiw's blog:



Stephen Pinker's theory is downright ridiculous. It just proves what a little speculation can do, but from here to make a serious point out of it, it is a long way. In this way, we could start killing everybody with IQ lower than a certain threshold. according to Pinker, in 200 years, we will have a smarter population. Those who are not pretty enough too - why not? I doubt that evolution is that fast and takes such a precise route.

going back to the female discrimination issue: somehow, i feel there are women who have made political careers out of this issue and are unwiling to drop it. They have no other issue to replace it with. It is an easy speach to learn. Etc. Not that women ar enot discriminated, but if they learn to whine and to reply on the society to fix the problems in their life, where will we be?


(1) current job is much less labour intensive, therefore, even footage for both men and women (lets not count on small finger type faster on the computer).
(2) women could not reliablely count on the support of men for regeneration, at least in north american, therefore, provide additional motivation to creat stable environment=college is one "easier" way.
(3) In addition, the women worklife is limited due to child care. She must earn the stable environment=reasonable pay job afer college, in order to provide 1-2 years child support (prefer with company benefit. Just look at the job search behaviour, most of the women prefer large company with good benefit)
(4) Men, on the other hand, got more choice for different kind of job selection, including "join the army" (30K signing up fee listed at some small town with airforce, navy ads of MBA rings a bell. specifically, targetted at high school). Computer gaming skill has been associated with "potential stock trading" for example. Poker game might be another. Significant change of male behaviour also may play a role: the engineering student could not repair a bike.. even refuse to get hands dirty. Lack of hands on experiences, might diminish the traditional advantage of male risk taking and "trail and error" advance explore/learning skills.
(5) As for higher earning bell curve for men, the beer drinking and golfing skill which women is lacking might play a key role in board room as well as corp. ladder. (short career life span of women might also contribute to the difference).
(6) The technology is moving in the fast lane (18-20 month per generation of new chips for example). Any "missing in action" more than one year (such as child bearing) may cause significant set back in pay as well.
(7) Overall, it is current environment (collective factors of influence) resulted the occurance as indicated by both professors. Try to adjust to a "balance" might take more than discussion and policy, but a major shift of "environemnt"-social behaviour...

--silly 1.4 cents.



You are making a large and entirely unjustified leap from the positive to the normative in responding to Pinker.



my point was that self-selection in the evolutionary sense does not happen in a few hundreds of years, but in a much longer interval. Besides, i realy doubt that evolutionary, you can condition in such a short span of time the birth of smarter kids. It is simply a speculation.

If one could assure such a fast change in the genetical characteristics of a group than one could decide to change our genetic map at will. Hence, horrible acts could be justified, sicne it would improve the prospects of humans as species. This is of course, bollocks.

If Pinker would base his argument on a learning process of the Jews (and transmission of these skills from generation to generation) rather than change in their genetical characteristics, maybe it would seem more plausible. So far, it seems just speculation. And why wouldn't one use more speculation in the argument then?


The number of Jewish refugees from Nazi-era Europe is small compared to the number of E. European Jewish immigrants to the US from the 1880's to 1924. It may even be small too compared to the number of survivors who immigrated after 1945. Has this group of people really influenced the average IQ of American Jews?


Posner: In the "ancestral environment"--the term that anthropologists use to describe the prehistoric period in which human beings reached approximately their current biological state--women who were "steady" would have tended to have the maximum number of children, while natural selection might favor variance in male abilities because variance would produce some outstanding men who would tend to reproduce more than other men (including the "steadies") in the polygamous conditions of prehistoric society.

Does anyone have any idea what he means by "steady"?

Jason Ruspini

Even if ability distributions are symmetrical and income-vs-performance is linear, the higher variance group might still earn more than the higher mean group depending on the risk appetites of employers.
You can quickly fire screw-ups, but upsides are much less limited, which could in itself produce a variance premium.


(Hitler may have had something to do with this as well, as it is plausible that the most intelligent European Jews saw the handwriting on the wall earliest and left Europe in the 1930s before it was too late.).

Godwin's Law has been invoked, and Posner is declared the loser of the argument.


Posner attempts to argue that discrimanatory practices in the middle ages have lead to high IQ's in today's Jews. I would have thought that any group would have "regressed to the mean" since the middle ages.

scott cunningham

Regression - by removing the lower end of the distribution, the mean is effectively pushed higher up the distribution. Regressing to the mean notwithstanding, mean IQ would still rise under Pinker's conjecture.


Well ... the mean would undoubtedly be higher for a few generations given Posner's hypothesis, but not six centuries later.


Personally, however, I would like to see a few of the top colleges abolish all preferences unrelated to academic merit--no athletic scholarships, no affirmative action, no favoritism for the children of professors or of major donors, and no legacy admissions. That would be a useful experiment in the benefits and perhaps costs of meritocracy. It would have the incidental effect of giving us a better idea of the extent of real differences across race and gender in academic capability.

As a very intelligent person (someone who others like to say "is good at taking standardized tests") I fully support this. But I am concerned that intrinsic merit would then be conflated with class. Access to books, tutors, hard-working parents, role models, a stable neighborhood, a healthy diet, etc. is not equal. Instead of pure merit, we would get a reflection of merit plus class. You would still see underrepresentation of the poor and to the extent there is an overlap of poverty with race, there would be a underrepresentation of certain racial groups. It would be unfortunate to conclude that certain racial groups were lacking in intrinsic merit because fewer of their members were born into the ideal conditions to create future scholars. It likewise would be wrong to conclude that everyone who scores well on a standardized test does so because of innate intelligence: some people have tutors and can afford prep courses. I sincerely doubt that Richard Posner is suggesting we promote equality of resources so that academia becomes a free market; it is not a free market, I suspect he would say, if half the competitors are subsidized by the government.

Half Sigma

"Since the distribution of income is skewed to the right (mean income higher than median income), I believe that one would expect a subpopulation with higher variance in ability to have a higher average income."

Absolutley correct! Take, for example, Howard Stern, he probably makes more money than all other radio talk show hosts combined, signficantly boosting the mean salary of the profession (but not the median salary).

Many professions are winner-takes-all.

Bernard Yomtov

Hitler may have had something to do with this as well, as it is plausible that the most intelligent European Jews saw the handwriting on the wall earliest and left Europe in the 1930s before it was too late.

This is offensive and ill-informed.

As David points out, the number of Jewish immigrants from Europe during the Nazi era was small relative to the number already here.

Also, there were any number of reasons why Jews might not have emigrated. Not least among these of course were the severe limits on immigration to the US or other safe countries (Some refugees were turned back of course - look up the SS St. Louis, or Switzerland's behavior) and the difficulty of emigrating from a country at war or occupied - Poland, Russia, France, etc.

The suggestion that the Jews could have simply hopped on a ship to the US or England is absurd. May I suggest, Judge Posner, that you retract it.


Interesting that no one has looked at the issues of "Pop-culture" and target markets. There appears to be five issues which might expalin the phenomenon:

1. gangs
2. guns
3. drugs
4. professional athletics
5. entertainment industries

Ask one question, "Who is the target market?" Certainly not women when it comes to lifestyle and professional pursuits. Why get an education when I can get rich selling drugs or becoming a rapper or athlete? The issue now becomes, how can academia compete when it vies against such for the imagination of youth. Just one thing, "MOTIVATION". But How? Academics can certainly be boring.

John Cash

You can call it evolution. You can call it bias. You can plug it into an equation and make it economics. I just call it mindset. It is the way things have been forever.

Since the earliest hunters poked their heads out of the cave, the male has been the risk taker and the female the home maker. The male caveman fought for prey all day while the female lay in the cave nursing child. If the male got killed by dinosaur, then the female moved on to another male for support. That's just how it is and always has been. It is the mindset.

Now, you can take the girl, dress her all up, and send her to college, law school or med school. That is fine. She'll do well. But, she cannot fight the mindset. Very few will become practicing lawyers or doctors. And very few will blow the charts off the gmat because they have given in to the mindset.

Achievement lag in black males? When you say "lag" I can only assume you mean as compared to whites. Hm, let me scratch my head a moment.

I've been to Congo. Black males behave socially the same there as they do in the United States. They are not remotely monogamous and father many children from many different females, staying around to raise none of them. Africa, overall, has some of the highest crime statistics in the world and specifically as related to black males and drug crimes.

So I guess their "lag" is not in relation to their native land, but rather as to us crusty whites.

But wait you say. Asians, Indians (from India) and other races are unbelievably successfull in the United States. Therefore, the plight of the black male is not due to racism.

Before I do this let me say that I am a 38 year old white male living in the heart of Illinois. I listen to Rush Limbaugh and I voted for George Bush and his daddy. I despise Jesse Jackson and Al Sharpton, but adore Harold Ford and JC Watts.

If racism is, as Jesse and Al would have us believe, the intentional deprivation of rights and benefits due to one's race, then it would appear the civil rights movemant worked and we have very few people in our society that behave as such.

But that is not racism. Racism is hidden. Most do not know they are racists. Most probably proud themselves of how far from being a racisist they are.

Throughout all of my driving years in Illinois I have never gotten a ticket for a broken headlight or taillight, even though I've driven by cops many times with such. Yet, you look at the driving abstract of almost any young black male and you'll see each one is loaded with such citations. Why is that?

And, I have never, I mean never, had a cop ask to search the trunk of my car or cabin of my vehicle. There must be some kind of mistake re this. Because as I drive the interestates of Illinois I often see troopers digging through black persons' trunks or rifling through their luggage right along side road. Makes quite a show.

Now wait a minute Jesse and Al, these cops arn't full of hate and out to intentionally get blacks. Their affliction is worse. It is bias, or the mindset. And they don't even know they have the disease.

Years and years of repeated lifestyle and social association have caused these cops to unconsciously think a jackpot will be found if they pull a young black male over and shake him down. They probably have many black friends and coworkers. They'd adamantly deny being racists because they are unaware of their bias.

You say, then, we need sensitivity training for the cops. But it is in all us. Doctors, lawyers, judges, school teachers, coal miners, and the list goes on and on. All unaware of the disease.

Gotcha, you say. Why aren't we that way with Asians, Indians and other races? See, the race card falls flat on its face!

Let me see. 200 hundred years ago did we chase down chinese in the bamboo forest of China? Catch them with nets? Chain them in boats and sail them to the U.S.? Enslave them and their descendants in hard labor for decade after decade? Then resent their freedom that neerly split our country in two? And spend the next 100 years subliminally uttering hate to our children and grandchidren that yellow people are lazy, poor, stupid and prone to criminal activity?

To compare a black male who is a 7th or 8th generation American, to males of other races that volutarily came here in the last half century or so is erroneous and pointless.

The mindset has set its crosshairs on the hispanic male next. The boarder states have seen decades of strife. The subliminal programing began there long ago. As more and more venture out across the country it grows like a cancer. That cop pulling over that van driven by an hispanic under the pretense of a broken taillight doesn't hate Mexicans. His social bias has conditioned him to "know" the van is full of illegal aliens, whom he loves. And so the mindset lives on.

Until we can openly talk about our biases, how we came about them, or even if we have any, racism will continue to fester. Can a white cop openly answer a question to his superior that he believes a black male driving through a white subdivision has likely committed a crime or is carrying drugs? Can a white judge admit he scews his application of sentencing guidelines more favorably for whites, especially white females, than he does for blacks?

Until they can we will accomplish nothing and the mindset wins.


Than we must expect inverse evolution now. Smarter and richer people will have higher value of their time and lower motivation to have children. If so polygamy must be promoted to solve the issue. :)

Bruce Hay

I wonder whether human evolution really selects for intelligence. We are herd animals, and people who think for themselves, question social conventions, etc. are, one would imagine, likely to be ostracized and, all else being equal, have a harder time finding mates. Of course this is offset to some degree, because intelligence has obvious survival value, and anyway really intelligent people might be able to fake conformity in order to be accepted by the crowd. Still, the fundamental point remains that in herd societies, a capcity for independent thinking is an ambiguous blessing.

Just look at the hosts of this blog. How long would they have lasted in a hunter-gatherer society, the way they run off at the mouth? We're lucky they were born now, and not in the ancestral environment.

The comments to this entry are closed.

Become a Fan

May 2014

Sun Mon Tue Wed Thu Fri Sat
        1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30 31