« Latin American monopolies | Main | The roots of structural unemployment »
Wages, height, and gender
A loyal MR reader asks:
We know women make less than men. We know shorter people make less than taller people. How much of the first is explained by the second?
Not much. Short men have less self-esteem (recall that male height at the time of high school predicts earnings better than adult male height), but short women, when they are young, feel no worse about themselves than do tall women. Next?
#28 in a series of 50.
Posted by Tyler Cowen on March 17, 2007 at 06:26 AM in Data Source | Permalink
Comments
The wage gap between men and women can largely be explained by a combination of factors such as: women tend to pursue fields that provide less money and more psychological pay, women are more likely to leave the work force to raise children, men are more willing to change jobs (giving up job security for pay) and a whole host of items that point to one thing. Women make less money because they value money less than men do (and ostensibly value other things like family, society, security and personal connections more).
Posted by: vc at Mar 17, 2007 8:30:51 AM
Anne Case and Christina Paxton of Princeton found that taller people, on average, scored better on cognitive tasks than people of a smaller stature. CEO's, politicians, heads of many organizations, all seem to have above average height.
As an aside, I am currently reading an interesting biography of St. Thomas Aquinas, who by the standards of the time was a giant.
Posted by: adrian at Mar 17, 2007 8:31:05 AM
Are you saying that women coming out of high-school all have low self-esteem and that hinders their money-making potential throughout their lifetimes?
Posted by: Xmas at Mar 17, 2007 9:28:16 AM
In the case of Aquinas being tall was a factor of his aristocratic background. Aristocrats in the 13th century WERE better than peasants - because they were better nourished all their lives they were taller, stronger, and smarter.
Posted by: Michael Tinkler at Mar 17, 2007 9:37:14 AM
Michael Tinker - It was his equally well nourished peers who gave him the nickname 'dumb ox,' so he was tall even by elite standards. Better nutrition results in increased height, for sure, Robert Fogel's 'Escape from hunger and premature death' shows in autistic detail how height has been rising over the last century with better health&nutrition. But at some point the increases stop, and so to reach the upper bands of the height distribution nutrition is not enough, you need the genes.
Posted by: adrian at Mar 17, 2007 11:06:41 AM
We're missing the point of the original question. To me the question says, if short people make less than tall people then it's natural that women make less than men, because they're shorter.
Posted by: vitalstatistix at Mar 17, 2007 12:06:52 PM
We're missing the point of the original question. To me the question says, if short people make less than tall people then it's natural that women make less than men, because they're shorter.
Yes, but Professor Cowen answered that. It's a matter of correlation versus causation.
He argues that the causation is that people with lower self-esteem make less money. Men who are shorter have lower self-esteem, but women who are short do not, since being short for a woman is considered attractive. Therefore, men who are short make less money than men who are tall, but women who are short do not make less money than women who are tall. However, since men are half the population, there is an overall correlation between height and income.
Under this argument, it is far from natural to conclude that the reason women make less is because they're shorter. Under this analysis, it does not follow at all, though it may follow from other reasons (such as personal choices, etc.)
Posted by: John Thacker at Mar 17, 2007 12:41:48 PM
I did not find Tyler's response convincing. The question was not about self-esteem, it was about pay. What should be examined is whether men of a certain height earn more than women of that same height. I would expect that to be the case.
Posted by: TGGP at Mar 17, 2007 4:47:29 PM
Tyler missed the important point. Self-esteem might be equal for women, but natural feeling of deserving to lead is not naturally felt by women, mostly because they are shorter. This submissiveness means lower workplace productivity.
Posted by: Dan at Mar 17, 2007 6:04:58 PM
"Women make less money because they value money less than men do (and ostensibly value other things like family, society, security and personal connections more)."
I think its actually the reverse. Men work harder not because they value money more, but because women value the money, and men want to attract mates and need paycheques to do so.
Posted by: Dan at Mar 17, 2007 8:55:02 PM
Not much. Short men have less self-esteem (recall that male height at the time of high school predicts earnings better than adult male height), but short women, when they are young, feel no worse about themselves than do tall women. Next?
Your link under 'not much' contradicts the theory implied in your very next sentence. You just sort of brushed off an interesting question. First of all there is an intelligence gap in America between men and women, both at the mean and, more significantly, at the tails of the IQ distribution. How much of the earnings gap can be attributed to this?
Second does this have anything to do with the height difference? Do men and women matched for height have more similar IQs?
Larger brains are more intelligent (PDF), so it is conceivable that this has some relationship to both the height differences and the sex differences.
Posted by: Jason Malloy at Mar 18, 2007 12:07:02 AM
Height and IQ are slightly correlated. When you control for IQ half or more of the income difference explained by height tends to disappear.
Posted by: Tino at Mar 18, 2007 2:07:36 AM
My understanding is that height is not correlated with income among women. If that is true, height is unlikely to explain any difference between men and women.
Posted by: Douglas Knight at Mar 18, 2007 12:32:00 PM
"Short men have less self-esteem" -
Indeed.
The famously petite Napoleon Bonaparte was noted for his lack of self-esteem.
Posted by: Martin at Mar 18, 2007 12:45:24 PM
I've never understood the height and IQ thing. If the study shows it to be true than maybe it is, but it contradicts all my life's evidence. I am very short for a white man and have a very high IQ. I went to what was at the time the best magnet school in the country as measured by test results and competitions, and the white males were blatantly, almost hilariously shorter than average. The american born asians also seemed to be shorter than the ABA average although I'm less certain of that. Whenever I went to math competitions it seemed like everywhere there were short white guys. Maybe it's a combination of poor perception by me, and some key exceptions to the correlation.
As for self-esteem, it's obviously tough for a short guy, but I do alright... going to a nerd high school probably helped a lot. Being blessed with a high IQ helps too! I'm going to test out the CEO theory also, or at least CFO... I'll probably get there by 35. I've never personally noticed discrimination against me at work. My business mentor of sorts is a very intelligent 5'7 bald Italian guy who just cashed out of his CEO position (although he cheated and built his own company).
Posted by: Bill at Mar 18, 2007 3:19:13 PM
In my opinion, shorter men do have a lower self esteem than do taller men. However, I do not think that is a factor in the difference in wages earned. From a woman's standpoint, I don't think height is a factor either. I think a short woman has just as much self esteem as a taller woman. Men are just a dominant figure in our society and they typically make more money than women do. That is just the way it is. I don't think it has nothing to do with the height difference between genders.
Posted by: Darren at Mar 18, 2007 3:37:48 PM
Here is wiki on 'heightism':
"A survey of Fortune 500 CEO height in 2005 revealed that they were on average 6 feet tall, which is 3 inches taller than the average American man. Fully 30% of these CEOs were 6 foot 2 inches tall or more; in comparison only 3.9% of the overall United States population is of this height. Equally significantly, similar surveys have uncovered that less than 3% of CEOs were below 5′7″ in height, and that 90% of CEOs are of above average height."
"Of the 43 U.S. Presidents, only five have been more than an inch below average height. Moreover, of the 54 US presidential elections only 13 have been won by the shorter candidate, "
Bill - When it comes to height there are other nutritional factors to consider. I would guestimate on the basis of IQ data that a good proportion of the 'short white guys' at those math comps were Ashkenazi Jews. In the Ashkenazi case nutrition may be a factor in a shorter average height, especially a possible low calcium intake due to a lactose intolerance rate of 60-80%. Asians, for example, are 95-100% lactose intolerant, and this is one reason why Japanese people still average 10cm shorter than Europeans despite an equal abundance of food. I would imagine that, on average, high IQ people within those ethnic groups would be tall group standards, although I have no evidence for that.
Posted by: adrian at Mar 18, 2007 3:54:01 PM
Yes, ashkenazi jews are a large part of it. I am not one, but a few of my short smart friends are. So were jews part of the studies correlating height and IQ? I'm still skeptical.
Also when you're talking about CEO there's a lot more leadership involved in both perception and reality, as compared to straight intellect. There's a huge obvious bias there for height, and I'd say that's the key factor for CEOs and US presidents. I wonder if the percentages are the same for CFOs and CIOs, I would guess they are less stark.
Fortunately, I don't have much natural affinity for leadership.
Posted by: Bill at Mar 18, 2007 5:25:28 PM
My understanding is that height is not correlated with income among women. If that is true, height is unlikely to explain any difference between men and women
It isn't true. Women differ in height by income group. Looking only at siblings from the same family, each inch of height is worth a 3.5–5.5 percent increase in women’s wages (PDF).
Posted by: Jason Malloy at Mar 18, 2007 6:59:17 PM
I believe it is all relevant. Times are changing but I believe steryotypes or generalizations are still relative because factors such a self confidence/image is/are projected in interviews and the workplace, thus possible affecting ones position/job and in turn ones salary.
Posted by: pk at Mar 18, 2007 9:02:10 PM
Leadership has always been correlated with physical size. But there was usually a short (Ashkenazi Jew?) guy behind the big lump. And it was the short guy with brains who pulled the strings and made good money too.
Hitler, Mussolini and Stalin were all (comparatively) little guys. Does this mean that in order for a short guy to make it in a leadership position, he has to be evil? Of course, none of them had Jewish advisors.
Posted by: Robert Speirs at Mar 19, 2007 10:06:18 AM
In todays business field these sentences makes sense. Overall as an economy the way one (any gender) carries themselves, is a direct effect of how far they will go. Image is key in success because first impressions last and stick with employers. So if a person goes into an interview with any doubt about their knowledge and or physical apperance this will be immediatly obvious to an employer. Going back to the height if someone interviewing for a position is shorter than the average size for a male than they could possibly feel inferior to anyone else throughout the interview, lowering their self confidence and making this obvious to the employer. This could especially be present if the employer seems to be a bigger man and the potential is smaller. All these taken into effect should not effect your future in deciding to pursue a goal because no matter what the limitations you can pursue any career that your qualified for.
Posted by: meagan h at Mar 19, 2007 4:29:13 PM
First of all there is an intelligence gap in America between men and women, both at the mean and, more significantly, at the tails of the IQ distribution. How much of the earnings gap can be attributed to this?
GAAAAAAH.
1. You don't even have to go past the abstract of the linked article to refute the "intelligence gap" claim, Jason Malloy.
"Males have only a marginal advantage in mean levels of g (less than 7% of a standard deviation) from the ASVAB and AFQT, but substantially greater variance."
In my world, a "difference" of less than 7 percent of a standard deviation between two samples isn't a difference.
I think it is back to the drawing board. How about selection effects, namely the disturbing tendencies for (a) poor black males to die young and violently or to be incarcerated and so not part of the workforce; (b) poor females to bear children at an early age and to drop out of the educational system prematurely, leading to dramatically reduced lifetime earnings potential. These are just two of a variety of selection mechanisms that undoubtedly affect the gender distribution of income quite independent of height.
Posted by: brianS at Mar 19, 2007 8:15:26 PM
There was a comment from before, and I have to disagree, women want the money more than the men do. I mean think about this, in a some of relationships women control the money. But I pretty much think its the truth for both parts.
Posted by: liz at Mar 19, 2007 9:21:07 PM
You don't even have to go past the abstract of the linked article to refute the "intelligence gap" claim
No, actually, the abstract, and the paper, report exactly what I said: a difference at the mean, larger differences at the tails.
There are a number of other datasets with the same finding.
Posted by: Jason Malloy at Mar 20, 2007 12:37:55 AM
the abstract, and the paper, report exactly what I said: a difference at the mean, larger differences at the tails.
well, I really don't want to be a jerk about this, so I apologize in advance, because I'm going to pursue this one step further. And I thank Jason Malloy for providing the link to the study in question. But my original comment stands upon further review of the paper. This study provides no persuasive evidence that there is a substantively significant intelligence (or generalized knowledge) difference between males and females.
the paper reports "no siginficant differences" on mathematics knowledge, that the women scored significantly higher on "the Word knowledge, Paragraph
comprehension, Numerical operations, and Coding
speed subtests of the ASVAB" and that the men scored significantly higher on "Science, Arithmetic, Auto and shop information, Mechanical comprehension, and Electronics information" on the ASVAB. I will leave what they mean by "significant differences" alone for a paragraph or two. Suffice it to say that the authors note that "Effect sizes range from small to large."
A bit of magic then comes in trying to pull a single "intelligence" factor out of these various tests of knowledge.
the authors extracted estimates of a general intelligence factor ("g") from confirmatory factor analyses of both the full battery of subtests and a subset of these tests (the AFQT battery of subtests). Let's ignore any general questions we might have about the appropriateness of estimating this "unobserved" thing ("g") for the purposes of this discussion and take the results at face value.
For the ASVAB model, the average "g" score was 8.66 (std dev. 3.2) for males, 8.45 (std dev. 2.8) for females. The "effect size" of the gender difference as assessed by the Cohen's d statistic was 0.068.
for the AFQT model, the average male score was estimated to be 15.08 with a std. dev. of 6.7, the average female score estimated to be 14.66 with a std. dev. of 6.0. They report a Cohen's d of 0.064.
I'm no expert in these particular test statistics, but these are awfully small effect sizes. I am not persuaded that these factor score differences say anything substantively significant.
The sample sizes are quite large in the study (1292 sibling pairs), so I suppose it is not shocking to find that the mean male and mean female scores on the subtests were frequently "significantly different". But what, exactly, does that mean?
The reported p-values in Table presumably are derived from treating the paired differences as random draws from a zero-mean normal under a null of no gender difference. So we are presented with "strong" evidence that the null hypothesis (that the mean difference is exactly zero) can be rejected with high confidence. This, to my mind, is not a very persuasive use of p-values.
As the study notes and you imply, "The present data suggest that a stronger difference in mental ability variance between the sexes could give rise to excess males at higher levels, balanced by an excess at lower levels." [my emphasis added].
I suppose that if this supposition were true and, further, selection mechanisms tended to remove more low-scoring males than females from the set of income earners, then we could observe that working men earn more than working women and that the difference was attributable to a gender difference in "intelligence" in favor of working men.
Posted by: brianS at Mar 20, 2007 4:15:06 AM
gay porno medical fetish pictures http://uk.geocities.com/gay791gay/gay-porno-medical-fetish-pictures.htm
[url=http://uk.geocities.com/pre682bi/free-gay-incest-rape.htm]free gay incest rape[/url]
<a href="http://uk.geocities.com/models183men/bondage-teens-boys.htm">bondage teens boys</a>
Posted by: vhnucwszbp at Mar 20, 2007 6:55:55 AM
[url=http://uk.geocities.com/models183men/boy-teens-in-leather.htm]boy teens in leather[/url]
Posted by: eygskuvlvn at Mar 20, 2007 6:57:09 AM
order poker bot
Posted by: uhpxaewzyv at Mar 20, 2007 6:57:24 AM
So if high school height affects lifetime earnings, is it height when you enter high school or leave high school? I am a male. I entered high school at 5' 1'' and 98 pounds and left high school 4 years later (yes I graduated on time and went to a top public college) at 6' 1'' 165 pounds. If entering high school, then I should be at the bottom. If at end of high school then I should be above average. I will reveal whether I am poor or above average after the answer.
Posted by: Murphy at Mar 20, 2007 3:16:07 PM
大家好，我是臺灣人，從臺灣一個人搬家來到美國，環境很陌生，感覺很孤單。以前在臺灣幾家知名的徵信社工作過，我是一個優秀的徵信工作者，希望早點找到適合自己的工作。希望通過貴站，認識更多的朋友。
Posted by: 謝文豪 at Apr 1, 2008 11:29:16 PM
