Friday, March 25, 2011

Skin color and the discrimination paradigm

Lighter skin color correlates with higher earnings among new immigrants to the U.S. This correlation holds up even if one controls for English language proficiency, education, occupation before migrating to the United States, and family background. It even holds up among immigrants of the same ethnicity, race, and country of origin. For some social scientists, only one possible explanation remains: discrimination.

The modern American credo blames underachievement by minorities on the majority, specifically through discrimination. The causal relationship may be direct, i.e., some people are consciously assigned to lower-paying jobs simply because of their skin color. Or it may be indirect, as Gunnar Myrdal argued almost seventy years ago in An American Dilemma:

The mechanism that operates here is the “principle of cumulation,” also commonly called the “vicious circle.”

[…] White prejudice and discrimination keep the Negro low in standards of living, health, education, manners and morals. This, in its turn, gives support to white prejudice. White prejudice and Negro standards thus mutually “cause” each other.

[…] If, for example, we assume that for some reason white prejudice could be decreased and discrimination mitigated, this is likely to cause a rise in Negro standards, which may decrease white prejudice still a little more, which would again allow Negro standards to rise, and so on through mutual interaction. If, instead, discrimination should become intensified, we should see the vicious circle spiraling downward.
(Myrdal, 1962, pp. 75-76)

Originally, this ‘American dilemma’ was supposed to explain underachievement by African Americans. Today, it is increasingly extended to immigrants, even newly arrived ones. This is the premise of a recent study by Joni Hersch (2008):

[…] most new legal immigrants to the United States have darker skin color than white U.S. natives and are on average shorter. This article considers whether skin color and height affect economic outcomes among new legal immigrants to the United States. (Hersch, 2008, p. 346)

To this end, Hersch consulted the New Immigrant Survey 2003. The survey involved interviewing a nationally representative sample of immigrants as soon as possible after they got permanent resident status. Among other things, the immigrant’s skin color was measured with a color scale: a series of increasingly darker hands numbered from one to ten. Interviewers were given the following instructions:

As you know, human beings display a wide variety of physical attributes. One of these is skin color. Unfortunately discrimination on the basis of skin color continues to be a reality in American life. Substantial evidence suggests that lighter skinned people fare better in a variety of social and economic settings than those with darker skins. In order to detect such discrimination, it is important that the NIS include a measure of skin color. We therefore ask interviewers to use the Scale of Skin Color Darkness as a guide to rate the skin color of each respondent on a scale of 0 to 10, where 0 is the lightest possible skin color (such as that of an albino) and 10 is the darkest possible skin color” (Hersch, 2008, p. 361)

After analysing the data, Hersch found the following:

[…] I find strong evidence that darker skin color is associated with lower wages, taking into account a wide array of demographic and productivity-related characteristics such as English language proficiency, education, occupation before migrating to the United States, and family background, as well as ethnicity, race, and country of origin, which are themselves highly correlated with skin color. Immigrants with the lightest skin color earn on average 17% more than comparable immigrants with the darkest skin color. On average, moving from the 10th percentile to the 90th percentile of the distribution of skin color within ethnic or racial groups would reduce wages by about 7%-9%. These magnitudes are roughly similar to the black-white disparity and Hispanic-non-Hispanic disparity reported in Altonji and Blank. (Hersch, 2008, p. 346)

She concluded: “The results indicate that any such discrimination is not merely ethnic or racially based nor due to country of birth. […] Skin color is not merely capturing the effects of ethnicity, race, or country of birth but also has an independent effect on wages.”

Why does skin color have greater explanatory power than race, ethnicity, or country of origin? One reason is that these other variables are often problematic. ‘Blacks’ include anyone with some ancestry from sub-Saharan Africa. ‘Hispanics’ encompass a wide range of populations. At one end, there are Amerindian groups in Guatemala and southern Mexico who use Spanish as a second language. At the other, there are Italian, German, and Jewish communities in Argentina who still maintain strong links with Europe.

Similar objections could be raised for country of origin. Hersch’s data show that 18.2% of immigrants from the United Kingdom self-identify as ‘non-white’ (Hersch, 2008, p. 355). There are similarly large non-European communities in France, Belgium, the Netherlands, Norway, and Sweden. In an increasingly globalized world, what does it mean to be ‘French’, ‘Dutch’, or ‘Norwegian’?

Interestingly, 1.7% of ‘Polish’ immigrants in the U.S. are actually Hispanic (Hersch, 2008, p. 355). This should be no surprise. Eastern European universities, with their low tuition, attract large number of Third World students. Many remain after graduation before eventually moving on to a third country, like the United States.

So to explain economic success among new immigrants, we should look at their skin color rather than their ‘race,’ ‘ethnicity,’ or ‘country of origin’. Does it follow, then, that color prejudice is the ultimate cause? Or do some immigrants fare poorly because they are less able? Hersch (2008, p. 378) reports some evidence for the second explanation:

The possible connection between skin color and ability has been examined using the 1982 GSS, which includes a 10-item vocabulary test as well as a measure of skin color for a sample of about 500 African Americans. Using these data, Lynn (2002) reports a positive correlation between lighter skin color and higher test scores.

But this kind of explanation strikes her as being faulty:

However, using the same data, Hill (2002) demonstrates that controlling for education and family background eliminates the relation between skin color and test scores. […] Available evidence in the scientific literature does not support a link between skin color and intelligence. In addition, the correlation between skin color and ancestry varies considerably, with low correlations in many populations of mixed ancestry (Parra, Kittles, and Shriver 2004). In the absence of genetic evidence or a high correlation between skin color and ancestry, it seems unlikely that inclusion of test scores as a measure of ability would greatly alter the skin color effects found in this article.

This is more than a bit disingenuous. First, saying that some immigrants are less able than others doesn’t necessarily imply a genetic explanation, even if they share the same ethnic background and supposedly differ ‘only’ in skin color. A Mayan immigrant from Guatemala may be as Hispanic as a Jewish immigrant from Buenos Aires, but the two individuals bring very different attitudinal and behavioral tools for survival in a modern market economy.

Second, a genetic explanation doesn’t imply a direct causal link between skin color and intelligence. Skin color is just an indicator of one’s ancestral gene pool, which may statistically differ from other gene pools for any number of heritable traits, including those that influence intelligence. It’s also disingenuous to claim that skin color is weakly correlated with ancestry by referencing a study done on a population that has been intermixed over many generations. If a British immigrant has brown skin, the chances are very good that he or she is not of European origin.

Finally, it’s disingenuous to treat discrimination as a default explanation that should be accepted unless there is strong evidence to the contrary. This is in contrast to evidence for discrimination, which is usually inferred. For instance, Hersch (2008, p. 368) notes:

An analysis shows that those with darker skin color are less likely to be self-employed, controlling for the same predetermined characteristics used in the regression equations reported below. While it might be tempting to interpret this finding as suggestive that immigrants with darker skin color avoid self-employment out of concern about customer discrimination, there is no information in the data regarding customer contact, and there is limited empirical evidence of customer discrimination in the literature generally, so it seems wisest to avoid making this leap.

The above interpretation doesn’t strike me as “tempting.” Self-employment is typically a refuge for those who have been excluded from the job market. I remember many young people who turned to self-employment in the 1980s and early 1990s because they had no alternative. All of the stable jobs were reserved for older workers with seniority. But for Hersch, self-employment shows you’ve had more freedom to choose your livelihood.

Another point deserves mention here. When I think of self-employed immigrants, I readily think of Sikhs, Lebanese, Chinese, Armenians, and other ‘middleman’ minorities. Yes, they’re lighter-skinned than most immigrants. But do they choose self-employment over wage labor because they face little or no discrimination? That’s not my impression.


The discrimination paradigm was first applied to a minority that had suffered centuries of unequal, prejudicial treatment. Bit by bit, and somewhat unthinkingly, it has been extended to other groups: Amerindians and Mexican Americans and, now, newly arrived immigrants. Discrimination is no longer supposed to exercise its crippling effects through generations of enslavement and Jim Crow. After all, we see these effects in people who’ve just arrived.

Well, why not look just a bit further ‘upstream’? These different patterns of economic behavior also exist in the immigrants’ home countries. They exist worldwide. They have nothing to do with whatever prejudices White Americans might have.

But why do these patterns exist? Why should lighter skin improve one’s ability to perform in a modern market economy? Surely that proposition is just as absurd.

Actually, no. Please hear me out before you start shouting. Success in a modern market economy depends on the ability to plan ahead, and that ability has been favored outside the tropics—in areas where people tend to be lighter-skinned. The non-tropical zone has a yearly cycle that makes planning necessary. This cycle began to impact human survival as early hunter-gatherers spread into the temperate and arctic zones. It became even more important when their descendents took up agriculture. The fall harvest had to last until early summer; otherwise you and your family would starve (Frost, 2010; Frost, 2011).

The same yearly cycle also made men necessary for family survival, particularly in winter when hunting used to be the only other means of sustenance. This is in contrast to the tropics, where year-round agriculture enabled women to provide for themselves and their children with minimal male assistance. In such circumstances, men spent more time and energy seeking additional mates (Frost, 2008).

Are these patterns of behavior genetic? Or are they learned from one’s cultural environment? From the standpoint of natural selection, the question is unimportant. What works works, and what doesn’t doesn’t. But the question does matter to curious humans. To find the answer, the first step is to admit that the question is legitimate. Are we ready to take that step?


Frost, P. (2011). Religiosity and the origins of civilization, Evo and Proud, January 21

Frost, P. (2010). Out of North Eurasia, Evo and Proud, May 27

Frost, P. (2008). Sexual selection and human geographic variation, Special Issue: Proceedings of the 2nd Annual Meeting of the NorthEastern Evolutionary Psychology Society. Journal of Social, Evolutionary, and Cultural Psychology, 2(4),169-191.

Hersch, J. (2008). Profiling the new immigrant worker: the effects of skin color and height, Journal of Labor Economics, 26, 345-386.

Myrdal, G. (1962). An American Dilemma, New York: Harper & Row.


BeyondAnon said...

And then mobile bandits came on the scene, leading to the rise of stationary bandits, because losing a small(ish) proportion of your income for some guarantee of protection was better than risking total loss, including, possibly, your life.

Harmonious Jim said...

It would be nice to know if the ability to plan ahead is also higher in other species among icebelt populations compared to sunbelt populations. My hunch is that in some cases it is. If so, it would lend support to your thesis.

Tod said...

"Discrimination is no longer supposed to exercise its crippling effects through generations of enslavement and Jim Crow. After all, we see these effects in people who’ve just arrived.".

A Mayan from Rios Montt's Guatemala is no stranger to discrimination.

Non white immigrants suffer less discrimination from whites in the US than they do from the ruling elite in their homeland.

Tod said...

I like Manning's ideas about oestrogenization. The Russian fox taming experiment produced lighter coats (Cochran and Harpending remark on this in their book).

The cave paintings from Grotte Chauvet are by all accounts incredibily good. By my way of thinking it's not obvious how such artistic ability relates to stable monogamous relationships, hunting, or foresight in provisioning. Maybe Europeans had been oestrogenized by 30,000 years ago ?

Anonymous said...

It became even more important when their descendents took up agriculture. The fall harvest had to last until early summer; otherwise you and your family would starve

This makes me wonder whether temperate agricultural cultures were selected for saving and long-term deferral of pleasure while tropical agriculturalists (i.e. South East Asia, not horticulturalists) were selected for a kind of ceaseless toil, with no breaks in work, but where planning was more day to day.

Anonymous said...

The arguments for skin color discrimination are not disingenuous, they are flat out dishonest.

They are utterly vitiated by a fundamental defect: they do not control for IQ.

This has been the problem with arguments concerning discrimination against blacks in the United States for half a century:

Is a black with an IQ of 100 (80, 120, 140, etc.) more or less likely to be academically/occupationally successful than a white WITH THE SAME IQ?

If outcomes are the same matched for IQ, then the United States does not suffer from skin color discrimination at all, but instead from discrimination against people of low intelligence – a category in which blacks are vastly overrepresented in comparison to whites.

In other words, progressives and social scientists have been deliberately coy on the issue of African/Caribbean/American black intelligence for almost a century, and it is time to state the obvious (currently forbidden by censorship laws):

Once blacks and whites matched for IQ were equally successful in the US, skin color discrimination had been eliminated.

Thereafter, the reason for black failure in the US must be attributed solely to absolutely devastating cognitive deficits – which have shown no improvement despite gargantuan government expenditures in a myriad of programs over more than half a century (involving everything from prenatal care, infant nutrition, self-esteem building, community empowerment, pilot schools, afterschool programs, personal computers for grade schoolers, and race quotas in hiring and education in the private and government sectors, the military, healthcare, education, etc. etc. etc.).

In other words:

Myrdal’s recommendations WERE followed in the United States, but have utterly failed to improve black cognitive ability/ performance.

His predictions were wrong. The reason is that human intelligence is grounded in fundamental mechanisms of genetic selection and evolution, not a contingent effect of short-term cultural developments.

Africans, mixed-race (b/wh) individuals, and those of Native American ancestry are on average significantly less intelligent than Europeans and Orientals, and are consequently less successful academically and occupationally.

Mystery solved.

Now the question is: how to break the code of silence and discuss these issues honestly and publicly at long last???

Kiwiguy said...

Hersh could look at Schneider & Jones' paper on immigrant earnings & IQ.

IQ in the Production Function: Evidence from Immigrant Earnings

Sister Y said...

It is precisely because skin color sometimes is a signal of something important (though not a hugely reliable one) that laws like Title VII are economically justifiable and necessary. They put everyone (forcibly) back in an initial-cooperation paradigm (a la tit for tat), allowing everyone to at least have the opportunity to be a cooperator, even though he may have outward indications of being a non-cooperator/planner/etc.

Peter Frost said...


When a bandit is in a long-term relationship of exploitation, he's less likely to kill or maim his "milk cow."

Harmonious Jim,

Perhaps. But non-human species don't have "culture," or at least not to the same degree as humans. They tend to adapt to winter environments by migrating or hibernating, not by building shelters and laying up stores.


A lot of people are ready. Unfortunately, they're waiting for someone else to make the first move.


Yes, many immigrants suffer color discrimination in their home countries. To varying degrees, this is so for Latin America, the Arab world, Ethiopia, and the Indian subcontinent.

But does this factor explain why they fare poorly as immigrants in the U.S.? If it were a factor, black Haitian immigrants should do better than black Jamaican immigrants (since economic and political power is disproportionately held by lighter-skinned Jamaicans). Yet we actually see the reverse.

In general, hunter-gatherers are excellent artists. But artistic excellence doesn't mean one has good planning abilities (my brother is an artist, and he definitely is not a planner).


Ceaseless toil? Not really, In general, tropical agriculturalists (in simple societies) work only to satisfy their own needs over the short term. Things changed with the advent of a landowning class that forced farmers to produce above and beyond their immediate needs.


I'll look at the paper. But Hersch acknowledges the relationship between IQ and earnings. She just says its due to factors like family background and education.

Sister Y,

I agree that skin color does matter to people, both consciously and subconsciously. But does color prejudice explain why darker-skinned immigrants do less well than lighter-skinned immigrants -- even when one controls for other factors? That's the question.

Kiwiguy said...


I'll look at the paper. But Hersch acknowledges the relationship between IQ and earnings. She just says its due to factors like family background and education.***

Right, this blogger 'Chuck' also offers an interesting analysis of Hersch's paper.

Colorism — So much for Franz Boas

Chuck said...


With regards to your reply to Kiwiguy:

"I'll look at the paper. But Hersch acknowledges the relationship between IQ and earnings. She just says its due to factors like family background and education."

The question is: does Hersh acknowledge the relationship between skin color and IQ.

In the paper, Hersh notes that:

"Available evidence in the scientific literature does not support a link between skin color and intelligence. In addition, the correlation between skin color and ancestry varies considerably."

This suggests that she doesn't. Which is why she deems that IQ isn't a worker skill that needs to be controlled for:

"Because inclusion of additional observables reduces the magnitude of the estimated skin color effect, it is worthwhile to consider what might be missing from the wage equations. Two possible omitted variables are attractiveness and some measure of ability as embodied in test scores."

As it is, both nationally and internationally, IQ correlates with color (for whatever reason). It is a factor, therefore, that needs to be controlled for -- and not dismissed.