Like many other people, I cringe whenever I’m routinely mistaken for another brown person. When I attended graduate school in the midwest, people repeatedly confused me with another desi woman in my class, Sheila, who looked absolutely nothing like me — the obvious difference being that Sheila was much lighter-skinned than I was. At least, to me, it was obvious. To other white people, it was apparently not. Never mind that Sheila was from India and a had a bourgeois Mumbai accent, whereas I was from southern California and talked like a valley girl. As far as other people at school were concerned, we were interchangeable.
And so, because of repeated instances like that, I had figured that brown folks were just more sensitive to skin tone differences than white people were. But apparently, that’s not the case. When I was at work yesterday, I caught this news blurb:
Light-skinned immigrants in the United States make more money on average than those with darker complexions, and the chief reason appears to be discrimination, a researcher says. Joni Hersch, a law and economics professor at Vanderbilt University, looked at a government survey of 2,084 legal immigrants to the United States from around the world and found that those with the lightest skin earned an average of 8 percent to 15 percent more than similar immigrants with much darker skin. “On average, being one shade lighter has about the same effect as having an additional year of education,” Hersch said.
While I don’t think her findings are entirely improbable, I’m curious as to what she defines as “one shade” of skin tone.
What’s also interesting that it seems the researcher compared skin tones within immigrant groups:
Hersch took into consideration other factors that could affect wages, such as English-language proficiency, education, occupation, race or country of origin, and found that skin tone still seemed to make a difference in earnings. That means that if two similar immigrants from Bangladesh, for example, came to the United States at the same time, with the same occupation and ability to speak English, the lighter-skinned immigrant would make more money on average.
So what does this mean? Contrary to my earlier beliefs, it seems that other people are able to distinguish between darker and lighter-skinned browns.
I also wonder why Hersch used immigrants as her subjects, and not second, third, or fourth generation Americans. Would the results be the same? I don’t know. I wasn’t able to find this particular study of Hersch’s online, and I usually prefer to link to original sources rather than the newspaper, but still, this is food for thought.
I am/was much lighter skinned, so people actually thought I was either hispanic or an eskimo(darn those MN winters with the fur rimmed hooded coats). Now in the workplace, people can’t guess my nationality either, and my boss actually thought I would check the “caucasion” box in the application forms!
the best (to my mind) critique i’ve seen of the within nation of origin skin color differentials i’ve seen is that the author is finding the effect of those who end up in blue collar jobs where people are outdoors vs. those who end up in white color jobs where they are indoors. e.g., this summer i was working in an office coding away, but two summers ago i was doing ecological field work, and the differential in skin tone was multiple “shades” for sure. also, there might be a “tall teenager” or “good looking” effect insofar as in cultures like bangladesh (and most cultures world wide) light skin is considered praiseworthy so these people have more self confidence and are willing to demand more and not settle for less.
read the original paper here. free PDF.
thanks, razib. for whatever reason I couldn’t find it earlier.
also, she did try to take outdoor work into account: Because outdoor work may cause skin color to darken, if jobs involving outdoor work are also lower paying, any estimated relation showing darker skin color and lower wages may be an artifact of the relation between outdoor work and lower wages. Note, however, that outdoor work may warrant higher wages as a compensating differential, which would serve to attenuate any negative effect of darker skin color on wages. There is no information on whether a respondentÂ’s job requires extensive work outdoors, so a direct measure of outdoor sun exposure is not available. As an alternative, I assign an indicator variable for occupations that are likely to require outdoor work.11 8 Since actual start date is not reported, I calculate tenure assuming interview date and start date both fall on the 15th day of the month. 9 Using data on immigrants to Israel, Friedberg (2000) finds the return to experience differs by whether the experience was acquired in the originating country or in Israel, with a much higher return to experience acquired in Israel. Because the sample studied here is comprised of new lawful immigrants, to a large extent their work experience acquired in the U.S. will be reflected in years of tenure. 10 Actual year of birth is missing for 739 observations but is reported as one of 10 categories spanning five years for 90 percent of the observations. I assign the midpoint for each of these ranges and calculate approximate age from the difference between interview year of 2003 and birth year.
also, let me add that i think the framing of this as discrimination in favor a white ideal (see some of the popular pieces) might be correct in the proximate sense, but is pretty eurocentric in the ultimate analysis. many dense agricultural societies seem to associate lighter skin with higher status or positive attributes* (we know about south asia as brown folk, but see china and japan). this predates european hegemony (one can even find it among the incas, who raided the amazonian lowlands to procure lighter skinned concubines). the validity of this generalization about skin tone is strengthened by the fact that other european traits, light eyes and light hair, were not valorized at all prior to european hegemony (brown haired japanese women used to dye it during the tokugawa period).
Here is a hypothesis for why lighter skinned immigrants might earn more. If the sample set is restricted to South Asians, then there might be historic social bias included in the skin-tone differential. For instance, lighter skinned historically “upper-castes” are the ones more privileged in most South Asian societies (even though caste, per se, does not have a religious denotation in Islam or Christianity, there are Punjabi vs. Sindhi biases even in Pakistan. Also, the upper castes in Bangladesh tend to be lighter in color than the working classes.) Immigrants from these classes are likelier to have access to have better education and softer skills in navigating the workforce. Thus, it is likelier that they make more money.
In fact, that makes more sense when you consider why the speaker did not include 2nd pr 3rd genners since I think this historical class bias is more likely to be attenuated among this group.
Further, to conclude causality from correlation is somewhat shortsighted in this case.
Good stuff, Naina, Razib. Wow! I’m heading out to buy some Fair and Handsome. Seems like the benefits are more than just “Womens are attracted“. Also got to remember the tip about “bathing twice a day….especially during Summers“.
Here is a hypothesis for why lighter skinned immigrants might earn more. If the sample set is restricted to South Asians, then there might be historic social bias included in the skin-tone differential. For instance, lighter skinned historically “upper-castes” are the ones more privileged in most South Asian societies
1) good that you are offering up a hypothesis
2) but i bet in the US case you would be wrong.
a) we know that pakistani americans are lower SES than indian americans, and the median skin tone for the former is assuredly lighter than the latter (the bangladeshis are even lower SES than pakistanis and on average rather dark, but their numbers are close to trivial).
b) of the various indian ethnicites in the USofA the skin color model should predict that punjabis are the most affluent and south indians the least, with gujaratis in the middle. i don’t know of any data for incomes and wealth across the various ethnic groups, but my own impression (i know fewer brownz than most here, so i am willing to be corrected) that the south indians are probably the most professional of the groups.
minor retraction: rather than the hypothesis being “wrong,” i guess it is better to say that the data is probably going to be pretty muddled with a host of confounding factors.
SES = Socio-Economic Status?
yes.
Perfect Razib. Muddled is most accurate. My main point was to illustrate that to infer causality would be wrong in this case.
See, she uses the word effect. That is what I object to.
So, running to get that facial bleach or Fair and Lovely cream might not yield better results, aside from making one more self-confident. It won’t help if you had a year less of school.
Razib has SES on the brain.
ses stuff for the brown american national groups (even includes sri lankans!): per capita income of indian americans $26,415 Sri Lankan – $26,330; Pakistanis – $17,685; Bangladeshi – $13,532
Oh the color !!! I think Dr. Kiran C. Patel is NOT light skinned. Well he gave $30 million grant to University of South Florida to have Patel Center for Global Solutions. Chalk that one up for exception, as Dr. Patel is extremely successful !!!
Razib,
Forget shades of brown for a minute. This is the most interesting observation from your link
50% of the H1-B work visas each year are taken by Asian Indians.
26,000 Asian Indians enter the US every year on student visas.
Foreign born populations: Asian Indians – 73%; Bangladeshi – 83%; Sri Lankan – 79%; Pakistani – 74%.
The question is: During comments, how come almost everyone (keyword: almost, not all) claim to be ABD (only 20-30 % should be ABDs in the comment section or maybe little higher in free for all internet), make clueless critiques about South Asia/ Indian subcontinent/ where ever they immigrated from – SA, Guyana or woes and pains of visa system and loops of immigration or challenges faced by foreign students.
I wonder, I wonder sometimes. It must be manifest destiny. I am Captain America. Memory is always short.
Razib Bhai, you may now get back to the topic.
That reasoning does not explain why desi-americans, who are the darkest-skinned immigrants to America after african immigrants, do better than average financially. What does explain that is the fact that desi immigrants to America tend to be the educated class of their countries, and in post-Civil Rights America merit trumps racial discrimination to a significant extent.
But look at the global picture: the dark African continent and the dark Indian sub-continent are the two most impoverished regions on the planet. The next darkest region, ASEAN/south-east Asia, is the next most impoverished region on the planet. The striking exception in ASEAN being Singapore, which has a heavy majority of light-skinned chinese. Chinese confucian Singapore is followed distantly by muslim Malaysia, which has a very large chinese minority of around 30%. Is it racial discrimination that explains this phenomenon?
Too late. Already signed up for a free sample, entered the ‘Hey Dude, win-an-iPod’ contest and downloaded the Kareena poster from the Fair and Handsome site. All the analysis is falling on deaf ears. I’ve got my ears covered – La-la-la-la-la.
Desis do better because of education in science and math related field. I am sure the hypothesis would hold if you norm the study for education.
Except for the fact that there is no real correlation between light skin and caste. Punjabi Jatt Sudras are significantly lighter than the south indian brahmins who seem to dominate the IT immigrants to America. Anyone doubt that brahmins are among the very darkest people on earth?
“Contrary to my earlier beliefs, it seems that other people are able to distinguish between darker and lighter-skinned browns. “
-You do realize that this may be a subconcious activity right?
1) Correlation is not causality. It might be: earn more –> read more junk —> spend more on fairness creams.
2) I think the more relevant construct is physical attractiveness (not controlled for in the study). In the past researchers have found that physical attractiveness (i) raises that person’s own expectations of success, (ii) positively influences teacher’s expectations about the student, (iii) helps in everyday social exchange, and (iv) generates favorable social perceptions.
After all Krishna was dark-skinned. And he had hordes of girlfriends (may be the flute made a difference). 🙂
I think Hersch has a misspecified model – he confounded skin color with attractiveness. There are many other features that give a person advantage.
3) About physical height: Malcolm Gladwell reports in Blink that most Fortune 500 CEOs of AmericaÂ’s corporate world are tall:
“In the US population, about 14.5 per cent of all men are six feet or over. Among CEOs of Fortune 500 companies, that number is 58 per cent.”
“Of the tens of millions of American men below 5’6″, a grand total of ten–in my sample–have reached the level of CEO, which says that being short is probably as much, or more, of a handicap to corporate success as being a woman or an African-American.”
Conclusion Hersch’s findings are not new. Both the findings – physical attractiveness (methinks that made the difference, not skin color) and physical height – are not something particular about immigrant populations. But the rhetorical strategy – i.e. positioning the paper in terms of discrimination – definitely helps in getting some media attention.
This happened to me just the other day…
Kush Tandon,
Do you understand how certain venues–online or not– may attract a higher %-age of a certain demographic, and that SM is not sampling uniformly from the brown population? Go to Sulekha, for instance, and you will find a different (and less articulate) crowd. Also, nothing wrong with critiquing where you came from, especially when, like in the case of Nirad C. or Sasthi Brata, the critique is not “clueless”. Even if it is, it can still be laudable if it manages to ruffle some bourgeois-nationalist (or for ABDs, bourgeois-ethnicist) feathers.
Well, if you hold everything else constant (like the researcher seems to do), you’re likely to find some effect in the one independent variable. What’s interesting is that it was such a clear and significant effect, though I agree that its unwise to conclude causation.
Holding so many elements constant and starting with a pool of just over 2,000 feels unsubstantial. I’m curious how many data points she had for each set of constants. If there were only 20 people in her groups of Bangladeshis with similar stats, I think its hard to really say anything — even if the results are similar in 200 sets of 20.
But I haven’t read the whole study, so lets give Hersch the benefit of the doubt and say her study captures a unique ‘effect’ of skin color (while one can assume the lighter subjects are generally better looking, this isn’t suggested the study — we should assume ‘beauty’ to yet another constant — as per: “Bipasha is so hot… eventhough she’s kinda dark”)
I think Naina, Razib, et al had it right in that most cultures glorify lighter skin tones, usually because it indicates social status. But the skin tone test is quickly replaced where other indicators of status are available, e.g. caste, wealth, education, perhaps even beauty.
I wouldn’t call this discrimination. I feel like the study assumes the conclusion. First, I don’t think the N is large enough, especially since there are so many controls. Just skimmed the study: there are 81 Indians, 74 “East Asian, South Asian, Pacific” and the word Bangladeshi doesn’t appear anywhere in the PDF. Then, even if she is correct about skin tone, discrimination requires something more than ‘innate preference.’
Ok, she mentions in the conclusion that attractiveness might be reason for the result… and that light-skinned black folk might be smarter because of white ancestry —- WHOAKAY! (to be fair, she says the ‘vast weight’ of genetic studies doesn’t support such a position)
Anyway, I don’t think the study is substantial enough to warrant much thought. Maybe it would have been more intersting if she compared across ethnicities? If she found that Indians were generally preferred to Jamaicans, but lighter Jamaicans were preferred to darker Indians… my brows would furrow.
Do you understand how certain venues–online or not– may attract a higher %-age of a certain demographic, and that SM is not sampling uniformly from the brown population?
Yes, that is why I said “maybe higher” but not significantly higher. Sure, the sampling is definitely not uniform but 15,000 hits/ day has to closely resemble uniform.
A lot 1.5 or even 2.0 genners have seen their parents go through visa uncertainties/ green card process/ life of foreign graduate student, up close etc.
Anyway, Bauji, I am a village idiot from Sakuti Tanda (a small village in UP), and I need to work @ MIT AI labs, because I need some AI.
Also, this was just a tangential observation. Let’s get back to the topic. Ta ta.
“Except for the fact that there is no real correlation between light skin and caste. Punjabi Jatt Sudras are significantly lighter than the south indian brahmins who seem to dominate the IT immigrants to America. Anyone doubt that brahmins are among the very darkest people on earth?”
Not really. If you were to fit tamil brahmins into South Asia’s colour spectrum, most would fall somewhere from the middle-to-light end. Most are around the same colour as a regular guju.
And I know plenty of Punjabis who are darker than the average south indian brahmin. Of course, out of all the brahmins, the tamil ones (and malayalee) are probably the darkest, but definately not among the very darkest people on earth.
this study seems pretty stupid. doesn’t it? i beleive it to an extent. i mean you guys’ve seen those studies that talk of attractive folks making more money than others by a significan percent.
i wonder if accent matter as well (the closest your accent to an american). maybe thats why…
But then, it’s only a study of racism, so the subjects other attributes weren’t measured so much of course. I mean the lack of controls is the same as not being able to tell one brown person from another. Better to stay mysterious, IMHO.
In certain IT jobs, accent do not matter. Your progress is based upon your ability to code. I guess skin color would also matter very little if you were a physician. However, desis that I have seen in upper management in fortune 500 companies generally tend to have caucasion characteristics. This study seems to be very discouraging to me cuz, I am goodlooking y’all…and still pretty much a ‘failure.’
I am very puzzled by this study. Indians – the darkest of all model minorities – are the highest earners in the US according to the latest census, beating jewish people and the japanese. Also, a lot of these Indians who are fueling the IT industry both in the US and in India stem from South India (Bangalore, etc.).
The fact that this study failed to mention this fact, leads me to believe that study is flawed concerning ‘skin colour’ and discrimination.
Although I believe that attractiveness does play a heavy role in success, but that goes beyond skin colour. It includes your facial traits, the way you dress, the way you talk, your self-confidence.
The average Indian has caucasoid features.
Based on personal experience, I would not be surprised if affluence increased in North Indian communities, and it would have nothing to do with skin color. Rather, it would have more to do with the fact that North Indian communities (Gujjus and Punjabis, in particular) are more likely to entrepeneurial than south indian communities, even if the data on professional qualifications might be similar.
There are a lot of factors contributing to success. In math and science related field, your knowledge is more easily quatifiable. Either you have it or your don’t people are paid accordingly. This is where the color privelege probably does not apply, and you see Indians in higher salaried position. In fields where math and science knowledge isn’t a premium (entertainment, Waste industry, Upper Management not associated with number crunching), skin color probably has a premium. Cisco, in earlier post I said caucasion characteristics, I meant ‘white’ or West European characteristics. By that I meant lighter skin colored, generally a sharper high bridge nose etc etc.
Just to second Naina’s comments about folks being confusing indians irrespective of accent, skin color, length of time in USA, national origin, etc. The so-called study runs counter to my (considerable) experience in the US as a brown guy and the innumerably many times I have had to discreetly question non-brownz to figure out which desi-dude they are actually referring to (“ravi said he would be done today? Huh, you know I think that was vivek in the meeting with us”).
Using the admittedly biased sample of the 2nd-tier leadership (and some first-tier as well) of the computer software sector in the US, one comes up with 20-30% desi and these are all seriously brown folks predominantly but not exclusively from the indian south. Maybe the technology sector is an outlier (no point having a light-skinned dude if he cant do the math!)?
Just a few samples:
Thomas Kurian @ Oracle http://www.oracle.com/corporate/pressroom/html/tkurian.html
Desh Deshpande http://web.mit.edu/newsoffice/2002/deshpande.html
Guess fair is more aesthetically pleasing, just as a straight nose and high cheekbones would be. Incidentally Naina was fawning over a fair Kashmiri Bhatt guy (with chiselled features) in a previous posting, who was quite clearly a deviation from the average desi-male look. So how can we question discrimination on skin tone if its not motivated but is instead instinctive and a very human bent for the aesthetically refined
I am curious about your statement that Indians who are in the Fortune 500 have ‘caucasian features’, as if they looked less like the average Indian. Who are you thinking really? Is Mittal too white for you? 🙂
If you are “failure” as you mention, it is not because you are dark-skinned or Indian and there is nothing you can do about it. You can and will, but the first step to success is not finding silly excuses to justify your shortcomings.
First of all, South Indian brahmins are certainly much fairer (and sharper featured) ON AVERAGE than other South Indians. Yes, exceptions exist and some are very dark (and some do not have sharp features). But ON AVERAGE you can clearly tell the difference. Many are more fair than the average non-brahmin northern Indian.
As for Jatts, it’s not really accurate to call them ‘sudras’. This was a label applied in the past by brahmins, khatris, etc. but it doesn’t make any sense on the ground…Jatts are actually THE dominant group in Punjab, socially, economically, and politically. They also own most of the land. If anything, they tend to look down on brahmins, khatris, etc. (who are mostly involved in service-oriented occupations).
I’m sorry if my comments are feeding the troll, I don’t want this thread to go off on a tangent.
And that is why tanning booths and the sunscreen lotion market are going bankrupt… 😉
Not according to this article:
At #41: I find it hard to take an article seriously that says things like the below:
By the way, pure Indian buffalo milk rocks! Too bad we can’t get it here. That’s a taste of India I do miss.
As for Jatts, it’s not really accurate to call them ‘sudras’. This was a label applied in the past by brahmins, khatris, etc. but it doesn’t make any sense on the ground…Jatts are actually THE dominant group in Punjab, socially, economically, and politically
Doesn’t matter. Most of the dominant agrarian castes throughout India have been classified as so-called shudras; as have many of the Hindu dynasties, e.g. the Mauryas and Marathas. You are confusing ritual status with economic (and now political) power. Of course if you called a group of Jatz ‘shudras’ you would probably get jumped. That’s one of the ironies of the caste system; the “ground reality” is indeed very different from textual presentations, including traditional ones.
Thank GOD they don’t have bufallo milk here, 8 years of drinking that stuff wasn’t all that pleasant. Thank God (or Cadbury) for bournvita.
Mittal was CEO of his own company and did not have to depend on others to get promoted. As for my ‘tounge in cheek’ comment about being a failure…heck one man’s failure is another man’s success. I am not a governor of a state, or a CEO of a fortune 500, but earn a very comfortable 6 figure income in a midwest state. That would be success to many. Failure or success is very subjective. Type A personalities tend to be achievers in society’s eyes but are never satisfied. Type B personalities do not achieve as much, but lead much more happier lives. I am a realist and have a good idea of general thinking patterns in desi and western society. I happen to percieve that there is a color privelege in both societies. You may see things differently, but that’s your reality…true or not.
Fuck a caste system bullshit
🙂
Jatts outwarriored the so-called kashatriyias and the know enough about making dollars to be on par with any so-called vaisya, and we can get our scripture on like anybody else
Preach on sahej. Just to take this ridiculousness to the next level, I would like to mention that jatts have the sharpest noses in the subcontinent according to a study done on noses in colonial india :). The nasal index turned out to be 68.8 and the nose was defined as leptorrhine. 🙂
http://www.geocities.com/Athens/Ithaca/1335/Anthro/sud_afr.html