Economic convergence between black immigrants and black natives

Alison Jane Rauh, a job candidate from the University of Chicago, has a new (job market) paper on this topic.  The abstract is full of information:

The number of black immigrants living in the US has increased 13-fold from 1970 to 2010, increasing their share of the black population from 1% to 10%. Black immigrants’ labor market outcomes surpass those of native blacks. This paper determines in how far the relative success of black immigrants is passed on to the second generation. While blacks of the second generation have equal or higher education and earnings levels than the first generation, the return on their unobservable characteristics is converging to that of native blacks. Race premia are put into a broader context by comparing them to Hispanics, Asians, and whites. Blacks are the only group that experiences a decrease in residual earnings when moving from the first to the second generation. Black immigrants do not only converge to native blacks across generations but also within a generation. For Asians and Hispanics, residual earnings decrease monotonically with age of immigration. For blacks, the residual earnings-age of immigration profile is upward sloping for those immigrating before the age of 15. Convergence across generations is mostly driven by low-educated second generation blacks that drop out the labor force in greater numbers than low-educated first generation immigrants do. Similarly, convergence within a generation is mostly driven by low-educated blacks who immigrate when they are young dropping out of the labor force in greater numbers than those who immigrate when they are older. A social interactions model with an assimilation parameter that varies by age of immigration helps explain this phenomenon. When making their labor force participation decision, immigrant men of all races, but not women, generally place more weight on the characteristics of natives the earlier they immigrate.

I take this to be a “peer effects are really really important” paper, namely that many of the virtues of immigrant culture are swallowed as the second generation assimilates.  I should note that the contents of this paper are interesting throughout, for instance: “Conditional and unconditional annual earnings of native black women are at 91% and 78% of white women, which points to a much more equal distribution than that of men (64% and 78%).”

Here is the abstract of another paper  (pdf) by Alison, entitled “Successful Black Immigrants Narrow Black-White Achievement Gaps”:

The number of black immigrants in the US quadrupled from 1980 to 2010, increasing their share of the black population from 4% to 10%. During that time period the black-white wage and employment gap widened substantially. This paper explores the extent native blacks differ from immigrant blacks. Additionally it determines in how far increased selective immigration masks an even greater deterioration in the economic condition of native blacks. In 2011, excluding black immigrants increases the white-black wage gap by 4% for men and 9% for women. It increases the employment gap by 13% and 19% for men and women respectively.

Here is the author’s home page.  I hope she gets a very good job.


*Posts mild sarcastic comment on what this implies for American immigration policy*
*Tells anecdotal story about black immigrant acquaintance from personal history who had immense success but whose children failed badly*
*Is Steve Sailer*

What do you think this implies for American immigration policy? Nothing? Really?

I'll bite. What does it imply for American immigration policy? I mean, I guess I can see a plausible case for "we should increase immigration from majority-black countries in the hopes that the improved labor-market outcomes from black immigrants gets more of a toehold in the wider black American experience." Is that what you're suggesting?

It implies a ton (namely that we should have less of it). I was just joking around because that's exactly how he'd respond.

It tells us a lot about how our pop culture promotes obviously destructive African American male tropes, which has the worst effect on American blacks, but also attracts other groups such as immigrant background males and dumb white people.

Another contributor is likely regression toward the mean: most African immigrants are legal immigrants, with many of them getting here through the higher education system, so the first generation represents an above-average selection from their home populations. Not surprisingly, their children are less above average than they are, just as the average NBA player's sons aren't as tall on average as he is.

I wonder how many, say the ones from West Indies etc. are visa overstayers? And if so, the assumption of having come through higher ed. is not a given. So also for asylees & family reunion categories.

I'm not very sure whether the above-average selection hypothesis holds or not. At least on IQ.

More likely it holds on motivation or tenacity perhaps.

You usually get regression toward the mean with chain migration: For example, Barack Obama Sr. gets into Harvard, so his brother Omar shows up in Massachusetts, followed by sister Zeituni who ends up in public housing.

Stevo - While blacks of the second generation have equal or higher education and earnings levels than the first generation, the return on their unobservable characteristics is converging to that of native blacks. Race premia are put into a broader context by comparing them to Hispanics, Asians, and whites. Blacks are the only group that experiences a decrease in residual earnings when moving from the first to the second generation.

Large regression to the mean on IQ isn't totally consist with increased education (it's not totally incompatible either, as the US does have a lot more educational resources than Africa or the West Indies or what have you).

But the question here is what residual earnings are exactly? They're obviously not total earnings because immigrant Blacks have more than native Blacks, and second generation immigrant Blacks, pace the above, have more than their parents.

"The amount of income that an individual has after all personal debts, including the mortgage, have been paid. This calculation is usually made on a monthly basis, after the monthly bills and debts are paid. Also, when a mortgage has been paid off in its entirety, the income that individual had been putting toward the mortgage becomes residual income."

Suggests that the "problem" may be more that second generation Black people tend to buy into the consumer society and become "big spenders", taking on debt to a greater extent than their parents do, while the increase in their educational capital and earnings, although absolutely present, is not enough to offset this.

The paper states that "Convergence across generations is mostly driven by low-educated second generation blacks that drop out the labor force in greater numbers than low-educated first generation immigrants do. Similarly, convergence within a generation is mostly driven by low-educated blacks who immigrate when they are young dropping out of the labor force in greater numbers than those who immigrate when they are older.". So low educated second generation blacks may not live frugally, work and accrue little debt to the same extent as first generation black migrants.

Factors as to why second generation Black people do this may be the greater integration of Blacks into American cultural life - they're expected to play the same consumer status games as Whites do, rather than being a living in their own little social world where it's acceptable to be a "typical coupon clipping Chinaman" (to quote Eddie Huang) - and perhaps secondarily also for the same reason that car salesmen target black men to pitch higher prices to (machismo, self esteem and pride aren't compatible with meek, frugal living over the long term).

I vaguely recall that General Colin Powell's bestselling autobiography was surprisingly lacking in tales of West Indian immigrant bourgeois frugality that I had come to expect from "Hey Mon with the Hedleys" from "In Living Color."

For example, Powell's extended family's main plan for acquiring the down payment for a house in New York was that when one of them had a dream about what number would be the jackpot in the numbers racket, they all plunged heavily ... and she was right! They hit the jackpot and used the money to buy a house.

Interestingly, Powell was intimidated in class terms by his wife's parents who were southern black upper middle class (e.g., Jim Crow era high school principals). I suspect that when Powell was secretary of state he knew exactly where national security adviser Condi Rice was coming from because Powell's wife was from the same class.

Hey Mon with the Hedleys:

"What do you think this implies for American immigration policy?"

Back in 2001, I wrote:

Conservative Establishment types assume that "assimilation" works automatically, sort of like "osmosis," to solve any minor cultural problems created by heavy immigration. They tend to have this 19th Century picture of assimilation in action: The huddled masses stumble off the docks in Boston, only to notice the Cabots and Lodges glowering at them for their uncouth behavior. Chastened, the newcomers resolve to begin the long struggle to become good Americans just like those august families.

Of course, these days, the descendants of those Puritans are working as program officers for foundations telling the newly-arrived immigrant how much better the culture back home in, say, Guatemala was than the culture created by their Boston Brahmin ancestors - who, as you will recall, were so criminally insensitive as to glower.

There is, however, one vibrant, self-confident culture in modern America: hip-hop. ... Hip-hop is doing what whites no longer have the will or means to do ...

The American promise in reverse, in that instead of immigrants traditionally arriving to avail themselves of greater opportunities, they arrive only to see their children be absorbed into a Borg-like culture of entertainment and entitlement.

For successful case studies of the children of black immigrants not falling prey to the allure of African American culture, consider top officials in the Obama Administration, such as the upbringing of Attorney General Eric Holder. His parents are from Barbados' mulatto middle class (like singer Rihanna) and they raised him in New York in a West Indian mulatto middle class bubble where he was carefully shielded from lower class African American and their culture.

Seems to have worked.

Valerie Jarrett was born in Iran where her M.D. father was conducting medical research and spoke Farsi as a child with her native servants.

Seems to have worked.

Of course, their's boss's upbringing was even more extreme: he was sequestered thousands of miles out in the Pacific Ocean away from any sizable black community. And even though Obama claimed to have transferred to Columbia U. to get in touch with Harlem, his most exhaustive biographer David Maraniss reports that he completely failed to make any black friends during his years in New York, where his most serious girlfriend was the daughter of the #2 man in the Australian CIA and the stepdaughter of the chief corporate counsel for Rio Tinto Mining (!). His main American friend was the son of U.S. diplomat, and his main social network was the "Pakistani mafia" of millionaire Marxist South Asians.

Seems to have worked.

I'd prefer the hip hop idiots to the destructive examples you cite.

When Keeping It Real Goes Wrong

I'm surprised to see a paper allowing for a natural demonstration of just how deep seated racism is in American society.

The interesting thing to compare would be the wages of such immigrants to Canada over the same time scale - and whether such effects are so clearly illustrated.

Ah, the Canada fallacy, so common and yet to stupid. Canada's black population has grown substantially in the past decade or so due to increased immigration. It now nears 3 percent. That is a massive difference to the United States, as are most differences between the two nations.

Here's new NYPD top cop William Bratton's 2006 interview with Toronto journalist Linda Frum on the color of crime in Toronto:

Hmm, so when they arrive people don't realize they are black and are not racist towards them, but then over time or as they have children, people get suspicious that maybe they are black and start racisming them more and more over time? Is that your theory?

How about this :

The parents have advantage due to their home country and almost certainly were somewhat smarter, savvier or wealthier than their peers as they managed to emigrate. But the systematic discrimination against black youth in our country manages to replicate its effects on the second generation regardless of whether they were children of native born citizens or immigrants.

It seems far more likely than the racist view (see below) that blacks have lower IQ and somehow first generation is benefiting from some sort of immigration bonus.

There are some ways to test your hypothesis.

1. The USA is huge. If there is systemic discrimination against blacks, it is unlikely to be identical everywhere. Therefore, in some states, the second generation should fare much better than in other states. (I do realize that there are some confounding factors there). On the other hand, if the phenomenon is homogeneous across the USA, it speaks against the systemic discrimination hypothesis - the cultures of, say, Maine, Oregon, Louisiana and Puerto Rico are just too different from one another to be identical when it comes to discrimination.

2. The discussion in itself makes a precarious presumption that blacks are one big group. In practice, black people can come from a variety of cultural, religious and ethnic backgrounds. Haitians and Somalis have culturally pretty much nothing in common. Again, does the effect hold across these groups? If not, it would again speak against the discrimination hypothesis, as the racist peers would be unlikely to distinguish between a Dominican and a Chadian.

There is a vaguely similar situation in the UK, where Pakistani immigrants are, on average, much poorer than the average Brit. However, at the same time, Sikh immigrants are the single most successful ethnic group in Britain.

Not sure I can totally buy into #1. The US is huge but we have a federal government that oversees it all. A major factor in this (according to the paper) is incarceration and that has been proven to overwhelmingly target black males. We have national level laws that target black males like crack having far harsher sentences than cocaine. Police departments across the US have been proven time, and time again to target young black males more than any other ethnicity or group. New York City that is generally regarded as being very liberal, has a long, well document history of targeting black men.

#2 is addressed in the paper. There's a concept called "fade to black", where actual nationality and cultural differences fade into just being regarded as "black", by employers, the criminal justice system, etc. One of the most interesting things they suggest is that the first generation may get a boost from having an accent and differentiating themselves from native blacks. The second generation not having the accent just blends in thus being better targets for prejudice.

I agree with Tyler. Read the article! It's fascinating. It's especially hard to discount with the number of "Asians have a culture of hard work, Blacks have a culture of laziness" going on in this comment thread. If you guys were hiring managers you would be making decisions that exactly mirrored the prejudices that create these outcomes!

Your last anecdote makes the cultural argument. I"d bet a good many of those non-Sikh Pakistanis are Punjabi, just like the Sikh immigrants, i.e. racially indistinct. (Same with the rest probably, too). So the culture that one assimilates into has a big effect.


If only the immigrants were not forced by racism to sell crack! Anyway, those are some bold assertions you are making, I would appreciate some (serious) links.

Also, if blacks with accents are discriminated against less or not at all, does that mean it's not racism after all?

1. Should be some variance though I would expect them to generally converge to the direction of the local black population.

2. Do 2nd and 3rd generation blacks make much of this distinction? Do the cultural differences extend to the metrics they're measuring in the study?

There's also the culture factor. There's plenty of evidence that suggests that African American culture has adverse effects on academic and financial success (though many commentators on this blog are likely to blame biology or genetics instead.) Possibly due to racism, children of black African immigrants are more likely to assimilate into the African American American subculture than any other.

People put the cart before the horse on Asian success as well. A culture that really, really, believes in education outperforms. Film at 11.

Well it is not "more likely than... that blacks have lower IQ" because that is a demonstrable fact and not something you can wish away.

Anyway, this is an interesting model of racism you are building where it acts slowly over time to make a person drop out of the labor force. What kind of mechanism are you positing? Exposure to racist media changes peoples' self-image of themselves, or something?

Your racism is really just proving my point Cliff. Wikipedia says it better than me " the term "black people" is not an indicator of skin color but of socially based racial classification".

So your fact would seem to indicate that society, through prejudice can somehow push down IQ scores.

Basically racists say "hey, race and IQ are all we have to work with, to make a racial connection. Give us a break!" They kinda miss the circularity of their pursuit.


You used the term first, not me. You really can't complain about it. Sure, it's not a genetic classification. But "blacks" tend to fall within a few genetic populations, such as East and West African. All these groups have lower than European/East Asian IQs. There's no doubt about that. If your only defense is blaming the scientists for racism then you are on very weak ground.

By the way, looks like you are a liar:

"The term "black people" is an everyday English-language phrase, often used in North America to refer to Americans and Canadians of Sub-Saharan African descent.[1][2] Outside North America, the term "black people", or close translations of it, is also used in other socially based systems of racial classification, or of ethnicity for persons who are perceived to be dark-skinned relative to other "racial" groups – or else who are defined as belonging to a "black" ethnicity in the country.

Different societies, such as Australia, Brazil, the United Kingdom, the United States, and South Africa apply differing criteria regarding who is classified as "black", and these have also varied over time. Often social variables, such as class and socio-economic status, affect classification, so that relatively dark-skinned people can be classified as white if they fulfill other social criteria of "whiteness," and relatively light-skinned people can be classified as black if they fulfill the social criteria for "blackness" in a particular setting.[3] As a result, in North America, for example, the term "black people" is not necessarily an indicator of skin color but of a socially based racial classification related to being African American, with a family history related to institutionalized slavery.[4][5] In other regions, such as Australia and Melanesia, the term "black" has been applied to, and used by, populations with a very different history."

Is it racism? Or is it the culture the immigrants assimilate into?

How would you parse the difference?

My theory: The parents are highly motivated and hardworking. The native levels of hard work / hardship make working in USA easy and hence they do. And thrive.

The next generation grows up in the relative prosperity of USA and lacks the edge and motivation needed to work as hard.

You should read the paper! She took on this exact theory and found it did not hold up.

There goes my theory. :)

Thanks! I will. I read the abstract Tyler posted a couple of times but didn't understand the jargon fully.

My mother has been a home health worker for coming on 20 years and talks about the differences between migrants and natives and how they interact with family. She says if she calls a family for the first time and their English isn't that great or a child has to translate she knows the patient will be well taken care of by the family about 80% of the time. Every subsequent generation in the US then does a worse job looking out for each other until you get the DAR crowd who keep grandma in a bed in the back room just to collect her social security check. Basically her opinion is that the more "American" you are, regardless of race, the more likely you are to be a jerk to your family members (both elderly and young). It's not economic theory, but the US changes you.

It's amazing the kind of contortions people will engage in to avoid politically incorrect conclusions. Yeah, um, America just like... ruins people. And the people who have been here the longest are the worst. Founding stock WASPS, especially, just a bunch of degenerate shiftless morons, not like these noble Somali cab drivers.

Even white Polish grandma is taken care of better by Generation#1 than Generation#2. It's not all about Somalia.

You got a citation for that, dude? It sounds like a convenient little just-so story you made up because you need it to be true. As someone pointed out downthread, second generation Asians and Hispanics do better than their immigrant parents.

You can be "doing better" economically & yet doing worse looking after old gramma. Different dimensions.

Would she have been better taken care of had the family stayed in Poland? I'm not so sure.

Most of the alleged issues where older people are not taken care of well are due to demographic age pyramid trends (people have fewer children than did when Polish grandma was a gal) and due to economic and work trends which do not vary as strongly by nation as people think.

The kind of "Oh, Polish grandma lives with us, so our culture is morally superior in its compassion" ideas that second generation migrants bandy around are probably at least 90% gibberish, coming more from a traditional cultures where anaemic housing markets and low personal wealth prevented independent nuclear family living, rather than some particular culture of respect and care for elders.

+1 on the observation.

Much of the lower average income of American blacks is explained by lower average IQ. What is the average IQ of African immigrants and of their children? It may be that cultural advantages dissipate in a generation while IQ disadvantages persist.

Some reversion to the mean? Perhaps black immigrants have particularly high IQ compared to their racial baseline, but their children on average aren't going to be as smart. I doubt it, but it's a possible mechanism that would explain some of the results.

What about the effect of differential fertility across groups?

It is well known that outside of very specific religious groups (such as the Mormons), education is negatively correlated with the # of kids. This is a statistically demonstrable phenomenon.

If the smartest people of the first generation spent their effort and energy on business or study, and left fewer kids than those who did otherwise, the second generation will be worse off.

Note that this would hold irrespective of ethnic composition of the group. So, do the Kosovar migrants have the same problem?

Understanding regression toward the mean is a huge step toward understanding how American society works. It's particularly important whenever the topic of IQ comes up.

Without having read the paper, my theory is regression toward the mean. Black immigrants are more strongly selected than others -- supposedly they are the most highly educated demographic in America -- so their children experience a steep regression toward a low IQ mean, level of conscientiousness, etc.

Oops, didn't read down here before posting my comments. I agree it's a possible mechanism, but I'm not automatically convinced of my "just so" intuition here without some evidence.

Regression towards the mean is a very strong phenomenon, when the original group is non-random.

If you took the top 10 000 mathematicians in the USA, and compared their mathematical results against the results of their kids, you would see some drop in the second generation.

Yet, the observation isn't nearly as emotionally loaded, and no one really suspects a pervasive anti-mathematician discrimination to be the cause.

Okay, sure. But there are some weird means being chosen to regress to here.

Let's take this as a simplified "black immigrants kids do worse than their parents" (note that that's not exactly what the study says). And we say, "regression to the mean."

But hispanic and asian immigrant's kids do better than their parents. Okay, sure, but that's regression to the mean as well, because hispanic and asian immigrants are below the mean asian and hispanic population's labor success level, while black immigrants are above the mean black labor success level.

So it's all "regression to the mean." Except that it's specifically regression to a RACIAL mean. They aren't regressing to the overall population mean. They aren't regressing to an overall children-of-immigrants mean. They're regressing to separate black, asian, and hispanic means. Which is why we are perfectly justified in suspecting a pervasive racial discrimination to be part of the cause (not, I would say, all of the cause).

"Which is why we are perfectly justified in suspecting a pervasive racial discrimination to be part of the cause"

Or genetics?

Could it be cultural & not discrimination? A Hispanic kid that grows up computes the internal mean he aspires to by subconsciously looking at other Hispanics?

Alternatively, could Hispanics & Asians be better at child-rearing in ways that can transmit traits (e.g. hard work, reading) more effectively to next generations?

> "Or genetics?"

A few things argue against genetics. One is that if we believe that intelligence or some more general trait that favors success in the labor market is strongly heritable, then by definition the children of immigrants should inherit their parents' more desirable traits. One does not genetically inherit traits from an amorphous mass of people not closely related to you, but with the same skin color. This would argue against regression to the mean, in other words.

Additionally, note that the children of immigrants are in fact more educated and successful than the mean -- it's just that after controlling for education level and family income, they do more poorly than their parents based on other factors. That argues that in fact their parents have been successful in instilling their children with desirable traits, but that the world at large sees them as less desirable than their immigrant parents, despite their educational attainment.

> "Could it be cultural & not discrimination?"

I don't think that you can disentangle the cultures of American ethnic minorities from discrimination. Obviously, the culture highlighted by, say, the Wire is one that depresses the labor outcomes of its people. But just as obviously, that culture is in part a product of things like the very high incarceration rate of black men. And just as obviously as that, the culture also perpetuates some of the discrimination. It's a giant tangled hairball.


You seem not to understand regression to the mean. It is well understood in the context of population genetics and very easily demonstrated. Take a look at any two very smart people- on average their children will be less intelligent because they regress to the mean of the genetic population. The same is true of height, although height is more heritable so there is less regression. Yes, African Americans have a lower mean than Americans of European descent, so studies show more regression among high-IQ African Americans.

Also, while second generation Hispanics have better economic outcomes, third and fourth generation Mexican immigrants actually regress towards first generation outcomes.


"second generation Hispanics have better economic outcomes, third and fourth generation Mexican immigrants actually regress towards first generation outcomes."

Say, you look at studies on this topic from the last decade. Is there a risk that what you are measuring is only a proxy for the economic trends of the times they grow in? Are these generations in the studies clustered around certain decades?

> This would argue against regression to the mean, in other words.

I don't think so. If two smart people have a child, the child is likely to be smarter than average, but less so than the parents.

So thinking that regression to the mean explains the observed phenomena doesn't even require any special beliefs about some baseline African IQ. It just requires thinking that African immigrants are more different from non-immigrant Africans in America than Hispanics or Asians are from their non-immigrant populations. Perhaps because immigrating from Africa is harder (because immigrating a long way from a poor place is harder), so only the relatively higher IQ do it.

"Say, you look at studies on this topic from the last decade. Is there a risk that what you are measuring is only a proxy for the economic trends of the times they grow in?"

That's always a problem, but somehow it only becomes an issue when someone doesn't want the study's finding to be true.

Arthur Jensen did a classic study of regression toward the mean in black and white children. IQ tests were given to more than 900 white sibling pairs and more than 500 black sibling pairs who went to school in the same California school district. He found that black and white children who had IQs of 120 had siblings whose mean IQs were 99 and 113, respectively. Similarly, blacks with IQs of 70 had siblings whose mean IQ was 79, while the siblings of whites with 70 IQs had a mean IQ of 85. Clearly, blacks and whites regress toward different IQ means. From IQ 50 to IQ 150 there were no significant departures from this linearity.

Jensen found the same thing in a study of black and white parent-child pairs, too, but the sibling test is more powerful because siblings share the same environment growing up.

Cliff et al,

Now you're back to arguing with a confused idea of what mean you're regressing to. You're hypothesizing that immigrant blacks are high performing compared to the American black mean due to genetic factors. But their children regress to the American black mean. And you appear to be confusing that as being analogous to the case where high performing (stipulated as "due to genetics") American blacks children regress to the mean of their genetic pool.

But if a Somali immigrant has a child with another Somali immigrant, it's fantasy-land to imagine that their child has a significant genetic mean with two parents who have a long history in America. In order for regression to the mean, on a genetic level, to have any meaning here, you have to show that the Somali-born immigrant is high-performing relative to the Somali mean, not the American black mean. Which, you know, maybe he is, but this assumes facts not in evidence. If in fact the Somali immigrant is performing at the Somali mean (and we have weak evidence that he is, because he's not performing better than other immigrants -- though this is not conclusive, as obviously immigrants are not randomly selected), then if his children regress to the American black mean, that argues strongly the the mechanism is not genetic.

Another possibility which would support a genetic regression to mean would be if children of one immigrant parent overwhelmingly had as the other parent an American native. But again, assumes facts not in evidence.

Add to this the opposite trend for children of hispanic or asian immigrants, and, frankly, it just doesn't make sense. We have to believe here that immigrants from African select for genetically high-performing individuals, but immigrants from Latin America or Asia select for genetically low-performing individuals, for this to be genetic. It's an extraordinary claim, and it doesn't even have ordinary evidence.

Malcolm Gladwell wrote an interesting chapter in his besteller "Outliers" about how hard his mulatto middle class Jamaican ancestors strove to avoid regression to the mean by only marrying fair skinned spouses. Various aunts and great-aunts couldn't resist sexy black men and their descendants fell out of the middle class, while Malcolm's mom married a white mathematician, and now Malcolm's a famous author.

Of course, reading Malcolm's explanation of his ancestors' successful mating strategy is like reading the Declaration of Independence translated into Newspeak.

Among Hispanic and Asian immigrants, most of the advance in income from the immigrant generation to the next comes from learning English.

For example, I used to employ a computer programmer who arrived from Seoul when he was 15. His father had been a pharmacist in Korea, but couldn't learn English well enough to pass the pharmacist exam here. So, he wound up owning a chain of dry cleaners in the black ghetto. The son was handsome and personable, an excellent corporate manager or high paying technical sales prospect, except that his thick accent was a problem in advancing beyond programming. I'm sure his accent-free born-in-America children are headed toward highly successful careers.

Almost all non-Hispanic black immigrants to the U.S. already speak English when they arrive from Jamaica, Barbados, Nigeria, Ghana, South Africa or wherever, so their children don't get much of boost over their parents from growing up speaking English, since their parents spoke fine English. This is very different from most Asian and Hispanic immigrants.

The exceptions might be Haitians and refugees, although refugees often are from the better educated strata of society. (French speaking African emigrants tend to go to France or Quebec.)

". Black immigrants are more strongly selected than others"

Apparently the biggest Black Immigrant group was from Africa. Does anyone know what fraction of these were refugees / asylees?

If this is indeed a sizable cohort then I question the applicability of the "strong selection" hypothesis.

It is in the article.

The refugee fraction is getting bigger all the time, but if you go back far enough to measure second generation performance, most black immigrants tended to be West Indian bourgeoisie or African star students.

And a lot of refugees come from the strata of society that know how to manipulate complex American bureaucratic requirements.

Yes, that'd make sense.

As a third generation West Indian (on my dad's side), I would say that assimilation effects are real. Of course, there's the fact that someone willing to pack up and leave their country for a different place is going to be much more likely to be bright (or at least hard-working) than average. There's also the fact that people will treat immigrant Blacks differently than native Blacks, full stop. Being that I am fully assimilated, people don't know much about my immigrant roots until I just come out and say it. It's amazing how "different" I become once I don't become one of "those people". Granted, there's some regression to the mean in terms of effort, but unless the second and third generation kids get a first generation type to vouch for them, it's much harder for them to get over that hump.

My experience is that the treatment you describe is true for both white and black Americans, wonder if that's your experience too.

Why does it take disclosure for second and third generation immigrants? Presumably because they have adopted the appearance of their closest racial group. Now consider the appearance of various racial groups. Is it any surprise that second and third generation black immigrants who choose to assimilate in to American Black culture fare worse than their parents?

It doesn't have to be about race either, anyone who adopts broken English and gang inspired clothing is likely to encounter prejudice.

America let's in ever dumber immigrants.
America is ever more balkanized and the dominant ethnic group ever more hated.

Enjoy being ruled by the Han, losers.

Anecdotally, at my ivy league university every single one of the black classmates I knew were first or second generation immigrants or foreign.

Right, Henry Louis Gates and Lani Guinier protested about that in 2004: most affirmative action at Harvard benefits either foreign elites, immigrants, or people with a white parent.

They stopped complaining so publicly with the rise of Barack Obama, who is obviously Exhibit A. Of course, Barack has done more with his affirmative action than Michelle Obama, who is a genuine African American, descended from American slaves through all four grandparents. In any sort of system where affirmative action is intended as compensation for American slavery and Jim Crow, Michelle ought to be the beneficiary, while Barack would not.

By the way, Dr. Rauh writes about the Gates/Guinier complaint about affirmative action in her paper, which has lots of good stuff in it.

It would be interesting to see a breakdown of this same effect by the communities and cultures the immigrants settle into. I absolutely believe there is significant contribution to this problem by the culture of the poor black community. I'd hypothesize that the effect is much less pronounced where these immigrants settle into non black communities.

By state, the highest test scores and lowest incarceration rates for blacks are found in Hawaii. A few notable features of Hawaii: many blacks got there via the military, which requires scoring above a minimum on the AFQT to enlist and which kicks out troublemakers. The black community in Hawaii is tiny and well-integrated. Hawaii has accepted interracial marriage for generations. Thus, many people in Hawaii who declare themselves to be black have a non-black parent, such as Barack Obama.

You can see somewhat similar patterns in a few other states where blacks perform relatively well, such as the Dakotas or the Vermont/New Hampshire region: military connections, high degree of interracial marriage, and few black role models.

Do ex-military blacks have less likelihood of a criminal history than the average black?

Do you know if there's data on this? I can see effects that might bend it either way.

Charming claims about meritocracy, racial IQ, and "the culture of the poor black community." Your Social Darwinist predecessors would be pleased!

A more obvious (and less racist) thing to look at would be that growing up in a culture where you're marginalized and harassed for your skin color is very different from entering such a culture as an adult. Malcolm X wrote a couple of interesting chapters on the subject; no doubt there's more to be found.

People who show up in the U.S. for the first time in their 20s or 30s generally don't join the neighborhood youth criminal gang.

Their sons who grow up here might join it, though. For example, this has been a pattern in Los Angeles for generations.

Yes but that same factor exists for Hispanics, right? Yet, why are 2nd Gen Hispanics doing better? Or are they not?

Applying for law schools ten years ago, I was surprised by a particularly candid table offered by Yale. You could drag your finger across a chart of GPA, LSAT score, and race (!!) to find your likelihood of admission.

As a white applicant, I stood less than a 5% chance. As a black applicant with the same numbers, it was greater than 95%.

I changed my name from Charles to Chartrawn on my application, but to no avail.

The University of Chicago had a similar thing circa 1998, listing the last year's undergraduate law school applicants' success by race with median and some other GPA/LSAT numbers. The difference in what was required was frankly shocking, and the page mysteriously vanished two days after I wrote about it on a law school listserv.

I urge people to read the paper. Read Table 2 and I think you will find data that makes much more sense than the econometric goobledygook.

A few observations:

--Looking at recent earnings data **conditional on employment** we see (for males) White = 62k, Black Native = 41K, Black Immigrant = 42K, Black Gen2 Age Adjusted = 49k.
--Looking at recent earnings data **conditional on employment** we see (for females) White = 39k, Black Native = 33K, Black Immigrant = 34K, Black Gen2 Age Adjusted = 44k

So there are a few story lines here. One is that the Gen2 blacks are doing better than the Gen 1 blacks if they are working. Looking at the unconditional data (Gen1 black males 34k, Gen2 black males age adjusted 37k) versus the conditional data tells you that there is a big difference in labor force participation between Gen1 and Gen2 males. (for women the participation rate appears to be about the same) . Is this cultural or racism? I'll let you decide.

Another story line is that Gen2 immigrant black women actually outearn white women. So if there is racism story here it seems to apply to males much more than women.

I would guess that the 2nd generation black women are filling up all the quota slots that 2nd-generation black men and our native black women are too stupid and lazy to do. There is a huge un-fillable demand for blacks who are not actively destructive, to fill up government mandated quotas.

"the return on their [blacks of the second generation] unobservable characteristics is converging to that of native blacks. "

What exactly does that mean?

Also, WTF is "Race premia"? I grok "earnings" but what are "residual earnings"?

Reading econ paper abstracts always makes me feel stupid.

As usual, just looking a the raw data is a much better idea than reading some econometric analysis of it and someone's interpretation thereof. Check out Table 2 in the paper.

Here's Dr. Rauh's conclusion:

8 Conclusion
This paper adds to the existing literature on three dimensions. First, it lays out the fact that black immigrants have become a large part of the black population in the last decades. Second, it demonstrates that the characteristics of immigrant blacks are very different from those of native blacks. Third, it determines how much immigrants blacks converge to native blacks across and within a generation. To put these patterns into perspective, they are compared to those of other races/ethnicities.
The share of black immigrants among the black population in the US has increased from 1% in the 1970s to 11% in 2011. Black immigrant males’ earnings and wages are higher than those of native blacks but the premium is small once we condition on being employed. Employment, education, incarceration, and marriage outcomes confirm that we are dealing with two completely different subsets of the population. Black immigrants are much more likely than native blacks to be employed, married, highly educated and not incarcerated.

In summary, first generation black immigrants undoubtedly have better labor market outcomes than black natives. They directly pass on some characteristics such as high human capital investments to the next gener- ation. The transmission of labor force attachment seems to be weaker, suggesting that the second generation is converging to natives. Black immigrants do not only converge to black natives across generations but also within a generation. For Asians and Hispanics, earnings premia decrease monotonically with age of immigration. For blacks, the earnings-age of immigration profile is upward sloping for those immigrating before the age of 15. Convergence across generations is mostly driven by low-educated second generation blacks that drop out of the labor force in greater numbers than low-educated first generation immigrants do. Similarly, convergence within a generation is driven by the fact that black first generation immigrants who arrive at an early age have a weaker labor force attachment than immigrants who arrive in the US when they are older. A social interactions model with an assimilation parameter that varies by age of immigration helps explain this phenomenon for men. When making their labor force participation decision, male immigrants generally place more weight on the characteristics of natives the earlier they immigrate.

Basically, the second generation of black immigrants do okay unless they don't get much education (i.e., aren't very bright). Then the males are in danger of not participating in the (legitimate) work force, just like African American males.

I may be interpreting this wrongly, but it appears that black males who were born abroad do worse the younger their parents brought them. For example, if you arrive at 17, you aren't that likely to join the Bloods or Crips. It just seems stupid.

But if you arrive at 2 and you grew up your whole life hearing from the other boys at school about how the Bloods or Crips were badass, well, if you want to prove your manhood, you join their local gang affiliate.

But maybe I've got this backwards.

Dr. Rauh seems to blame the age effect on accent, with white people positively discriminating in favor of foreign black accents. But of course the stereotype that a black with a prissy-sounding Anglo-Nigerian accent is probably a grad student type who won't mug you is, according to all the evidence Dr. Rauh assembles, a pretty accurate one. Whereas a jet black youth who tries to have an accent like some gangsta probably isn't a grad student type. If he sounds like a gangsta during a job interview, maybe that's what he is.

"Here is the abstract of another paper (pdf) by Alison, entitled “Successful Black Immigrants Narrow Black-White Achievement Gaps”:

Barbara Bush was referring to this in part when she made her notorious post-Katrina remarks in 2005 about how New Orleans blacks would be better off in Houston.

Houston, as the center of the oil economy, attracts black immigrants, while New Orleans, with its weak economy does much less so.

New Orleans developed a notably toxic African-American culture pre-Katrina with homicide rates well above the already high African American norm.

The truly interesting question is whether a high black immigrant city like Houston can develop a less dysfunctional overall black culture, or whether the glamorous gravitation of hip-hop will keep any sizable black community in American from escaping culture driven dysfunction.

She finished undergrad in 2011?!

That was exactly my reaction! Who finishes a PhD in 3 years??

Love to see some comments from someone who actually read the paper!

It's quite interesting that the unprocessed data paints a different picture than the processed data and the text.

First generation blacks earn about exactly the same as native blacks for both males and females, so there is no "mean" to revert to except the starting point.

Second generation blacks both male and female make more money both conditional on employment and unconditionally than first generation blacks.

It may well be true that second generation blacks underperform second generation immigrants from other parts of the world, but they certainly don't underperform first generation blacks.

I dont think that is at all clear from the abstract or the paper.

The effect she's picking up is that a certain fraction of black immigrants' sons raised in America decide to prioritize keeping it real just like their favorite rap songs tell them to, and thus wind up hustling on the streets or in prison or both. In contrast, the ones who don't get sucked in by the glamorous dysfunctions of African-American culture do okay.

A lot of black immigrant families that have been here a couple of generations, like Attorney General Eric Holder's, think like this:

Here is part of a soliloquy by a Haitian professor of French and Creole Literature in Miami in Tom Wolfe's "Back to Blood." The Haitian professor is thinking about his fair daughter whom he dotes upon and is sending to a private college versus his dark son whom he ignores and who is getting into African-American gang culture at his slum public school:

"Now he looked directly at Ghislaine. He smiled ... to cover up the fact that he was trying ... objectively ... to assess her face. Her skin was whiter than most white people's. As soon as Ghislaine was old enough to understand words at all, Louisette had started telling her about sunny days. Direct sun wasn't good for your skin. The worst thing of all was to take a sunbath. Even walking in the sun was too much of a risk. She should wear big-brimmed straw hats. Better still, an umbrella. Little girls couldn't very well go around with parasols, however. But if they had to walk in the sun, they should at least have straw hats. She must always remember that she had very beautiful but very fair skin that would burn easily, and she should do anything to avoid sunburns. But Ghislaine figured it out very quickly. It had nothing to do with sunburns ... it had to do with sunbrowning. In the sun, skin like hers, her beautiful whiter-than-white skin, would darken just like that! In no time she could turn Neg ... just like that."

Another factor is regression toward the mean in the quality of immigrants. Especially among African immigrants, the pioneer is often a smart student like Barack Obama Sr. But the exceptional pioneer tends to pull along in his wake more mediocre family members like the President's drunk-driving Uncle Omar and public housing resident Aunt Zeituni.

Back when Thomas Sowell was writing in 1978 about how West Indian immigrants were hot stuff, emigrating from the West Indies to America had been mostly a process for West Indians of better social class. Over time, however, they tend to drag along with them lower ranking relatives, so the influx becomes more mediocre: regression toward mediocrity.

I'm the first member of my Ghanaian family to be born in the USA. I guess I'm kind of the prince of the family, lol. The cultural pressure for the children of Black immigrants to assimilate into what I believe is a regressive African-American culture is real. Especially for the males. I briefly succumbed but now I'm a business school graduate making good money.

Haven't married yet but based on the relative success of my last relationship with a fellow Ghanaian American, my first, I think I'm going to go that route. If I can avoid mixing with African-Americans for one generation atleast my family will be better off in the long term.

There's no evidence to suggest that black immigrants fare better than traditional American blacks. Tens of thousands of black immigrants have been deported for criminal offenses over the last decade, and I've come across numerous African and Caribbean immigrants engaged in low wage, low skilled labor(cab drivers, custodians, and peddlers of bootleg merchandise). African-Americans are easily the black group that dominates every major facet of American existence(science, technology, sports, entertainment etc.)

Comments for this post are closed