American Hispanics are doing better than we had thought

by on February 17, 2016 at 12:50 am in Data Source, Economics, Education, Uncategorized | Permalink

Many of the more successful individuals start identifying as “white,” which biases the measured results downward for the Hispanic category:

Because of data limitations, virtually all studies of the later-generation descendants of immigrants rely on subjective measures of ethnic self-identification rather than arguably more objective measures based on the countries of birth of the respondent and his ancestors. In this context, biases can arise from “ethnic attrition” (e.g., U.S.-born individuals who do not self-identify as Hispanic despite having ancestors who were immigrants from a Spanish-speaking country). Analyzing 2003-2013 data from the Current Population Survey (CPS), this study shows that such ethnic attrition is sizeable and selective for the second- and third-generation populations of key Hispanic and Asian national origin groups. In addition, the results indicate that ethnic attrition generates measurement biases that vary across groups in direction as well as magnitude, and that correcting for these biases is likely to raise the socioeconomic standing of the U.S.-born descendants of most Hispanic immigrants relative to their Asian counterparts.

Here is the NBER paper by Brian Duncan and Stephen J. Trejo, via Sam Bowman.

1 duderino February 17, 2016 at 1:05 am

I wonder if its the ones who are primarily European (Spanish) who are identifying as white, rather than the more obvious Mestizos. It would mimic the dynamics of skin tone and economic class in Mexico if so.

2 anon February 17, 2016 at 2:15 am

The paper looks at 2nd and 3rd generation. “Ethnic attrition” arises mostly from intermarriage with whites. Skin color differences would mean less and less across generations.

3 Doug February 17, 2016 at 6:39 am

Take a random sample of Ivy League Hispanics. Now see if you can tell the difference between them and a group of Mayflower descendants.

4 Art Deco February 17, 2016 at 12:03 pm

There are few Castillans in the United States. Puerto Ricans and Cubans are caucasoid to a much greater degree than Chicanos or Central Americans. Cubans generally perform satisfactorily in America. The social metrics adhering the Puerto Ricans are some of the worst of any sizable immigrant group. I tend to doubt that identifying as white understates the economic performance of Hispanics.

5 stalin February 17, 2016 at 1:28 am

A joke told to me by a Mexican:

Q. Why are there no rich Mexicans?
A. When a Mexican gets rich, he becomes Spanish…

6 Steve Sailer February 17, 2016 at 1:33 am

“I used to be a Mexkin. but I’m makin’ money now so I’m gonna be a Spaniard.” — Golfer Lee Trevino at the 1968 U.S. Open

http://www.si.com/vault/1968/06/24/609159/eyes-rightbut-wrong

7 Thor February 17, 2016 at 5:22 pm

“Lee Trevino”. Hmmm. I remember seeing the name. I assumed he was Chinese-Italian.

8 Steve Sailer February 17, 2016 at 1:58 am

There’s a further complication that needs to be borne in mind: The federal government has allowed individuals to self-identify as more than one race since the 2000 Census, but you still can’t self-identify as more than one “ethnicity.” You have to pick either Hispanic or Not Hispanic.

The lack of any government recognition of the reality of mixed ethnicity makes the data lumpier and cruder.

Growing up in Los Angeles in the 1960s, I knew three families with Spanish surnames that were either of mixed ethnicity (one set of parents were Mexican and Swedish and the other Central American and Icelandic) or not Latin American in the slightest (my friend’s Spanish-surnamed father was a Yale man who claimed to be the descendant of a Spanish Armada admiral shipwrecked in Ireland).

9 A Scientist February 17, 2016 at 8:27 am

It strikes me that you seek a particular granularity. You want race, in a little more detail, but are not ready for loose genomic relationships.

A Hispanic can easily be some combination of Spaniard (who himself might be a bit Moorish) and Native American (of one origin or another).

Why on earth should government expect citizens to self-score that accurately, or at all?

I say use cheek swabs for health, and SAT scores for career guidance. All specific, no tropes required.

10 ladderff February 17, 2016 at 10:16 am

Fine! If that deal were on the table, I am quite sure Steve would be the first to take it. It isn’t. Don’t you know that colorblind policies are racist? Where have you been?

11 A Scientist February 17, 2016 at 10:59 am

No, Steve is stuck in the old model.

http://www.vdare.com/articles/the-race-faq

Phrenology is to neurology as Racism is to genomics.

(Perhaps some of the activists you dislike are stuck too, with Steve, in a misguided struggle.)

12 A Scientist February 17, 2016 at 11:05 am
13 TMC February 17, 2016 at 11:19 am

Glad you followed up with a link to Nature to verify this was PC crap.

14 A Scientist February 17, 2016 at 11:32 am

Would you believe “Why Steve Sailer is wrong” by Tyler Cowen on May 29, 2009?

http://marginalrevolution.com/marginalrevolution/2009/05/why-steve-saile.html

15 TangoMan February 17, 2016 at 2:11 am

The unacknowledged counter-trend here is that the performance of the Non-Hispanic White category gets influenced to some degree when it includes the subset of formerly Hispanic. What is the threshold where successful Hispanics decide to identify as white? Is it at the high end of the Hispanic range but still below the white range, is it above the high end of the Hispanic range and also above the low end of white range?

A way to engineer convergence is to bleed off top performers from one group and add then to another group such that the 2nd group sees its mean level decline from the addition of the 1st group’s top performers. If this is what is going on, to whatever degree of effect, that’s an interesting story in itself.

16 TGGP February 17, 2016 at 2:24 am

One earlier study which avoided that problem is the one from Generations of Exclusion, as it began far enough back that the census wasn’t keeping track of them as a distinct group, and followed up with the descendants of the original participants.

17 anon February 17, 2016 at 2:26 am

The paper looks at 2nd and 3rd generation. Do the least successful Hispanics intermarry with blacks, and do their descendants begin to identify as black? That would have the opposite effect on measures of Hispanic achievement.

18 JC February 17, 2016 at 3:07 am

Hispanic is for Latin American descendants only or Spaniards are Hispanics as well? So Enrique Iglesias is Hispanic but a white French-American is white?

19 mulp February 17, 2016 at 3:51 am

If the United States is no longer discriminating against non-whites, why would anyone stop identifying as a non-white ethnicity?

Scalia commented that he thought his service on the Supreme Court inspired young Italians to believe they were more than mafia thugs.

When Scalia talked of seeking the original intent of the words when written, I find it odd that he was blind to the intent considered it reasonable for him to be denied any public office because he was not a Protestant. Born in New Jersey, he would have been prohibited from holding office until the State constitution was changed in 1844. In fact, no one on the Supreme Court since Obama has been president could have held any public office in half the original States, and many would not have been allowed to vote circa 1800.

20 prior_test February 17, 2016 at 5:28 am

I’m confident, that with only the rarest of exceptions, that every single commenter here fully agrees with examples that prove ‘The arc of the moral universe is long, but it bends toward justice.’ http://quoteinvestigator.com/2012/11/15/arc-of-universe/

And that the American commenters are fully aware of their own history, of course.

21 Rich Berger February 17, 2016 at 7:26 am

Still embittered about the 19th century? You’re even older than you seem.

22 The Honorable Roger Taney February 17, 2016 at 9:20 am

It’s a good thing that the State Constitutions do not constrain whom the President may appoint.

23 Cliff February 17, 2016 at 11:01 am

Probably because they are actually white. On their college and scholarship applications they probably put Hispanic

24 Art Deco February 17, 2016 at 5:30 pm

Supposedly, the ancestry of Cubans is about 75% caucasoid, that of Puerto Ricans 65% caucasoid, and that of Mexicans 35% caucaosoid. This fellow contends that Mexicans living in America are 57% caucuasoid:

http://www.mexidata.info/id1442.html

25 Miguel Madeira February 17, 2016 at 6:56 am

«Many of the more successful individuals start identifying as “white,” which biases the measured results downward for the Hispanic category»

«Hispanic is for Latin American descendants only or Spaniards are Hispanics as well? So Enrique Iglesias is Hispanic but a white French-American is white?»

I see some confusion between “race” and “ethnicity” in these points – “hispanic” is not a category in the same classification as “white”, “black”, “native american”, “asian”, etc.; is a category in the same classification as “italian”, “arab”, etc.; “hispanics” could be “whites”, “blacks”, etc.

26 A Scientist February 17, 2016 at 8:33 am

The popular and political conceptions of race and ethnicity are indeed confused, and lag genomics by approximately a generation.

27 TallDave February 17, 2016 at 10:04 am

That seems to be accurate in my experience, most first-generation Hispanics seem to self-identify as Mexican or Cuban or Puerto Rican or etc, and Asians are if anything more distinct. The kids who are born or raised here, though, are just American, and will probably intermarry.

Mildly amusing aside: whenever my wife goes to the local ethnic supermarket, people come up to her and speak Spanish. She laughs and tells them she’s Filipino.

28 anon February 17, 2016 at 10:15 am

Approx 3M Spanish speakers in the Philippines.

29 Jonathan February 17, 2016 at 12:19 pm

Tyler,

Why the scare quotes around white but not not hispanic?

30 JWatts February 17, 2016 at 5:18 pm

Scare quotes is a bad name. They are often used to emphasize when the word is in question. Which is exactly how Tyler was using it.

31 M February 17, 2016 at 1:50 pm

For comparison on the General Social Survey, by comparison, for American born, those who self identify as their family from Mexico (variable: Ethnic) and as White (variable: Race), do not do any seem to do better than those who identify as having family from Mexico and Other, in the income (Conrinc) and education (Degree) variables. On the education variable, those who identify as ethnically Mexican and race Other, do better if anything, than those who identify as ethnically Mexican and White.

(Slightly more self identify as White, at around 58% of the sample, vs 42% as Other. Sample size 756 for Conrinc, 1071 for Degree).

Same is true for Puerto Ricans, although only Education (Degree) has a decent sample size for them.

There’s a variable Racehisp as well, introduced to the GSS more recently than the Race variable above, that captures the Hispanic category separately. But that’s absolutely no use for looking at ethnic Mexicans who identify as White vs Hispanic, as out of the 259 sample size of USA born ethnic Mexicans under that variable, 255 identified as Hispanic and only 2 as White.

Possible may give some perspective on “Ethnic attrition”. Ethnic attrition seems really unlikely to me, in the sense of people swapping out of the Hispanic category as they get richer. It does seem likely as they accumulate more ancestors from non-Hispanic ethnic groups, but then, that’s not “Hispanics are doing better than we think”, really.

32 Steve Sailer February 17, 2016 at 4:31 pm

A running theme in Joseph Wambaugh’s long series of LAPD cop novels going back to Serge Duran in “The New Centurions” in 1971 are characters who are “ambiguously Latino.” The former LAPD detective usually includes at least one character in each book whom readers could quite reasonably argue over whether that person is Latino or not. Over the years, Wambaugh has gotten a lot of mileage out of exploring the many subtleties of the question.

33 IVV February 17, 2016 at 6:09 pm

I’m the son of a white American (French-German-Irish-English, with a lot of that mixing actually in Europe) and a Mexican-American (Spaniard-French-Purépecha, where the Spanish is Moorish and the Moorish likely Sephardic). Officially, I’m Hispanic, but if you went by my name, skin color, lack of Spanish knowledge, etc. everyone would say I’m a non-Hispanic white–even in Latino groups, people wonder why I’m there, until I mention my history. My brother and sister, it varies. My brother is darker complected than myself, but people thought he might be Assyrian (the Moorish genes do show; Scandinavia didn’t really like him crossing borders when he visited). My sister speaks Spanish and focuses on her Mexican heritage much more, but her skin color is a light pink so she passes, too. We’ve all got graduate degrees.

So yeah, Americans treat us all as white. That’s what’s going to happen with the Mexicans in the United States. It happened to the Irish, too, after all. Just a matter of time.

34 Gafiated February 18, 2016 at 6:32 am

That was before the census had a separate category for “Non-Irish Whites”.

Comments on this entry are closed.

Previous post:

Next post: