No, I am not reporting on the fact that net Mexican immigration to the United States is now zero. I am among the minority who find this unfortunate, though happy that the economy in Mexico is vibrant enough to attract folks back home or keep them from leaving. I’d like to point out another under-reported and happy idea: that immigrants to America have always and continue to assimilate remarkably well.
For example, this recent paper by economists Ran Abrimitzky, Leah Platt Boustan and Katherine Eriksson, finds that:
The panel data thus allow us to examine the assimilation of those migrants who remained in the US long term, rather than returning to Europe. We find that immigrants who remained in the US held higher-paid occupations even upon first arrival to the US, and their age-earnings profile was rather similar to US natives. The difference between the repeated cross-section and the panel data is driven by the change in composition of the repeated cross-section with years in the US: return migrants drop out over time. Thus the larger degree of convergence in the repeated cross-sections relative to the panel reflects negative selection of return migrants. Our paper further shows that it is important to account for differences in migration patterns across sending countries. While permanent migrants from some countries performedbetter than US natives upon first arrival, migrants from other countries performed worse than natives. Moreover, these differences in performance across countries are shown to be persistent across generations.We further examine the assimilation in the marriage markets of first and second generation migrants from this age of mass migration and find high rates of endogamous marriage and persistence in endogamy across generations. We find evidence that cultural distances between sending country and the US are correlated with migrants’ performance in theUS labor markets upon arrival. The panel data thus allow us to examine the assimilation of those migrants who remained in the US long term, rather than returning to Europe. We find that immigrants who remained in the US held higher-paid occupations even upon first arrival to the US, and their age-earnings profile was rather similar to US natives. The difference between the repeated cross35section and the panel data is driven by the change in composition of the repeated cross-section with years in the US: return migrants drop out over time. Thus the larger degree of convergence in the repeated cross-sections relative to the panel reflects negative selection of return migrants.Our paper further shows that it is important to account for differences in migration patterns across sending countries. While permanent migrants from some countries performed better than US natives upon first arrival, migrants from other countries performed worse than natives. Moreover, these differences in performance across countries are shown to be persistent across generations. We further examine the assimilation in the marriage markets of first andsecond generation migrants from this age of mass migration and find high rates of endogamous marriage and persistence in endogamy across generations. We find evidence that cultural distances between sending country and the US are correlated with migrants’ performance in the US labor markets upon arrival.
Lot’s to digest here: but while there is some variation in immigrant labor market performance based on country of origin, immigrants do very well once they come here, and certainly do not seem to be falling behind their native counterparts. Note what is excellent about this study – it tracks the same group of people over time rather than looking at the cross-section. As an illustration, consider two ways of discussing the evolution of income in America:
(1) Between 1980 and today the average income of the median worker in America (in real terms) was flat.
(2) How well did a guy who was at the median in 1980 do over the next 30 years?
Note that these are radically different questions. Almost all of the discussion of the evolution of income in the United States looks only at repeated cross-sections of data over time as in question (1). But in 1980, my dad was below the median person and over the last 30 years he has done well for himself. I was barely born in 1980 and now I am in the pool of households counting in the statistics. The structure of households changes over time. Family size changes over time. Marriage and divorce changes over time. Migration changes over time. All of these things impact the cross-section in ways that suggest that we can show dramatic increases in measured median income while every single person is worse off over time, just as we can show stagnation in measured median income while every single person is better off. We can illustrate shortly. The point of (2) is that the right question is to track how well particular people do over time, or at least to track how well people in that position do over time (for example, how did a 40 year old with a college degree do in 1980 versus today), but this is not at all what we do. The aforementioned paper DOES track the same people over time, and thus we have a much better picture of what is happening to those particular people than survey cross-section data can tell us.
But that study was for immigrants coming to America during the open borders era when the large majority were of European origin. Aren’t things different today? Here are Tyler Cowen’s older musings on the qualitative observations (more empirics to follow):
On the other side of the ledger, here are a few relevant factors:
1. The slower influx of Mexicans (100,000 a year vs. a former 500,000 a year) means that assimilation will from now on proceed more rapidly, and certainly more rapidly than the critics had been predicting.
2. The effect of Latino communities in lowering crime rates and revitalizing neighborhoods and cities has been stronger than might have been expected twenty years ago.
3. The notion that Latino migrants to the U.S. might help seed and sustain a broader Latin American economic and democratic boom has become a reality, and this was not obvious twenty years ago.
4. The idea that “the New World” will become a major trading bloc to rival “Chinese Asia” is a more important idea than it might have seemed twenty or even ten years ago. The United States needs extensive Latin connections to maintain its status as active leader of that bloc.
5. Outsourcing is more of a force than we had thought, and the possibility of outsourcing raises the (relative) gains from allowing immigration. I will write more on this in the future, so I’ll leave the details for now.
Overall, the arguments on immigration have changed quite a bit in the last ten to fifteen years, but those changes have cut in both directions.
Here is a paper from the American Political Science Association:
SamuelHuntington argues that the sheer number, concentration, linguistic homogeneity, and other characteristic ofHispanic immigrants will erode the dominance of English as a nationally unifying language, weaken the country’s dominant cultural values, and promote ethnic allegiances over a primary identification as an American. Testing these hypotheses with data from the U.S. Census and national and Los Angeles opinion surveys, we show thatHispanics acquire English and lose Spanish rapidly beginning with the second generation, and appear to be no more or less religious or committed to the work ethic than native-born whites.Moreover, a clear majority of Hispanics reject a purely ethnic identification and patriotism grows from one generation to the next. At present, a traditional pattern of political assimilation appears to prevail.
Here is Jason Riley:
Dowell Myers, a demographer at the University of Southern California, calls it the “Peter Pan Fallacy.” “Many of us assume, unwittingly, that immigrants are like Peter Pan,” says Mr. Myers, “forever frozen in their status as newcomers, never aging, never advancing economically, and never assimilating.” In this naïve view, he says, “the mounting numbers of foreign-born residents imply that our nation is becoming dominated by growing numbers of people who perpetually resemble newcomers.”
The reality, however, is that the longitudinal studies show real socio-economic progress by Latinos. Progress is slower in some areas, such as the education level of adult immigrants, and faster in others, such as income and homeownership rates. But there is no doubt that both assimilation and upward mobility are occurring over time.
With respect to linguistic assimilation, which is one of the more important measures because it amounts to a job skill that can increase earnings, the historical pattern is as follows: The first generation learns enough English to get by but prefers the mother tongue. The children of immigrants born here grow up in homes where they understand the mother tongue to some extent and may speak it, but they prefer English. When those children become adults, they establish homes where English is the dominant language.
There’s every indication that Latinos are following this pattern. According to 2005 Census data, just one-third of Latino immigrants in the country for less than a decade speak English well. But that proportion climbs to 75% for those here 30 years or more. There may be more bilingualism today among their children, but there’s no evidence that Spanish is the dominant language in the second generation. The 2000 Census found that 91% of the children of immigrants, and 97% of the grandchildren, spoke English well.
If American culture is under assault today, it’s not from immigrants who aren’t assimilating but from liberal elites who reject the concept of assimilation. For multiculturalists, and particularly those in the academy, assimilation is a dirty word. A values-neutral belief system is embraced by some to avoid having to judge one culture as superior or inferior to another. Others reject the assimilationist paradigm outright on the grounds that the U.S. hasn’t always lived up to its ideals. America slaughtered Indians and enslaved blacks, goes the argument, and this wicked history means we have no right to impose a value system on others.
But social conservatives who want to seal the border in response to these left-wing elites are directing their wrath at the wrong people. The problem isn’t the immigrants. The problem is the militant multiculturalists who want to turn America into some loose federation of ethnic and racial groups. The political right should continue to push back against bilingual education advocates, anti-American Chicano Studies professors, Spanish-language ballots, ethnically gerrymandered voting districts, La Raza’s big-government agenda and all the rest. But these problems weren’t created by the women burping our babies and changing linen at our hotels, or by the men picking lettuce in Yuma and building homes in Iowa City.
Keep the immigrants. Deport the Columbia faculty.
Here is Linda Chavez:
Take Hispanic dropout rates. A snapshot looks bad: 42% of Hispanics, according to the Current Population Survey, had not finished high school in 2005. But nearly half of the people counted aren’t dropouts in the usual sense; they’ve never dropped in to an American school. They are immigrants who completed their schooling, such as it was, before coming here in their late teens or 20s. Granted, low education levels will make their climb up the economic ladder slower — 60% of Mexican-born adults have not completed high school. But the earnings of Hispanic immigrants will improve as they gain work skills and experience, and the evidence is strong that they will do so. Mexican-born men, for example, had higher labor force participation rates than native-born male workers, 88% compared with 83%, and lower unemployment rates than native workers, 4.4% compared with 5.1% in 2006. Labor force participation rates of illegal aliens are higher yet, a whopping 94%.
More importantly, the children of Hispanic immigrants are graduating from high school. The high school completion rate for young, U.S.-born Hispanics is 86%, only slightly lower than the 92% of non-Hispanic whites. Hispanic immigrant children who do enroll in school after they come here are as likely as American-born Hispanics to earn a high school diploma (although half of Mexican immigrants 15-17 years-old do not enroll in school).
Hispanics are more likely than either whites or blacks to continue their education at two-year institutions; in 2000 they represented 14% of all students enrolled in two-year institutions. Only 12% of U.S.-born Hispanics earn four-year degrees compared with 26% of non-Hispanic whites. Nonetheless, the economic returns on education are substantial for Hispanics. As a 2006 study on Hispanics by the National Research Council of the National Academy of Sciences reported, “We consistently find that, after adjusting for the levels of human capital (e.g., schooling and English language proficiency), Hispanics do almost as well as whites with respect to both employment and labor market earnings,” which the authors note is not the case for blacks, who still lag behind whites even after adjusting for observable measures of human capital
Is America Better Situated to Absorb Immigrants Today than in the Past? Don Boudreaux argues yes:
Historically, most immigrants settled in cities — think, for example, of Manhattan’s Little Italy and San Francisco’s Chinatown.
And the resources and amenities available in metropolitan areas today are far greater per capita than they were just before Uncle Sam abandoned his open-immigration policy in the 1920s.
Consider that in 1915 the typical dwelling in America housed 5.63 persons; today it houses fewer than half that number — 2.37 persons. Combined with the fact that today’s typical dwelling has about 25 percent more square footage than its counterpart had back then, our ability to absorb immigrants into residential living spaces is today more than twice what it was a century ago.
What about workers? A measure of ability to absorb workers is capital invested per worker. Today, the amount of capital invested per worker is nine times greater than it was just after World War I. Because a worker’s productivity rises when he has more capital to work with, and because his pay is tied closely to his productivity, workers today produce and earn more than workers did during the open-borders era.
Don’t lose sight of our labor market’s great flexibility. In addition to absorbing millions of immigrants in the 19th and early 20th centuries, it easily absorbed the 46 million women who entered the work force during the second half of the 20th century.
Here is Ed Lazear on Mexican assimilation (they are not assimilating as well as other ethnic groups):
Mexican immigrants assimilate more slowly than other immigrants as reflected in English fluency. They also have lower levels of education, lower wages, and live in more concentrated areas than other immigrants. The source of the problem seems to be U.S. immigration policy. By admitting large numbers of Mexicans, relative to other groups on a family rather than job basis, the United States selects a group of immigrants from Mexico who are already at a disadvantage.
So, Mexicans seem to be assimilating just fine, albeit not as well as other ethnic groups may be. Check out who had a hard time assimilating:
These authors find that there is some negative assimilation pressure, but not from the immigrants’ character themselves, but rather from the changing nature of a globalized labor market:
assimilation immigrants assimilation assimilation assimilation immigrants immigrant assimilation immigrants