Though the title of Gregory Clark’s new Foreign Affairs essay (“The American Dream Is an Illusion“) is regrettable — my guess is that it was written by an editor hostile to Clark’s argument — the essay itself is compelling and important. Those who are familiar with Clark’s The Son Also Rises and A Farewell to Alms will quickly grasp the premise. Drawing on a wide range of data sources, Clark has chronicled the pace of social mobility over centuries across a number of different countries, and his central finding is that “social mobility rates are extremely low” and that “seven to ten generations are required before the descendants of high and low status families achieve average status,” both in egalitarian countries like Sweden and in more laissez-faire countries like the United States.
In his new essay, Clark applies this insight to immigration policy. Specifically, he posits that the apparent success of immigrant assimilation in earlier eras largely reflects the fact that “immigrants who quickly assimilated to their new society in countries such as the United States were often positively selected from the sending populations.” The poor immigrants who made their way to the U.S. from Scandinavia and central and eastern Europe were generally literate women and men well-equipped for life in a modernizing society. Of course, not all immigrants fell into this category. Clark discusses Americans of French origin, including those descended from French settlers in Louisiana and from more recent French Canadian immigrants. While Irish and Italian Catholic immigrants faced more intense discrimination than people of French origin, he notes that their descendants have achieved more or less average social status while people of French origin have not. According to Clark, this reflects the fact that French who arrived in the United States “were overwhelmingly drawn from the lower classes of Acadia and Quebec, as a result of demographic patterns and selective migration,” and “the effects of this lower social status have persisted across generations,” despite intermarriage.
And Clark maintains the same pattern is recapitulating itself among more recent immigrants. Visa restrictions helped ensure that immigrants from some regions (sub-Saharan Africa, the Arab world, South Asia, and East Asia) had skills that were of value in the U.S., and this effectively limited immigration to people who were from groups with above average social status in their native countries. Immigrants who did not face these restrictions, because they arrived as refugees or as unauthorized immigrants, “entered the United States with low social status and have struggled to achieve upward mobility since.” A similar pattern obtains in Europe, where the descendants of guest workers drawn from rural populations have found it difficult to climb the economic ladder.
One of the most persistent myths surrounding the immigration debate is that if the U.S. placed a heavier emphasis on skills in shaping its immigration policy, the share of immigrants from Latin America would plummet. (Mark Krikorian offers a version of this thesis in a recent article for NRO.) Clark illustrates why this isn’t necessarily the case. Clark observes that “migrants from Mexico and Central America tend to be negatively selected from their home populations: they are often the people who found themselves in such desperate economic circumstances at home that they preferred to live as illegal immigrants in the United States,” which helps explain why the social status of descendants of migrants from Latin America tends to be low.
But a skills-based immigration policy would create more opportunities for skilled Latin American immigrants who have something to lose, and who would not be willing to live as unauthorized immigrants. Consider a recent Pew Global Attitudes Project poll of Mexicans, which found that 34 percent of Mexicans would move to the U.S. if given the opportunity, and half of them (17 percent) would do so without authorization. It seems reasonable to bet that the 17 percent who would not do so without authorization are drawn from Mexico’s more educated classes. Mexico’s educational attainment rate is low by the standards of affluent market democracies, yet it is increasing: while only 12 percent of Mexican 55-64 year-olds have a post-secondary education, 22 percent of 25-34 year-olds have one. There is a fairly large pool of educated Mexican immigrants to draw from, should the U.S. choose to do so.
Clark, however, offers a different strategy: he calls for increasing the immigration of educated Latin Americans from countries like Argentina, Brazil, Chile, and Peru, as doing so would “bolster the overall social status of the Latino population in future generations, and their representation in higher-status positions in the society.” While this seems like a perfectly sound idea, it’s not clear to me that it would help the U.S. forestall the emergence of “a substantially poorer and less educated Latino underclass,” particularly if, as seems likely, the descendants of skilled immigrants are more likely to intermarry than the descendants of less-skilled immigrants, a phenomenon that reflects the larger rise of assortative mating (in which people choose partners with similar levels of educational attainment) and that contributes to ethnic attrition (in which people cease identifying with a given ethnic group, usually because they are of mixed ancestry and their connection to the group has attenuated). Indeed, a mestizo underclass might come to see itself as racially distinct from Latinos descended from middle- and upper-class social groups, a phenomenon that is arguably already taking hold.
As for immigrants of Asian origin, it is important to note that the channels for skilled immigration Clark identifies are not the only channels that Asian immigrants use. Many Asian immigrants arrive via family unification visas, and there is a large number that has arrived via the diversity visa lottery. As a general rule, the relatives of skilled Asian immigrants will also tend to be skilled, but this isn’t always or necessarily the case. It is not uncommon for a capable immigrant to invite less-capable relatives to join her in her adopted country. There might in fact be considerable social pressure for her to do so.
And as David Nakamura of the Washington Post reports, the Obama administration is contemplating an executive action that could dramatically increase legal immigration, despite the fact that large majorities of Americans consistently oppose such an increase:
The proposal outside groups are pushing centers on changing the way the government counts the number of foreigners who are granted green cards, which allow them to live and work in the United States. Under the law, 226,000 green cards are reserved for family reunification and 140,000 for employment in specialized fields, numbers that Congress established in 1990.
The government has traditionally counted each family member against the limit when granting visas to foreign siblings of U.S. citizens. The spouses and children of permanent U.S. residents and foreign workers have counted against the limits as well. Advocates are calling on Obama to count only the principal green-card holder in each case, while allowing the rest of the family members in, which would reduce huge backlogs in both categories.
More than 4.4 million people are waiting for green cards, according to the State Department.
Asian American advocacy organizations have focused on such changes because, other than Mexico, the countries with the longest waiting lists of people trying to join relatives in the United States are the Philippines, India, Vietnam and China — with delays stretching as long as two decades.
This executive action would have a profound effect on the future composition of the U.S. population, and the future composition of the Asian origin population. It likely means that the Asian American population, which now has a median household income higher than that for the U.S. as a whole (partly because this population is concentrated in high-wage, high-cost regions), would grow poorer and less capable of self-support.