The Associated Press offers a brief summary of a new report from the Pew Research Center that, among other things, observes that Asian immigrants now outnumber (slightly) immigrants from Latin America:

Asians recently passed Hispanics as the largest group of new immigrants to the United States. The educational credentials of these recent arrivals are striking. More than six-in-ten (61%) adults ages 25 to 64 who have come from Asia in recent years have at least a bachelor’s degree. This is double the share among recent non-Asian arrivals, and almost surely makes the recent Asian arrivals the most highly educated cohort of immigrants in U.S. history.
Compared with the educational attainment of the population in their country of origin, recent Asian immigrants also stand out as a select group. For example, about 27% of adults ages 25 to 64 in South Korea and 25% in Japan have a bachelor’s degree or more.2 In contrast, nearly 70% of comparably aged recent immigrants from these two countries have at least a bachelor’s degree.
Recent Asian immigrants are also about three times as likely as recent immigrants from other parts of the world to receive their green cards—or permanent resident status—on the basis of employer rather than family sponsorship (though family reunification remains the most common legal gateway to the U.S. for Asian immigrants, as it is for all immigrants).

This reflects an interesting shift. We know that the labor market position of less-skilled workers in the U.S. has deteriorated in recent years, yet the place premium for less-skilled workers, i.e., the wage differences for identical workers in the U.S. and other countries, has been high enough to outweigh that fact. Now, however, a combination of factors — modest wage growth in some of Latin America’s emerging economies, more aggressive immigration enforcement in the U.S., the housing bust, etc. — have depressed migration levels from Latin America. It is important to remember that the place premium alone isn’t enough to drive migration, as people are risk-aversive and reluctant to disrupt their existing social networks. This is why “chain migration,” in which an immigrant sponsor members of his extended family who in turn sponsor her own extended family members and so on in a chain, has been so important in recent decades.

A number of reformers have called for shifting U.S. immigration policy from its current heavy emphasis on family reunification to an emphasis on skills. The 2009 “Breaking the Immigration Stalemate” report from Brookings and the Kenan Institute at Duke represented one such effort. Back in 2007, Yuval Levin wrote an excellent, comprehensive piece for Commentary on “Fixing Immigration,” and he addressed family reunification in detail:

[B]y far the largest share of new legal permanent residents—about 60 percent in 2005—are relatives of naturalized American citizens or (in far fewer cases) of other legal permanent residents. Indeed, “family unification” is easily the foremost organizing principle of American immigration policy today. Once an individual has become an American citizen, he or she can petition on behalf of relatives—not only spouses and children, but also parents and siblings. Parents of new citizens account for almost 10 percent of new permanent residents each year, and siblings for another 7 percent. Once naturalized as citizens, each such spouse, child, sibling, and parent can bring other relatives over, creating a so-called “chain immigration” that accounts for a large and growing portion of newcomers to America.

There is an obvious problem with this emphasis, however:

It is easy to see why family unification appeals to legal immigrants themselves. And keeping spouses and their children together makes good humanitarian, social, and cultural sense. But what sense is there in extended family unification? What is the great benefit to America in admitting parents and siblings of adults admitted as immigrants, and then further admitting the parents and siblings of those new immigrants (and of their spouses), all the while leaving less room for new arrivals who might meet important economic needs or be better suited to become new Americans?

And so Yuval calls for reforming this practice:

New immigrants should be permitted to petition only for spouses and for unmarried children below the age of twenty-one. Keeping parents and their young children together is not only humane and ethical, it is good social policy, helping to channel the energies of immigrants toward building stable families and communities. But to permit more extended family relationships to define the character of American immigration is absurd. Although the narrowing of family-based immigration categories would need to be carried out over time, allowing those now in the process to adjust, and clearing at least some of the backlog, new extended-family petitions could be curtailed immediately and eventually ended.

He also calls for ending the diversity visa lottery, among other things. If we shifted these roughly 220,000 slots from extended family unification and the diversity visa lottery, we could immediately increase the amount of employment-based immigration by a considerable margin. This wouldn’t necessarily mean that the source countries would change dramatically. But the composition would change, e.g., a larger share of the Latin American and Asian influx would have advanced degrees and a high level of English language proficiency.