Heaven’s Door, by George Borjas, is by far the best introduction I have seen to the economics of immigration. Born in Cuba, Borjas came to America as a child. (He is also my colleague at Harvard, so skeptics should feel free to discount my enthusiasm for his book.) In 1970, he reports, America’s foreign-born workers earned as much as natives. By 1998 male immigrants typically earned only 77 percent of what natives earned. That makes the wage gap between immigrants and natives three times as large as it was in 1910.10 Immigration from Mexico has had a central part in this change. Mexico’s share of the foreign-born population has risen from 8 percent in 1970 to 28 percent today. Mexican-born men in the United States earn less than half what non-Latino whites earn, so as Mexico’s share of the immigrant labor force grows, the economic gap between immigrants and natives inevitably widens.11 America now accepts more legal immigrants from Mexico than from all of Europe. If one adds illegal immigrants, Mexicans are more numerous than Asians. The Bush administration hopes to expand Mexico’s share even further. It has said that it wants to legalize many of the three to five million “undocumented” Mexicans currently in the United States, as well as increase the number of unskilled Mexicans admitted on a temporary basis.12
Only half the Mexicans living in America have attended secondary school, and only a third have graduated. Only one in eight claims to speak English very well.13 Italians and Poles had similar handicaps in 1910, but Americans are far better educated today than they were a century ago, so the gap between Mexicans and American-born workers is wider. In view of their educational and linguistic disadvantages, Mexican-born workers actually do surprisingly well. Roger Waldinger, a sociologist at UCLA, finds that most California employers prefer Mexicans to the American-born workers who apply for unskilled jobs, because they believe Mexicans are more reliable and disciplined than natives. But while having the right attitude is often enough to get an eight-dollar-an-hour job, it is seldom enough to get one for sixteen dollars an hour.14
Since 1970, immigration has increased the number of unskilled job applicants faster than the number of skilled job applicants. First-year economics predicts that increasing the relative number of unskilled workers will depress their wages, because employers will not need to raise wages to attract applicants for unskilled jobs. Nonetheless, those who favor an expansive immigration policy often deny that the increase in the number of unskilled job applicants depresses wages for unskilled work, arguing that unskilled immigrants take jobs that natives do not want. This is sometimes true. But we still have to ask why natives do not want these jobs. The reason is not that natives reject demeaning or dangerous work. Almost every job that immigrants do in Los Angeles or New York is done by natives in Detroit and Philadelphia. When natives turn down such jobs in New York or Los Angeles, the reason is that by local standards the wages are abysmal. Far from proving that immigrants have no impact on natives, the fact that American-born workers sometimes reject jobs that immigrants accept reinforces the claim that immigration has depressed wages for unskilled work.
Back in 2001, I blogged:
The left wing New York Review of Books just ran a terrific two part series on immigration by distinguished Harvard sociologist Christopher Jencks called "Who Should Get In." It's striking, although it really shouldn't be surprising, that starting from nominally opposite ends of the political spectrum, he and I reach almost identical conclusions about what's in the best interests of American citizens. Jencks' last sentence is, "Fifty years from now our children could find that admitting millions of poor Latinos had not only created a sizable Latino underclass but—far worse— that it had made rich Americans more like rich Latin Americans."