Author pleads for better class of immigrants

Wednesday, September 8, 1999


George Borjas was 8 years old when Fidel Castro's tanks rolled into Havana. Three years later, his family, having lost their small factory making men's pants, fled to Miami.

Borjas, 48, is now a professor of public policy at Harvard University, specializing in immigration. He's also an advocate of a policy that would have prevented his own family from coming here. He makes his case in his new book, "Heaven's Door" (Princeton, $27.95).

The book is neither an argument to close the nation's doors nor to throw them wide open. It's an argument to screen applicants for age, language and work skills. In contrast, the current immigration system, set up in 1965, is based almost entirely on whether applicants have family members in the United States.

The 1965 reforms began what Borjas calls the Second Great Migration -- the first being the influx of 1885-1914. In the new wave, he writes, a high proportion of immigrants have been unskilled. On balance they do not take the jobs of Americans; the jobs multiply to employ the hands willing to work. But more of those jobs are at the bottom end.

And that's good for a lot of people. The San Diego couple who hires a nanny at $200 per week can afford to work more hours. The manufacturer who would have moved to Mexico now has affordable employees. In Seattle, we employ foreigners to make our lunches and stay up all night in convenience stores.

There's a cost to this. In the past 25 years, the gap between the top and bottom in incomes has widened dramatically. Immigration, says Borjas, is not the only reason for this, but he calculates that it "may account for nearly half the decline in the relative wage of high school dropouts."

Borjas argues that the United States would be better off attempting to select a better-educated class of applicants, as Canada does. This, he allows, would change the ethnic mix of immigrants. It would be attacked for being discriminatory. But all systems of screening applicants are discriminatory, he writes, including the one we have now.

Under Borjas's proposal, for example, the United States would end up admitting more Canadians and fewer Mexicans -- simply because the average Canadian immigrant has 13.8 years of schooling, and the average Mexican 7.6. The average Canadian immigrant earns 24 percent more than the average American; the average Mexican earns 39 percent less.

Huge differences also exist in the different groups' use of welfare. After 10 years in the United States, 5 percent of East Indians are receiving some kind of government assistance; 9 percent of Britons, 20 percent of Haitians, 21 percent of Russians, 34 percent of Mexicans and 58 percent of Dominicans.

Such differences are normal. They are based, Borjas says, on "ethnic capital" -- the group's know-how, attitudes and values handed down through generations. Among immigrants, half of the first generation's earning-power differential is passed to their kids, and half of the second generation's differential is passed to their kids. "Ethnicity matters in economic life, and it matters for a very long time," he writes.

Any change in the mix of these groups would change the U.S. labor market, helping certain industries and hurting others. To take a local example, without Mexican immigrants who would pick the apples? Well, Borjas argues, maybe high-wage America shouldn't be growing apples.

He writes, "Does it make sense for the United States to be in the business of exporting hand-harvested fruits and vegetables?"

Oops. He just lost Central Washington. (But there are alternatives: Growers in the Canadian Okanagan employ French-Canadian youths.)

After piling up all his evidence, Borjas allows that immigration policy "cannot be based on the evidence alone." All policies have winners and losers.

This book assumes that the people who count are native-born Americans, because they make the political decisions. It proposes a policy that arguably would make them better off economically.

And while screening could pay Americans a dividend, Borjas allows that it would still miss two-thirds of the difference between the highly productive immigrant and the merely average. That's because two-thirds of the difference between individuals is due to "unknown factors."


P-I reporter Bruce Ramsey can be reached at 206-448-8391 or