Let's Do Immigration Right
How to avoid the mistakes we made when we argued about free trade.

(FORTUNE Magazine) – THE BOILING DEBATE over the economics of immigration may give you an eerie sense of déjà vu, and no wonder: Its superheated rhetoric recalls the polarized and exaggerated arguments over open trade and globalization in the 1990s. Foes of immigration try to brush off legitimate macroeconomic studies like the 1997 National Research Council report that shows immigration adds $10 billion a year to the economy, and the work of academics like Giovanni Peri and Gianmarco Ottaviano, who found that immigration raised average wages by as much as 2.5% in the 1990s. Instead, the critics often point to real problems that immigration aggravates, like bloated state budgets and reduced opportunities and wages for low-skilled minorities. But then they pin the blame entirely on the worker influx--not unlike trade critics who rightly complained of disturbing economic inequality in Mexico, but wrongly implied that NAFTA was the primary culprit, not simply a policy that failed to cure it.

Meanwhile, supporters of immigration have been repeating mistakes from the trade debate too. Many of us who fought for market opening back in the '90s made blanket statements about its benefits for jobs and the economy while pooh-poohing or ignoring its harsh impact on particular communities and groups of workers. Supporters also tend to gloss over the degree to which significant increases in immigration can depress wages and even cost jobs of low-skilled U.S. workers. Harvard's George Borjas and Larry Katz have found that between 1980 and 2000, predominantly low-wage immigration from Mexico depressed the wages of U.S. high school dropouts by 7.7% compared with those of their college-educated peers.

While there is no shortage of cases where increased immigration hurt a specific group of low-skilled workers, few are as vivid or devastating as what befell African-American janitors 25 years ago in Los Angeles. After seeing steady gains through the work of their union, SEIU Local 399, the janitors were making a solid $12 an hour in 1983 (equivalent to about $24 an hour today). Then non-unionized companies using workforces 94% made up of illegal immigrants earning less than $4 an hour stole away the best contracts. The result, according to a Government Accountability Office report, was that unionized black janitors saw their ranks collapse from 2,500 in 1977 to 600 in 1985--with only 100 still making top wages. Yet the woes of such groups may get lost in the wash in large economic studies.

Acknowledging immigration's impact on low-skilled workers is not a call to close U.S. borders, deny our heritage as a nation of immigrants, or ignore immigration's compellingly positive effect on prices and productivity. Rather, it is recognition that, as with aspects of trade, we need to offset the harm that tends to concentrate on those who are already most vulnerable to economic change.

For low-income workers affected by immigration, buffering the costs could mean raising the minimum wage or expanding effective programs for at-risk minority youths, like the Job Corps, which takes disadvantaged kids out of their neighborhoods for intensive training and education. Business advocates as well as advocacy groups for Hispanics and African Americans might also propose boosting the earned income tax credit both for individuals and for families with more than two children. Today this subsidy doesn't provide extra help for larger families or offer more than a few hundred dollars to the childless working poor. Enhancing it might keep these folks out of poverty, compensate for wage losses they may suffer from greater immigration, and provide a stronger incentive for them to stay in the workforce. Addressing real harms to vulnerable workers is a far better course than either turning our backs--or shutting America's doors.

GENE SPERLING is a former National Economic Advisor, Senior Fellow at the Center for American Progress, and author of The Pro-Growth Progressive (2005).