IN the prosperous suburbs of San Diego, Calif., many families hire illegal aliens as live-in maids and nannies. The going rate is $150 a week -- not even the minimum wage.
These householders benefit directly from immigration in that they get cooking, house-cleaning, and child care cheap. Some owners of garment factories and farms that hire migrants, illegal or legal, benefit in other ways.
But the net economic benefit to native workers and taxpayers of the United States of the relatively large flow of immigrants is tiny, according to research by George Borjas, an economist at the University of California, San Diego. ''I was shocked,'' he says.
In a new National Bureau of Economic Research paper, Mr. Borjas calculates that the benefit probably runs about $6 billion: only 0.1 percent of the $6 trillion economy he assumes, or about $26 per native-born person per year. Under different assumptions, the benefit could be as high as $20 billion -- still small.
Some economists argue that since new immigrants must eat, sleep, and be housed, this demand soon creates new jobs, income, and tax revenues that quickly overwhelm any special costs in the way of higher spending on social welfare, education, or other government services. But Borjas couldn't find an economic study that mathematically balanced benefits and costs. So he set out to do it.
The issue is highly relevant politically. Two Republican presidential candidates, California Gov. Pete Wilson and Pat Buchanan, have made an issue of curbing immigration. Earlier this month, Mr. Buchanan called for at least doubling the Border Patrol, a double security fence along the most porous 70-mile section of the US-Mexico border, a five-year moratorium on virtually all immigration, and denial of automatic citizenship to children of illegal immigrants born in this country.
Borjas, who is one of several economic advisers to Governor Wilson, found one serious and substantial economic impact of immigration: It causes a large redistribution of wealth from labor to capital. Native workers, those who compete with immigrants for jobs, lose about $114 billion, or 1.9 percent of gross domestic product. Borjas finds, contrary to some other studies by economists, that immigrants do depress the wages of native workers. Users of immigrants -- ''capitalists'' in Borjas's terminology -- gain about $120 billion, or 2 percent of GDP.
''That is a huge transfer of wealth,'' Borjas says. Immigration, he adds, is one reason working-class Americans, especially those with high school or less education, have seen their incomes decline after inflation and the well-to-do with higher education have prospered during the 1980s and early 1990s.
Businesspeople often say immigrants tend to take the poor-paying, crummy jobs that native-born Americans won't accept, for example, as stoop laborers picking field crops.
What these people are really saying, says Borjas, is that Americans won't accept those jobs at the low wages offered. Without immigrants, the wages would have to be hiked sufficiently to attract natives or, if that is uneconomic, the businesses would go out of business. ''Maybe we shouldn't be growing tomatoes,'' he says. ''Maybe we should import them.''
Indeed, Borjas holds that natives tend to move out of areas where immigrants choose to live, partially because of low-wage pressures, resulting in what has been called ''the new white flight.'' If immigrants didn't lower wages of native workers, there would be no net benefit to the nation from immigrants, he says.
In recent years, Borjas and other economists have studied the characteristics of immigrants in some detail. They found that recent immigrants will remain economically disadvantaged throughout their working lives; that this disadvantage may be partially transferred to their offspring; and that recent immigrants are more likely to need welfare than are natives.
These trends arise from a decided decline in the educational attainment of successive immigrant waves relative to native-born Americans in recent decades. The typical new immigrant had 11.1 years of schooling in 1970, compared with 11.5 years for the typical native worker. By 1990, the numbers had changed to 11.9 for new arrivals in the US and 13.2 years for natives.
Borjas says a more selective immigrant policy, boosting the average skills of immigrants, would increase the economic benefits.