The U.S. Takes the Wrong Immigrants

By George J. Borjas

The Wall Street Journal
(Copyright (c) 1990, Dow Jones & Co., Inc.)

Just as countries compete in a world-wide market where goods and services are exchanged, they also compete in the market for people. The U.S. accepted about 600,000 legal immigrants annually during the 1980s, not including the 2.5 million persons who applied for amnesty under the provisions of the 1986 Immigration Reform and Control Act. Including illegals, more immigrants were admitted during the 1980s than in any other decade in American history.

By presenting a specific set of economic opportunities, and by pursuing an immigration policy that prevents the entry of some persons but encourages the entry of others, the U.S. makes a particular type of offer in the "immigration market." The evidence is clear that the American offer is becoming progressively less attractive to the world's most talented people.

Since the immigration reforms of 1965, U.S. immigration law has encouraged family reunification and discouraged the arrival of skilled immigrants: 75% of legal immigrants in 1987 were granted entry because they were related to an American citizen or resident, while only 4% were admitted because they possessed useful skills. As a result, the skills of immigrants entering the U.S. have declined sharply over the past few decades relative to the skills of natives. Newly arrived immigrants admitted in the late 1950s had about half a year more schooling than natives did and earned about 8% less per hour. Immigrants admitted in the late 1970s had 0.7 fewer years of schooling and earned about 17% less than natives.

The U.S. competes for immigrants with several other countries, but its main rivals are Australia and Canada. Among them, these three countries account for two-thirds of the world's legal immigration. But since 1965, Australia and Canada have succeeded better in drawing the most talented immigrants. While the predicted lifetime earnings -- the best indicator of the value of the technical and intellectual skills that immigrants bring with them -- of the immigrants who entered the the late 1970s were 30% lower than those of American natives, immigrants to Canada who arrived in Canada during that period are projected to earn only 13% less than Canadian natives. Late 1970s immigrants to Australia will likely earn about the same as Australian natives.

By contrast, those immigrants who arrived in the U.S. in the early 1960s will have lifetime earnings just 7% below those of natives -- a figure exactly to the performance of contemporary immigrants to Australia and only very slightly worse than the 3% less that early 1960s immigrants to Canada will earn.

The flagging performance of immigrants to the U.S. is due in part to the changes in their national origins. The national origin groups that dominate the immigrant flow today do relatively less well in the labor market than the groups that dominated earlier flows. For instance, the predicted lifetime earnings of immigrants from Canada, Germany or Britain are about 20% higher than those of natives; immigrants from India or Korea earn about 7% less than natives; and immigrants from the Dominican Republic, Jamaica or Mexico earn 30% to 40% less than natives.

One ready explanation for these disparities is that the skills of workers originating in advanced, industrialized economies are more easily transferable to the U.S. labor market than are the skills of persons from less developed countries. But a subtler cause is at work too: It is the most skilled workers who wish to leave such countries as Sweden and Britain.

Because of the wage structure and redistributive income policies in the European social democracies, the income gap between the highly skilled and the less skilled is small. Highly skilled workers are not well rewarded and the less skilled are protected from poor labor market outcomes. This creates an incentive for those highly skilled people to emigrate. In poor countries, on the other hand, the wage gap tends to be very large, and it is the less skilled who have the most incentive to leave.

The problem with unskilled immigrants is not, as is often supposed, that they reduce the living standards of native workers. A 10% increase in the number of immigrants decreases the average native wage by at most two-tenths of 1%, and has little effect on the unemployment rate of practically all native groups, including the black poor.

In fact, even a large and unexpected increase in the supply of immigrants -- such as the arrival of 125,000 Cubans into the Miami area during the 1980 Mariel boatlift -- does little harm to native earnings and employment. The trend in the wages and unemployment rates of both black and white natives in Miami over the 1979-1985 period differs little from the trends observed in a number of comparable cities, none of which received anything like the Miami influx.

But though unskilled immigration doesn't deprive natives of jobs or wages, it does have other costs. The poverty rate of newly arrived immigrants admitted in the late 1960s was five percentage points higher than that of natives, while the poverty rate of immigrants admitted in the late 1970s was 18 percentage points higher. Similarly, immigrant households admitted in the late 1970s are about two percentage points likelier to receive public assistance than immigrants who arrived in the early 1960s.

The differences in poverty and welfare recipiency rates among national origin groups are huge. Only 7% or 8% of the immigrants from Britain or Germany are living below the poverty line, but 14% of Koreans, 26% of Mexicans, and 34% of immigrants from the Dominican Republic are. The native poverty rate is 12%. Similarly, only 5% of German and British households are on welfare, but 10% of Filipino households, 17% of Cuban households and 26% of Dominican households receive public assistance. Eight percent of native households live on welfare.

The shift toward a more unskilled immigrant flow increases the burden of income transfer programs. There really is a fundamental conflict between the welfare state and immigration. Before welfare benefits became widely available in the 1960s, prospective immigrants to the U.S. would make their decision based on a comparison of the economic opportunities available to them here and in their country of origin. Now they compare welfare opportunities too. Without welfare, the U.S. could open its borders entirely, knowing that those who were needed would stay, while those who weren't would probably leave.

National income and tax revenues are substantially lower than they would have been if the U.S. had attracted more skilled immigrants. If the people who immigrated in the late 1970s had been as skilled as those who came in the early 1960s, national income would be at least $6 billion higher and tax revenues would have increased by $1.5 billion per year.

In both Canada and Australia, visas are now allocated through a point system, which grades visa applicants in terms of educational attainment, age and occupational background. The presence of relatives in the country is only one factor among many. Canada and Australia also "sell" visas to persons who have sufficient financial resources to open businesses and create employment opportunities for natives. But it may well be, for instance, that the skilled persons who choose Australia or Canada would have preferred the more favorable tax policies or labor market conditions provided by the U.S. If American law were different, they might have come here instead.

Of course, a visa allocation system based on the applicant's ability to pay discriminates against people who lack these financial resources. Similarly, a point system discriminates against people who lack the favored skill characteristics. Any visa allocation system, however, is bound to lead to inequities, and the inequities that would be created by a merit-oriented immigration policy would be no more egregious than those associated with present or previous immigration policies. Throughout the first half of the century, the national origins quota system flatly prohibited the entry of Asians; current policy prevents the entry of most persons who do not have relatives in the U.S.

Immigration policy is inherently discriminatory. It selectively picks and chooses from the many applicants. It may stress national origin, or skills, or financial resources, or family ties, or any combination of characteristics that Americans deem economically and politically desirable, and consistent with the country's values and beliefs. Because there are only a limited number of visas, the policy has to restrict or prohibit the entry of many classes of persons. Inevitably, difficult choices must be made.

Mr. Borjas, a professor of economics at the University of California, Santa Barbara, is the author of Friends or Strangers: The Impact of Immigrants on the U.S. Economy, (Basic, 1990).