THE UNITED STATES needs a radical reform of its immigration laws. We need to shift from a policy based on nepotism to one that evaluates applicants on the basis of skills. The realities of the 21st century will demand that change: We will be in sharp competition not only with Japan, but with dynamic technological competitors in a United Europe and the newly industrialized countries on the periphery of Asia.

Engraved on the base of the Statue of Liberty is Emma Lazarus's poem "The New Colossus." It reads, in part, "Give me your tired, your poor, your huddled masses yearning to breathe free." These inspiring words once matched the needs of the nation as they welcomed millions of immigrants to New York Harbor. In the late 19th and early 20th centuries, America was transformed from a largely self-sufficient agrarian economy into a mighty industrial power with the most technologically advanced agriculture and an industrial sector second to none.

But we can never return to such a policy, for two reasons. First, legal immigration of several million predominantly low-skilled workers from less developed nations would have devastating effects on the earnings of lower-skilled workers born in the United States, particularly those already disadvantaged as a result of minority status. And it would only exacerbate the increase in income inequality found in Census Bureau data over the past decade -- a development largely due to widening wage differentials between more highly skilled and lesser-skilled workers.

Second, the demands of modern society have changed dramatically. Economic growth and development, and subsequent increases in the quality of life, will be determined by the technological sophistication and innovative skills of the workforce -- not its manual skills. The world economy is becoming one market and only the most efficient and innovative will prosper.

Yet in one area of international competition -- immigration -- the United States is purposely restraining itself, despite a natural advantage. Millions of people in countries around the world see the United States as the land of seemingly unlimited opportunity for personal freedom, economic advancement and a better future for their children.

Facing the New Reality Nonetheless, we give the advantage to our competitors. Here's how: In fiscal year 1989, legal immigration totaled about 600,000 individuals (excluding the 490,000 illegal aliens receiving amnesty). Of these, over 435,000 were relatives of people already in the United States as citizens or resident aliens, and another 84,000 had been admitted as refugees. Of the kinship admissions, about 218,000 were the immediate relatives (spouse, minor children or parents) of adult U.S. citizens who can enter without any numerical limits.

The remainder entered under four of the six "preference" categories in immigration law reserved for other relatives: adult unmarried children of U.S. citizens (over 13,000 in 1989); spouses and unmarried children of resident aliens (nearly 113,000); married children of U.S. citizens and their families (nearly 27,000); and siblings of U.S. citizens and their families (over 64,000). Immigrants entering under the kinship or refugee categories are not screened for skill level, professional status or likely economic success. The unskilled and the highly skilled compete on an equal footing.

The other two preference categories (the administratively cumbersome third and sixth) are in principle reserved for professional and skilled workers. These visas are limited by law to 54,000 annually; thus one might infer that about 10 percent of immigrants are highly skilled. But things are not always what they seem. Nearly 60 percent of the 54,000 occupational preference visas in 1989 were used by the spouses (generally lesser-skilled individuals) and dependent children of applicants. So it would seem that fewer than 22,600 immigrants, or less than 4 percent of non-amnesty immigrants, were skill-tested in 1989.

But again, appearances are deceiving. The sixth preference for "skilled workers and other workers in labor shortage" is a vehicle for the entry of cooks, baby sitters and other low-skilled workers. About half of the recipients of sixth-preference visas in 1989 (excluding spouses and children) were for non-professional service occupations.

So how many immigrants admitted annually are really selected because of their high skills or professional occupation? Probably between 15,000 and 20,000 -- or 3 percent of all immigrants (excluding amnesty cases) and 6 percent of immigrants declaring an occupation.

Of course, one might ask: Does it really matter under which bureaucratic label an immigrant enters, since surely some immigrants admitted under other criteria have professional skills? The answer is yes. Using 1988 data (to avoid the confounding effect of the amnesty cases in 1989), of the nearly 45,200 immigrants reporting professional and technical occupations, nearly 8,000 entered under the "professional" (third) preference, over 1,700 entered under the "skilled worker" (sixth) preference and about 35,500 entered under all other categories. That is, only 11 percent of those entering under other categories who declared an occupation were professionals; the actual proportion may be even smaller because many of those who do not declare an occupation do in fact enter the labor market.

However, we can be more successful in the international competition for skilled immigrants by changing the focus of national policy. Even the current system of occupational preferences, with all its faults, does result in the selection of more highly skilled immigrants. In 1988, the occupational preferences provided nearly one-half of all immigrating mathematical and computer scientists, nearly four out of 10 natural scientists, over one-third of the engineers, one-third of the university professors and three out of 10 nurses.

To be sure, the unrestricted immigration of the immediate relatives of U.S. citizens (spouse, aged parents, minor children) should continue for obvious humanitarian reasons. This group currently numbers about 220,000 per year. There should be a generous policy for bona fide refugees, as distinct from "economic refugees" and people who just want to get out of a dangerous place. The annual number in this category will vary with political circumstances in other countries, but about 100,000 per year is realistic for the foreseeable future. For all others, perhaps 400,000 to 600,000 per year, the relevant question asked by the immigration authories should be: "What can you contribute to the American economy?"

In that regard, we can imitate our most effective competitors, Canada and Australia. They rely on a "point system" whereby visa applicants are awarded points on the bases of characteristics known to enhance the successful economic adjustment and productivity of immigrants. For the United States, these would include schooling, professional experience, prime working age and fluency in English. Points should also be awarded if an accompanying spouse has professional skills. Applicants receiving more than a threshold number of points would be given visas for themselves, their spouses and minor children. This threshold could be varied over time to maintain desired levels. Furthermore, a more explicit recognition of the importance of immigrant entrepreneurship should be made by creating an investor category -- a temporary visa would be granted for a large job-creating investment in the U.S. which would become a permanent visa if the enterprise survives for at least five years.

The critics may say a skill-based point system is racist. It is simply not true that a skill-based system will favor Europeans over Third World peoples. In fact, the American experience with occupational preferences, and the Canadian and Australian experiences with their point systems, appear to "favor" Asians -- Indian doctors, Filipino nurses, Japanese economists, and so on.

Setting New Priorities Congress is aware that current law does not satisfy the needs of the American economy. In this session of Congress both houses have considered major immigration reforms. The Senate has adopted a point-system approach, while the House last week passed a bill focusing on a targeted-labor approach.

The Kennedy-Simpson bill (S.358), which passed the Senate last year, would be a first step in introducing an effective point system. While increasing the number of visas for kinship immigration and the occupational preferences, S.358 would also introduce a point system for "selected immigrants" with an annual ceiling of 54,000 visas. The spouses and unmarried children of selected immigrants would not count against this total, but rather would have to apply under the sharply expanded (from 72,200 to 148,000 visas per year) kinship category (second preference). Points would be awarded on the basis of age, education, training, work experience and occupational demand. In addition, up to 6,800 visas would be made available annually for those investing at least $1 million in businesses that would create at least 10 full-time jobs for individuals who are not family members.

Kennedy-Simpson is not flawless. Kinship criteria will still dominate immigration policy. The point-allocation system does not give sufficient weight to very high levels of skill, and it ignores the importance of fluency in English. A select set of countries will be given more favorable treatment in determining which immigrants qualify. Moreover, the bill retains the "occupational preferences" and provides points for occupations in "high demand" (a targeted-labor approach) under the mistaken view that the Labor Department knows better than the labor market how workers should be allocated among particular jobs.

In the House, a measure introduced by Rep. Bruce Morrison (D-Conn.) was approved last Wednesday. This too is a useful first step. It would replace the current occupational-preference visas with an "employment-based" program for 65,000 visas annually (75,000 from FY1997 onward). Their spouses and children would apply for visas under an expanded family preference category.

Rather than using a point system to evaluate skills, however, the Morrison bill provides for an administratively complex system for targeting immigrants to particular occupations and regions of the country that the Department of Labor deems to have "labor shortages." This targeted approach will enlarge the federal bureaucracy, but is less likely than the Senate system to augment the nation's skills.

Moreover, the House bill expands immigration opportunities based on kinship. As in Kennedy-Simpson, it would sharply increase from 70,200 to 150,000 the numerical limit on the immigration of the spouse and unmarried children of permanent resident aliens, including several million illegal aliens who are receiving amnesty under the 1986 legislation.

When the House and Senate versions go to conference later this month, the committee should produce a compromise bill that includes a large number of visas issued under a skill-based point system and restricts kinship visas to only the immediate relatives of U.S. citizens.

The United States has the opportunity to regain its position as a technological and industrial world leader and at the same time improve the standard of living of its population, particularly those from disadvantaged backgrounds and with few skills. We need to utilize all the tools on the public workbench for this purpose. Immigration is too important an activity to ignore because of mythologies that no longer serve the national purpose.

Barry Chiswick is chairman of the department of economics at the University of Illinois at Chicago.