Richard Posner, University of Chicago Law School:

"I approach the issue of immigration reform (theoretical reform – neither Becker nor I are considering the political obstacles to radical changes in immigration law) somewhat differently. I begin by asking: why restrict immigration at all? The only answer I consider fully compatible with a market-oriented approach to social issues is that the immigrant might reduce the net social welfare of the United States, if for example he was unemployable or on the verge of retirement, or was a criminal, or was likely to require highly expensive medical treatment, or if he would impose greater costs in congestion or pollution than he would confer benefits, with benefits measured (crudely) by his income before taxes and by any consumer surplus that he might create. I assume that the welfare of foreigners as such does not enter into the U.S. social welfare function; but immigrants who create net benefits in the sense just indicated contribute to the strength and prosperity of the nation.

The problem of the "undesirable" immigrant—the immigrant who wants to free ride on the services and amenities that the United States provides its citizens—could be solved by means of a two-stage process. In the first stage, the prospective immigrant would be screened for age, health, IQ, criminal record, English language capability, etc.; the screening need not be elaborate. If the would-be immigrant "passed" in the sense that he seemed likely to add more to U.S. welfare than he would take out, he would be admitted without charge. If he flunked the screening test, an estimate would be made of the net cost (discounted to present value) that he would be likely to impose on the U.S. if he lived here and he would be charged that amount for permission to immigrate.

An alternative, less revolutionary, approach to screening out free-rider immigrants would be, first, to deny immigrants access to Medicaid and other welfare programs until they had lived in the United States for a significant period of time, and, second, to auction off a certain number of immigrant visas to the highest bidders. Immigrants willing to take their chances without access to welfare programs (not that all access could be denied—no one could be refused emergency medical treatment on a charity basis), and immigrants willing to bid high prices in an immigration auction, would be likely to be productive citizens, in the first case, and to cover any costs they would impose on the nation's health or other welfare systems, in the second case.

Either the more or the less revolutionary alternative would impose significant transition costs, but that would be true of any radical change in immigration policy. The obvious cost (though not really a cost, rather a redistribution of income) would be that by increasing the supply of labor, an immigration policy that made it easy for employable workers to enter the U.S. labor force would reduce wages in the labor markets that the immigrants entered. A closely related but subtler consequence is that the downward effect of large-scale immigration on wages (a short-run effect, in all likelihood) would complicate the process of determining the correct fee to prevent free riding: an immigrant who might be able to pay his way at the existing wage level might be unable to do so if the wage level fell as a result of massive immigration. Similarly, congestion and pollution externalities might increase at an increasing rate with massive immigration, requiring a further adjustment in the fee charged the "undesirables."

Either approach seems to me preferable to a flat fee for all would-be immigrants. A flat fee would not do away with the need to screen, since some would-be immigrants might impose net costs on the U.S. that were greater than the fee; that is why Becker's approach includes screening. The flat fee would exclude two types of immigrant that should, in a market-oriented approach, be admitted. One type would be "undesirables" willing and able to compensate the United States for the expected costs that they would impose–and so they would not be free riders after all; a very wealthy person on the verge of retirement would be an example of such an "undesirable." The second type would be highly promising would-be immigrants (for example, persons with a high IQ) who for some reason—perhaps because they reside in extremely poor countries—simply could not pay the down payment on the fee.

The fee would, it is true, increase government revenues, which may seem a plus. But it would do so at the usual cost of distorting the allocation of resources, in this case by excluding immigrants in the second class.

I note two complications. First, it may be desirable to adhere to the current policy of granting asylum to foreigners who are escaping persecution, even if they do not seem likely to be able to pay or to earn enough to cover the costs they'll impose on this country. My reason is not sentiment, but the fact that people who are persecuted tend to be either nonconformists or members of particularly successful minorities, and in either case they, or at least their children, are likely to be productive citizens even if their U.S. employment prospects are dim. Second, the United States in formulating immigration policy may have to worry about "brain drain," and, what may be more important, "leadership drain," from poor or unstable countries. For example, it would be highly unfortunate if all the Iraqis who have the ability and motivation to build a democratic, free-market society fled to the United States. Thus it may sometimes be in our national interest to exclude persons who would otherwise be highly desirable immigrants, in order to shore up forces or tendencies in their own countries that promote U.S. interests. However, I do not know how to mesh this concern with either my or Becker's proposals." [February 25, 2005; The Becker-Posner Blog has the post]