[Peter Brimelow writes: George Borjas is the outstanding immigration economist in the U.S., which has gotten him into trouble. This is from Afterword to the just-published paperback version of his latest book.]
It has been a year since Heaven's Door was published. And it has been an interesting year, both in terms of the reactions to my book and in terms of how the debate over immigration policy has evolved.
I was not surprised that my book generated interest among participants in the immigration debate and among political commentators. After all, the book addresses topics that are highly contentious and controversial—and these topics evoke a great deal of emotion among most Americans.
I was surprised, however, at the nature of the reaction—as represented by the reviews and comments published in major newspapers and magazines. To a large extent, the reviews can be easily allocated to two extreme camps: the favorable reviewers tended to like the book a lot, while the unfavorable reviewers disliked the book intensely. As with immigration itself, there seemed to be no middle ground, no subtleties over the type of book that I had written and the type of policies I had proposed. It was often difficult to imagine that the reviewers were referring to the same book. For example, Peter Skerry, who reviewed the book for the Washington Post, characterized Heaven's Door as a "tour de force in the economics of immigration," while Sylvia Nassar, who reviewed it for the New York Times, concluded that it was "impressively researched." In contrast, Jagdish Bhagwati, who reviewed it for the Wall Street Journal, declared that the "economics arguments...are critically flawed."
In retrospect, and given the fact that the immigrant experience forms an important part of the American psyche, I should have expected those extremes. I thought that Heaven's Door made a nuanced argument that could not be easily tagged as fitting into either extreme of the immigration debate. After all, the book stresses that there are both benefits and costs associated with immigration. The book also argued that if the United States wished to increase the economic benefits from immigration, the country should pursue a different type of immigration policy, a policy that would favor the entry of skilled immigrants. To be sure, I left little doubt as to whether I think the United States should consider economic considerations in setting immigration policy (it should!). Nevertheless, the subtleties quickly got lost in the mythological mist that enshrouds any discussion of U.S. immigration.
In reflecting over the reaction to Heaven's Door, it is clear (at least to me) that the book elicited the most negative responses from reviewers writing for conservative and libertarian publications, such as the Wall Street Journal and Reason. These publications typically support calls for unrestricted immigration to the United States. The editorial page of the Wall Street Journal, for example, is famous (or, depending on one's point of view, infamous) for repeatedly proposing a new constitutional amendment: "There Shall Be Open Borders."
From my perspective, the reaction exhibited by these publications represented a "circling of the wagons" mentality. Rather than evaluating the overall merits of the idea that some restrictions on immigration are probably better than none, and that it might be worth discussing which types of restrictions might be most beneficial, the response was to attack almost every fact reported in the book, to raise questions about the credibility of the findings, and to dismiss the whole enterprise as my misguided attempt to reflect on immigration policy when I failed to fully understand the (obvious?) implications of the fact that the United States is "a nation of immigrants."
Surprisingly, the reviews re-ignited the debate over an empirical observation that I thought had been settled beyond dispute. In particular, some reviewers doubted my claim that there has been a drop in the relative skills of immigrants who entered the United States in recent decades. The empirical validity of this finding, which I first published in the academic literature in 1985 and which has been re-examined by many social scientists in the past decade, is well established. In fact, a panel of scholars assembled by the National Academy of Sciences in 1997 reviewed the evidence, conducted additional empirical research, and concluded that the skills of new waves of immigrants "have been declining relative to that of native-born Americans. This decline appears across a number of measures, including education levels and wages."
Some reviewers claimed that there might not have been a drop in the relative skills of legal immigrants who entered the country in the past few decades. They argued that the trends revealed by the Census data were contaminated because the Census enumerates both legal and illegal immigrants (as well as such nonimmigrants as foreign students and business executives temporarily working in the United States). If the presence of illegal immigrants in these data biases the direction of the trends in immigrant skills, the evidence then provides little information about what happened to legal immigration and what should be done about it.
I doubt that the detractors have a valid point. As I showed in Heaven's Door (see Figure 2-4), the downward trend in the relative skills of immigrants remains even if one simply examines the trend among non-Mexican immigrants, on the presumption that a large fraction of the illegal immigrants are from Mexico and are relatively unskilled. Moreover, the panel assembled by the National Academy of Sciences itself conducted an empirical analysis that analyzed data collected by the Immigration and Naturalization Service (INS) and concluded that "the decline in the relative skills of the foreign-born over the last few decades is not due exclusively to illegal immigrants or nonimmigrants. The data suggest that the relative skills of legal immigrants have also been falling over the period."
A recent study conducted by Guillermina Jasso, Mark Rosenzweig, and James Smith (and published in a volume that I edited for the National Bureau of Economic Research) reexamined the INS data and seems to suggest that the absolute skill level of legal immigrants exceeded that of natives for much of the 1972-95 period.  It is worth pointing out, however, that the INS data are particularly ill-suited for documenting trends in immigrant skills. After all, the data do not report the immigrant's actual earnings or educational attainment; instead, one makes a guess about the person's economic potential by using information on the type of job the immigrant held before moving to the United States. For instance, if the immigrant was a medical doctor prior to migrating, one assumes that the immigrant will earn the average salary of doctors in the United States. Needless to say, there is a great deal of measurement error in this measure of the immigrant's potential skills. After all, the immigrant's pre-migration job may have little to do with what that person ends up doing in the United States—particularly if the occupation requires re-licensing (as in the case of medical doctors). Moreover, the Jasso-Rosenzweig-Smith study is itself inconclusive. The relative economic status of the immigrant population depends on the type of statistical fine-tuning that is made to adjust the data for the fact that the Census Bureau changed the way it coded occupations between 1970 and 1980. 
Some reviewers also claimed that it was "absurd" to conclude, as I do, that the net economic benefits from immigration are small, probably less than $10 billion a year. This estimate comes from a simple application of the widely used textbook model of a competitive labor market. This is the same model that is typically used to analyze the economic consequences of such government policies as minimum wages and payroll taxes. The market for ideas provides what is perhaps the most convincing argument in favor of my estimate. The immigration area, after all, is highly contentious. If it were that simple to show that the gains from immigration are huge, there is an audience ready and willing to buy such numbers. My estimates are so "absurd" that not a single academic study has concluded that they are higher—and some studies have concluded that they are lower.
The mischaracterization of the evidence was perhaps most apparent when it came to the issue of whether immigrants had a negative effect on the wages of workers already in the United States. The Wall Street Journal review, written by Jagdish Bhagwati, an economic theorist specializing in international trade issues, shows this tendency in a very striking way. Mr. Bhagwati dismissed my evidence ("Mr. Borjas's larger economic claim, that immigration drives down wages, does not survive scrutiny") by claiming that economists Gordon Hanson and Matthew Slaughter showed that U.S. wages are not driven down by immigration. The Hanson-Slaughter work actually demonstrates that immigration does not seem to create wage differentials between states that receive many immigrants and states that do not, a point addressed at length in Heaven's Door (see pages 73-78). The concluding paragraph of the Hanson-Slaughter study explicitly describes how their work relates to the question at hand: "The evidence that state-specific endowments shocks do not trigger state-specific wage responses does not imply that the United States overall had no wage response to increased immigration" (the emphasis is theirs).
The arguments that Mr. Bhagwati and other reviewers used to quickly dismiss the evidence on the adverse labor market impacts of immigration were further undermined by the unexpected entry into this debate of the authority on U.S. economic policy. In February 2000, Alan Greenspan, the chairman of the Federal Reserve, made some remarks linking immigration and inflation that are destined to change the nature of this debate. In Mr. Greenspan's words:
Imbalances in the labor market perhaps may have even more serious implications for potential inflation pressures. While the pool of officially unemployed and those otherwise willing to work may continue to shrink, as it has persistently over the past seven years, there is an effective limit to new hiring, unless immigration is uncapped. At some point in the continuous reduction in the number of available workers willing to take jobs, short of the repeal of the law of supply and demand, wage increases must rise above even impressive gains in productivity…In short, unless we are able to indefinitely increase the rate of capital flows into the United States to finance rising net imports or continuously augment immigration quotas, overall demand for goods and services cannot chronically exceed the underlying growth rate of supply.
Mr. Greenspan's pronouncement restates the obvious: the laws of supply and demand inevitably imply that a large increase in the supply of workers lowers wages and reduces inflationary pressures. Mr. Greenspan, in effect, partly credits the large immigration flows of the 1990s's for the U.S. economy's low inflation rate. Under Mr. Greenspan's astute interpretation, the fact that immigrants lowered wages need not be seen as an adverse impact of immigration; instead, it is a favorable consequence for it holds inflation in check. Many of those who, in the pre-Greenspan days, had argued strenuously that immigration had little impact on the economic opportunities of native workers now began to argue just as strenuously that immigration fueled economic growth by keeping wages and prices down.
Many of the unfavorable reviewers were so busy questioning the credibility of the facts reported in Heaven's Door that they seemed to miss the whole point of the book. The United States will inevitably attract many more immigrants than the country is willing to admit. As a result, choices have to be made. Current immigration policy benefits some Americans (the newly arrived immigrants as well as those who employ and use the services the immigrants provide) at the expense of others (those Americans who happen to have skills that compete directly with those of immigrants). Before deciding how many and which immigrants to admit, the country must determine which groups should be the winners and which should be the losers. Few of the detractors were willing to take on this question.
Finally, there is an amusing, if unintended, irony in some of the most extreme reviews—particularly those written from a libertarian perspective. Stuart Anderson wrote a review for Reason magazine that is worth singling out. Mr. Anderson dismissed my proposal for a skills-based point system by repeating my book's warning that the system would be difficult for bureaucrats to administer, and then added sarcastically: "As if bureaucrats are well suited to handle any labor market decisions" (his emphasis). I wonder if Mr. Anderson saw the incongruity in his statement. At the time he wrote the review, he was a high-ranking immigration apparatchik himself—the "director of immigration policy" for the Senate Immigration Subcommittee. His boss, Senator Spencer Abraham, led the charge—presumably with Mr. Anderson's advice—to alter the U.S. labor market for skilled workers by greatly expanding the number of H1-B visas granted to temporary high-tech workers. Does Mr. Anderson truly believe what he writes about bureaucratic incompetence?
In the year that followed the publication of Heaven's Door, the immigration debate was dominated by three unrelated events. The first was the media obsession with the saga of Elián Gonzalez, the six-year-old boy who was rescued from the Florida seas on Thanksgiving Day in 1999. His mother died during her attempted escape from Cuba, and Fidel Castro soon made the return of the boy a cause célèbre that dominated the airwaves and the front pages of newspapers for months. Regardless of how one feels about the merits of the case, the Elián Gonzalez saga made an important contribution to the future debate over immigration policy. It illustrated the awesome and arbitrary power that the INS has over the lives of many foreign-born persons residing in the United States. The INS has a great deal of discretion in deciding whether particular aliens should be allowed to remain in the country. And once the agency chooses to pursue a set of clearly defined goals, it seems to be able to achieve those goals with relatively little interference.
There is now, in fact, a healthy debate over the agency's competence in managing U.S. immigration policy. And Congress is seriously considering proposals that would break up the INS into two separate agencies. One agency would be responsible with border protection and enforcement of immigration regulations, while the other would provide naturalization services to legal immigrants already in the United States.
There is much to be said for the proposition that administrative incompetence rules at the INS. Consider, for instance, the agency's record during the Clinton years. At the beginning of Doris Meissner's tenure as head of the INS in 1993, the agency reported that there were just over 3 million illegal aliens residing in the United States. By the year 2000, there were 6 million. [VDARE note: More recent estimates suggest 12 million. But hey, who's counting?] This huge increase in the illegal alien population occurred despite much higher expenditures on border enforcement. The INS budget tripled from $1.5 billion to $4.8 billion during that time. One could easily argue that there has been a tremendous waste of resources at the agency in recent years—billions of dollars spent on enforcement with little to show for it.
The failure of the INS in enforcing existing immigration laws extends far beyond the agency's flawed stewardship of the U.S. borders. The INS has a huge backlog of persons who are already living in the United States and who are waiting to regularize their immigration status and get their green cards. This backlog increased from an "equilibrium" of about 121,000 persons in 1994 to over 1 million by 2000. The agency, it seems, cannot even keep up with the legal immigrant flow.
Finally, political events in Mexico will inevitably alter the nature of the dialogue over immigration policy in the United States. Over 7 million Mexican immigrants now live in the United States, and an additional 130,000 Mexican legal immigrants and 150,000 illegal immigrants enter the country annually. Soon after his election on July 2, 2000, President Vincente Fox proposed that the two countries share an open border within 10 years. There would then be free movement of people and goods.
The Mexico-U.S. wage gap is among the largest wage gaps found between any two contiguous countries in the world. The World Bank reports that an American manufacturing worker earns 4 times the salary of a manufacturing worker in Mexico and 30 times the salary of an agricultural worker. These income differences ensure that an open border would increase the number of Mexicans immigrants.
The new immigrants would probably resemble the Mexican immigrants already in the United States, whose poverty rate exceeds 33 percent. Firms that employ less skilled labor would gain from having access to an even larger pool of workers. But, as Mr. Greenspan's reiteration of the laws of supply and demand suggests, Mr. Fox's proposals would further erode the economic opportunities available to the most disadvantaged Americans. An open border between Mexico and the United States would surely benefit Mexico and Mexican immigrants. But would it be good for the United States?
In the end, the answers to many of the concerns that drive the debate over immigration policy hinge on the question that I pose at the beginning of Heaven's Door, a question that has yet to become a central issue in the modern immigration debate: What do the American people want immigration to do for the United States?
 See, for example, David Warsh, "Rainbow Didn't Just Happen," Boston Globe, October 3, 1999; Robert J. Samuelson, "Ignoring Immigration," Washington Post, May 3, 2000; and Alan B. Krueger, "Work Visas Are Allowing Washington To Sidestep Immigration Reform," New York Times, May 25, 2000.
 Peter Skerry, "How Immigration Re-Slices the American Pie," Washington Post, October 28, 1999; Sylvia Nassar, "A Gloomy View of Immigration," New York Times, October 10, 1999.
 Jagdish Bhagwati, "A Close Look at the Newest Newcomers," Wall Street Journal, September 28, 1999.
 James P. Smith and Barry Edmonston, The New Americans: Economic, Demographic, and Fiscal Effects of Immigration, Washington, D.C.: National Academy Press, 1997, p. 185.
 James P. Smith and Barry Edmonston, The New Americans: Economic, Demographic, and Fiscal Effects of Immigration, Washington, D.C.: National Academy Press, 1997, p. 196.
 Guillermina Jasso, Mark Rosenzweig, and James P. Smith, "The Changing Skills of New Immigrants to the United States: Recent Trends and Their Determinants," in George J. Borjas, editor, Issues in the Economics of Immigration, Chicago: University of Chicago Press, 2000, pp. 185-235. Ironically, two of the authors of this study (Jasso and Smith) are responsible for the section in the National Academy of Sciences report that uses exactly the same data to reach seemingly different conclusions.
 The INS has also commissioned the New Immigrant Survey, which provides detailed information on skills and earnings for the cohort of legal immigrants admitted in July and August of 1996. The mean education of this cohort is only slightly lower than that of natives; see Guillermina Jasso, Douglas S. Massey, Mark R. Rosenzweig and James P. Smith, "The New Immigrant Survey Pilot (NIS-P): Overview and New Findings about U.S. Legal Immigrants at Admission," Demography 37 (February 2000): 127-136. Like the Census data, the New Immigrant Survey suffers from a particular type of inclusion bias. The INS has a huge backlog of persons (436,000 persons in 1996) who are already living in the United States and who are waiting for their green cards. Many of these immigrants probably entered the country illegally, probably have lower-than-average skills, and would be included—and rightly so—in Census data. The data collected by the New Immigrant Survey may well have looked quite different if the INS had chosen to substantially reduce the backlog in the summer of 1996.
 Compare the evidence presented in Figures 5.1A and 5.1B (page 194) of the Jasso-Rosenzweig-Smith study. The unadjusted data show that the average occupational earnings of new immigrants are below those of native workers in 12 of the 14 years between 1978 and 1991; are above native earnings in 1992 and 1993; and then fall again below native earnings in 1994 and 1995. Once the authors make a particular (and somewhat arbitrary) adjustment for the change in occupational coding, the data suggests that the occupational earnings of new immigrants is at least as high as that of natives.
 Gordon H. Hanson and Matthew J. Slaughter, "The Rybczynski Theorem, Factor-Price Equalization, and Immigration: Evidence from U.S. States," National Bureau of Economic Research Working Paper No. 7074, April 1999, p. 29.
 Alan Greenspan, "The Revolution in Information Technology," Remarks before the Boston College Conference on the New Economy, March 6, 2000.
 For example, in 1994, Frank Sharry, the Executive Director of the National Immigration Forum, a pro-immigration advocacy group, claimed that immigrants "increase wages and mobility opportunities for many groups of U.S. workers" (Frank Sharry, "Admitting Newcomers is in the Nation's Best Interests," The Plain Dealer, Cleveland, Ohio, April 1, 1994). In 2000, he invoked Alan Greenspan to argue that immigrants "ward off higher inflation and higher interest rates" (Frank Sharry, "Immigrants' Contribution to America," Business Week, April 3, 2000).
 Stuart Anderson, "Muddled Masses," Reason Magazine, February 2000.
 My feelings over the Elián Gonzalez case are made amply clear in "Let Elián Remain Free," New York Times, January 12, 2000.
 For a more detailed discussion, see George J. Borjas, "Mexico's One-Way Remedy," New York Times, July 18, 2000.
June 10, 2001