“What distinguishes undocumented migrants is that they value being in the United States so highly that they are willing to take great personal risks, perhaps pay high fees to smugglers, and flout the law to get here.”

The immigration issue in the United States has created strange political bedfellows. Cosmopolitans on the left, business interests on the right, and libertarians who sympathize with both of these groups generally support more-open immigration. Opponents of a more-liberal policy include labor1 interests on the left, traditionalists or “law-and-order” advocates on the right, and the populists who occupy the political landscape where those interests intersect.2 Political compromise—if it is to be had at all—usually results in giving all parties a piece of what they want, perpetuating the byzantine nature of U.S. immigration policy. There are two policy areas in which advocates on both sides have compromised and appealed to a “common sense” perspective: policy on high-skill guest workers and on undocumented immigrants.

On the issue of high-skill immigration, the conventional view is that high-skill workers are in short supply, and that well educated guest workers offer greater benefits to the American economy than other kinds of immigrants because they advance scientific and technological frontiers. High-skill migrants are also considered non-threatening and unlikely to go on welfare. It seems like common sense to design special visa programs targeted towards these workers (as has been done in the United States) or give high-skill migrants special consideration in the visa application process (as in New Zealand and Canada).

The conventional view of undocumented immigration, by contrast, is that it is a threat to the social fabric, domestic labor markets, and even national security. When it comes to the question of undocumented immigrants, erstwhile proponents of liberal immigration regularly concede ground to the opposition and demand well-guarded borders, crack-downs on employers of undocumented immigrants, and deportation of offenders.

But the late economist John Kenneth Galbraith, himself an immigrant, once warned us about the problem with convention. He is credited with noting that “the conventional view serves to protect us from the painful job of thinking.” But immigration is an important enough issue that we need to go beyond convention and, however painful is the process, actually think. The conventional view of both high-skill and undocumented immigration is misleading. The economic case for giving high-skill immigrants an advantage in the immigration process is weak. Moreover, undocumented immigrants, far from being undesirable, signal through their efforts to enter and stay in the country that they are a valuable addition to American society.

High-skill migrants do not deserve special treatment

Over the years, Congress has served up an alphabet soup of high-skill immigration visas: F-1s for students, Optional Practical Training (OPT) status for recent graduates, H-1Bs for newly hired college educated guest workers, L-1s for college educated guest workers who are current employees abroad, and O-1s for guest workers with “extraordinary abilities.” Each visa has its own requirements and limitations, but what they have in common is that they are not available to low-skill foreigners who want to come to the United States and work. In the view of most people, deliberately tailoring access to other areas of American public life in this way would be wrong. We have done away with literacy tests at the voting booth, for example. Nevertheless, the practice is wildly popular in the case of immigration policy.

Figure 1. Labor Demand and Labor Supply: A Labor Shortage.

Figure 1. Labor Demand and Labor Supply: A Labor Shortage.



Advocates of visas that are available exclusively to well-educated workers usually justify their proposals with allusions to labor shortages in high-skill labor markets.3 Proponents of high-skill visas rarely present a rigorous explanation of what they mean by “labor shortage,” but the general claim is that technology firms, R&D labs, and universities are starving for talent that the labor market is not providing. When economists use the term “labor shortage,” we usually refer to a situation in which the number of workers demanded at the current wage exceeds the number of workers who are willing to supply their labor at that wage. The main reason shortages exist in a free market is that a government has imposed a price ceiling that is below the free-market price. The shortage of rental housing in New York City, for example, is due to government-imposed ceilings on rents. Figure 1 shows a case in which a ceiling on wages causes a shortage. The equilibrium price is W0. At a wage or salary of W0, the amount demanded and supplied both equal Q0. But if the government imposes a wage ceiling of WC, the amount demanded will be QD and the amount supplied Qs. There will be a shortage of QD minus Qs. In the absence of ceilings, prices tend to adjust. That’s true for the labor market also. Because wage ceilings in the United States are currently rare to non-existent, it is hard to believe that there really is a shortage in the economists’ sense.

Figure 2. Labor Demand and Labor Supply: An Inelastic Supply of Labor.

Figure 2. Labor Demand and Labor Supply: An Inelastic Supply of Labor.



More often than not, these advocates seem to be asserting that the high skill labor market is simply characterized by what economists would call an “inelastic” long run supply of highly skilled workers. Such a case is shown in Figure 2. At a wage rate of W0, there is no shortage: the amount of labor demanded equals the amount supplied. But with such an inelastic supply of labor, workers are not very responsive to changes in wages, so that the amount of labor supplied is almost fixed. Notice, in Figure 2, that if the demand for labor increased, the wage rate would increase a lot, but the amount of labor supplied would increase only a little. If inelastic labor supply is all that is meant by “labor shortage” then intervention cannot be justified on so-called “market failure” grounds since a market with inelastic supply is still a properly functioning market. Either some other market failure (perhaps a failure downstream) or some non-economic objective needs to be invoked to justify the special treatment of high-skill immigrants.

Many prominent economists, including John R. Hicks (1959), Walter Oi (1962), Richard Freeman (1975, 1976), and Austan Goolsbee (1998) have agreed that in the short run, labor supply curves for highly skilled workers are relatively inelastic.4 They note that it is difficult to substitute specialized workers across different fields and that in many cases training new workers takes a considerable amount of time. These factors imply that new high-skill workers in a given field cannot be brought online immediately, even with the enticement of higher wages. The argument that labor supply is unresponsive to wage increases in the long-run is a different claim entirely. This claim implies that high wages will never entice students and workers to change their plans and enter new fields. Because we find it so implausible and, also, virtually lacking in evidence, few economists believe that in the long run the supply of high-skill workers is inelastic.5

Whether a proponent of the high-skill labor shortages view is suggesting that a proper labor shortage exists (Figure 1), or that students and workers do not respond to wage signals (Figure 2), one implication of an increase in demand for these workers is similar: a substantial long-run increase in the amount of labor supplied will not occur. In the case of a labor shortage, an increase in demand will not lead to an adjustment in the wage or the employment level. If demand is inelastic in the long run, we would expect to see wage increases but no subsequent large increases in the amount of labor supplied.

Available labor market data on a variety of high-skill occupations suggest that the worst fears of the proponents of high-skill visas generally do not hold true. Recently, my colleagues and I have analyzed earnings and employment data for electrical engineers, petroleum engineers, programmers, and information technology (IT) occupations. We found that these markets have inelastic short run supply, but that in the long run workers do respond to wage fluctuations.6 We reached this conclusion by investigating how educational institutions and labor markets responded to large exogenous swings in demand, like the shock to oil and gas production in the 2000s, or the bust of the dot-com bubble in the year 2000. In each case, wages responded first, increasing substantially in the case of increased demand (as in the market for petroleum engineers), and falling or stagnating when demand was weak (for example, after the dot-com bust). After a short lag, graduation rates for students in these field responded to wage signals, with increasing enrollments and graduations when wages were growing and higher rates of exit when wage growth faltered. Eventually, as a result of changes in graduation rates and the rate at which workers persisted in these professions, employment in high-skill occupations responded to wage rates. This repeated pattern suggests that neither labor shortages nor long-run labor supply inelasticities pose a problem for employers of high-skill workers.

Our work is not unique in concluding that the long-run labor supply of high-skill workers is not highly inelastic. Ryoo and Rosen (2004 p. S110) found in a widely cited paper that in engineering “supply and enrollment decisions are remarkably sensitive to career prospects.”7 Several decades before, in some of the earliest work on high-skill labor markets, Richard Freeman found that physicists and lawyers responded to price signals, albeit with a lag.

These conclusions are important because they go to the heart of the argument for visas available exclusively to high-skill workers. The evidence suggests that the labor market for these workers functions much as we would expect any labor market to function. So why should these workers be treated differently in immigration policy than other workers?

Undocumented immigrants are exactly the sort of people we should be welcoming

Most Americans are broadly pro-immigration because they have been convinced of the benefits of a “melting pot” and because they are conversant enough in their own family histories to know that their parents, grandparents, or great grandparents were likely immigrants too. However, a common addendum to this support for immigration is a deep disapproval of undocumented (or, more pejoratively, “illegal”) immigrants.8 In addition to the obvious violations of American immigration law, undocumented immigrants are accused of violating that old grade school taboo—don’t cut in line!—thus reaping the benefits of immigration that law-abiding immigrants sometimes have to wait years to enjoy. Once appeals to legality and fairness have been exhausted, criticisms of undocumented immigrants quickly descend into the realm of the parochial, the protectionist, and even the prejudiced: “they’ll change American culture,” “they’ll take our jobs,” “they don’t learn English,” or “they’re dangerous.” In a nutshell, Americans tend to embrace immigrants, but do not think undocumented immigrants are the sort of people we should be welcoming into the country. An important question for evaluating these claims is whether or not the pool of undocumented immigrants is composed mainly of the sort of people most of us want to have here.

For more information, see Immigration, by George J. Borjas in the Concise Encyclopedia of Economics and An Economic Case for Immigration, by Benjamin Powell, June 7, 2010, Library of Economics and Liberty.

When economists think about the composition of immigrant populations, they tend to focus on what is known as “self-selection.” (For a seminal treatment of self-selection, see Borjas, 1987.)9 Immigrants are not chosen randomly from around the world to come to the United States. Instead, individuals around the world consider the costs and benefits of migration relative to staying in their home country. They weigh these costs and benefits against the various constraints they face—the money available for travel, the details of immigration law, and their own skills—and decide to migrate or not depending on their evaluation of what option is best. They are therefore said to “self-select” into migration. Self-selection processes shape who decides to migrate and who does not. If, for example, low-skill workers in Mexico face higher net benefits from migrating to the United States than higher-skill workers in Mexico and if the number of low-skill Mexican workers exceeds the number of high-skill Mexican workers, we would expect most Mexican immigrants to be relatively low-skill.

So the question at hand is, what sort of people self-select into undocumented migration to the United States? Undocumented migrants are primarily from Latin America and, therefore, have many of the same characteristics as other Latin American migrants. What distinguishes undocumented migrants is that they value being in the United States so highly that they are willing to take great personal risks, perhaps pay high fees to smugglers, and flout the law to get here. This is presumably not the case for non-immigrants and legal immigrants, who consider these costs to be greater than the benefits of being in the United States without documentation. Therefore, an immigrant’s status—whether he is documented or undocumented—reveals a great deal of information about that immigrant and how much he or she values living in the United States. High costs of migration mean that migrants have “skin in the game,” and that only immigrants who enjoy the highest net benefits put in the effort to enter the country.

An undocumented status may communicate other information as well, of course. Willingness to flout immigration law may signal a more fundamental tendency towards criminal activity. Although this might apply in some cases, most undocumented immigrants try hard to stay in this country and therefore try to avoid criminal behavior. Career criminals can break the law in their home country, after all. What would be the point of a risky migration to an unfamiliar country to engage in activities that could easily get you deported? For migrants in the illegal drug business the profits may be worth the risk, but for the vast majority of ordinary undocumented immigrants these concerns simply do not apply.

If undocumented immigration status signals a high value of migration and a strong desire to live in the United States, then amnesty for undocumented immigrants who have come here despite an increasingly militarized border and escalated enforcement efforts is a far more palatable policy option than it is often portrayed to be. Maintaining the immigration restrictions that create undocumented immigrants in the first place is obviously an inefficient way to reveal this information: undocumented individuals cannot live completely fulfilling lives while their legal status is in limbo. However, given the reality of a large undocumented population, the process of migrant self-selection should allay concerns about the wisdom of an amnesty. After all, how many native born Americans would go to such lengths to get into the United States? These are largely people who are worth embracing as fellow Americans.


A market-based immigration policy is practically by definition a liberal immigration policy, but even within the framework of relatively open borders, many pro-immigrant voices insist on differentiating between potential migrants on the basis of education level or legal status. The conventional wisdom holds that well educated immigrants deserve greater access than less educated immigrants and that undocumented immigrants should be assimilated only after immigrants with valid visas or green cards (if at all). Both views can be challenged using the insights of economics. High-skill labor markets, much like other markets, function according to the laws of supply and demand and, therefore, do not need a differential boost from the government. In the case of undocumented immigrants, self-selection mechanisms suggest that a migrant’s undocumented status communicates important information about the high value that they derive from living in the United States. If more people thought about these issues as economists tend to, the conventional wisdom would shift towards support for a more broadly welcoming policy that treats immigrants of all backgrounds equally.


The reader may think that this refers to unions, but it is broader: many labor protectionists are not necessarily associated with unions.

In reality, the political fault lines around immigration are even more complicated than this. Prominent libertarians in public life such as Ron and Rand Paul, as well as libertarian intellectuals such as Hans Herman Hoppe and Anthony de Jasay have expressed opposition to immigration liberalization. At the same time, some conservatives see a melting pot as an essential element of the American tradition or as a component of faith-based conceptions of social justice and, therefore, often use the language of cosmopolitans on the left to advocate open immigration. Such conservatives range from George W. Bush with his “compassionate conservatism” to Catholic Charities U.S.A. and other Catholic non-profits. These secondary fractures in the political landscape serve only to reinforce how complicated the politics of immigration can be.

Rothwell, Jonathan, and Neil G. Ruiz. 2013. “H-1B Visas and the STEM Shortage”.

See: Hicks, J.R. 1959. “Supplementary Note A: The Theory of Wage Differentials” in Essays in World Economics. The Clarendon Press: Oxford, UK. pp. 247-250; Oi, Walter. 1962. “Labor as a Quasi-Fixed Factor.” The Journal of Political Economy. 70(6): pp. 538-555; Freeman, Richard. 1975. “Supply and Salary Adjustments to the Changing Science Manpower Market: Physics, 1948-1973.” American Economic Review. 65(1): pp. 27-39; Freeman, Richard. 1976. “A Cobweb Model of the Supply and Starting Salary of New Engineers.” Industrial and Labor Relations Review. 29(2): pp. 236-248; Goolsbee, Austan. 1998. “Does Government R&D Policy Mainly Benefit Scientists and Engineers?” American Economic Review. 88(2), pp. 298-302.

A notable exception is Paul Romer (2001), a macroeconomist whose work on economic growth focuses on research and development and technological change. See: Romer, Paul. 2001. “Should the Government Subsidize Supply or Demand in the Market for Scientists and Engineers?” Innovation Policy and the Economy. Vol. 1, pp. 221-252.

Salzman, Hal, Daniel Kuehn, and Lindsay Lowell. 2013. “Guestworkers in the High Skill U.S. Labor Market: An Analysis of Supply, Employment, and Wage Trends”. Economic Policy Institute Briefing Paper #359; and Salzman, Harold, Leonard Lynn, and Daniel Kuehn. forthcoming. “Dynamics of Engineering Labor Markets: Petroleum Engineering and Responsive Supply,” in The U.S. Labor Market for Engineers and the Global Economy. NBER and University of Chicago Press. Kuehn, Daniel and Harold Salzman. forthcoming. “Setting the Stage: Background on Engineering Graduates,” in The U.S. Labor Market for Engineers and the Global Economy. NBER and University of Chicago Press.

Ryoo, Jaewoo, and Sherwin Rosen. 2004. “The Engineering Labor Market.” Journal of Political Economy. 112(1): pp. S110-S140.

“Undocumented” is often preferred to “illegal” because most undocumented immigrants have not committed a crime: they have only committed an administrative violation.

Borjas, George. 1987. “Self-Selection and the Earnings of Immigrants.” American Economic Review. 77(4): pp. 531-553.


*Daniel Kuehn is an adjunct instructor and doctoral student in American University’s Department of Economics. His work on high-skill labor markets and high-skill immigration policy has been funded by a generous grant from the Alfred P. Sloan Foundation.