Border Security, and International Law
Committee on the Judiciary
U.S. House of Representatives
This is what Americans imagine our employment-based immigration categories to be. Even many of those concerned about the harmful impacts of low-skilled immigration into a modern society still often take for granted that higher-skilled workers are beneficial – hence the title of this oversight hearing, “The Need for Green Cards for Highly Skilled Workers.”
But like everything else in immigration policy, skills- and employment-based immigration is not what it seems. Once we peel away the misconceptions, we find that the “highly skilled” workers in question aren’t really that highly skilled, and the “need” for them has little to do with the national interest and much to do with firms seeking cheaper and more compliant workers. In fact, the employment-based immigration category with the highest standards, the category that really does select for the best and brightest around the world, is never fully used, precisely because there are so few people in the world who have such extraordinary abilities.
First a few numbers. In FY 2007, 162,176 foreigners were granted legal permanent residence in the United States in the five employment-based categories. More than half of these – 85,030 – are in the third employment-based category, or EB-3, for skilled workers, professionals, and others, though a majority (48,275) of those green cards are actually for the spouses and children of such workers. It’s this category that is at the center of the discussion about the “need” for “highly skilled” workers.
Research shows that, contrary to the claims of tech-industry lobbyists, these workers are not the best and brightest, the cream of the crop, the global elite of talent. Dr. Norman Matloff, a professor of computer science at the University of California, Davis, recently calculated the premium paid to foreign workers in science, technology, engineering, and mathematics whose employers are petitioning for green cards. He did this by computing the ratio of their salaries to the prevailing wage for that occupation. In a market economy, if these foreign workers were indeed outstanding talents they would be paid accordingly, with wages far above the prevailing wage.
They’re not. In his report (“H-1Bs: Still Not the Best and the Brightest,” May 2008, http://www.cis.org/sites/cis.org/files/articles/2008/back508.html), Dr. Matloff called the ratio the “Talent Measure” – the higher the number, the greater the premium employers were paying for the worker’s talents compared to the wage paid to other workers in the same field with comparable experience. Dr. Matloff found that the Talent Measure was at or near 1.0 for virtually all the professions and tech firms he studied – i.e., they are average workers in their fields. (By definition, the ratio cannot be lower than 1.0, since employers are barred law from paying below the prevailing wage.) He concluded that “the vast majority of the foreign workers – including those at most major tech firms – are people of just ordinary talent, doing ordinary work. They are not the innovators the industry lobbyists portray them to be.”
What’s more, Dr. Matloff found that the large majority of these foreign workers are hired at the two lowest levels of ability, according to the Labor Department’s classification, and thus unlikely to be contributing much to innovation. In fact, most of the large tech firms had virtually no workers in the highest skill level, where innovators are most likely to be found, despite the fact that it is these very firms which argue that innovation depends on their ability to import foreign workers.
And looking at H-1B visas, the “temporary” visas that serve as a stepping-stone to employment-based immigration, paints the same picture. Software expert John Miano has looked at the employer applications for H-1B workers, and found that the overwhelming majority are not highly skilled for their occupations and are paid well below the median for comparable American workers. And he concluded in "Low Salaries for Low Skills: Wages and Skill Levels for H-1B Computer Workers," April 2007, http://www.cis.org/sites/cis.org/files/articles/2007/back407.html :
The newly available data on skills suggest one of two things is happening, neither of which is consistent with the claims of employers pushing for the expansion of the program. Either the H-1B program is used primarily to import relatively less-skilled, entry-level, or trainee workers (and thus is of dubious value to the American economy), or employers are lying about these workers’ skills in order to suppress their wages.
In other words, unless tech companies are engaged in a massive conspiracy to lie to the government, the current skilled immigration flow is not made up mainly of Einsteins, but rather ordinary workers for their fields.
So what should our skills-based immigration program look like? The first thing to keep in mind is that in today’s America “skilled” doesn’t mean what it did a century ago. Then, a high-school graduate was unusual, and a college graduate was rare indeed; in 1910, only 13 percent of American adults had graduated high school and fully one-quarter had no more than five years of schooling. At the same time, only 2.7 percent of Americans had college degrees. Today, with Americans having attainted dramatically higher levels of education, any foreigner asking to be admitted based on high skills would need to demonstrate even greater levels of accomplishment – acquired abroad, without subsidy from the American taxpayer – to justify admission.
Also, the admission of large numbers of technical workers would have a perverse long-term effect – it would decouple American business from the American educational system, since companies could simply import workers from abroad. Business is the single most important lobbying group at the federal, state, and local level, and if it’s true that American students are not being adequately trained for the technical jobs of tomorrow, mass skilled immigration frees American firms from the need to pressure lawmakers and schools for whatever educational reforms might be needed to address this problem.
Thus there’s no reason any employer should be permitted to make an end run around our vast continental labor force of more than 150 million people unless the prospective immigrant in question has unique, remarkable abilities, and would make an enormous contribution to the productive capacity of the nation.
Perhaps the simplest way to approach this would be to admit anyone who scores above 140 on an English-language IQ test. A more likely approach would be to keep part of the current system, but limit skilled immigration to a portion of the first employment-based category (EB-1) – specifically, “aliens of extraordinary ability” and outstanding professors and researchers. Congress intended this to be the real cream-of-the-crop category, intended “for the small percentage of individuals who have risen to the very top of their field of endeavor.”
(The EB-1 category also gives green cards to multinational executives or managers, people who are not necessarily, as anyone who reads the business pages knows, the best and the brightest.)
These two groups – “aliens of extraordinary ability” and outstanding professors and researchers – accounted for about 11,000 green cards last year (including spouses and children). This is the real Einstein immigration, and these are the only foreigners who should be granted permanent residence based on skills or employment. We could do without a numerical cap altogether, so long as standards for admission are set sufficiently high, but to prevent “bracket creep,” as it were, it might be best to cap such immigration at 15,000 per year. After all, how many geniuses are there in the world?