American immigration laws heavily restrict highly skilled immigration. The employment-based green card, largely designed for highly skilled workers, has an annual cap of 140,000 green cards but it imposes enormous fees and country-of-origin regulations that make the system costly to use for both immigrants and their prospective American employers. Worse, the government’s interpretation of unclear statutory language guarantees that fewer than half of these green cards issued actually apply to workers while the rest are allocated to their family members.
Fortunately, the employment-based green card is not the only way for highly skilled immigrants to work in the United States. The H‑1B visa provides another avenue. The H‑1B is a temporary visa that allows American firms to hire skilled foreign workers in specialty occupations. Some 99 percent of H‑1B visa workers have a bachelor’s, masters, PhD, or professional degree. The number of H‑1Bs issued annually for American firms is capped at 85,000 – 65,000 from abroad and 20,000 for foreign graduates of American universities. When the economy is growing, these few H‑1B slots frequently fill up within days of becoming available. In 2012, 61 percent of H‑1B visas went to workers in computer-related occupations.1 Importantly, H‑1Bs are uncapped for research occupations at non-profit research institutes affiliated with colleges and universities.
The duration of the H‑1B visa is three years but it can be renewed for an additional three year term. Unlike with other guest worker visas, H‑1B holders can apply for a green card while they are working in the United States if they find an employer willing to sponsor them. If the green card approval process takes longer than the maximum six year duration of the H‑1B visa then the worker is allowed to work until the green card is approved or denied.
Liberalizing the employment-based green card and H‑1B visa would likely reap large economic benefits for the United States. The current contributions of highly skilled immigrants reveal the economic potential of liberalization. Skilled immigrants currently prove to be very innovative if patents are used as a proxy measurement for innovation.
In 2006, the World Intellectual Property Organization recorded that 24.2 percent of international patent applications from the United States had at least one non-citizen inventor compared to just 7.3 percent in 1998.2 That high rate of patenting undercounts the patent contributions of immigrants by excluding those who became U.S. citizens. Between 1998 and 2006, immigrants from China, Taiwan, India, Canada, Germany, and the United Kingdom accounted for almost a fifth of all patents that were filed or co-filed by immigrants.
Jennifer Hunt and Marjolaine Gauthier-Loiselle found that a 1 percentage point increase in college graduate immigrants as a share of the population increases patents per capita by 9 to 18 percent.3 Patents filed by immigrants are not lower in quality compared to patents filed by U.S.-born inventors, according to data provided by Harvard Business School. Looking deeper into the immigration system, Kerr and Lincoln found that a 10 percent increase in workers on the H‑1B visa in a particular American city corresponded with a 0.3 to 0.7 percent increase in total patents approved from that city. The authors estimate that a large increase in H‑1B visas could have a long-run effect on innovation that can have a significant positive impact on economic growth.4
A 2005 World Bank working paper focused on the patents filed by foreign-born student researchers – a research sector largely without H‑1B visa caps. It estimated that a 10 percent increase in the number of foreign graduate students would raise patent applications by 4.7 percent, university patent grants by 5.3 percent, and non-university patent grants by 6.7 percent.5 In 2011, 76 percent of all patents issued to the top ten patent producing universities had at least one foreign-born inventor. Foreign-born inventors were involved with 87 percent of patents in semiconductor device manufacturing, 84 percent of patents in information technology, and 79 percent of drugs or drug compounds.6
Highly skilled immigrants can boost innovation in specific sectors of the U.S. economy. A working paper by Moser, Voena, and Waldinger looked at how some specialized immigrants who fled Nazi Germany affected chemistry patents in the United States. The sudden emigration of German Jewish chemists to the United States during the 1930s resulted in a 70 percent increase in patents in the chemistry sub-fields populated by the emigrants. The initial increase in patents by the Germans led to an increase of patents by U.S.-born chemists as co-inventors beginning in the 1940s. These patent effects as well as the direction of U.S. chemistry research were affected by the Germans through the 1970s.
Not all patents are productive, but a higher rate of patenting tends to increase the number of productive patents that contribute to increases in total factor productivity (TFP). In a recent paper, Peri, Shih, and Sparber attempted to measure the TFP impact of immigrants who worked in the science, technology, engineering, and mathematics (STEM) professions across 219 American cities from 1990–2010. They found that an increased number of H‑1B workers were responsible for between 10 and 25 percent of aggregate TFP growth during that time period.7
In another paper, Peri found that immigrants of all skill levels tend to specialize in tasks that require fewer communication skills in English, allowing U.S.-born workers to specialize in more communication-intensive occupations while decreasing employment competition between the two groups of workers. The resulting immigrant-induced task specialization is strongly associated with TFP growth from 1960 to 2006 across American states.8 The TFP effects of task specialization are distinct from those caused by immigrant patenting.
A third major benefit from highly skilled immigrants is their high rate of entrepreneurship. The founding of new firms is an important contributor to innovation and job growth in the United States. In 2013, immigrants were nearly twice as likely to start a business as U.S.-born Americans. Between 1995 and 2005, 25.3 percent of all technology and engineering firms established in the United States had at least one immigrant founder. Immigrants from India, China, the United Kingdom, and Taiwan tended to be the most entrepreneurial. In Silicon Valley, 43.9 percent of technology and engineering startups had at least one immigrant co-founder between 2005 and 2012.9
Several conclusions can be drawn from this research. The impact of immigrant workers on TFP are notoriously difficult to measure and papers attempting to do so are open to methodological challenges. However, much research finds that immigrant workers increase TFP and there is no major study or academic research that has found that immigrants reduce it. Increased patents and innovation are likely the main way by which immigrants affect TFP while task specialization is an additional factor. Studies on immigrant innovation through patents are generally convincing as an increase in the supply of scientists and engineers has historically increased the supply of research and development in the United States.10 One influential paper by Jones estimated that as much as 50 percent of U.S. productivity growth between 1950 and 1993 could be attributed to growth in the share of scientists and engineers – two sectors likely to expand if skilled immigration was liberalized.11
Furthermore, highly-skilled immigrants are very entrepreneurial, contributing to innovation and productivity growth through the creation of new firms in the high-tech sector. New firms are a major source of job growth and innovation as they often take risks that larger, established firms are unwilling to bear.
The contributions of highly skilled immigrants to innovation, productivity growth, and entrepreneurship are likely great in proportion to their numbers. However, the restrictiveness of America’s immigration policy has severely limited the potential of our economy to benefit even further. There are several policy reforms that could increase high-skilled immigration and thus the overall economic gains.
First, the current H‑1B visa and green card caps for skilled foreign workers should be either eliminated or, at a minimum, increased as much as possible. The family members of employment-based green card applicants should be exempted from the cap, instead of counted against it.
Second, American firms should face a lower regulatory threshold to hiring foreign-born workers. However, political pressure will demand a labor market test to decrease the possibility that immigrants will push U.S.-born American workers out of the workforce. The primary regulatory method of controlling for that, a process called labor certification, is too onerous and expensive. Replacing labor certification with a simple fee or tariff, so long as it is not exorbitant, would incentivize American firms to seek out American employees first without imposing an uncertainty-producing web of expensive and wasteful government regulations.
Third, the immigration process should be simplified so that firms and immigrants can apply for a visa or green card without having to hire a lawyer and incur costly legal expenses. Such fees can cost tens of thousands of dollars and are a tremendous source of legal uncertainty in the high-tech employment market.
Fourth, immigrants on the H‑1B work visa should be allowed to change jobs more easily rather than be tied to a single employer. Under the current system, an H‑1B worker can only change jobs after complying with a confusing set of requirements that could result in the migrant losing his or her H‑1B visa. The benefits of highly skilled migration would be greater if the workers can move to any U.S. firm willing to hire them. Related to this, family members of the H‑1B worker who are currently on the H‑4 visa should be allowed to work in the United States. Highly skilled foreign workers are likely to be married to highly skilled people so allowing them to work would increase the number of skilled workers. But even if H‑4 visa holders are not highly skilled, allowing them to work legally would increase the potential income of families headed by a highly skilled worker, increasing their likelihood of coming to the United States. The government is currently writing regulations to allow some H‑4 visa holders to work, but work authorization should extend to all of them.
Fifth, the government should create a startup or entrepreneur visa. Many skilled immigrants who want to start American firms instead of working on an H‑1B visa are thwarted because an immigrant entrepreneur cannot sponsor himself. Even if a skilled foreign worker partners with an American who wants to sponsor the worker on an H‑1B visa in a startup, government regulations mostly prohibit that because the law requires that any sponsoring firm be successful enough to guarantee the worker employment for the duration of the visa – a requirement few startups can meet.
The innovative and productivity benefits of allowing more highly-skilled immigrants are most readily apparent in patenting and entrepreneurship. And while their direct effect on TFP as workers is more difficult to measure, it is likely positive with no evidence that it is negative. In sum, liberalizing highly skilled immigration has many upsides and no downsides for American productivity, innovation, and entrepreneurship.
Notes
1http://www.uscis.gov/sites/default/files/USCIS/Resources/Reports%20and%20Studies/H‑1B/h1b-fy-12-characteristics.pdf
2http://people.ischool.berkeley.edu/~anno
/Papers/Americas_new_immigrant_entrepreneurs_I.pdf
3https://www.aeaweb.org/articles.php?doi=10.1257/mac.2.2.31
4http://www.people.hbs.edu/wkerr/kerr_lincolnjole10_h1b.pdf
5https://openknowledge.worldbank.org/bitstream/handle/10986/8957/wps3588.txt?sequence=2
6http://www.renewoureconomy.org/wp-content/uploads/2013/07/patent-pending.pdf
7http://ideas.repec.org/p/nor/wpaper/2013009.html
8http://ideas.repec.org/a/tpr/restat/v94y2012i1p348-358.html
9http://www.kauffman.org/~/media/kauffman_org/research%20reports%20and%20covers/
2012/10/then_and_now_americas_new_immigrant_entrepreneurs.pdf
10http://www.jstor.org/stable/2138581
11http://web.stanford.edu/~chadj/SourcesAER2002.pdf
The opinions expressed here are solely those of the author and do not necessarily reflect the views of the Cato Institute. This essay was prepared as part of a special Cato online forum on reviving economic growth.
