June 17, 2013
“If it ain’t broke, don’t fix it” is a phrase introduced into the American political lexicon in the 1970s by Bert Lance, the Director of the Office of Management and Budget under President Jimmy Carter. New York Times columnist William Safire later called the phrase a “caution against obsessive reform,” and it is a caution that Congress would be smart to heed as it considers comprehensive immigration reform, particularly if such reform includes the implementation of onerous restrictions on employers’ ability to hire temporary foreign workers holding H 1B or L-1B visas.
In its current form, the Senate’s immigration reform bill (S. 744) would prohibit so-called H-1B-dependent employers from placing foreign workers holding H 1B visa status (as temporary workers in specialty occupations) at third-party client sites (except for H-1B-dependent employers which are nonprofit institutions of higher education, nonprofit research organizations or health care providers petitioning for physicians or other health care workers). Other employers could outplace an H-1B worker only upon the payment of a $500 STEM education and training fee. An H-1B-dependent employer is defined as an employer with 25 or fewer full-time equivalent employees (FTEs) and more than 7 H-1B workers; at least 26 but not more than 50 FTEs and more than 12 H-1B workers; or at least 51 FTEs and at least 15 percent H-1B workers. The bill would also prohibit the outplacement of L-1B workers (intracompany transferees with specialized knowledge) if 15 percent or more of the employer’s full-time equivalent workforce is made up of L-1Bs. Otherwise, employers would be prohibited from outsourcing or outplacing an L-1B worker unless: (1) the worker is controlled and supervised by the petitioning employer; (2) the placement is not an arrangement to provide labor for hire; and (3) the employer pays a $500 fee. The bill also places new limits on the total percentage of H-1B and L-1 workers an employer can have in its U.S. workforce, increases the mandated salaries and visa fees for H-1B workers, and creates a new “H-1B-skilled-worker-dependent” definition which would bring more U.S. employers under the purview of many of these new restrictions.
The provisions restricting or prohibiting outplacement of H-1B and L-1B workers (and the other restrictions mentioned above) are presumably aimed at stemming the tide of outsourcing jobs, especially information technology (IT) jobs, to countries such as India. The only problem is that the impetus behind these proposed restrictions stems from a profound misunderstanding about the prevailing software development business model that has been the main focus of such efforts. Moreover, these restrictions are meant to fix a problem that has to a large degree already fixed itself, as many U.S. companies—having learned the hard way that there are hidden costs to outsourcing—are beginning to “insource” or “reshore” their IT operations and other services to offices in the United States. Most important, the restrictive provisions won’t work, and could actually have the counterproductive effect of driving more jobs offshore.
First, let’s define our terms. While historically, a “job shop” was a machine shop or other small-scale manufacturing facility engaged in fabricating small quantities of custom-made parts produced according to customer specifications, in more recent years the term (sometimes also called a “body shop”) has come to refer to enterprises that import foreign workers into the United States to work directly at third-party worksites as labor for hire, directly displacing U.S. workers. The urgent need to address problems related to the Y2K bug at the turn of the millennium required resources well beyond what domestic IT consulting companies could provide, and helped create a market for job shops to bring large numbers of foreign computer programmers into the United States to provide Y2K remediation services to U.S. customers. In 2004, in a delayed reaction to the Y2K situation, Congress enacted the L-1 Visa Reform Act to address the job shop problem by controlling the practice of stationing L 1B workers at client sites.
However, as we have discussed elsewhere,1 it quickly became apparent that job shops really were no longer a significant problem. Accordingly, both U.S. Citizenship and Immigration Services (USCIS) and Congress redirected their focus to the labor market impact of the onshore/offshore outsourcing model in the software development industry. Under this business model, software development work occurs largely abroad, but workers are temporarily deployed to the U.S. to perform onsite services, such as fact-gathering and testing. This led USCIS to move beyond the narrow aims of the L-1 Visa Reform Act and take steps (first initiated in the agency’s field offices through adjudication and ratified by a decision of the Administrative Appeals Office) to curb offshore outsourcing by limiting the L-1B “specialized knowledge” visa category as a whole, and by placing limits on third-party site placements of both L-1B and H-1B workers. (The agency’s overly restrictive and essentially incomprehensible redefinition of what constitutes “specialized knowledge” for L-1B visa purposes is another important topic, and will be the subject of a later blog post—but basically, a new requirement that knowledge be narrowly held within the petitioning organization was constructed, thus limiting the percentage of employees that could possess a particular form of specialized knowledge.)
The motivation behind efforts to curtail offshoring of U.S. jobs is beyond the purview of this article. There is little doubt that globalization has led to the offshore outsourcing of services such as software development to lower-priced markets. But it is not clear that imposing burdensome restrictions on outplacing temporary foreign workers at client sites will stem this tide. On the contrary, restrictions on the placement of H-1B and L-1B workers at third-party sites in the United States fail to account for the distinction between “job shopping” and a true outsourcing business model, which occurs not only in the technology sector but in many other industries as well. For example, when a U.S. business contracts with a global IT technology consulting firm to develop a new software solution to a key business challenge, the majority of the consulting firm’s development work typically occurs abroad. The firm will temporarily deploy employees from its offshore development centers directly to client sites to interface with the client, gather the necessary requirements, and engage in software testing and implementation in order to deliver services and ensure the quality of the final product. Another example, outside of the IT industry, would be an auditor engaged in reviewing a client’s worldwide operations. The auditor may enter the United States on an L-1 visa, but need to work primarily at the client’s site, since this is where the necessary information is located. Preventing H-1B or L-1B employees from working under such common business models—consisting of legitimate outplacement arrangement—hurts the U.S. businesses that need these workers’ services.
Why can’t IT consulting businesses hire American software engineers? Despite a persistently high unemployment rate overall in the United States, unemployment rates in science, technology, engineering and mathematics (STEM) fields remain low. According to a 2012 report, the unemployment rate for U.S.-citizen STEM workers with Ph.Ds is only 3.15 percent, and is 3.4 percent for those with Master’s degrees. In some STEM occupations, the unemployment rate is virtually nil: 0.1 percent for petroleum engineers, 0.4 percent for computer network architects, and 0.5 percent for nuclear engineers. Overall unemployment for U.S. workers in STEM fields (including those with Bachelor’s, Master’s and Ph.D degrees) is just 4.3 percent (compared to a national unemployment rate of approximately 8 percent). Keep in mind that economists typically consider a 4 percent unemployment rate to represent “full employment” (insofar as the remaining unemployed are typically in transition between jobs, a phenomenon known as “frictional unemployment”).
The fact is that the U.S. educational system is still not producing sufficient numbers of U.S.-born STEM professionals. While enrollment of U.S.-born students in STEM programs is rising, foreign students still make up 30 percent of the total enrollment in U.S. graduate science and engineering programs, and this percentage has held steady since 1990, according to a report from the Congressional Research Service—or perhaps as much as 40 to 45 percent according to other reports. Data from the Bureau of Labor Statistics indicate that between now and 2020, there will be approximately one million job openings in computing professions that require at least a Bachelor’s degree, but our educational system is projected to produce less than half the number of U.S. graduates needed to fill those positions. Even at the Bachelor’s level, where the supply of STEM degrees and demand for STEM workers is fairly equal, employers are hard-pressed to find U.S. workers to fill computer science jobs. The reason for this is that U.S.-born students who do obtain STEM degrees tend to be attracted to non-STEM career paths for a variety of economic, social and cultural reasons, as detailed in a 2011 report from Georgetown University’s Center on Education and the Workforce. For example, a little understood aspect of the nature of IT consulting positions in particular is that they require constant relocation, from project site to project site, of the type that native-born American workers are usually unwilling to stomach. On a broader scale, the competencies, skills and talents that STEM graduates possess are in demand across the economy, diverting U.S.-born STEM workers into nontraditional STEM occupations and “making what seems like plenty, not enough to go around,” according to the Georgetown report.
Ironically, what is also only beginning to be understood is that “the offshoring of services is slowing down because most of the work that can be done remotely has already gone, and because firms are becoming more aware of the disadvantages of sending work to the other side of the world,” according to one report in The Economist. Offshore outsourcing is also coming to an end of its own accord for other reasons, including—according to another report in The Economist—productivity improvements that have wiped out jobs that might otherwise have been outsourced; a huge labor turnover in offshore locations that can lead to problems with quality; and cost advantages that are disappearing due to a growing middle class and associated rises in wages in countries such as India and China. See also the similar conclusions in a 2012 report by KPMG International, “The Death of Outsourcing,” or an earlier report by McKinsey & Company, “IT services: The new allure of onshore locales.” Don’t believe the likes of The Economist or McKinsey? Even The Atlantic, in a 2013 article called “Why We Can All Stop Worrying About Offshoring and Outsourcing,” has argued that “[t]he old math no longer applies when it comes to where multinationals choose to open shop,” and that “[i]conic American companies like Apple, Google, Caterpillar, Ford, Emerson, GE, and Intel” are “reshoring” overseas work back to the United States and adding plants and jobs in the United States. In addition, as the New York Times reported last month, many Indian outsourcing firms have been expanding their operations in the United States, and Indian companies have invested $5.9 billion to set up offices closer to their U.S. clients.
The only solution the Senate bill offers is for companies to apply for permanent residence on behalf of their foreign employees. Foreign workers on H-1B or L-1 visas for whom a labor certification or immigrant visa petition has been filed would be considered “intending immigrants” and would not be counted in the workforce calculations that are used to determine whether an employer is H-1B-dependent or is required to pay extra visa fees (as much as $10,000 per H-1B employee in some cases). The problem is that this is counter to the foreign outsourcing business model, under which employees brought to the United States on temporary work visas are international assignees, not intending immigrants. Compelling employers to sponsor for permanent residence employees who never intended to immigrate permanently to the United States may increase the immigration of skilled foreign workers, but it is an odd way to save American jobs (granted that a certain number of conversions to permanent resident status for certain employees is sensible).
Numerous reports and studies have concluded that immigrant entrepreneurs create jobs for American workers and strengthen the U.S. economy, and that foreign-born STEM workers in particular increase employment and wage opportunities for U.S.-born workers in both STEM and non-STEM fields. For every foreign-born student graduating with an advanced degree from a U.S. university who stays in the United States and works in a STEM field—usually on an H-1B visa—an average of 2.6 jobs are created for U.S. workers. The Senate bill recognizes the economic benefits of immigration by including provisions that would loosen restrictions on foreign students seeking permanent residence, increase the annual cap on H-1B visas, create a new entrepreneur visa, and increase the overall supply of employment-based green cards. Undoing these positive steps by simultaneously enacting restrictions that hamper the ability of U.S. companies to fully utilize the skills of those foreign workers they hire on H-1B and L-1B visas would be unwise.
As I (Austin Fragomen) stated in 2003, in testimony before the Senate Judiciary Committee in hearings on what later became the 2004 L-1B Visa Reform Act, there are larger economic forces surrounding outsourcing and offshoring that cannot be combated by imposing visa restrictions in the United States. Limiting U.S. employers’ ability to leverage the skills of their temporary foreign workers, and assessing huge fees on businesses that seek to employ such workers, may be designed to encourage employers to hire Americans instead. But imposing further burdens on global companies that make the United States an even less attractive locale for business operations and investment could very well be counterproductive, and drive more jobs offshore (especially so long as U.S. workers with the requisite skills cannot be found in sufficient numbers), reversing the nascent trend toward reshoring overseas jobs back to the United States. Burdensome restrictions could also spur the development of new technological solutions that would eliminate jobs altogether. Finally, many U.S. industries have long-standing relationships with foreign IT companies—such as the banking industry, which relies on trading systems that were developed by foreign IT companies that continue to service and upgrade systems installed a decade or more ago—and would be negatively impacted if the restrictions on the H-1B and L-1 visa categories in the Senate bill become law.
Congress, please—beware of obsessive reform that does not account for unintended, long-term consequences that could hurt the American economy.