Temporary Migrant Workers Or Immigrants?The Question For U.S. Labor Migration
Employment-based U.S. immigrant and nonimmigrant work visa data from 1987 to 2017show that the number of permanent immigrant work visas has remained relatively constant over time but that the number of temporary work visas has increased sharply. That is, the labor migration system has shifted from one in which permanent immigrant workers annually made up approximately 20 percent of new migrant workers to one in which they make up less than 10 percent. Major legislative reforms do not explain the change; this article examines available government data showing how the labor migration system involves mostly nonimmigrant, temporary migrant workers who have few options to remain permanently in the United States and raises questions about the implications for the future legal landscape of immigration.
immigration, labor migration, visas, work visas, lawful permanent residents, employment-based, family-based, green cards, IMMACT90, immigration reform, guestworkers, temporary migrant workers, temporary labor migration
The United Nations (UN) International Labour Organization (ILO) estimates that 234 of the world's 258 million total international migrants (90.7 percent) are migrant workers, defined as working age (fifteen and older) and "either employed or unemployed in their current country of residence" that is not their home country. Those 234 million migrant workers make up 4.7 percent of all workers in the world and 4.2 percent of the global population age fifteen and older (ILO 2018).
The United States—where according to UN estimates nearly fifty million migrants live (UNDESA 2019)—hosts nearly one-fifth of all international migrants, and is indisputably a nation of immigrants, and perhaps the nation of immigrants. According to the Bureau of Labor Statistics in the U.S. Department of Labor, migrant workers make up 17.4 percent of all workers in the U.S. labor force (BLS 2019) and in 2017 were more likely to be employed than the native born. Migrants had a labor-force participation rate of 66.0 percent and an unemployment rate of 4.1 percent; the corresponding rate for native-born U.S. workers was nearly 4 percentage points lower than that of migrant workers and an unemployment rate that was 0.3 percentage points higher (BLS 2019).
Although no official definition of labor migration is yet agreed to, the International Organization [End Page 18] for Migration defines it as the "movement of persons from one State to another … for the purpose of employment" (IOM 2019). The legal and administrative system of visas and adjudications, recruitment rules, annual limits, inspections, and enforcement that regulate the movement of persons into the United States for work effectively make up the U.S. labor migration system—that is, they are the pathways that facilitate the ability of migrant workers to come to the United States for employment.
This article focuses on U.S. labor migration and the visas that specifically authorize the employment of foreign citizens in the U.S. labor market, including the visas issued to the spouses and children who accompany those who are issued work visas. A number of major changes to the labor migration system were ushered in by the Immigration Act of 1990; this article examines how the U.S. labor migration system has evolved since its enactment with respect to the composition of the authorized migrant workforce in the U.S. labor market, and raises questions about what the implications are for the future legal landscape of immigration.
migrant pathways into the united states
People migrate to the United States in a number of ways, either lawfully or without authorization. The available lawful pathways include obtaining or adjusting to lawful permanent resident (LPR) status—also commonly referred to as obtaining a permanent immigrant visa or green card—either through the family-based (FB) or employment-based (EB) preference categories, a humanitarian pathway as a refugee or asylee, or the diversity visa (DV) lottery. All persons in LPR status enjoy nearly all of the same labor rights as U.S. citizens, including being able to work for any employer and in any position except for those that explicitly require citizenship. Persons in LPR status can become naturalized citizens after five years, or after three years if they are married to a U.S. citizen, if they apply to United States Citizenship and Immigration Services (USCIS), a subagency of the U.S. Department of Homeland Security (DHS), and meet certain requirements.
The other major pathway into the United States is as temporary visitor, student, trainee, diplomat, exchange visitor, or employee by acquiring a nonimmigrant visa or status. Nonimmigrant visas are temporary, meaning that the foreign-born person to whom the visa is issued must depart the United States after the visa expires unless they adjust to LPR status by acquiring either one of the immigrant visas described or another valid nonimmigrant status. Many nonimmigrant visa classifications authorize the visa holder—sometimes referred to as the beneficiary—to be employed in the United States. Employed nonimmigrants are also often referred to as temporary foreign workers, temporary migrant workers, or guestworkers, but no one definitive term has been agreed upon.
The other route involves migrants who are present in the United States but who do not have an authorized immigration status; such individuals are sometimes referred to as unauthorized, undocumented, or irregular migrants, and sometimes (derogatorily) called illegal migrants or illegal immigrants. Unauthorized migrants either entered into the United States without inspection by the appropriate authorities—often referred to as entering without inspection in government documents and data—and may have done so in a clandestine manner. Other unauthorized migrants may have originally entered the United States lawfully with a nonimmigrant visa or through the Visa Waiver Program and after an inspection by government authorities, but then lost their immigration status. The loss of status may have occurred because of a violation that led to the revocation of their visa or status or the simple expiration of a nonimmigrant visa that was temporarily valid for a set time, usually a period of authorized travel, employment, or study. Unauthorized migrants make up 4.5 percent of the U.S. workforce (Krogstad, Passel, and Cohn 2019). This article, however, focuses on migrant workers who are authorized to be employed in LPR or nonimmigrant status.
The Pew Research Center estimates the size of the population of migrants who live in the United States under different immigration statuses and the shares of the total they represent. In 2017, the number of lawful immigrants—those with an authorized immigration status—came [End Page 19] to 35.2 million, 77 percent of all migrants in the United States. Of these lawful immigrants, 20.7 million (45 percent) are naturalized U.S. citizens and 12.3 million (27 percent) are LPRs. The unauthorized immigrant population stood at 10.5 million in 2017–23 percent of all migrants in the United States—and 2.2 million were "temporary lawful residents" holding nonimmigrant visas, some 5 percent of the total (Radford 2019).
changes brought about by the immigration act of 1990
The Immigration Act of 1990 (IMMACT90), enacted on November 29, 1990, is the last time the U.S. Congress crafted a major overhaul to immigration laws that passed both houses of Congress and was signed by a sitting president.1 The Immigration and Nationality Act of 1965 twenty-five years earlier was the previous major reform (for more, see Zamora 2015). IMMACT90 kept the basic framework of the 1965 act but made numerous amendments to the provisions on legal immigration, including remaking the permanent immigrant visas that grant lawful permanent residence, updating some of the temporary, nonimmigrant visa classifications that require departure after a period of employment, and creating new nonimmigrant visas that authorize employment. Many aspects of the new-look immigration system that resulted from IMMACT90 became the basis of the modern U.S. labor migration system and remain largely unchanged today. This section summarizes some of the major changes, mostly focusing on those relevant to labor migration (for more about the reforms ushered in by IMMACT90, see Donato and Amuedo-Dorantes, this issue, 2020).
Lawful Permanent Residents And Permanent Immigrant Visas In IMMACT90
IMMACT90 established new preferences for both family-based and employment-based visas that grant lawful permanent resident status that can lead to citizenship—known as immigrant visas—and created a new category called the diversity visa (see table A1 for a listing of permanent immigrant visa preference categories). The humanitarian visa system was left virtually unchanged. The DV was created with a lottery system for foreign citizens in countries with historically low levels of immigration to the United States as a way to diversify those who receive immigrant visas.2 All potential beneficiaries of FB visas must be sponsored by either a U.S. citizen or an LPR; in most cases, U.S. employers must sponsor a person for an EB visa.
IMMACT90 updated the overall annual numerical limit (cap) on worldwide immigration. This after 1995 was intended to be set at 675,000 visas per year, which includes a limit of 480,000 total FB visas and of 226,000 for the subset in the preference categories—and 140,000 EB visas and 55,000 DV visas. However, IMMACT90's overall cap, which includes a complex system of caps for the various preference categories—where unused numbers can be assigned to other preferences3—is always surpassed because IMMACT90 did not set a cap on the number of immediate relatives of U.S. citizens who can acquire LPR status. These are defined as spouses of U.S. citizens, children of U.S. citizens if they are unmarried and under twenty-one years old, and parents of U.S. citizens if the citizen is twenty-one or older. This aspect of the legislation was not a change, however; the cap exemption for immediate relatives also existed under the 1965 Act (Chishti, Hipsman, and Ball 2015).
The existing FB preferences were reformed to include four preferences for other relatives of U.S. citizens and LPRs. Each preference has its own cap as well as per-country limit (often referred to as the per-country ceiling) of 7 percent; this rule was an update to the previous caps of about twenty thousand per country per [End Page 20] year (Chishti and Yale-Loehr 2016). The per-country ceiling means that no more than 7 percent of all immigrant visas issued in a preference may go to the nationals of any single country (for more on limits in FB preference categories, see Kandel 2018).
Employment-Based Immigrant Visas and Preferences in IMMACT90
EB permanent immigrant visa categories were reformed into five new preferences (employment-based first preference through fifth preference, EB-1 through EB-5), the intent being to increase the number of immigrants who could work in the United States after being sponsored by an employer, but especially skilled and educated immigrants (Chishti and Yale-Loehr 2016). The five EB preferences are as follows:
EB-1 for priority workers, which includes three subcategories: persons with extraordinary ability in the sciences, arts, education, business, or athletics; outstanding professors and researchers who are recognized internationally; and multinational managers and executives.
EB-2 for professionals holding advanced degrees or baccalaureate degrees and at least five years of experience, and persons of exceptional ability in the sciences, arts, or business.
EB-3 for skilled workers with at least two years of professional experience, professionals with jobs that require at least a baccalaureate degree, and unskilled workers (also referred to as "other workers" in the statute and government documents) capable of filling jobs that are not seasonal and that require less than two years of training or experience.
EB-4 for "certain special immigrants," which includes a number of categories including ministers of religion and other religious workers, former employees who worked on the Panama Canal, Iraqi and Afghan interpreters-translators, retired employees of international organizations and their spouses and children, and special immigrant juveniles (who have been abused, abandoned, or neglected by a parent).
EB-5 for immigrant investors who invest in a commercial enterprise in the United States and create or preserve ten permanent full-time jobs for qualified U.S. workers.
The total annual cap on employment-based immigrant visas in the Immigration and Nationality Act of 1965 was 54,000, a limit that was nearly tripled with IMMACT90's updated cap of 140,000. DHS (2019) data show that an average of just over 58,000 EB visas were issued in each of the four years preceding IMMACT90, and an average of 146,000 were issued between 2010 and 2017. Despite this 160 percent increase resulting from the increase in IMMACT90, relative to the total number of immigrant visas issued every year—including FB, EB, DV, and humanitarian visas—EB visas accounted only for approximately 12.2 percent of all 1.1 million immigrant visas issued in 2017 (Witsman 2018), relative to 9 percent of the total in 1990 (Chishti and Yale-Loehr 2016).
Athough EB-1 through EB-5 are capped annually at 140,000 combined, each preference has its own cap. EB-1, EB-2, and EB-3 are set at 40,040 (28.6 percent of the total); EB-4 and EB-5 are each capped at 9,940 (7.1 percent). The annual cap may sometimes change overall or within the preferences because any unused FB visas in one fiscal year become available to be used as EB visas in the following year, and any unused visas in one EB preference then become available for other preferences.4 Spouses and children who accompany a principal EB immigrant (the main EB immigrant listed on a petition) are referred to as derivatives (because they derive from the original immigrant petition) and count against the EB annual limit. Every year, roughly half of EB visas are issued to principal immigrants and the other half to derivatives. In addition, the annual EB numbers are also limited by the per-country ceiling prohibiting more than 7 percent of EB immigrant visas [End Page 21] being issued to the nationals of any one country.
The vast majority of persons issued one of the 120,120 EB visas available in EB-1 through EB-3 each year must have been sponsored by a U.S. employer seeking to hire them for a specific job opening. In addition, all employers seeking to hire immigrants through EB-2 and EB-3 are required to obtain an approved labor certification from DOL, certifying that no U.S. workers were available for the position that will be filled by the new EB immigrant, after the employer has advertised the job opening and recruited U.S. workers. The exceptions to these requirements are persons with extraordinary ability applying through EB-1, who are allowed to self-petition for a visa without an employer and without a specific job offer, as well as those applying through EB-2 who may apply for a national interest waiver, that, if successful, exempts them from the job offer or labor certification requirements.
Each of the EB-4 categories has its own unique requirements; some require a job offer but none require an approved labor certification from DOL (for more, see USCIS 2019a). The EB-5 preference does not require a job offer or labor certification, but instead has regulations that govern which types of investments are permitted and the minimum investment amount. In general, to qualify for EB-5, a foreign citizen must make a minimum qualifying investment of $1.8 million, but the minimum is $900,000 if made in a qualifying high-unemployment or rural area (a targeted employment area or TEA).5
In terms of overall numbers and where EB visas fit in, as figure 1 shows, the number of immigrant visas issued that are capped—the FB and EB preferences and DV—have fluctuated at times but remained relatively stable near their annual limit for at least the past decade. The family-based immediate relative preference, however, has grown from 235,484 in 1992—the first full fiscal year after IMMACT90 was enacted—to 516,508 in 2017, more than doubling over thirty years. [End Page 22]
Nonimmigrant Visa Classifications And Temporary Work Visas In IMMACT90
In the year immediately before IMMACT90, only a few major nonimmigrant visa classifications that authorize temporary work existed, among them E, H, J, and L. IMMACT90 kept many of these existing classifications in place with few changes, but made significant changes to the major nonimmigrant visas that authorize employment and created a handful of new temporary work visas (Leiden and Neal 1990):
The rules for nonimmigrants working on ships and as longshore workers with D visas were updated.
The E visa for investors was expanded in terms of eligibility.
An annual cap of sixty-six thousand was created for the H-2B visa for nonagricultural seasonal jobs (H-2B was created four years earlier by the Immigration Reform and Control Act of 1986).
The rules in the H-1B classification for professional workers in specialty occupations were updated and an annual cap of sixty-five thousand was established.
Key definitions establishing who can be an intracompany transferee under the L-1 visa were updated.
The O visa was created for individuals with extraordinary ability or achievement in the sciences, education, business, athletics, or in the arts (O-1), and for individuals accompanying them (O-2), as well as spouses and children accompanying beneficiaries of O-1 or O-2 visas.
The P visa was created for internationally recognized athletes and members of internationally recognized entertainment groups and their essential support personnel (P-1), performers in a reciprocal exchange program and their essential support personnel (P-2), artists and entertainers coming to be part of a culturally unique program and their essential personnel (P-3), and spouses and children who are accompanying beneficiaries of P-1, P-2, or P-3 visas.
The Q-1 visa was created for workers in international cultural exchange programs designated by USCIS.
The R visa was created for temporary religious workers (R-1) and their spouses and children (R-2).
Although new annual limits were imposed on H-1B and H-2B in 1990, no other nonimmigrant visa programs were capped by IMMACT90 or other legislation. The total number of non-immigrant work visas issued in 1989—the year before IMMACT90's enactment—was 296,598. By 1999, a decade later, the number had more than doubled to 624,899.
Two other nonimmigrant visa classifications that authorize employment were created a number of years later and are part of the contemporary U.S. labor migration system. The CW-1 nonimmigrant visa, the CNMI-Only Transitional Worker visa classification, began in 2012 and allows employers in the Commonwealth of the Northern Mariana Islands (CNMI) to hire migrant workers, many of whom are employed in the construction industry, and the CW-2 visa can be issued to the spouses and children of CW-1 workers. The TN visa was created through the North American Free Trade Agreement (NAFTA) between the United States, Canada, and Mexico in 1994. TN visas allow Canadian and Mexican citizens to work in the United States in professional occupations that include accountants, engineers, attorneys, pharmacists, and nurses, among others.
work visas over the past three decades
Some of the trends in the issuances of permanent immigrant and temporary, nonimmigrant work visas in the thirty years since IMMACT90 was enacted merit attention.
EB Immigrant and Nonimmigrant Visas
The available data on EB permanent immigrant and temporary, nonimmigrant work visas issued that were selected for this analysis are represented in figure 2. Those data come from the Department of Homeland Security and the Department of State for the years 1987 to 2017, as well as other sources, such as the General [End Page 23]
Accountability Office and the Pew Research Center. The total number of EB immigrant visas (including EB-1, EB-2, EB-3, EB-4, and EB-5) that grant LPR status is shown and labeled in the tables and figures throughout as EB-LPR to distinguish that they are EB immigrant visas that grant LPR status. Those data represent immigrant visas issued in all five EB preferences to both principals and derivative spouses and children.
Shown in figure 2 are thirty-six nonimmigrant visa classifications, subclassifications, or programs within a visa classification that authorize temporary employment or allow spouses and children to accompany the principal nonimmigrant worker, and that currently exist or existed between 1987 and 2017. The main rationales for temporary, nonimmigrant work visas include filling labor shortages, managing inevitable migration, cross-border communing, facilitating youth exchange programs and admitting foreign students, allowing intra-corporate transferees, fulfilling trade agreement provisions, and facilitating foreign investment in countries of destination (Costa and Martin 2018). The nonimmigrant work visas in the United States authorize temporary employment in a wide range of occupations and business activities, including in professional occupations in the medical and engineering fields and information technology jobs, as well as in industries such as agriculture, construction, landscaping, and food processing, and in service sectors as maids and domestic servants, and professional sports teams.
The temporary, nonimmigrant visa classifications for spouses and children on this list either permit the spouse beneficiary to be employed (E-2 and L-2, for example), or prohibit employment (O-3), or prohibit some spouses to work but not others (for example, some H-4 [End Page 24] spouses of H-1B workers may apply for work authorization but the rest may not).6
A number of nonimmigrant visa classifications were excluded from this list, either because they do not authorize employment, usually authorize employment for a very short duration, or because they authorize employment that is not typical in the U.S. labor market or for which no EB visa counterpart is in place. This includes, for example, short-term business visitors, diplomats employed by foreign governments or international organizations, and an estimated share of J-1 visa program categories that do not permit employment. Data are unavailable for migrant workers employed through the Optional Practical Training program while on F-1 visas before 2004, for example, and many years of data are unavailable for the numbers of workers in specific work programs within the J-1 Exchange Visitor program, though data are available for the total number of visas issued for all years.7 The number of TN statuses issued to Canadian nationals at points of entry into the United States is also not available because the U.S. government does not publish it.
These data do not represent the total population of migrant workers with EB or nonimmigrant visas who are currently authorized to be employed, or who were authorized to be employed at a particular point in time. EB visas do not expire unless the immigrant violates the terms of the EB visa or their LPR status in some way, and some share become naturalized U.S. citizens. The total current population of permanent immigrants working in the United States with EB visas has not been estimated by the U.S. government. The total population of migrant workers employed with nonimmigrant visas is difficult to calculate by specific visa classification and no time series with this information exists because of a lack of reliable government data and estimates, and as a result of the varying durations that the different visa classifications authorize (for example, H-2A visas authorize farm employment for less than a year and H-1B visas authorize employment for up to six years in occupations requiring a college degree).8
DHS publishes an estimate of the total non-immigrant population by broad category but not by individual visa classification (Baker 2018). Daniel Costa and Jennifer Rosenbaum (2017) estimate the number of migrant workers who were employed with nonimmigrant visas at some point in 2013, by visa classification. In total, 1.42 million migrants were employed for all or part of 2013, accounting for approximately 1 percent of the U.S. labor force at the time. They did not, however, estimate the total stock or population or the number of full-time-equivalent (FTE) jobs filled by migrants with nonimmigrant visas. The Organization for Economic Cooperation and Development, for the first time, estimates that the number of FTE jobs filled by migrants with nonimmigrant work visas in the United States was 1.6 million in 2017, adding 1.06 percent to the U.S. labor force (OECD 2019). This article does not attempt to calculate any new population estimates; the numbers that follow reflect newly issued EB immigrant visas and nonimmigrant work visas issued to migrant workers in the corresponding fiscal year or years.
The 1990 Immigration Act spurred a sharp increase in nonimmigrant work visas, though [End Page 25] employment-based permanent immigrant visas increased but leveled off.
As noted, figure 2 shows the number of work visas issued to principal EB immigrants in all five EB preferences combined (EB-LPR), including their spouses and children, and work visas issued to principal nonimmigrant beneficiaries and their spouses and children. Unfortunately, only a few years of data on EB immigrant and nonimmigrant work visas are available before IMMACT90 became law in November 1990; data for EB visas before 1986 and nonimmigrant visa issuances before 1987 are not publicly available from DHS or State. Table 1 shows that in the four complete fiscal years before IMMACT90 (1987 to 1990), the number of EB visas was relatively constant but the annual share of permanent EB visas decreased from 19 to 15.6 percent, and the total share over the four years was 17.2 percent (EB-LPR share). In other words, EB permanent immigrant visas accounted for just over one-sixth of all new work visas (temporary and permanent) issued to migrant workers in 1987, annual totals remaining mostly constant, fluctuating by no more than 1,208. During the same period, the annual number of nonimmigrant work visas issued was steadily increasing each year, with nearly seventy thousand more issued in 1990 than in 1987.
Table 2 shows the same information as table 1, but for each of the remaining years of the 1990 decade. Because IMMACT90 was enacted during the second month of the government's 1991 fiscal year, 1991 issuances are excluded because the government was in the process of implementing a new system of permanent immigrant and temporary nonimmigrant work visas that year, possibly making the numbers unreliable due to the transitional nature of the year. Thus, excluding fiscal 1991, the annual share of permanent work visas peaked at 30.2 percent in 1993 and reached its lowest point, 8.3 percent, in 1999. Although this fluctuation in the share of EB visas is large, after an initial spike in the share in the first half of the 1990s, the downward trend during the second half of the decade is clear. Overall, the EB-LPR share for the 1992 to 1999 period is 18.7 percent, which is 1.5 percent higher than the total for the four years before IMMACT90 became law. Despite IMMACT90's nearly tripling of the number of EB visas available, in the years immediately afterward, the increase in share relative to the increase in nonimmigrant work visas was minimal.
The years immediately after IMMACT90 (1992 through 1999) are notable for kicking off a rapid rise in the number of nonimmigrant work visas issued, which increased by 92.2 percent, from 325,155 in 1992 to 624,899 in 1999. The number of EB visas fluctuated quite a bit, ending the decade with a low of 56,678 in 1999 [End Page 26]
[End Page 27] and peaking at 147,012 in 1993, with an average over the eight years of almost 102,000 per year.
The decade of the 2000s saw a further continuation of the upward trend for nonimmigrant work visas that was set in motion in the early 1990s. As table 3 shows, the number of nonimmigrant visas in the selected classifications issued in 2000 was 727,234 and surpassed one million in 2007 and 2008. The number of nonimmigrant work visas issued dropped to 866,765 in 2009, coinciding with the start of the Great Recession, and the average for the decade stood at 869,226.
In the first six years of the 2000s, the EB-LPR share ranged between 10.2 and 22.1 percent of all work visas issued. However, in the last four years of the decade, the shares leveled off at roughly 13 to 14 percent of all work visas, a trend that would continue until 2013. The number of EB visas issued each year fluctuated between 81,727 and 246,877 between 2000 and 2005. The sharp decline in 2003 was due to the government's immigration functions being reorganized on the creation of DHS in 2003, and the increase in 2005 had to do with a temporary recapture of unused EB visas from the previous year that was mandated legislatively by the Real ID Act of 2005 (Kandel 2018). The totals were stable at roughly around 160,000 during the next three years of the decade and dropped to 140,903 in 2009.
During the 2010 to 2017 period (2017 is the final year for which complete data are available at the time of writing), the number of EB visas issued was relatively stable in the 140,000 to 150,000 range and the number of nonimmigrant work visas again increased greatly (see table 4). The first full fiscal year of the Great Recession was 2009, which as noted saw a drop of more than two hundred thousand over 2008, and in the three following years—2010 through 2012—of the economic recovery, the number of nonimmigrant work visas slowly grew but remained below the one million issued in 2007 and 2008. The number of nonimmigrant work visas once again reached one million in 2013 and continued to increase steadily in the following years, peaking at 1.58 million in 2017. The number of EB visas issued remained relatively stable, ranging from 137,855 to 151,596 in all but one year, 2013, when 161,110 were issued.
The average number of EB immigrant and nonimmigrant work visas issued during the three periods—the 1990s after IMMACT90, the 2000s, and from 2010 to 2017—illustrate the broader overall trends: issuances of EB visas increased slowly until stabilizing around the annual cap in the 140,000 to 150,000 range, and the number of nonimmigrant work visas issued has increased dramatically. As tables 3 through 5 show, the number of EB visas averaged 101,767 over the 1992 to 1999 period, 156,999 during the 2000s, and 145,553 between 2010 and 2017. The number of nonimmigrant work visas averaged 442,645 between 1992 and 1999, 869,226 during the 2000s, and 1,149,131 between 2010 and 2017.
The share of EB permanent immigrant visas since 1992 has gradually declined, peaking at 30.3 percent in 1993 and bottoming out at 8.7 percent in 2017, when fewer than one in eleven work visas granted the recipient with LPR status. The total share between 1992 and 1999 was 18.7 percent, declining to 15.3 percent during the 2000s, and 11.2 percent in the 2010 to 2017 period.
nonimmigrant migrant workers and permanent immigrants
The data trends shown reveal a shift toward a U.S. labor migration system that brings in many more workers temporarily than permanent immigrants who can eventually become naturalized citizens. What does it mean to have a labor migration system structured to increasingly be made up of temporary workers rather than permanent immigrants?
Temporary, Nonimmigrant Work Visas, Labor Standards, and Worker Rights
More than one million nonimmigrant visas have been issued every year to migrant workers, their spouses, and children since 2013. As noted, these nonimmigrant, temporary migrant workers are hired by U.S. employers to temporarily fill jobs in a wide range of occupations. Employers have more than two dozen nonimmigrant visa classifications they can choose from—each of which has its own distinct purpose and history. The most common nonimmigrant visas that authorize employment are the H-2A, H-2B, H-1B, J-1, L-1, F-1 (via the Optional Practical Training program or [End Page 28]
The first temporary migrant worker program in the United States was the Bracero program, which was negotiated as a bilateral agreement between the United States and Mexico in the early 1940s. Since then, cases have been numerous of abuse and exploitation of migrant workers employed with temporary visas that have come to light through the media, reports from advocates and labor unions, and government audits (Galarza 1956; Meissner 2004). Many of the abuses occur because of the structure of the visa programs, which have few rules and inadequate protections and oversight by federal or state labor authorities, and largely because in nearly all cases, employers own and control the visa status of temporary migrant workers. That means that if a migrant worker with a nonimmigrant visa gets fired, they lose their visa status and can become removable from the United States (Bauer and Stewart 2013). That many temporary migrant workers pay hefty fees to obtain their temporary jobs in the United States (CDM 2013) also means that they are afraid to speak up and complain to their employer or government authorities if their wages are stolen or other workplace violations take place. For example, complaining could mean getting fired, which can result in the inability to earn back the money invested to obtain the temporary job and visa.
The enactment of IMMACT90 did not bring with it a restructuring of this basic setup for nonimmigrant visas that authorize work, and even in just the last decade, there have been numerous scandals and shocking tales of worker abuse have been uncovered in virtually every nonimmigrant visa program. For example, BuzzFeed News published a series of reports on the H-2A nonimmigrant visa for temporary migrant workers in agricultural occupations and the H-2B nonimmigrant visa for other low-wage, nonagricultural occupations. The reports revealed documented cases in which temporary migrant workers were "deprived of their fair pay, imprisoned, starved, beaten, raped, and threatened with deportation if they dare complain." The report also recounted cases of employers going to great lengths to avoid hiring local workers in favor of temporary migrant workers, and how DOL inspectors tasked with oversight of both programs had failed to penalize employers to the fullest extent after discovering that they broke the law, even allowing many lawbreaking employers to keep hiring new temporary migrant workers (Garrison, Bensinger, and Singer-Vine 2015a, 2015b; Bensinger, Garrison, and Singer-Vine 2016).
In regard to the H-1B visa, a program dominated by information technology staffing companies and prominent technology firms, examples of abuses include temporary migrant workers whose wages were stolen and who were held in debt bondage (Smith, Gollan, and Sambamurthy 2014) and grade school teachers who were victims of human trafficking (Stockman 2013). In the J-1 visa, nonimmigrant exchange visitors participating in cultural exchange programs have been trafficked for sex work (Mohr, Weiss, and Baker 2010), robbed of wages, charged exorbitant fees, and forced to rely on soup kitchens (Stewart 2014), and gone on strike to protest major U.S. corporations for providing them with poor working and living conditions (Preston 2011; Jordan 2013). In July 2019, a federal court approved a $65.5 million settlement to be paid to ten thousand former workers in J-1 status by the sponsor agencies that recruit migrant workers through the State Department's J-1 au pair program. The lawsuit alleged that the sponsor agencies colluded to keep wages low and violate minimum wage and overtime laws (Slevin 2019).
In the L-1 visa—for managers, executives, and specialized knowledge employees of multinational firms and which (like H-1B) is used frequently by information technology staffing firms—the Department of Labor found a group of temporary migrant workers from India who were being paid $1.21 an hour in California for computer-related jobs (Avalos 2014). This amount represented what their wage would have been in Indian rupees, but the median wage for the job in the area was $19 or $45 dollars per hour according to survey data, depending on the exact duties of the position (Costa 2014).
A Mother Jones report found that foreign [End Page 30] graduates of U.S. universities on F-1 student visas who were employed through OPT were trapped "in virtual servitude" while working in the tech industry with the universities that sponsor OPT as "willing partners" (Swaminathan 2017). Numerous cases have been revealed in which domestic workers employed with A-3 and G-5 visas were victims of human trafficking, sometimes at the hands of high-profile diplomats (Ramchandani 2018). In Los Angeles, a bakery owner was ordered to pay $15.3 million in 2016 for exploiting eleven workers from the Philippines who slept on the floor and were paid $2 per hour. The workers were employed with E-2 visas, a visa intended for use by foreign investors, but which also allows investors to bring employees to work for them (Poston 2016).
Disney lobbied for the creation of the Q-1 visa in IMMACT90, to allow the company to staff its theme parks with "cultural exchange" workers. According to Professor Kit Johnson (2018), the visa has saved Disney more than $19 million per year because "Q-visa workers are paid less, on average, than their U.S. citizen counterparts; they require no health insurance or pension plans and are exempt from certain employment taxes; they pay to live on-site; and they reduce turnover costs in that they tend to leave their positions infrequently."
In regard to the TN visa, which allows Canadian and Mexican citizens to work in the United States in professional occupations under NAFTA, a 2018 report by a migrant worker advocacy group reported on numerous cases of economic coercion, wage theft, fraud, and discrimination against college-educated TN workers (Mauldin 2017). A news report in 2019 detailed how U.S. farms have recruited college-educated veterinarians from Mexico, but ultimately provided them with low-wage jobs on dairies requiring little or no education (Perez 2019).
Few rules are in place to protect temporary migrant workers in most visa programs or the U.S. workers employed alongside them. The Department of Labor—the federal agency tasked with protecting workers in the U.S. labor market—only has a specific mandate to oversee the H visa programs. And though DOL sometimes investigates and finds wage violations committed by employers against temporary migrant workers, as it did in the case of the L-1 Indian tech workers just cited, investigations like that one involved violations of the federal and state minimum wage laws, not any specific L-1 visa program rules. In fact, outside the H visas, virtually no rules are intended to protect the temporary migrant employees, other than the basic labor standards that apply to all workers in the United States.
Changing jobs or employers is difficult in most U.S. temporary work visa classifications, but the terms and conditions of some nonimmigrant visas permit migrant workers to change employers relatively easily—in particular the J-1, F-1/OPT H-1B, and TN visa—though the rules vary even among these visas. In the J-1, which is managed by the State Department, sponsor organizations partner with the State Department to manage oversight and compliance. Those private organizations act as middlemen between the J-1 workers and U.S. employers, and ultimately must sign off on a job change for a J-1 worker. In the F-1/OPT context, universities play a key role and ultimately approve employment for OPT workers.
The H-1B process for changing employers is straightforward, but the worker must find a new employer willing to apply for an H-1B visa. The TN visa perhaps offers the easiest way to change employers—the migrant simply applies for a new TN visa or status listing the new employer and awaits a decision from USCIS (and may renew their status with the same employer in the same fashion). Nevertheless, migrant workers in these visa programs have still been subjected to substandard workplace conditions and been the victims of fraud and even trafficking, which suggests that the ability to change employers is not a panacea for protecting temporary migrant workers.
Wage rules in temporary work visa programs are also a key aspect of some programs. H-2A, H-2B, and H-1B have a number of rules that govern how employers should treat their temporary migrant workers, including payment of a prevailing wage higher than the federal minimum wage, designed to protect U.S. wage standards. No other visa program, however, has a similar rule. Some research suggests that even the wage rules in the H-2A, H-2B, and H-1B visa [End Page 31] programs are inadequate. For example, temporary migrant workers are sometimes paid less on average than similarly situated U.S. workers in the H-2B (Costa 2016, 2017) and the H-1B programs (Costa and Hira 2020; Hira 2015), which can result in downward pressure on local wage rates for similar occupations. A report by Lauren Apgar (2015) using Mexican Migration Project data reveals that the employment outcomes of Mexican temporary migrant workers in H-2A and H-2B were "as poor as, or even worse than, those experienced by unauthorized Mexican immigrants" and that "both groups are disadvantaged when compared with LPRs." Part of Apgar's explanation for this result is that H-2A and H-2B workers are tied to their employers and cannot easily switch jobs to take advantage of new skills they pick up. As a result, H-2A and H-2B workers who are technically "legal" don't benefit from a wage premium for being legally authorized workers vis-à-vis unauthorized immigrants.
In addition, U.S. worker employment can be directly affected. Media reports recount numerous instances when employers have laid off their U.S. worker employees, sometimes hundreds at a time, and replaced them with much lower-paid temporary migrant workers on H-1B visas (Preston 2016; Hira 2015; Kight 2019). The replaced U.S. workers were often forced to train their replacements (Preston 2016; Hira 2015; Kight 2019; Whitaker 2017). This can occur because of the legal frameworks of the visa programs, which in most cases do not require employers to first seek available U.S. workers before they are allowed to hire a temporary migrant worker. Only the H-2A and H-2B visa require that; as a result, in most cases replacing U.S. workers with lower-paid workers with non-immigrant visas is legal. Even in the case of H-2A and H-2B, U.S. workers have been replaced despite the rules (see, for example, Garrison, Bensinger, and Singer-Vine 2015a).
Nonimmigrant Work Visas and Comprehensive Immigration Reform
Temporary, nonimmigrant work visas have been considered one element of the three-legged stool that some politicians, advocates, and commentators believe is required to come to an agreement on comprehensive immigration reform (CIR), the other two being border and interior immigration enforcement and a legalization and path to citizenship for the unauthorized immigrant population (see, for example, Napolitano 2009). The business community has prioritized temporary work visa programs as a necessary component of CIR, without which they would possibly not support CIR legislation. Advocacy groups and labor unions have pointed out how temporary work visa programs keep migrant workers indentured and can be used to undercut wages and labor standards (Marshall 2009), and have pushed for additional rights for temporary migrant workers, including the ability to change employers and obtain LPR status (Parker and Greenhouse 2013).
Since at least 2006, coming to a political agreement on temporary work visas in the CIR context has been difficult. New York Senator Chuck Schumer, who was part of a team of eight senators who drafted CIR legislation in 2013, noted at the time that "this issue has always been the deal breaker on immigration reform" (Parker and Greenhouse 2013). Until major stakeholders and lawmakers come to a consensus on temporary work visa programs, CIR-type legislation that includes other important reforms—such as a legalization and path to citizenship for the unauthorized immigrant population—will face significant hurdles before it can become law.
New EB Immigrants and LPR Status
The characteristics of the migrants who are issued EB permanent immigrant visas and become LPRs are not well known because the Departments of Homeland Security and of Labor do not publish microdata on EB beneficiaries, though aggregate data about countries of origin, destination states, and the number of beneficiaries who adjusted their status or were new arrivals are available. Some longitudinal survey microdata about the characteristics and outcomes of EB immigrants are available through the New Immigrant Survey and have been analyzed for this volume but are quite limited (see Gelatt, this issue, 2020; Rosenzweig and Jasso 2013). Some additional useful information, however, is available from the Department of Labor's Office of Foreign Labor Certification [End Page 32] (OFLC), which processes employer requests for permanent labor certifications for jobs they wish to fill with permanent immigrants, who could then eventually be issued a new EB visa. That information is submitted to the department's Program Electronic Review Management (PERM) system and made public. The OFLC publishes annual reports offering basic analyses of the PERM data. However, those data do not reveal which of the applications eventually resulted in the issuance of an EB visa; they show only the first step in the application process. Finally, some conclusions about the EB population can be inferred by the nature of the respective EB preference.
The vast majority of EB immigrants are educated, skilled, or wealthy.
Almost all 120,120 visas available every year in the EB-1, EB-2, and EB-3 preferences can only be used to fill jobs that require skills, training, and advanced education, but may go to persons with extraordinary abilities who are not required to have a job offer at the time they petition for an EB-1 visa (many of whom are likely to have advanced degrees). The sole exception is the EB-3 other workers (OW) category of visas, which are available for filling permanent, year-round jobs that require little or no education and training. The EB-3 OW category is capped at ten thousand by statute but was reduced temporarily to five thousand by the Nicaraguan Adjustment and Central American Relief Act of 1997.9 Homeland Security data for the five most recent years for which data are available as of this writing (2013 through 2017) show that an average of just under 2,700 EB-3 OW visas have been issued per year, a relatively low number given the current annual cap of five thousand.
PERM data from OFLC contain information regarding the minimum education requirements for the job certifications they review for prospective EB immigrants. In 2018, 46 percent of jobs submitted for permanent labor certification required at least an advanced degree, 40 percent required a bachelor's, 11 percent required less than a bachelor's, and 3 percent were listed as Other (OFLC 2018). Little is known about the education levels of the approximately ten thousand EB-5 immigrants every year, but it is reasonable to infer that they are quite wealthy, considering that before 2019 they had to be able to invest at least $500,000 or $1 million to qualify. (Future EB-5 immigrants will have had to pay $900,000 or $1.8 million under the current EB-5 regulations.)
Most EB immigrants were previously employed in the U.S. labor market with nonimmigrant work visas before adjusting to LPR status.
Every year, roughly half of the approximately one million permanent immigrant visas issued every year in the FB and EB preference categories combined are issued to beneficiaries who are already in the United States and usually either hold a nonimmigrant work or visitor visa (Witsman 2018, figure 1). When those nonimmigrant beneficiaries obtain LPR status, they are considered to have "adjusted" their status to that of lawful permanent resident, as opposed to being "new arrivals" who are issued new immigrant visas abroad, which then authorize their admission into the United States in LPR status. In regard to the EB preferences, the proportion who are already in the United States and adjust is much higher. DHS data on LPRs list the numbers of EB immigrants who adjusted or were new arrivals; as table 5 shows, the vast majority of new EB immigrants over the past ten years for which data are available—87 percent—were adjustments of status rather than new arrivals.
The main exception in the EB preferences is EB-5, where the shares and the trend are reversed. According to DHS data, 9,877 total EB-5 visas were issued in 2017, 1,630 of which were adjustments of status (16.5 percent) and 8,247 were new arrivals (83.5 percent) (DHS 2019). A reasonable explanation for this is that EB-5 beneficiaries are generally wealthy foreign citizens residing abroad who have decided to use a large investment in the United States as a pathway to permanent residence in the United States, [End Page 33] rather than working in the United States temporarily for an employer who is seeking to obtain LPR status for them.
In terms of the specific nonimmigrant visa classifications that new EB immigrants are adjusting from while in the United States, unfortunately little is known because DHS does not publish its data on nonimmigrant visas held by migrants who adjusted to LPR status through EB preferences. PERM data represent only the initial step in the process of employers seeking to hire an EB immigrant and the visa classifications of workers seeking to adjust to LPR status as EB immigrants. In 2018, PERM data show that 68 percent of applications for labor certifications were for jobs held by nonimmigrants in H-1B, 7 percent by those in L-1, and 7 percent by foreign students on F-1 visas (OFLC 2018).
The only known available microdata revealing the specific visa classifications of nonimmigrants who completed the adjustment to LPR status through an EB preference were acquired from DHS via a Freedom of Information Act request by Lazaro Zamora, formerly of the Bipartisan Policy Center, for fiscal years 2010 through 2014. Zamora (2017) published research based on these data revealing that "about 43 percent of all individuals that adjusted to LPR in EB immigrant categories between FY2010 and FY2014 were H-1B or L-1 principal visa holders (not including their spouses and children)." That finding underscores that nearly half of all EB immigrant visas went to temporary migrant workers who were also educated professionals: the H-1B visa requires the nonimmigrant beneficiary to hold at least a university degree related to the field in which they work or equivalent experience, and L-1 beneficiaries—although in many cases a university degree is not explicitly required—work as either managers or executives for international firms (in the L-1A subclassification) or have specialized knowledge that makes them valuable to an international firm. As a result, in both L-1 subclassifications, most beneficiaries are likely to have a university degree.
Zamora graciously shared the microdata, enabling calculations of how many total EB visas were issued to beneficiaries who were employed with H-1B or L-1 visas, as well as H-4 and [End Page 34] L-2 (spouses of workers with H or L visas). In total, 469,687 EB visas were issued to nonimmigrants who held H-1B, H-4, L-1, or L-2 visas at the time they adjusted their status. These EB visas accounted for 76.8 percent of the 611,941 total EB visas issued via adjustment of status to persons in a nonimmigrant status in DHS's dataset for 2010 through 2014, of which 40.1 percent were issued to H-1B and H-4 nonimmigrants alone. The remaining 142,254 EB visas were issued to migrants in one of 111 nonimmigrant visa statuses or other statuses, including one nonstatus.10
A tiny fraction of EB immigrant visas that were issued between 2010 and 2014–503 total, or 0.08 percent—were issued to migrant workers in the two main nonimmigrant work visa programs for low-wage jobs, the H-2A and H-2B. This is strong evidence that skilled and educated migrant workers working temporarily in the United States are much more likely to adjust to LPR status than those working with non-immigrant work visas in agriculture or other low-wage jobs that do not require a university education.
Nearly one million migrant workers have been approved for EB immigrant visas but are waiting in the green card "backlog" while in a nonimmigrant visa status and face challenges in the labor market as a result.
The statutorily mandated per-country ceiling prevents more than 7 percent of EB visas from being issued to nationals of any one country in the preference categories. But a majority of workers who have been approved to receive an EB visa hail from just two countries, and as discussed, most EB visa applicants are already in the United States working under a nonimmigrant visa. According to PERM data, 68 percent of labor certifications filed for EB visas in 2018 were for workers in the United States who were employed under H-1B visas (OFLC 2018). Available reports on H-1B reveal that in most years most new H-1B visas are issued to Indian nationals (63 percent in 2017), Chinese nationals receiving fewer but always the second most (14 percent in 2017) (USCIS 2018).
Unsurprisingly, the most recent PERM data show that more than half (52 percent) of the applications submitted for permanent labor certifications for EB visas in 2018 were for workers who are Indian nationals and that 11 percent were for Chinese nationals—accounting for nearly two-thirds of all new labor certification applications. No other nationality accounted for more than 4 percent of the total (OFLC 2018). The high shares of labor certifications and petitions for EB visas for Indian and Chinese nationals has long been a trend (for more, see the OFLC's annual reports).
More than a hundred thousand new H-1B petitions are approved every year, and the total population of H-1B workers employed in the United States is nearly half a million (Costa and Rosenbaum 2017). That large numbers of employers are applying to obtain EB visas for tens of thousands of their H-1B employees every year, combined with the 7 percent per-country ceiling and that those workers are mostly from India and China, is what has resulted in what is known as the green card backlog. The 7 percent ceiling means that only 8,408 of the 120,120 visas available in EB-1, EB-2, and EB-3 are available each year for Indian nationals, and that another 8,408 are available for Chinese nationals. USCIS reported that as of April 20, 2018, 34,824 Indian nationals had an approved EB-1 petition, 216,682 an approved EB-2 petition, and 54,892 an approved EB-3 petition—a total of 306,601—but who are in the backlog awaiting for an immigrant visa number to become available to them as a result of the per-country ceiling (Kandel 2018). The total EB backlog for Chinese nationals was 67,031. Those numbers, however, reflect only the principal prospective immigrants, meaning that their derivative spouses and children are not counted. Using multipliers provided by USCIS, the Congressional Research Service estimates that 826,867 people in the United States are waiting in the EB backlog with an approved EB immigrant petition (Kandel 2018).
Prospective immigrants living abroad who have an approved EB visa but must wait for a [End Page 35] visa number to become available because of the per-country ceiling are also backlogged. The State Department reports that this population numbered 112,189 as of November 1, 2017; the top three countries are China, India, and the Philippines. The total combined backlog is 939,056.11
The number of nonimmigrants in the United States and persons abroad waiting in the EB backlog means extensive wait times for EB visas, in at least one case extending beyond the average life expectancy of a person living in the United States. According to the State Department visa bulletin, which is updated monthly to let prospective immigrants know when a visa number might be available in a particular preference category, EB visa numbers for Indian nationals waiting in EB-2 or EB-3 are available only to applicants who filed an immigrant petition in mid-2009 or earlier (known as a priority date) (U.S. Department of State 2019). EB-2 and EB-3 visas are available immediately for all other nationalities except Chinese, who need to have a priority date of November 2016 for EB-2 or January 2016 for EB-3, or November 2007 if the visa is in the EB-3 OW category. However, for Indian nationals who apply now for visas in EB-2 or EB-3, the wait times are much longer: David Bier (2018) estimates that Indian nationals will have to wait 151 years for a visa in EB-2 and seventeen years in EB-3.
The long wait times lead to insecurity and exclusion for those in the backlog as well as an employment relationship ripe for exploitation. For example, during the time that migrant workers in a nonimmigrant status are in the backlog, they cannot easily switch between jobs and employers by virtue of their status, which in turn makes it more difficult to improve their wages and working conditions (as discussed). The backlog also makes migrant workers more vulnerable to abuse by employers because the employer is in charge of the sponsorship process to obtain a permanent labor certification and responsible for filing an immigrant worker petition on the worker's behalf with Homeland Security. The worker's reliance on the employer to secure LPR status can make it more difficult for them to complain to their boss or labor standards enforcement authorities in the case of workplace abuses or labor violations, given that it could jeopardize their ability to obtain LPR status (Levy 2019; Misra 2020). In addition, migrant workers in the backlog who have children in a nonimmigrant status also face the possibility that their children will pass the age of eligibility for obtaining their own derivative LPR status during the wait in the backlog, resulting in their adult children becoming removable (Hauslohner 2019).
Considering that the share of visas issued to temporary migrant workers since 1990 has dwarfed those issued to permanent immigrant workers who arrive in the United States under LPR status, what does it mean to have a labor migration system tilted so heavily toward temporary workers? It means two things, at least. First, employers will continue to have increased access to migrant employees who can be legally underpaid and are virtually indentured to their bosses, and thus unlikely to complain about wages and working conditions. Why? Complaining could result in the loss of their temporary immigration status and subsequent removal from the United States. Second, the vast majority of the 1 percent of the U.S. labor market of temporary migrant workers will never become permanent residents or naturalized citizens, which will affect their ability to integrate into the United States and prevent many of them from earning higher wages.
Permanent immigrants with LPR status, on the other hand, are not tied to a particular employer, may switch jobs at will, and are not in fear of losing their ability to remain in the United States on the basis of workplace issues and their employment relationship. Another benefit of LPR status is the ability to access the same labor standards as U.S. citizens—such as by being able to approach state or federal labor authorities without fearing removal—and to earn higher wages. Apgar (2015) for example, [End Page 36] finds that permanent immigrants earned more than both nonimmigrants and unauthorized immigrants. Other research finds that earnings for LPR immigrants who then go on to become naturalized citizens see additional income gains (Pastor and Scoggins 2012).
Considering the legal framework for U.S. labor migration, especially the existence of annual numerical limits for EB visas but not for most nonimmigrant work visas, it is highly likely that the current trajectory for temporary and permanent work visas will continue if no major reforms are enacted. It is worth considering whether this outcome is what Congress intended when drafting and voting on IMMACT90. For example, was it foreseeable that the number of temporary, nonimmigrant work visas would grow from more than three hundred thousand in 1990 to more than 1.4 million in 2017? Did Congress intend to tilt the system so far in the direction of temporariness, and was the EB immigrant cap of 140,000 per year designed to exclude so many temporary migrant workers from ever obtaining LPR status? Or did members of Congress simply fail to predict that the U.S. workforce would someday include more than 1.6 million temporary migrant workers—and accordingly misalign the size of the EB immigrant cap?
If the U.S. labor migration system maintains its current legal framework and trajectory, employers will be able to continue using the immigration system to hire more migrant workers as the population and workforce grows, but fewer and fewer of those workers will ever have the opportunity to become permanent immigrants and eventually naturalized citizens and to benefit from the additional income gains and security that LPR status brings. This possibility raises important questions about the inclusion, integration, and political participation of temporary migrant workers (Cook-Martin 2019) that Congress should consider the next time it makes major reforms.
In terms of U.S. labor migration, what the legal landscape in the twenty-first century looks like will depend heavily on whether Congress decides to keep a system in place under which migrant workers are temporary, indentured, and often underpaid, and nearly all must return home someday, never having an opportunity to participate in American political life. Congress has the option, on the other hand, to update the immigration system in a manner that allows a much larger share of migrants to arrive in the United States with full labor and employment rights in the labor market, and the knowledge that they can eventually become naturalized citizens. A system that facilitates an immediate or quick transition to LPR status would allow migrants to confidently make long-term investments in themselves and their new home and to benefit from full participation in American society. [End Page 37]
[End Page 38]
[End Page 39]
Daniel Costa is director of immigration law and policy research at the Economic Policy Institute, United States.
1. Immigration Act of 1990, Pub. L. No. 101-649, 104 Stat. 4978 (November 29, 1990).
2. The diversity visa is not discussed in detail here because it does not exist specifically for employment purposes, though it does authorize employment (for more, see American Immigration Council 2017).
3. For example, any unused FB visas in one fiscal year become available to be used as EB visas in the following year.
4. More specifically, an unused visa number in EB-1, EB-2, or EB-3 becomes available for next preference category (EB-1 would go to EB-2, for example); an unused visa number in EB-4 or EB-5 then becomes available for EB-1.
5. The minimum investment amounts in EB-5 were recently increased to $1.8 million and $900,000 from $1 million and $500,000 respectively, by a regulation published by DHS on July 24, 2019, at 84 Fed. Reg. 35750 (see USCIS 2019b).
6. The reason that spouse beneficiaries of nonimmigrant work visa classifications have been included in this analysis is that many classifications for spouses permit them to obtain work authorization from USCIS, and because they are nonimmigrant counterparts to the spouses who are issued EB visas through their principal immigrant spouse.
7. Based on the share of J-1 visas issued in programs that authorize employment in years that data are available and analysis in earlier publications (Costa 2011; Costa and Rosenbaum 2017), I have estimated that 80 percent of J-1 visas authorize employment every year.
8. In general, although the U.S. government collects a significant amount of information on nonimmigrant visas from employers who hire temporary migrant workers, those data are nevertheless inadequate and recorded inconsistently across federal agencies. As a result, little is known about the temporary migrant workforce (for more, see Costa and Rosenbaum 2017).
9. The reduction of EB-3 OW visas by five thousand per year will remain in place until all of the adjustments under NACARA have been offset (see NACARA, P.L. 105-100 ).
10. The one included nonstatus refers to those who entered without inspection (EWI) and are technically unauthorized. EWIs accounted for 15,313 of all EB visas beneficiaries between 2010 and 2014.
11. This total combines the April 2018 total for the USCIS backlog with the November 2017 total for the Department of State backlog.