Third International Metropolis Conference, Israel, November 1998
COMPARATIVE EXPERIENCES WITH TEMPORARY WORKERS:
THE UNITED STATES*

COUNTING TEMPORARY WORKING MIGRANTS

Data on the entry of temporary migrants into the United States are collected or made available by the Immigration and Naturalization Service (INS) and the Department of State (DOS). The INS temporary admission figures are frequently seen in the literature, but they are misleading and it is preferable to avoid or to nuance their presentation.¹ At the same time, the DOS data do not include information on certain categories of admission, but then both sources miss some migrants. Reliable data or estimates for the resident population of temporary workers are currently nonexistent. See table 1 for statistics corresponding to the discussion below.

State Visa Issuances

Visas are issued by the U.S. Department of State to individuals abroad. From 1992 to 1996 the number of visas issued grew 16 percent from 5.4 to 6.2 million. Individuals apply for their visa through U.S. embassies and, when the application is approved by the appropriate agency, DOS issues the visa for admission to the United States. These figures are for individuals, however, they are over counts to the degree that a few individuals who are issued a visa consequently fail to enter the United States. These figures also do not include visa "waivers" which permit individuals from qualified countries to enter the United States on a temporary basis without a nonimmigrant visa.²

These figures do not include nonimmigrant visas obtained in the United States through a nonimmigrant visa "change of status" which occurs when an individual is granted a new and different nonimmigrant visa within U.S. borders. For example, in FY 1995 there were more than 26,000 changes of status to H-1B, out of a possible 65,000 H-1B approvals, which are not captured in the visa issuance data.³ Most were from the foreign student (F) nonimmigrant visa.⁴ With no regular and accurate information on change of status, it is not possible to examine whether or not increasing proportions of speciality workers (H-1B) were previously foreign students (F), which might signal changes in the academic market and graduate training.

Nonimmigrant Admissions

The temporary or NIV admission data collected by the INS do not count individuals, but rather they count the multiple entries made by individuals over the course of the year. Each time the INS admits a tourist, businessperson, worker, etc., they are counted in the admissions’ data, which is consistent with its administrative role to inspect each entrant. The number grew 19 percent from 20.8 million in 1992 to 24.8 in 1996. Admissions’ data do not include parolees, withdrawal, stowaways, and refugees, but they do include admissions on a nonimmigrant basis under the temporary visa-waiver program. The visa-waiver program originally extended only to countries that offered a reciprocal waiver of visas to U.S. citizens. It is limited to admissions in the visitor for pleasure and for business classes of admission, with admission not to exceed 90 days. In 1996, there were 25 countries in the Visa Waiver Pilot Program with 11,192,932 admissions (INS 1997:108).

Several interesting observations emerge from the data in Table 1. First, some classes like temporary visitors (tourists, business travelers) and treaty traders make several entries and exits from the United States during a year — their admission numbers are four to five times as large as the number of individual visas issued. Second, international representatives, students, workers, exchange visitors, and intracompany transferees are less mobile — their admissions’ numbers are two to three times as large as the number of individual visas issued. Third, these latter groups are more stable in the United States, note that the person-year population of students is actually greater than the number of individual visas issued, an indicator that students and especially their families tend to reside in the United States throughout the academic and school year with few trips home.

Person-year Population

Temporary workers by definition come and go, but some sizable proportion effectively reside in the U.S. for at least one year and more. Yet the number of persons working in the U.S. at any given time is not known. In the case of H-1Bs and other workers the effective year-round population must be sizable. If 65,000 H-1Bs enter the U.S. and spend up to the three to six years permitted, then the number working in the U.S. at any point during the year might reach 180,000 to 360,000. However, we simply have no administrative data for this population.

Ideally, there should be a number of reliable estimates of duration of stay and of the average or person-year population. This is the "average" population, reflecting just the portion of a year that all temporary individuals spend in the United States. For example, this is the full-time equivalent contribution of working visas to the labor force — if there were 100 workers with an average stay of one-half year, they would contribute 50 person-years’ worth of labor in that year.

In 1994, the INS was able to estimate an average duration of stay for temporary workers that yields an estimate of the person-year population shown in Table X shows for selected working visas. The person-year population calculated here includes only those who travel outside of the U.S. during the year and return. In 1994, the only year for which an estimate is available, a year-round average population of 1.4 million temporary migrants (with tourists) was estimated, including some 48,815 temporary workers and trainees, 89,651 exchange visitors, and 20,765 intracompany transferees. These figures suggest that a small, numerical role is played by temporary migrants when compared to the total U.S. labor force; however, they underestimate the stock of the temporary population.

Occupations of Temporary Speciality Workers (H-1B)

The DOS visa issuances and INS admissions’ data do not tell us about the occupations of temporary workers. Data from the U.S. Department of Labor gives us an indirect measure of the occupations that H-1B workers are likely to fill.

The DOL data come by the way of Labor Condition Attestations (LCA) filed by employers who petition to have a job filled by a nonimmigrant. The first attestation system was enacted as part of the Immigration Nursing Relief Act of 1989.⁵ The Immigration Act of 1990 created attestations for crewmembers (D), a very small student-worker program (F), and for specialty occupation workers (H-1B). In principle, LCAs are faster to process than the more carefully verified certification that the jobs of certain classes of employment-based permanent residents must clear before a visa is issued. The employer must attest that certain wage and employment conditions prevail before they are permitted to fill a job(s) with a nonimmigrant worker. Once approved by the DOL, the employer may turn to the INS to have a nonimmigrant visa issued to a qualified foreign person.⁶

Table 2 shows, nevertheless, that something can be learned from the DOL data on applications for H-1B speciality nonimmigrants. Notice that there is substantial growth in the average wage of Computer-related occupations (24%) from 1992 to 1997, as might be expected for an occupation with growing employer demand and spot-shortages of qualified U.S. workers. More interesting is the11 fold growth in applications for H-1B workers. Although the number of applications has no direct relationship to actual individual workers — recall that applications may include numerous applications by the same employer and that multiple jobs may be requested on each application — they do reflect increased employer search activity / demand for such workers. The movement of the wage and applications data loosely mirror the increase in visa issuances in H-1Bs. They suggest that the demand for computer-related (foreign) workers is greater today than in other occupations.

Adjusters From Temporary to Permanent Status

How temporary are "temporary" workers? Not only do they stay several years in the United States with a temporary visa, a sizable proportion stays on and gains permanent status. Temporary nonimmigrants are permitted to "adjust" their status in-country to that of legal permanent resident. Working temporaries tend to adjust within the same occupation, about one-third of the H-1B adjusters report science or engineering occupations, while 90 percent of intracompany transferees (L) report executive or administrative occupations upon adjustment. In contrast, temporaries for whom work is most often secondary, e.g., foreign students (F) and exchange visitors (J), do not have typical occupations upon adjustment (Lowell 1996:8).

We might expect an increase in the numbers of those adjusting following the Immigration Act of 1990 because it increased the number of permanent LPR slots available. In the year immediately following the Act’s implementation in 1992, the increase in numbers also had to do with the relief of a backlog that had built up in previous years. Table 3 shows that there was more than a doubling of the number of those adjusting from temporary specialty worker (H) to any permanent (LPR) status between 1991 and 1992. Note that those adjusting from intracompany L status increased almost fourfold across the same two-year span. Following a decline in the number of adjusters FY 1994 and 1995, after the initial backlogs had been worked off, the numbers increase again. Perhaps, the increase has to do with the growing number of specialty workers (H-1Bs) and intracompany transferees (L’s) who were admitted in 1992, and four years later in 1996, have become eligible to adjust status.⁷

Table 3. Adjustments to Legal Permanent Residence from Selected Nonimmigrant Classes.

Thus, we can see a visa-status trail leads from temporary to permanent residency. There are no data that tell us what proportion of temporary nonimmigrants make the transition from temporary NIV to legal permanent resident LPR status. But while the data do not permit us to directly track adjusters, indirect estimates suggest that the proportion of adjusters as of FY 1994 are substantial (North 1996:74).

Nearly four-tenths of temporary speciality workers (H) may adjust into some LPR immigrant category, but just 4 percent of exchange visitors (J). Of course, these figures do not tell us about the longer chains that exist — such as a "change in status" from one temporary visa to another (e.g., F student to H-1B speciality worker), followed by the adjustment into permanent status. One crude lower-bound estimate based on available data suggests, for example, that about one-fifth of the foreign students moved both from F directly, and indirectly through H-1B, to achieve any permanent status in FY 1996 (Lowell, 1996:8). The estimate of direct adjustment show above, therefore, is too low.

Occupations of Status Adjusters and the Immigration Act of 1990

Two outcomes of the Immigration Act of 1990 should be evident in the adjustment patterns of temporary migrants and in the overall occupational make-up of those admitted into permanent status. Because the Act increased the number of admissions for skilled employment-based workers, there should have been an increase in the relative numbers of those admitted in professional categories.⁸ Because the Act fosters temporary to permanent visa-adjustments, by permitting H-1Bs to intend to stay in jobs that are permanent in nature, we might expect an especially strong growth in the occupations represented by these adjusters.

These are to some extent "counterfactual" possibilities that can be best addressed by asking what would admissions look like if the 1990 Act had not been implemented. Table X addresses this question by taking the linear trend from FY 1972 to FY 1991 and extrapolating it to FY 1996 by admission category (with projections bounded by category maximums). The actual FY 1996 numbers are then compared to the "what if" projections to get an idea of how much growth occurred after the Immigration Act of 1990 went into effect in 1992 (Greenwood and Ziel, 1997).

Source: Greenwood and Ziel, 1997.

The upper-left quadrant of the table 4 shows that for "new" principal immigrants — those with no reported prior work experience in the United States — there was an actual decline in numbers from a predicted 152,598 to 118,376. The 1996 figure was only 78 percent of those predicted for new immigrants. Conversely, among principles who adjusted from a prior temporary status, the numbers were 252 percent larger than predicted with the greatest growth in speciality and technical occupations which include computer scientists. At the same time, among only those adjusters from H-1B status there was really substantial growth in the number of those reporting health professional occupations (498 percent), while adjustments from L status saw substantial growth in the executive and managerial occupations (353 percent).

Clearly, there has been substantial growth in the numbers of those in professional, scientific and engineering occupations following implementation of the Immigration Act of 1990. Growth in executive, administrative, and managerial occupations occurred only for those adjusting from temporary to permanent status. The bulk of the speciality occupations (H) adjusts into professional and especially health-related occupations, while the bulk of intracompany transferees adjusts into executive and managerial occupations. Temporary migration is clearly affecting the skill-composition of the permanent admission system.

Footnotes

¹ Observers tend to omit or to be very careful in discussing, for example, the INS multi-count of illegal border entries or apprehensions. Temporary admissions data are little more valid.

² Established by the Immigration Reform and Control Act of 1986, extended by the Immigration At of 1990 though 1994, and further extended by subsequent legislation.

³  In 1995, these would be added to the 51,832 visa issued to H-1B nonimmigrants (number that year unknown). Additionally, individuals are not counted against the 65,000 legislative limit when they are granted an extension of time beyond the time allotted on their first H-1B visa (recall an initial three years is allowed with up to six years maximum with the same or another U.S. employer). In FY 1998 as many as 47,500 extensions were petitioned for and, assuming that most all were approved, these logically might be added to the 65,000 initial visas issued in that year (as a measure of employer demand for jobs to be filled by H-1Bs).

⁴  Special tabulations from data provide by the INS cited in Lowell, 1996:7.

⁵  The nonimmigrant nursing program (H-1A) was discontinued in 1995.

⁶  The DOL collects information on the employer, on each petition made by an employer (who may petition for multiple persons to fill a job description), and on the occupation of the job being petitioned for and the wage offered. Along with the admissions data just discussed, the INS collects petitions for nonimmigrant workers. Because the two data systems are not linked in any fashion, and because the INS does not gather basic job information, the number of temporary admissions by occupational category is not known.

⁷  The different temporary classes have distinct propensities to adjust into specific LPR categories of admission. As of FY 1994, 47 percent of foreign students (F) adjusted into the exempt spouse category and 44 percent into the skilled EB category (EB-3). In contrast, about one-half of both speciality workers (H-1B) and exchange visitors (J) adjusted into the skilled EB-3 category; while about one-half of intracompany transferees (L) adjust into the priority EB-1 category.

⁸  Immediate occupational increases would reflect those in the employment-based backlog in the first two years following the Act’s implementation in 1992.

Back to Events Index Page