1990, 2000, America First, American workers, bachelor's degree, banking, computer related fields, Disney, employment, foreign workers, H1B, H1B cap, H1B dependent employers, H1B visas, immigration, India, innovation, Insourcing, IT, masters, outsourcing, prevailing wage, salaries, science, skills, STEM, Temporary work visa, unemployment, US workers, wage estimate
“The results are mixed. In banking and computer science, dependent h1b employers pay much less compared to natives than non-dependent h1b employers, who in fact pay more than natives. In the computer science data, the H1B coefficient is now positive, revealing that dependent employers account for all of the negativity found in the original equation.” (Sperry, 2017, p. 19)
Excerpts from: “Are H1B Visa Workers Paid Less than Similarly Employed Natives? ” by Sperry, Will (34 pages), Haverford College. Department of Economics, 2017:
The H1B visa is under scrutiny over fears of Americans losing their jobs or experiencing lower salaries, especially in STEM occupations. Some have suggested that companies prefer the H1B workers because, among other reason, they are paid less than similarly employed Americans. This paper compares the salaries of H1B workers in America with similarly employed native workers in banking, computer science, sciences and architecture and engineering occupations in 2015. Combining the 2015 Occupational Employment Survey by the Bureau of Labor Statistics and the 2015 H1B disclosure data published by the Department of Labor, I regress salary with an H1B dummy variable, a variable for location and occupation, as well as other relevant controls, in order to test if H1B holders are in fact paid less than similar natives. Overall, I find that H1B holders paid between $6000 and $16000 [less] in the STEM jobs, but paid $2000 more in banking jobs. I also find that observations where the prevailing wage came from an alternative survey source were correlated with lower salaries, and that H1B employers who are dependent on H1B workers have even lower salaries than H1B employers in total. Finally, I use an incomplete skill level variable to estimate how much of the salary gap is due to differences in skill level between H1B workers and Americans. By comparing H1B and H1B dependent employers, I find evidence that much of the salary gap is due to skill level, but some of it may be due to tampering of prevailing wage data.” (Page 1 – original is missing [less] but based on the thesis, this appears to be the missing word.)
(Note that by “Indians” Sperry clearly means Asian Indians, i.e. from India)
The H1B visa is a temporary work visa offered by the US government that focuses on skilled immigrants, especially in the computer science and technology fields. Created originally in 1952, the visa underwent many changes and has been the subject of multiple presidential acts.
In 1990, the annual cap was set at 65000 workers per year, and this number would continue to rise until it reached over 200,000 in the early 2000s, when it would be restricted again to 80,000 workers per year. Workers are sponsored for a visa by their employing company, and are able to stay in the US for a period of three years, with a possible three-year extension afterward. As of 2014, 57% of visa went to Indians, with 65% of all visas going to jobs in computer related fields. The visa holders are mainly in the 25-34 age bracket, and around half have a bachelor’s degree, with around half holding a masters.
Recently, the H1B visa has again come to the attention of the government as fears of outsourcing grow (especially in computer related fields, with 20% of all employees in the US being immigrants). For example: a lawsuit against Disney which gained national attention, in which former employees claimed that Disney had made them train their H1B replacements before letting them go in order to save on IT costs. This lawsuit, which resulted in a hearing in front of Congress, represents a growing advocacy movement in opposition to the visa.
Opponents of the visa argue that, especially in the STEM fields, foreign workers are preferred to natives because companies can pay them a lower salary.i By moving into the American labor market, they argue that H1B workers are depressing wages, increasing competition for employment, and making STEM fields less attractive for American students. Technically, H1B workers must be paid a prevailing wage, which is determined by skill level and occupation, and is designed to ensure that H1B workers are not underpaid compared to natives with similar skills in similar jobs. However, there are three aspects of the application system which critics assert makes it possible to ignore the prevailing wage. First, employers are free choose from a number of prevailing wage sources for the lowest wage estimate available. Secondly, the US government admits that it is not able to adequately check for the accuracy of the prevailing wages used in applications. Finally, critics suspect that employers misrepresent the jobs that visa holders will do, filling out their applications with lower paying jobs, and then having them do higher level work…” (pp.2-3)
“To conclude, I find that H1B employees are underpaid in the STEM fields, but not in banking. I find that dependent employers pay less than other h1b employers, and that most of this difference is due to dependent employers hiring lower skilled employees and paying less of a premium above the prevailing wage. However, there is still a difference in salary, and prevailing wage, when accounting for skill level, suggesting that some other factor, potentially illegal, is causing dependent employers to pay lower salaries. In addition, I find some evidence that employers use private or alternative prevailing wage surveys to procure lower estimates, and pay a lower salary, although the evidence is not nearly conclusive. From these results, it seems clear that H1B workers are less skilled than their native counterparts in STEM occupations, and that most are not crucial to the innovation of American industries. I cannot conclude that H1B employees are paid less than similar Americans due to the omission of a skill level variable, but my results do suggest that H1B dependent employers may be finding ways to pay their H1B workers less than the actual prevailing wage, especially in the computer science occupations…” (p. 25) Original (34 pp.) found here: https://scholarship.tricolib.brynmawr.edu/bitstream/handle/10066/19254/2017SperryW.pdf https://scholarship.tricolib.brynmawr.edu/handle/10066/19254 This item’s license is described as http://creativecommons.org/licenses/by-nc/4.0/
“Are H1B Visa Workers Paid Less than Similarly Employed Natives? by Sperry, Will (34 pages)
Advisor: Kontorovich, Vladimir
Department: Haverford College. Department of Economics
Issue Date: 2017
Access Restrictions: Open Access
Permanent URL: http://hdl.handle.net/10066/19254
NOTE THAT H1B VISAS ARE DUAL INTENT AND CAN BE CHANGED TO GREEN CARDS AND THEN CITIZENSHIP. BASED ON THE CENSUS DATA, ONE SUSPECTS THAT ONCE THEY GET CITIZENSHIP THAT, AT LEAST IN THE CASE OF ASIAN INDIANS, THEY WILL HAVE A HIGHER MEDIAN INCOME THAN AMERICANS, SINCE EAST INDIANS HAVE A HIGHER MEDIAN INCOME AS A WHOLE. THIS COULD BE, AT LEAST IN PART, BECAUSE OF MINORITY SET ASIDES. This is from 2010, since we were unable to find again the more recent one in a timely manner: https://www.census.gov/newsroom/pdf/cspan_fb_slides.pdf