A new proposed rule from the Department of Labor meant to protect American workers is instead set to do the opposite.
Last September, President Donald Trump directed the Department of Labor (DOL) to revise how it calculates minimum, or “prevailing,” wage requirements across a host of work visa programs, especially the H-1B and employment-based green cards. These requirements determine how much American employers can pay foreign workers—and therefore whether or not those employers can use immigrants to undercut American labor.
Finally, a reason to check your email.
Sign up for our free newsletter today.
The new rule, published last week, offers two possible changes to the prevailing wage standard. In the rule’s primary proposal, which we term “Blind Benchmarking,” DOL simply tinkers with its long-standing framework, preserving two loopholes that allow employers to undercut Americans. The alternative option, “Experience Benchmarking,” would bar employers from obtaining H‑1B visas if they pay foreign workers less than comparable American workers. If they actually want to protect American workers from unfair foreign competition, the administration should adopt this alternative.
The DOL classifies jobs for which employers sponsor foreign workers into four “wage levels,” each with a required minimum salary. Under the Blind Benchmarking approach, DOL would simply raise these thresholds to prevent employers from paying Americans less than foreign workers.
But Blind Benchmarking suffers from two flaws that hinder its ability to achieve this goal. First, Blind Benchmarking determines wage levels based on job description rather than the qualifications of the foreign candidate. This encourages employers using the H-1B and employment-based immigrant programs to game the system by strategically writing job descriptions that secure a lower wage requirement.
Second, Blind Benchmarking does not use data that allow for apples-to-apples comparisons between foreign and American workers. Federal law directs DOL to set prevailing-wage requirements on education and experience, meaning the minimum pay of an H-1B software developer with a master’s degree and ten years of experience should reflect what Americans with the same job and qualifications typically earn. But DOL’s only data source for Blind Benchmarking, the Occupational Employment and Wage Statistics (OEWS), provides no information on education or experience.
Because of these flaws, we estimate that under DOL’s Blind Benchmarking proposal, 17 percent of H-1B visas will go to workers earning less than similarly qualified American workers in the same occupation and labor market.
Consider two illustrative examples of where this rule fails. In 2024, NVIDIA sponsored a 31-year-old, foreign-born computer science Ph.D. for a role as a computer and information research scientist in California, with a salary of $225,000. We estimate that this salary is more than $22,000 above the earnings for a typical American with the same qualifications in the same occupation and labor market. Yet under the DOL’s proposal, this worker would be ineligible for a visa.
The same year, the IT outsourcer Cognizant sponsored a 44-year-old computer systems engineer for a job in College Station, Texas, at a salary of $78,000. That is $18,000 less than a typical American worker with the same qualifications, occupation, and employment location. Under DOL’s proposed rule, this worker would still be eligible for an H-1B visa.
Fortunately, DOL’s alternative proposal, Experience Benchmarking, would solve the core problems with the current system. Experience Benchmarking sets sponsored workers’ prevailing wages based on their precise qualifications rather than the requirements listed in their job description. This change would eliminate incentives for employers to game the system by strategically listing lower job requirements.
Experience Benchmarking also uses additional data to make direct comparisons of foreign workers with comparable American workers when setting prevailing-wage requirements. A foreign worker with a master’s degree, five years of experience, and a sponsorship for a chemist job in Philadelphia would have to be paid at least what the typical chemist in Philadelphia with the same degree and experience earns. If the worker’s salary offer is lower, he or she would be ineligible for a visa.
As the Department argues in its proposed rule, Experience Benchmarking “essentially ends the practice of wage arbitrage,” in which employers exploit skilled visa programs to reduce labor costs. It would also better align skilled visa programs with their real purpose: admitting top talent with sought-after skills. A sponsored foreign worker commanding a large wage premium over comparable American workers suggests he or she offers rare and valuable skills, for which an employer is willing to pay handsomely, and from which America will benefit.
Critics of the U.S. immigration system have justifiably raised the alarm about how the H-1B and other employment-based pathways aren’t preventing wage arbitrage. While we disagree with hardline restrictionists who oppose high-skilled immigration, they are correct that the prevailing wage system is deeply flawed. Experience Benchmarking, unlike DOL’s main proposal, offers a way forward that actually fixes the system’s long-standing problems.
If DOL continues merely to fiddle with its broken system, it will fail American workers. Experience Benchmarking would represent a departure from how DOL has historically managed the prevailing-wage system, but it would help ensure that the new rule actually achieves its objective.
Photo by Mark Felix / AFP via Getty Images