Submitted electronically via https://www.federalregister.gov/documents/2025/09/24/2025-18473/weighted-selection-process-for-registrants-and-petitioners-seeking-to-file-cap-subject-h-1b
Business and Foreign Workers Division
Office of Policy and Strategy, USCIS
U.S. Department of Homeland Security
5900 Capital Gateway Drive
Camp Springs, MD 20746
RE: Department of Homeland Security, Weighted Selection Process for Registrants and Petitioners Seeking To File Cap-Subject H–1B Petitions, Notice of proposed rulemaking, CIS Docket No. 2820-25, DHS Docket No. USCIS-2025-0040, RIN: 1615-AD01
The Economic Policy Institute (EPI) is a nonprofit, nonpartisan think tank established in 1986 to include the needs of low- and middle-income workers in economic policy discussions. EPI conducts research and analysis on the economic status of working America, proposes public policies that protect and improve the economic conditions of low- and middle-income workers—regardless of immigration status—and assesses policies with respect to how well they further those goals. EPI submits these comments to the U.S. Department of Homeland Security (DHS) and its subagency, U.S. Citizenship and Immigration Services (USCIS), in response to their Notice of Proposed Rulemaking (NPRM) regarding the visa allocation process for H-1B visas. The proposed rule takes the current equal weighting random allocation process —often referred to as the H-1B “lottery”— for visas when more petitions than the number of visas available are filed when the application window opens, and modifies it to allocate H-1B visas to employers based on a weighted selection process that would generally favor the allocation of H-1B visas to higher skilled and higher paid applicants.
EPI has researched, written, and commented extensively on the U.S. system for labor migration, including in particular the H-1B program and other temporary work visa programs. EPI published a lengthy piece of research detailing the need to improve the way the U.S. Department of Labor (DOL) sets the H-1B wage levels,1 which we coauthored and have annexed to this comment as a separate PDF file. The report, however, did not address the flawed manner in which H-1B visas are allocated; a problem that goes hand-in-hand with H-1B prevailing wage levels that have been set artificially low for many years.
In 2016, President Donald Trump campaigned on reforming the H-1B program and immediately promised to do so after becoming president-elect. Yet virtually no substantive action was taken until just weeks before the 2020 election. At that time, the U.S. Department of Labor (DOL) and USCIS/DHS issued three rules, a DOL Interim Final Rule (IFR) updating the H-1B prevailing wage levels,2 a USCIS IFR modifying the definition of H-1B “specialty occupation,”3 and an NPRM, on the H-1B allocation process by lottery.4 While DHS and DOL have the requisite legal authority to make these regulatory changes, the timing and process of their issuance made them susceptible to procedural legal challenges. On December 1, 2020, a federal court in California struck down both the DOL prevailing wage IFR and USCIS IFR.5 During his 2020 campaign, President Biden’s immigration plan stated he would “reform temporary visas to establish a wage-based allocation process and establish enforcement mechanisms to ensure they are aligned with the labor market and not used to undermine wages,”6 but the Biden administration never took any steps to propose an updated allocation scheme for H-1B visas to implement that campaign promise.
EPI generally supports the main substance of this NPRM and believes it improves the current allocation process for H-1B visas, which is random and susceptible to companies that have learned how to “game the system” to unfairly obtain large numbers of H-1B visas at the expense of other companies seeking to hire H-1B workers. We do not believe that it is the best method for allocating visas because high shares of petitions will still be awarded to firms paying their H-1B workers at below-median wage rates, but it is reasonable and preferable to the status quo. If the NPRM or final rule issued pursuant to this NPRM is challenged in court based on DHS’s interpretation of the statute that establishes how H-1B visas are allocated, then DHS should consider alternate methods to achieve the same goal of allocating visas by wage levels. These comments will suggest one alternative way to do it.
This proposed rule will incentivize H-1B employers to pay their migrant worker employees at fairer wage rates that are commensurate with local U.S. wage standards and result in rewarding employers that recruit and hire higher-skilled workers and pay higher salaries with additional visas and workers. That in turn will improve the program by reducing the number of H-1B workers who are underpaid according to U.S. wage standards—but without reducing the overall number H-1B visas that are issued—and will also better protect the wages of similarly situated U.S. workers. The proposed rule will also increase the overall skill mix and quality of the pool of H-1B workers, thereby boosting the impact of the H-1B program on the U.S. economy. As a result, the proposed rule helps address a major critique EPI has long held about the program—that it is exploited by firms that use it to legally undercut U.S. wage standards—and which Members of Congress from both major parties have attempted to address through proposed bipartisan legislation to reform H-1B, which includes a preference allocation system for H-1B visas, one that is more detailed and includes additional factors based on other policy priorities, unlike the one in the proposed rule based on a weighting of wage levels).7
It must also be noted at the outset of these comments that actions taken both recently and in 2020 by federal agencies with respect to wages for migrant workers in temporary work visa programs have been inconsistent and confusing. While DOL took action with its 2020 proposal and IFR that—if it had not been struck down—would have raised wage rates for underpaid migrant workers in the H-1B, H-1B1, and E-3 visa programs, DOL almost simultaneously issued a new wage rule for the H-2A program that would have cut wages for the migrant farmworkers in that program.8 DOL has now in 2025 again issued another rule for H-2A visas that will drastically cut the wages of farmworkers and lead to billions of dollars in wages being transferred from poor farmworkers to farm owners and operators, and lead to reduced opportunities for U.S. farmworkers, including the 30% of farmworkers who are U.S.-born citizens.9 This is troubling and misguided, especially considering that DHS at the end of the first Trump administration had determined that farmworkers were part of the U.S.’s critical infrastructure workforce, and the Department of State has designated H-2A workers as “a national security priority”—because of their contribution to stabilizing the food supply chain during the Coronavirus pandemic.
Both DHS and DOL should issue regulations that lead to improved labor standards and higher wages for all work visa programs, and not treat workers differently based on their education levels, occupations, and nationalities. All temporary migrant workers deserve to be paid fairly for their labor and no work visa programs should operate essentially as employment law loopholes that permit employers to legally underpay migrant workers.
The first major section of these comments addresses the flaws in the current H-1B program, and the second section specifically addresses elements of the NPRM.
The H-1B program is an important avenue for attracting skilled, talented workers from abroad—but is deeply flawed and in desperate need of reform
The H-1B program provides temporary, nonimmigrant U.S. work visas for college-educated workers and fashion models from abroad. While no one can deny the importance of attracting skilled, talented workers to the United States, the reality is that some of the biggest beneficiaries of the H-1B program are outsourcing companies that have hijacked the system—using it to pay low wages, replace thousands of U.S. workers with much-lower-paid H-1B workers, and to send decent-paying technology jobs abroad.10 Outsourcing companies, however, are not the only abusers of the system: The vast majority of employers who use H-1B visas are legally allowed to pay their H-1B workers at wage levels below the local median wage for the occupation.
The major structural, programmatic flaws in H-1B are the following:
U.S. employers do not have to recruit U.S. workers before hiring H-1B workers
Employers and corporate lobby groups claim that they use the H-1B primarily to bring in the “best and brightest” workers from abroad to fill labor shortages when they can’t find willing and available U.S. workers, especially in science, technology, engineering, and math fields (STEM), but the reality is that:
- Under current law, employers are not required to recruit and hire U.S. workers or prove they are experiencing a labor shortage before hiring H-1B workers.
- “H-1B-dependent” employers—those filling 15% or more of their U.S. jobs with H-1B workers—are required to recruit U.S. workers first, but they all get around the requirement by exploiting a cheap and easy loophole: they can hire an H-1B worker without recruiting U.S. workers if the H-1B worker holds a master’s degree or receives an annual salary of over $60,000.
- For comparison, $60,000 per year is $56,810 less than the national average wage for all workers employed in computer occupations.11 Analysis of the FY 2024 DOL Labor Condition Application disclosure data shows that employers exempt themselves from recruiting U.S. workers for effectively all of H-1B positions (over 99%).12
U.S. employers can legally underpay H-1B workers
For years, corporate lobbyists and other H-1B proponents have claimed that H-1B workers cannot be paid less than U.S. workers because employers must pay H-1B workers no less than the “prevailing wage.” That is true only as a tautology. The reality is:
- Employers have the option of paying the prevailing Level 1 “entry-level” wage or Level 2 wage, which are between 20-40% less than the local median wage (Level 3) that employers pay workers in the occupation in the local region.
- As DHS notes in Table 12 of this draft rule, the five-year average of cap-subject beneficiaries who were paid at the two lowest wage levels—both of which are set below the local median wage—was 83% for fiscal years 2020-2024.
- While the wage level is supposed to correspond to the H-1B worker’s education and experience, in practice the employer gets to choose the wage level and the government doesn’t verify whether the two reasonably correspond to each other.
H-1B workers are often exploited and some arrive to the United States in debt after paying hefty recruitment fees
H-1B workers sometimes pay hefty fees to labor recruiters, which means that many arrive virtually indentured to their employer, fearing retaliation and termination if they speak out about workplace abuses or unpaid wages. And widespread abuses have been documented—even human trafficking and severe financial bondage.13
H-1B workers do not have sufficient job mobility between employers
The H-1B visa itself is owned and controlled by the employer; an H-1B worker who is fired or laid off for any reason becomes deportable unless they can find new employment within 60 days. This arrangement results in a form of indentured servitude.14 During waves of mass layoffs in the tech industry, numerous stories have been publicly reported about the precarious situations and impossible choices that thousands of H-1B workers are left to face.15 Thus, H-1B workers have good reason to fear retaliation and deportation if they speak up about wage theft, workplace abuses, or other working conditions like substandard health and safety procedures on the job. While H-1B workers have the ability to switch jobs if they can find another employer willing to petition for a new visa for them (sometimes referred to as “portability”), and have 60 days to find a new employer if they are fired, these avenues are not straightforward enough and inadequate to mitigate the power that employers have over the right of their H-1B workers to remain employed in the United States.16 These protections should be improved.
H-1B workers are not allowed to self-petition for lawful permanent residence
The ability of H-1B workers to become lawful permanent residents and remain in the United States is entirely at the whim of their employers. Even after working for an employer for six years in H-1B status, the employer has the power to decide if an H-1B worker can remain in the country—in many cases after an H-1B worker has established firm roots in the United States. That power keeps H-1B workers from complaining and asserting their employment rights. To remedy this wrong, H-1B workers should be allowed to petition on their own for permanent residence after a short provisional period—no longer than 18 months—and independent from their employer.
Outsourcing companies are using the H-1B program to underpay H-1B workers, replace U.S. workers, and send tech jobs abroad
As we have documented in the annexed report and numerous other commentaries and testimonies, roughly half of the top 30 employers of H-1B workers in most years are not innovative high-tech firms like Apple and Google.17 Some of the biggest users of the H-1B visa are staffing firms that specialize in information technology (IT) and accounting and that pay H-1B workers the lowest wages legally allowed, and outsource their H-1B employees to third-party firms. Some of those firms also have a business model dependent on sending jobs offshore where labor costs are lower.
Typically in this scenario, H-1B workers do computer and engineering work at the office of a U.S. employer but are employed by an outsourcing company, some of which are based abroad or have major operations abroad.18 The many reported cases of U.S. workers being laid off and replaced by H-1B workers have all been facilitated by this arrangement. In multiple incidents, the H-1B workers have been hired with annual wages of around $30,000 to $40,000 less than the workers they have replaced. Before they are laid off, the U.S. workers are often forced to train their own H-1B replacements as a condition of their severance packages; this is euphemistically known as “knowledge transfer.” Major, profitable U.S. employers like Disney and Toys “R” Us—as well as public employers and institutions like the University of California and regulated utilities like Southern California Edison and Northeast Utilities—have laid off thousands of U.S. workers who were forced to train their own replacements. Eventually, many of the outsourced jobs that are filled by H-1B workers get moved offshore.19
Contrary to the popular narrative proffered by corporations that support expanding and deregulating the H-1B visa program—the staffing firms that use H-1B visas are not using them to keep technology jobs in the United States—instead they are using them precisely to facilitate the offshoring of as many of those jobs as they can. That is in fact, the business model of those firms. News reports, including from Bloomberg and the New York Times, have shown that outsourcing companies “game the system” in order to obtain a high share of H-1B visas, which leaves fewer available for the firms that directly employ H-1B workers.20
Allowing outsourcing companies to hire H-1B workers lets employers utilize the immigration system to degrade labor standards for skilled workers—as a result, they should be barred from obtaining H-1B visas
The outsourcing/staffing model of employment generally may increase the incidence of employment law violations by separating the main beneficiary of the labor provided by H-1B workers—the third-party firm that hires the outsourcing firm, i.e. the “lead” employer—from the H-1B workers who perform the work. Firms that rely on outsourced H-1B workers are a textbook (if extreme) example of what former DOL Wage and Hour administrator David Weil calls a “fissured” workplace, where the relationship between the worker and the lead employer is fissured, or broken, via the use of a temp agency or subcontractor21 (in this case the H-1B outsourcing firm). Research shows that fissuring leads to a wage penalty for workers who are subcontracted, employed as temps, and work for staffing firms,22 in part because the subcontractor keeps a percentage of the wages earned by the workers. It is also common knowledge that employers use this model to avoid paying for benefits like health care, retirement funds, and to avoid liability for labor violations. Because the staffing and outsourcing model contributes to the fissuring of the labor market, it should not be allowed as part of the U.S. immigration system—not in H-1B or in any other temporary or permanent immigration programs.
Analysis of the NPRM, “Weighted Selection Process for Registrants and Petitioners Seeking to File Cap-Subject H-1B Petitions”: Updating the H-1B lottery
We now turn to the NPRM, which proposes to update the random lottery allocation system for cap-subject H-1B petitions. The H-1B statute at 8 USC §1184(g)(3) requires that H-1B visas or statuses “be issued … in the order in which petitions are filed for such visas or status.” However, the practical realities of the H-1B annual numerical limit or “cap” and the way that USCIS receives petitions for H-1B visas, renders this impossible to implement in practice—leaving USCIS little choice other than to propose a rational alternative that is consistent with the broader goals and intent of the H-1B statute.
There is no feasible way for USCIS to meet the statutory requirement of allocating H-1B visas “in the order in which petitions are filed.”
At present, when demand for H-1B visas immediately exceeds the annual limit of 65,000 for cap-subject petitions and annual limit of 20,000 visas reserved for foreign graduates of U.S. universities who have obtained at least a master’s degree—USCIS allocates the visas by a random lottery where each petition has an equal chance of being approved, regardless of occupation, region, or that salary that will be paid to the H-1B worker. This process is utilized because, since fiscal year 2014, USCIS has received far more H-1B petitions that are subject to the annual cap in the first five days of the eighteen-month application window, than the 85,000 available slots for that fiscal year. Since so many petitions are submitted to USCIS nearly simultaneously, it is impossible for USCIS to determine the order in which the petitions were filed, a necessary prerequisite to adequately comply with the H-1B statute’s requirement that H-1B visas or statuses be allocated to employers “in the order in which [those] petitions [were] filed for such visas or status.” Therefore, USCIS allocates them at random via an electronic lottery, which appears to be a choice of convenience and not a well-reasoned response to a phenomenon unanticipated by the statute. The process has been in place since FY 2008.
However, considering the ambiguity of 8 USC §1184(g)(3) and the fact that the letter of the statute cannot be adhered to in practical terms, other interpretations about how to allocate H-1B visas can be equally reasonable so long as they are consistent with the intent of the H-1B statute.
The prevailing wage and its enforcement is the most important protection for labor standards in the H-1B program
DHS should measure the merits of any allocation scheme based on how well it advances the administrative efforts to meet the H-1B program’s intent, “to help employers who cannot otherwise obtain needed business skills and abilities from the U.S. workforce.”23 As we’ve described above, and elsewhere, one of the most effective ways for the H-1B program to meet this intent is to employ a labor market test that includes good faith recruitment from the U.S. domestic supply and a requirement that employers hire qualified applicants before an H-1B application is approved. Absent a labor market test, under current law, DHS and other agencies involved in H-1B governance, DOL, the State Department, and the Department of Justice (DOJ), should craft rules to move the program closer to fulfilling its intent.
The current allocation scheme, based on a random lottery, undermines the government’s efforts to fulfill the program’s intent. Replacing it with a better scheme, as proposed by the NPRM, would improve program outcomes: better addressing labor shortages and increasing the overall skills of the H-1B worker population.
As DHS drafts its final rule, the agency should consider and rank alternate allocation schemes based on how well their outcomes meet the program’s purpose. Allocation schemes that do a better job at filling pressing labor shortages should be ranked higher. Since there is no direct way to measure the severity of labor shortages, proxy measures must be used. The best proxy for this purpose is the migrant worker’s wage premium compared to comparable U.S. workers. The larger the wage premium, the more likely it is that the migrant worker brings specialized skills that cannot otherwise be obtained from the U.S. workforce. Migrant workers earning larger premiums are more likely to complement, rather than compete with, U.S. workers.
The prevailing wage is the most important worker protection in the H-1B program and its effective administration is essential. The H-1B program laws and rules specify that the prevailing wage for each position is calculated using three attributes: (1) occupational classification; (2) worksite geography; and (3) level. The employer selects these three variables based on the position description they are trying to fill, not the actual worker’s skills. In virtually all Labor Condition Application (LCA) filings, employers use the Office of Foreign Labor Certification (OFLC) Occupational Employment and Wage Statistics (OEWS) data or a private wage survey to determine the position’s prevailing wage, rather than requesting that one be provided by the National Prevailing Wage Center, as employers are permitted to do.
Allowing employers to interpret the key factors that determine the prevailing wage introduces errors. First, employers interpret the occupation and level differently, even in cases when they position is the same. Further, granting employers substantial front-end discretion over choices when they have significant conflicts of interest invites widespread non-compliance, which can only be remedied through strong back-end enforcement. But the government’s back-end enforcement has been nonexistent. DHS’ worksite visits and Requests for Evidence may have improved compliance with (2) worksite geography. However, no government agency ever ensures that the employer’s choice of (1) occupational classification and (3) level, accurately reflects the position being filled, or the H-1B worker’s actual activities in the position. Any employer that misclassifies the occupation or level, or both, significantly undermines program integrity, regardless of whether the misclassification is intentional or inadvertent. Public disclosure data indicate that some employers are likely exploiting these vulnerabilities on a mass scale, stealing wages from H-1B and U.S. workers alike.
Examples from existing H-1B data demonstrate the failure of the government’s prevailing wage regulations in achieving their raison d’être, which is protecting workers, labor standards, and the integrity of the labor market.
Take this first example which comes from one of Microsoft’s H-1B employees, which can be found at DOL LCA Case Number I-200-23087-882196, and which was selected and approved by USCIS for an H-1B visa in the FY 2024 lottery.24
Example of Microsoft H-1B position at Level 1 prevailing wage (entry-level)
| Employer | Microsoft |
| Date of Birth | 1989 |
| Age | 35 |
| Education | DOCTORATE DEGREE |
| Degree | AERONAUTICS & ASTRONAUTICS |
| Job Title | APPLIED SCIENCES |
| Prevailing Wage Level | I (Entry) |
| Prevailing Wage | $90,438 |
| Wage Rate of Pay From | $126,220 |
Source: USCIS Form I-129 petition data obtained by Eric Fan, Bloomberg, through Freedom of Information Act request from the U.S. Department of Homeland Security, see BloombergGraphics / 2024-h1b-immigration-data at GitHub. LCA data comes from LCA Programs disclosure data files for fiscal year 2024, Office of Foreign Labor Certification, Performance Data, Employment and Training Administration, U.S. Department of Labor; DOL LCA Case Number I-200-23087-882196.
Microsoft, one of the top ten H-1B employers, placed a 35-year-old worker with a doctorate degree into a position it classified at Wage Level 1. Based on their age, the H-1B visa recipient likely has at least 7 years of experience in addition to holding a doctoral degree. Yet, they are filling an entry-level position, one that, according to DOL’s prevailing wage guidance, would be appropriate for an “internship” or a “worker in training.”25 This appears to be an example of wage-level misclassification that can only be remedied through tighter rules and stronger enforcement.
The second example is for one of Deloitte’s H-1B positions, which can be found at DOL LCA Case Number I-200-23258-349652, which was certified by DOL in FY 2023.26
Example of Deloitte Consulting H-1B position at Level 1 prevailing wage (entry-level)
| Employer Name | Deloitte Consulting LLP |
|---|---|
| Job Title | Senior Consultant |
| SOC Code | 15-1299.08 |
| SOC Title | Computer Systems Engineers/Architects |
| Prevailing Wage Level | I |
| Prevailing Wage | $61,838 |
| Worksite City | Philadelphia |
Source: LCA Programs disclosure data files for fiscal year 2023, Office of Foreign Labor Certification, Performance Data, Employment and Training Administration, U.S. Department of Labor. DOL LCA Case Number I-200-23258-349652.
Deloitte Consulting, another one of the top ten H-1B employers, classified its Senior Consultant job in Philadelphia as an entry-level position, one that, according to DOL’s prevailing wage guidance, would be appropriate for an “internship” or a “worker in training.” Classifying a Senior Consultant at Wage Level I directly conflicts with DOL guidance that states that job titles with “‘senior’ (e.g., senior programmer)” should be classified at Wage Level III. Yet, the DOL certified the LCA despite the employer’s apparent wage-level misclassification.
The Deloitte case also highlights problems with the wage surveys used to source prevailing wages. In this case, the corresponding prevailing wage for Standard Occupational Code (SOC) 15-1299.08 (Computer Systems Engineers/Architects) is extremely low by any benchmark. The National Association of Colleges and Employers (NACE), a professional association of university-based career services officers and business campus recruiters, conducts regular wage surveys of recent bachelor’s degree graduates across the country. For the class of 2024, the average starting salary for computer and information sciences majors was $88,907.27 The DOL H-1B prevailing wage of $61,838 is $27,000 less, offering Deloitte a 30% discount over the market rate for recent graduates.
This demonstrates how lax DOL oversight and guidance on employers’ occupational code choices can lead to market distortions. Had Deloitte selected SOC 15-1252, Software Developers, which has a similar occupational profile, the prevailing wage rate it would have been required to pay was 30% higher.
Given these operational realities, DOL and DHS should make it a priority to implement a comprehensive back-end compliance system that includes random audits, to validate that the occupation and prevailing wage level accurately match the H-1B worker’s actual activities in the position. Given DOL’s expertise in labor standards enforcement and occupational duties, and the Wage and Hour Division’s (WHD) primary role in enforcing the promises made by employers on LCAs, DOL should take the lead on such compliance efforts, with DHS possibly partnering in support of DOL. Any enforcement of prevailing wage levels, including random audits, will require additional funding and staffing for DOL. As a result, the administration should make a request to Congress for new funds to hire additional WHD investigators. (A recent report shows that the number of WHD investigators is at 611, an all-time low,28 likely due in part to the administration’s funding and staffing cuts at DOL.)
Accurately identifying prevailing wages is also critical to ensuring the effectiveness of the allocation scheme. The government has recently announced significant efforts, such as the forthcoming DOL proposed rule to update prevailing wage levels, as well as Project Firewall, an enforcement initiative, to improve program integrity. While we believe such initiatives may be a positive step, they remain only aspirational—and may be difficult to implement given significant DOL funding and staffing cuts, and we worry about DOL following through, given DOL’s actions on other fronts that we are critical of—thus our analysis is based on the current state of program implementation.
A successful H-1B allocation scheme depends on accurately identifying and labeling prevailing wage levels and occupations
The key factor for the allocation scheme is determining the foreign worker’s wage premium, which is calculated as the difference between the H-1B worker’s wage and that of a similarly situated U.S. worker. Identifying the H-1B wage is straightforward since the proposed rule requires employers to submit the proffered wage in their electronic registration, rather than just the wage level selected at the LCA level. The imprecision of the employer’s choice of both occupation and level makes benchmarking against similar U.S. workers problematic since the agency cannot ensure they are, in fact, similar.
The proposed rule requires employers to map their proffered wage to a wage level based on the SOC and geography. This step eliminates the most problematic element in the prevailing wage process, i.e. the employer’s interpretation of the position’s level. However, the employer’s interpretation of the occupation and the possibility of its misclassification remain problematic. The interpretation of occupations is non-standard, and different firms would likely map the same exact position to various occupations. For example, firms may interchange a relatively low wage six-digit level occupation, such as 15-1211 Computer Systems Analysts, which at the national level has a Level III wage of $103,790, with the higher wage of 15-1252 Software Developers, which has a Level III wage of $133,080 at the national level.29 An employer paying an H-1B worker $104,000 would be Wage Level III if they select the position of Computer Systems Analyst, but Level I if they identified the position as Software Developer.
The agency can mitigate the adverse effects of occupational misclassification on the proposed allocation scheme by having employers use a four-digit, instead of a six-digit, SOC to identify the wage level for registration purposes. Employers would select the six-digit occupation with the highest median wage within its four-digit SOC family, and then map their proffered wage to the corresponding wage level. For example, applications with any computer occupation (15-12XX) would map their wage levels to 15-1221, Computer and Information Research Scientists. Correcting occupational misclassification this way is analogous to the proposed rule’s handling of multiple worksite locations to, “prevent gaming of the weighted selection process…”30
Using a wage level prioritization allocation for H-1B visas is reasonable and consistent with the H-1B statute, and it should be the preferred method for allocation
The program statute states that the visa is open to any “specialty occupation” and that prevailing wages should be adjusted based on “the occupational classification in the area of employment.”31 Congress wrote this language in recognition that the wage structures of occupations vary, e.g., accountants are typically paid lower wages than software developers, and that the cost of living varies across geographies, e.g., the California Bay Area has a much higher cost of living than Cleveland, Ohio. The allocation scheme should favor registrations that offer the highest wage premium—as compared to the highest wage amount—measured after controlling for occupation and geography.
USCIS’s proposed rule in 2020 and final rule in 2021 would have achieved this by allocating H-1B petitions first to employers who file petitions that pay Level 4 wages—the highest H-1B wage level—and then to petitions with the lower wage levels in descending order (Level 3, Level 2, and then Level 1).32 Since higher wages are a valid proxy for higher skills in this context,33 the proposed allocation system by wage levels would have resulted in the highest-skilled and best paid applicants with in an occupation being selected for H-1B petitions from the available annual pool of petitions. In fact, this formulation continues to be our preferred allocation scheme for H-1B petitions, although we would modify it to include the requirements from the current NPRM that employers list the proffered wage level on their registration filing that would then be mapped to the appropriate wage level, and when a position is located in multiple geographic locations, using the lowest wage level among the locations as the proffered wage. Had it gone into effect, the rule would have resulted in the selected H-1B workers being paid at higher wage levels, and those higher wages would have helped safeguard U.S. wage standards in major H-1B occupations like information technology and other computer occupations. This is essential given that workers in those occupations have seen virtually no real wage growth since the late 1990s.34 Wage levels are an appropriate proxy for skill since they account for wage variations by geographic location (higher versus lower cost areas) and occupation.
The statutory language at 8 USC §1184(g)(3) is ambiguous and silent as to how visas should be allocated if they cannot be issued in the order that they were filed, and it offers no additional insight into what Congress meant by “filed.” The filing of an H-1B petition is the most reasonable explanation of what “filed” means, but since allocating H-1B visas in the exact order in which they were filed is rendered impracticable by virtue hundreds of thousands of petitions being submitted in one day or over the course of a few days, creating an allocation scheme based on wage levels is more reasonable than creating a random lottery. In addition, a wage level-based allocation scheme is reasonable because it is consistent with the intent of the H-1B visa statute, which is to provide “American businesses” with “highly skilled, specially trained personnel to fill increasingly sophisticated jobs for which domestic personnel cannot be found”35 (emphasis added). Since employers are not required to test the labor market for available U.S. workers before hiring H-1B workers (as discussed in section I), prioritizing higher-skilled, higher-paid H-1B workers also ensures that the employers facing a true shortage of talent—as indicated by the higher wage levels they are willing to pay—have a better chance at obtaining an H-1B visa, which also furthers the goals set out in the statute.
The proposed weighted lottery is reasonable and an improvement from the status quo but will not achieve an ideal outcome
With all of that being said, including our preferred proposal for an allocation scheme discussed in the preceding section, we nevertheless generally support DHS’s proposed weighted allocation scheme because it is reasonable and a measurable improvement over the current scheme in meeting the H-1B program’s intent. It will shift the total number of H-1B petitions that are selected where employers will be required to pay wage rates that are higher than the local median wage for the occupation (i.e., Level 3 and Level 4 wages).
Under the current random allocation scheme, DHS shows in Table 13 of the NPRM that petitions at every wage level have a 29.59% probability of being selected. However, given that so many more petitions have wages promising to pay Level 1 and Level 2 wages, the random lottery results in an outcome where 83% of H-1B beneficiaries selected are at Level 1 and Level 2, as shown in Table 12, with only 17% selected at Level 3 or 4. DHS notes that moving to a weighted lottery would result in the probability of the wage levels being selected shifting towards an increasing the likelihood that petitions at Level 3 and Level 4 are selected: “15.29 percent [probability] for level [1], 30.58 percent for level [2], 45.87 percent for level [3], and 61.16 percent for level [4].”36
Assigning greater weight to petitions at Level 3 and Level 4 will bring implementation of the H-1B program closer to the intent of the original statute, while still retaining a high degree of randomness and permitting significant numbers of petitions at Level 1 and Level 2 to be selected. An allocation process that uses wage levels, rather than a prioritization scheme that uses individual salary levels, will also likely result in H-1B petitions being allocated among a broad mix of occupations and regions, which we believe is a preferred outcome—while also prioritizing petitions for the most highly-skilled workers in each of the occupations.
However, we believe DHS should consider and model other alternatives to H-1B petition allocation that they believe are reasonable, lawful, and help align the H-1B program with the intent and goals of the statute. Other allocation methodologies have also been proposed publicly by advocates; we would welcome discussion, modeling, and analysis of those proposals by DHS.
But given the information that we have available to us now, as noted in the preceding section, we believe that the wage level prioritization in DHS’s proposed rule in 2020 and final rule in 2021 would be preferable. By first allocating H-1B petitions to employers who file petitions that pay Level 4 wages and then to petitions with the lower wage levels in descending order, the result would be that the highest-skilled and best paid applicants within an occupation would be selected for H-1B petitions.
Proposals for a prioritization scheme based solely on compensation levels will favor a small number of occupations and regions, and should be rejected
Some advocates have called for an H-1B allocation proposal that ranks petitions by actual salary level—rather than a wage level that accounts for skill, occupation, and geography—with the latest being published in September 2025 by the Institute for Policy.37 The author of the proposal, Jeremy Neufeld, argues that DHS’s weighted lottery will result in IT outsourcing firms gaining 8% more H-1B visas, that F-1 visa graduates would receive 7% fewer visas, and that the H-1B median salary would increase by just 3%.
Given our understanding of Neufeld’s proposal and modeling, his analysis does not exactly match DHS’s current proposal, calling into question his results. Neufeld appears to have utilized the wage levels selected by employers at the LCA level—rather than what DHS proposes—which is to take the “proffered” wage; i.e. the actual salary that the employer intends to pay the beneficiary—and use that to determine the appropriate wage level listed on the registration that will be weighted, according to the OEWS wage levels for the occupation and region. We know from our own reviews of H-1B LCA and petition data that in many cases the actual salary proffered by employers would correspond to a higher wage level according to OEWS wage levels by occupation and region. (In other words, an employer may select Level 2 but pay a wage that actually corresponds to Level 3 or 4.) Why this occurs reveals how the administrative process is flawed, but the fact remains that any modeling of DHS’s weighted lottery would require mapping proffered wages to wage levels to achieve a reliable result.
We believe that Neufeld’s proposal based on actual salary levels would have certain consequences, not all of which are accounted for in his September 2025 publication or an earlier publication that discusses multiple proposals for an H-1B allocation scheme, including one based solely on “compensation ranking.”38 In that report, Neufeld concedes that a compensation-based ranking of petitions would result in more H-1B petitions being awarded to employers in higher-cost regions of the United States. Those regions would mostly include the urban areas on both coasts of the United States, where the cost of living is higher.
The goal of the H-1B program is to fill labor shortages. Occupational shortages/surpluses are not measured by absolute wages, and a high absolute wage for a specific occupation is not a reliable indicator of a labor shortage. Instead, as a 2001 National Academies of Sciences, Engineering, and Medicine study explains, shortages are observable: “When demand exceeds supply in a particular occupation, compensation tends to rise relative to compensation in other occupations that require similar education, effort, and working conditions” (emphasis added).39
In his analysis, Neufeld does not state explicitly that a larger share of H-1B petitions would go to certain occupations. We believe those are most likely to be computer and engineering occupations, which will crowd out petitions for H-1B workers who would be employed in occupations like teachers, attorneys, accountants, and doctors (especially doctors employed in rural areas). In other words, a compensation-based ranking would mostly benefit technology companies like Google, Microsoft, Amazon, and Meta, that employ H-1B workers in computer occupations who reside on the coasts of the United States.
Ample evidence shows that computer occupations are not experiencing a labor shortage, so prioritizing their allocation would undermine rather than advance the H-1B program’s goals. Wage growth for computer occupations has been stagnant, technology firms have been laying off skilled software employees by the tens of thousands, and recent U.S. computer science graduates are facing the worst job market in a generation, with record levels of unemployment and underemployment, just as U.S. computer science students earn degrees in their highest numbers.40 The latter fact is especially salient since the vast majority of H-1B approvals, 83% according to Table 12 in the NPRM, are for Level 1 and 2 positions, the very jobs that the large surplus of computer scientists in the U.S. should fill. Neufeld’s proposed scheme would prioritize computer occupations—the very job market experiencing a labor surplus rather than a shortage.
Neufeld proposes accounting for the geographical outcome by “adjusting for regional price parity can address geographic heterogeneity without significantly reducing the economic benefits of compensation-based ranking,” and further suggests that DHS make age adjustments, because compensation-based ranking “may not account for the full future potential and lifetime contributions of younger workers who start at lower salaries but have high growth potential.” The age adjustment proposal is not expounded on further, so we are not sure how it would operate in practice.
We have two reactions to this. First, we believe that an allocation scheme solely based on compensation would be clearly inconsistent with the H-1B statute on its face, and thus require Congressional action (as opposed to a weighted or wage level lottery that can be implemented by DHS regulation). The H-1B statute at 8 U.S.C. § 1182(n)(1)(A)(II) clearly states that the H-1B prevailing wage should take into consideration “the occupational classification in the area of employment,” unless it is based on a wage that is already being paid to an employer’s similarly situated U.S. (non-H-1B) employee. Using wage levels that reflect occupation and region are clearly consistent with the statute, and thus more likely to be upheld if challenged in court. A compensation-based ranking scheme reflects neither, and while Neufeld attempts to account for the “area of employment” language in the statute with a suggestion of “adjusting for regional price parity,” he does not explain how that would work in practice—and in any case, it would likely insert an additional and unnecessary layer of complexity to the H-1B registration and allocation process.
Second, we believe that an H-1B program that permits a broader range of occupations is preferable, rather than one that mostly awards H-1B petitions to tech companies. Employers in various industries need a viable pathway to hire skilled talent from abroad and should thus not be closed off from the U.S.’s main visa program that could allow them to do so. We believe that all occupations and contributions of workers have value—and that the H-1B program was designed to permit employers to hire the most skilled and talented workers within a broad range of occupations and industries—not simply as a “computer visa” that almost exclusively benefits tech and information technology companies.
Early career foreign graduates of U.S. universities should have a pathway to employment in the United States that leads directly to a green card
Numerous other commenters, including Neufeld, raise the concern about wage-level-based allocation schemes crowding out younger and early career migrant workers who are just starting their careers. We agree that skilled and talented younger and early career workers, including many of those who graduate from top U.S. universities, deserve a pathway to employment in the United States. However, the H-1B program was created to help employers fill labor shortages with skilled workers, not as an early career development program for foreign students that leaves them in an indentured status for years, if not decades.
There are better proposals that would allow employers to recruit and hire foreign graduates of U.S. universities, including the Keep STEM Talent Act, which would put foreign students on a direct path to a green card if they have a job offer from a U.S. employer that pays the local median wage.41 Nevertheless, early-career workers are not excluded from the H-1B program and would have a legitimate opportunity for an H-1B visa under the weighted allocation, even if they were paid a Level 1 or Level 2 wage. Many early career graduates who are presently paid at Level 1 and Level 2 wages, especially those with advanced degrees, should be offered Level 3 and Level 4 wages, which would make them even more likely to be selected through a weighted allocation scheme. And employers who are currently underpaying their H-1B employees may also change their behavior and offer higher wages to increase their chances in the weighted lottery.
It should also be noted that advocates for more H-1B petitions being approved at below-median wage levels ignore current employment realities, wherein recent college graduates in computer occupations are experiencing unemployment rates that are “more than double the unemployment rate among recent biology and art history graduates, which is just 3 percent,” as the NY Times recently reported. The Times cited statistics from the Federal Reserve Bank of New York—noting that “[a]mong college graduates ages 22 to 27, computer science and computer engineering majors are facing some of the highest unemployment rates, 6.1 percent and 7.5 percent respectively”42—and pointed to examples of skilled graduates applying to hundreds of jobs with little luck. Entry-level H-1B jobs at wage levels 1 and 2 are exactly the types of opportunities that should be available first to U.S. workers, including both U.S.-born citizens and permanent immigrants with green cards. Yet, the H-1B program undermines their opportunities by allowing employers to underpay temporary migrant workers with a precarious immigration status that employers can exert an extraordinary level of control over.
The H-1B random lottery benefits outsourcing companies that pay low wages and game the system, making it more difficult for start-up and direct-hire firms to hire H-1B workers
The current random lottery allocation may give each petition equal odds, but it does not give each firm equal odds. The main beneficiaries of the current random lottery allocation process have been H-1B employers that are staffing firms, which are more likely to pay H-1B workers at the lowest wage rates.43 The H-1B staffing firms use an outsourcing model to send their H-1B employers to third-party worksites, and earn their profits by undercutting local wage rates for college-educated workers. These companies, as Bloomberg and the New York Times reported, “have obtained many thousands of the visas — which are limited to 85,000 a year — by learning to game the H-1B system without breaking the rules.”44 The “system” the Times is referring to here is the H-1B random lottery.
Real-world scenarios illustrate how the random lottery has favored outsourcing firms over those seeking truly skilled workers. The outsourcing firms have most of their workforce in low-cost countries such as India, and employ hundreds of thousands of workers. If an outsourcing firm seeks 1,000 approved H-1B petitions in any fiscal year, it will submit 3,000 petitions on behalf of its workers located in India. Since the odds have been roughly one-in-three, the random lottery rewards the outsourcing firm with the 1,000 approved petitions it sought. Since many of the workers the outsourcing firms seek to hire have relatively lower skills—as evidenced by being assigned wage rates that are almost exclusively at Level 1 and Level 2—they are largely interchangeable workers, and it matters little to the outsourcing firms which specific one thousand workers are selected. Compare this to a start-up firm seeking to hire a specific engineer with a truly special skill set that it is willing to pay a Level 4 wage to obtain; the start-up firm has only a 33% percent chance of winning the lottery. The outsourcing companies, on the other hand, flood the lottery with multiple applications to obtain thousands of visas per year, which crowds out start-up firms and employers with legitimate needs that employ workers directly and pay H-1B workers at higher wage levels.
It’s important to emphasize that the DHS rule that moved to a beneficiary-centric registration process to reduce lottery abuse perpetrated primarily by small IT staffing companies did nothing to eliminate the abuse described above by outsourcing firms. DHS must enforce the requirement that employers have a bona fide job for the prospective H-1B worker and begin rejecting and revoking petitions in cases where employers maintain large numbers of “travel-ready” and “visa-ready” workers for speculative positions.45
USCIS should consider alternate approaches in case of a legal challenge.
Some commenters and critics of the proposed H-1B allocation process may contend that the statutory language at 8 USC §1184(g)(3) does not authorize DHS to update the H-1B allocation process by wage level as detailed in the proposed rule, because DHS would no longer be issuing H-1B visas in the order in which the petitions were filed. While we disagree with this assessment as discussed herein, DHS should nevertheless explore alternative methodologies for wage-based allocation that it can propose in a new rulemaking if the proposed rule is the subject of a successful statutory challenge in federal court.
One simple new methodology, for example, could consist of having staggered filing deadlines for petitions by wage levels. 8 USC §1184(g)(3) is silent as to whether there can only be one filing period or whether there can be multiple. Therefore, USCIS could have a first filing period, where only petitions with jobs paying Level 4 are considered. Once all the Level 4 petitions are submitted and approved, then a second filing period at a later date could be set to receive only petitions with jobs paying Level 3 wages. After those are collected and approved, if there are any visas remaining under the H-1B cap, then a filing period for Level 2 wages would be next, and finally a filing period for Level 1. With a process like this, the cap-subject H-1B petitions in a given fiscal year would not all be submitted at once, thereby allowing USCIS to adjudicate and allocate petitions “in the order in which” they were filed, as the statute requires. If there end up being more petitions than available H-1B slots during a filing period for a particular wage level, USCIS could conduct a “mini-lottery” in order to randomly allocate the petitions within that wage level.
Fixing the H-1B program requires improving the H-1B prevailing wage methodology
As we have discussed in this comment, a reasonable and fair H-1B prevailing wage methodology that truly reflects market rates for H-1B occupations is essential to reforming the H-1B program and must exist in tandem with any allocation process that can adequately safeguard wages and working conditions. At present, the H-1B prevailing wage methodology is neither reasonable nor fair.
Section 4 of President Trump’s recent proclamation directs the Secretary of Labor to initiate a rulemaking process to revise the H-1B prevailing wage levels.46 We hope to see DOL’s proposal soon. The current methodology sets H-1B wages based on DOL survey data by occupation and local area, setting them at specified percentile levels in the distribution of surveyed wages in the OEWS, which have ben chosen by DOL. The current wage levels are set at arbitrarily low levels (i.e. percentiles), leading to the vast majority of H-1B workers being paid at wage rates that are below the local median wage for the occupations they fill (as discussed at length in this comment). DOL in 2020 finalized a rule to amend the H-1B prevailing wage levels, but it was blocked in federal court on procedural grounds, and the Biden administration DOL considered proposing its own updated H-1B methodology soon thereafter, even soliciting comments from the public on it, but the Biden DOL ultimately never introduced its own proposed rule.
DOL has the requisite legal authority to attempt again to modify the H-1B prevailing wage levels to appropriate rates that protect U.S. wage standards and prevent adverse effects caused by the H-1B program. For far too long, the H-1B wage levels have been set at an artificially low level that undercuts U.S. wage standards—and we have called on both Republican and Democratic administrations to fix the problem using their executive authority. It is reasonable for DOL to do so now, and we call on DOL to update the prevailing wage levels so that Level 1 is no lower than the local median wage for the occupation, i.e. the current Level 3 wage.
Fixing the H-1B program requires increased and improved labor standards enforcement and adequate funding and staffing at worker protection agencies like DOL
As we have already noted, adequate reform of the H-1B program that improves wages and working conditions for all workers requires robust labor standards enforcement, in particular adequate oversight from the Wage and Hour Division (WHD) at DOL. Unfortunately, the current outlook for labor standards enforcement is bleak.
Federal budgets for the past decade at least have heavily prioritized immigration enforcement over labor standards enforcement, with nearly 14 times as much being appropriated for immigration enforcement ($30.2 billion) as compared to the amount appropriated for labor standards and worker protection agencies ($2.2 billion) in 2023.47 The 2025 budget bill passed by Republican legislators through reconciliation has dwarfed this disparity, giving $170 billion to the administration to carry out its immigration enforcement activities, while failing to appropriate even one additional cent to worker protection agencies. This, at a time when worker protection agencies are already underfunded and understaffed.
Figure A shows that, in inflation-adjusted 2023 dollars, WHD—the primary agency responsible for enforcing the wage promises made by employers in H-1B LCA filings—had a budget in 2006 of $250 million, and in 2023, $260 million—an increase of just $10 million over nearly two decades. As Figure A also shows, this trend has been consistent with appropriations for two other key worker protection agencies, the Occupational Safety and Health Administration (OSHA) and the National Labor Relations Board (NLRB). At both OSHA and the NLRB, inflation-adjusted appropriations were significantly lower in 2023 compared with 2006.
Funding for federal worker protection agencies has declined or been flat since 2006: Annual appropriations for the Wage and Hour Division, the Occupational Safety and Health Administration, and the National Labor Relations Board, fiscal years 2006–2023, in constant 2023 dollars (in thousands)
| Fiscal Years | Wage and Hour Division | Occupational Safety and Health Administration | National Labor Relations Board |
|---|---|---|---|
| 2006 | $250,419 | $714,035 | $377,469 |
| 2007 | $254,506 | $715,567 | $369,605 |
| 2008 | $248,595 | $687,799 | $356,300 |
| 2009 | $274,244 | $728,662 | $372,958 |
| 2010 | $318,047 | $780,591 | $396,011 |
| 2011 | $308,315 | $756,706 | $383,080 |
| 2012 | $301,340 | $749,549 | $369,349 |
| 2013 | $281,455 | $700,089 | $344,976 |
| 2014 | $288,734 | $710,795 | $352,953 |
| 2015 | $292,467 | $710,647 | $352,534 |
| 2016 | $288,824 | $701,793 | $348,142 |
| 2017 | $282,799 | $687,154 | $340,880 |
| 2018 | $276,056 | $670,771 | $332,753 |
| 2019 | $272,931 | $664,132 | $326,831 |
| 2020 | $284,910 | $684,294 | $322,848 |
| 2021 | $276,624 | $664,833 | $308,361 |
| 2022 | $261,332 | $635,070 | $285,512 |
| 2023 | $260,000 | $632,309 | $299,244 |
Notes: All values have been adjusted to constant 2023 dollars using the Consumer Price Index (CPI-U) and reflect totals for the U.S. government’s fiscal year (October 1 to September 30).
Source: U.S. Department of Labor, Fiscal Year 2025—Department of Labor, Budget in Brief and Archived Budgets, fiscal years 2008–2024; National Labor Relations Board, Performance Budget Justification, fiscal years 2008–2024.
In addition to funding levels that have barely kept up with inflation at WHD—the primary agency tasked with protecting labor standards in the H-1B program—the number of WHD investigators that the agency employs is at an all-time low. Those investigators are also primarily responsible for ensuring that federal wage and hour laws are obeyed by employers across all 50 states and U.S. territories, and protecting the roughly 170 million workers in the U.S. labor market. Figure B shows that there were only 733 WHD investigators at the end of 2023 to enforce all federal wage and hour laws, 79 fewer than in 1973, the first year for which data are available, and 499 fewer than the peak year of 1978 when there were 1,232 WHD investigators. By May 2025, just four months into the Trump administration, the number of WHD investigators had dropped to 611, a new historic low. While the number of WHD investigators is now half of what it was at its peak in 1978, the number of workers those investigators have a mandate to protect has tripled, and the number of establishments subject to WHD enforcement has quadrupled.48
Number of federal wage and hour investigators is at a historic low: Number of Wage and Hour Division investigators, U.S. Department of Labor, 1973–2025
| Year | Investigators on staff at year’s end |
|---|---|
| 1973 | 812 |
| 1974 | 869 |
| 1975 | 921 |
| 1976 | 964 |
| 1977 | 980 |
| 1978 | 1,232 |
| 1979 | 1,087 |
| 1980 | 1,059 |
| 1981 | 953 |
| 1982 | 914 |
| 1983 | 928 |
| 1984 | 916 |
| 1985 | 950 |
| 1986 | 908 |
| 1987 | 951 |
| 1988 | 952 |
| 1989 | 970 |
| 1990 | 938 |
| 1991 | 865 |
| 1992 | 835 |
| 1993 | 804 |
| 1994 | 800 |
| 1995 | 809 |
| 1996 | 781 |
| 1997 | 942 |
| 1998 | 942 |
| 1999 | 938 |
| 2000 | 949 |
| 2001 | 945 |
| 2002 | 898 |
| 2003 | 850 |
| 2004 | 788 |
| 2005 | 773 |
| 2006 | 751 |
| 2007 | 732 |
| 2008 | 731 |
| 2009 | 894 |
| 2010 | 1,035 |
| 2011 | 1,024 |
| 2012 | 1,067 |
| 2013 | 1,040 |
| 2014 | 976 |
| 2015 | 995 |
| 2016 | 974 |
| 2017 | 912 |
| 2018 | 835 |
| 2019 | 780 |
| 2020 | 823 |
| 2021 | 782 |
| 2022 | 810 |
| 2023 | 733 |
| 2024 | |
| 2025 | 611 |
Note: Numbers represent Wage and Hour Division investigators on staff at the end of each fiscal year (the federal government's fiscal year runs from October 1 to September 30), except for 2022 and 2023. The 2022 number represents the investigators on staff at the end of November 2022. The 2023 number represents the investigators on staff as of December 7, 2023. The 2025 number represents investigators on staff as of May 2025. Data are not available for 2024.
Sources: Author’s analysis of Wage and Hour Division (WHD) data on number of investigators from unpublished Excel files provided by WHD staff members to the author. Source for 2020 and 2021 is Rebecca Rainey, “Wage-Hour Investigator Hiring Plans Signal DOL Enforcement Drive,” Bloomberg Law, January 28, 2022. Source for 2022 is Rebecca Rainey, "Wage Division Enforcement Declines Again in Wake of Hiring Woes," Bloomberg Law, Decemer 28, 2022. Source for 2023 is Jessica Looman, "Big Results for Workers in 2023," U.S. Department of Labor Blog, December 7, 2023. Source for 2025 is Jake Barnes, Janice Fine, Daniel J. Galvin, Jenn Round, Hana Shepherd, To Help U.S. Workers, We Need Labor Standards Enforcement, Not Mass Deportations, Data Brief, Workplace Justice Lab, Rutgers University, May 2025.
Having already cut at least half a billion dollars from DOL funding, requiring the agency to close regional offices that enforce wage and hour laws,49 the administration is seeking even further decreases in labor standards enforcement funding and staffing, with a reported goal of cutting 30% of funding from DOL.50 The administration has fired or forced out leaders and career staff at worker protection agencies and cut budgets almost indiscriminately, severely hampering the ability of federal agencies to protect wages and working conditions. And there is now a likely pause in labor and employment law enforcement because of the government shutdown.
If the administration is serious about creating new and improved rules for the H-1B program and enforcing them through new initiatives like Project Firewall, then they must allow DOL to play a lead role and have deconfliction measures in place with DHS to prioritize protecting worker interests over immigration enforcement goals. The administration must also call on Congress to drastically increase budgets for key worker protection agencies, reinstate staff at DOL that has been fired or laid off, and ramp up hiring of new staff. In addition, DHS should offer temporary immigration status protections coupled with work authorization to protect migrant workers who report their employers’ violations of labor and employment laws from retaliation, which will also encourage them to come forward and assist labor agencies in holding H-1B employers accountable when they break the law.
Conclusion
For years, migrant worker advocates, unions, academics, and both Democratic and Republican lawmakers have pointed out the need to change employer incentives by shifting away from the H-1B random lottery towards a true prioritization process in which visas are issued to employers seeking to hire and retain skilled workers by paying them fair wages that reflect market rates. We have urged the Administration and the Congress to explore alternatives to the lottery system that would directly prioritize wages and skills, and thus we support this regulatory effort to implement this change via a weighted lottery.
The H-1B visa program is the largest temporary work visa program in the United States and an important pathway into the U.S. labor market for skilled migrants from around the world—but it is a pathway that has serious deficiencies when it comes to the labor and employment rights of migrant workers and preserving U.S. labor standards. By issuing this proposed rule, DHS has taken an important first step towards fixing a system that has rewarded low-road employers with a business model that hinges on underpaying migrant workers. But as these comments suggest, even more should be done to improve the regulations that should act as safeguards to protect H-1B workers and similarly situated U.S. workers. H-1B workers should be paid fairly, have equal rights, and have an opportunity to become lawful permanent residents within a reasonable period of time. The administration should also take further action on other visas and in other areas to lift wage standards and improve working conditions for all workers, regardless of immigration status.
Regards,
Daniel Costa, Esq.
Director of Immigration Law and Policy Research
Economic Policy Institute
Ron Hira, Ph.D., P.E.
Associate Professor
Department of Political Science
Howard University
1. Daniel Costa and Ron Hira, H-1B visas and prevailing wage levels: A majority of H-1B employers—including major U.S. tech firms—use the program to pay migrant workers well below market wages, Economic Policy Institute, May 4, 2020.
2. See, Department of Labor, Employment and Training Administration, Strengthening Wage Protections for the Temporary and Permanent Employment of Certain Aliens in the United States, Interim Final Rule DOL Docket No. ETA-2020-0006, RIN: 1205-AC00, October 8, 2020; see also Daniel Costa, “EPI comments on DOL wage level methodology for H-1B visas and permanent labor certifications for green cards,” Economic Policy Institute (public comments), November 9, 2020.
3. U.S. Citizenship and Immigration Services, Department of Homeland Security, Strengthening the H-1B Nonimmigrant Visa Classification Program, CIS Docket No. 2658-20, DHS Docket No. USCIS-2020-0018, RIN: 1615-AC13, October 8, 2020.
4. U.S. Citizenship and Immigration Services, Department of Homeland Security, Modification of Registration Requirement for Petitioners Seeking To File Cap-Subject H-1B Petitions, CIS No. 2674-20; DHS Docket No. USCIS-2020-0019, RIN 1615-AC61, 85 Fed.Reg. 69236 (November 2, 2025)/
5. Michelle Hackman, “Federal Judge Strikes Down Trump’s H-1B Visa Rules on Highly Skilled Foreign Workers,” Wall Street Journal, December 1, 2020.
6. Presidential Candidate Joe Biden, immigration plan; previously available at https://joebiden.com/immigration/ (archived and on file with authors).
7. See, Sec. 104. H-1B Visa Allocation in S. 2928, H-1B and L-1 Visa Reform Act, 119th Congress (2025-2026) and H.R. 6993, H-1B and L-1 Visa Reform Act, 116th Congress (2019-2020).
8. Dave Jamieson, “Trump is hoping to deliver a parting gift to the agriculture lobby: an effective wage cut for farmworkers,” Huffington Post, November 9, 2020.
9. Employment and Training Administration, Adverse Effect Wage Rate Methodology for the Temporary Employment of H-2A Nonimmigrants in Non-Range Occupations in the United States, Interim Final Rule, U.S. Department of Labor, 90 Fed. Reg. 19365 (October 2, 2025). See discussion in Lauren Kaori Gurley, “Trump administration says immigration enforcement threatens higher food prices,” Washington Post, October 11, 2025; David Dayen, “Trump Labor Department Says His Immigration Raids Are Causing a Food Crisis,” The American Prospect, October 8, 2025.
10. Eric Fan, Zachary Mider, Denise Lu, and Marie Patino, “How Thousands of Middlemen are Gaming the H-1B Program,” The Big Take, Bloomberg, July 31, 2024.
11. U.S. Bureau of Labor Statistics, “15-0000 Computer and Mathematical Occupations (Major Group),”Occupational Employment and Wage Statistics, May 2024, U.S. Department of Labor.
12. Authors’ analysis of the LCA Programs disclosure data files for fiscal year 2024, Office of Foreign Labor Certification, Performance Data, Employment and Training Administration, U.S. Department of Labor.
13. See, for example, “Techsploitation,” Reveal News, The Center for Investigative Reporting, and Farah Stockman, “Teacher Trafficking: The Strange Saga of Filipino Workers, American Schools, and H-1B Visas,” Boston Globe, June 12, 2013.
14. Christopher Lapinig, “How U.S. Immigration Law Enables Modern Slavery,” The Atlantic, June 7, 2017.
15. See for example, Erika Werner, “High-skilled visa holders at risk of deportation amid tech layoffs,” Washington Post, February 24, 2023.
16. For a discussion about H-1B portability in practice, see Daniel Costa, “Is portability a panacea?
Changing employers in US temporary migration programmes,” chapter 8 in Christiane Kuptsch and Fabiola Mieres (eds.), Temporary labour migration: Towards social justice? edited volume, International Labour Organization, February 2025.
17. See for example, Daniel Costa and Ron Hira, H-1B visas and prevailing wage levels: A majority of H-1B employers—including major U.S. tech firms—use the program to pay migrant workers well below market wages, Economic Policy Institute, May 4, 2020; Daniel Costa and Ron Hira, “Tech and outsourcing companies continue to exploit the H-1B visa program at a time of mass layoffs: The top 30 H-1B employers hired 34,000 new H-1B workers in 2022 and laid off at least 85,000 workers in 2022 and early 2023,” Working Economics blog (Economic Policy Institute), April 11, 2023.
18. See for example, Senator Richard Durbin, “How American Jobs are Outsourced,” YouTube.com video, April 16, 2016.
19. See for example, Stef Kight, “U.S. companies are forcing workers to train their own foreign replacements,” Axios, December 29, 2019; Julia Preston, “Pink Slips at Disney. But First, Training Foreign Replacements,” New York Times, June 3, 2015; Julia Preston, “Toys ‘R’ Us Brings Temporary Foreign Workers to U.S. to Move Jobs Overseas,” New York Times, September 29, 2015; Michael Hiltzik, “How the University of California Exploited a Visa Loophole to Move Tech Jobs to India,” Los Angeles Times, January 6, 2017; Patrick Thibodeau, “Southern California Edison IT Workers ‘Beyond Furious’ over H-1B Replacements,” Computerworld, February 5, 2015.
20. See for example, Eric Fan, Zachary Mider, Denise Lu, and Marie Patino, “How Thousands of Middlemen are Gaming the H-1B Program,” The Big Take, Bloomberg, July 31, 2024; Julia Preston, “Large Companies Game H-1B Visa Program, Costing the U.S. Jobs,” New York Times, November 10, 2015.
21. David Weil, The Fissured Workplace: How Work Became So Bad for So Many and What Can Be Done to Improve It, Harvard, 2014.
22. A number of studies show a wage penalty for subcontracted/outsourced workers. For example, see Arindrajit Dube and Ethan Kaplan, “Does Outsourcing Reduce Wages in the Low-Wage Service Occupations? Evidence from Janitors and Guards,” Cornell University ILR Review. January 1, 2010); Deborah Goldschmidt and Johannes Schmieder, “The Rise of Domestic Outsourcing and the Evolution of the German Wage Structure,” The Quarterly Journal of Economics, Oxford University Press, vol. 132(3), 2017, pages 1165-1217; Andres Drenik, Simon Jäger, Pascuel Plotkin, and Benjamin Schoefer “Paying Outsourced Labor: Direct Evidence from Linked Temp Agency-Worker-Client Data,” Econometrics Laboratory, University of California, Berkeley, September 2020.
23. Wage and Hour Division, “H-1B Program,” U.S. Department of Labor, n.d. (last visited October 23, 2025).
24. The source for this table is USCIS Form I-129 petition data obtained by Eric Fan, a reporter with Bloomberg; see BloombergGraphics / 2024-h1b-immigration-data at GitHub, and the source for LCA data is LCA Programs disclosure data files for fiscal year 2024, Office of Foreign Labor Certification, Performance Data, Employment and Training Administration, U.S. Department of Labor.
25. See DOL prevailing wage guidance at Employment and Training Administration, “Prevailing Wage Determination Policy Guidance: Nonagricultural Immigration Programs,” Revised November 2009, U.S. Department of Labor.
26. Source for LCA data is LCA Programs disclosure data files for fiscal year 2023, Office of Foreign Labor Certification, Performance Data, Employment and Training Administration, U.S. Department of Labor.
27. Kevin Gray, “Average Starting Salary for Class of 2024 Shows Mild Gain,” National Association of Colleges and Employers, August 26, 2025.
28. Jake Barnes, Janice Fine, Daniel J. Galvin, Jenn Round, Hana Shepherd, To Help U.S. Workers, We Need Labor Standards Enforcement, Not Mass Deportations, Data Brief, Workplace Justice Lab, Rutgers University, May 2025.
29. Occupational Employment and Wage Statistics, 15-1211 Computer Systems Analysts and 15-1252 Software Developers, Industry: Cross-industry, Private, Federal, State, and Local Government, May 2024, U.S. Bureau of Labor Statistics, U.S. Department of Labor.
30. U.S. Department of Homeland Security, Weighted Selection Process for Registrants and Petitioners Seeking To File Cap-Subject H-1B Petitions, Notice of Proposed Rulemaking, 90 Fed. Reg. 45992-3 (September 24, 2025); see in particular this passage: “If the H-1B beneficiary would work in multiple locations, or in multiple positions if the registrant is an agent, the registrant would select the box for the lowest equivalent wage level among the corresponding wage levels for each of those locations or each of those positions and would list the location corresponding to that lowest equivalent wage level as the area of intended employment.”
31. S.358 – Immigration Act of 1990, U.S. Public Law 101-649.
32. U.S. Department of Homeland Security, Modification of Registration Requirement for Petitioners Seeking To File Cap-Subject H-1B Petitions, Notice of Proposed Rulemaking, 85 Fed. Reg. 69236 (November 2, 2020); Modification of Registration Requirement for Petitioners Seeking To File Cap-Subject H-1B Petitions, Final Rule, 86 Fed. Reg. 1676 (January 8, 2021).
33. Daniel Costa and Ron Hira, H-1B visas and prevailing wage levels: A majority of H-1B employers—including major U.S. tech firms—use the program to pay migrant workers well below market wages, Economic Policy Institute, May 4, 2020.
34. See Analysis of Bureau of Labor Statistics data by Hal Salzman and Khudodod Khudododov, “It Ain’t Pretty: Wage growth has been low or stagnant for many occupations. Coding skills offer little reprieve,” Figure published in Rachel Rosenthal, “Tech Companies Want You to Believe America Has a Skills Gap,” Bloomberg Opinion, August 4, 2020; Hal Salzman, Daniel Kuehn, and B. Lindsay Lowell, Guestworkers in the high-skill U.S. labor market: An analysis of supply, employment, and wage trends, Economic Policy Institute, April 24, 2013; Ron Hira, “Is There Really a STEM Workforce Shortage,” Issues in Science and Technology, Col XXXVIII, No. 4, Summer 2022.
35. 85 Fed. Reg. 69238, citing H.R. Rep. 101–723(I) (1990), as reprinted in 1990 U.S.C.C.A.N. 6710, 6721 (stating ‘‘The U.S. labor market is now faced with two problems that immigration policy can help to correct. The first is the need of American business for highly skilled, specially trained personnel to fill increasingly sophisticated jobs for which domestic personnel cannot be found and the need for other workers to meet specific labor shortages’’).
36. U.S. Department of Homeland Security, Weighted Selection Process for Registrants and Petitioners Seeking To File Cap-Subject H-1B Petitions, Notice of Proposed Rulemaking, 90 Fed. Reg. 46006 (September 24, 2025).
37. Jeremy Neufeld, The “Wage Level” Mirage: How DHS’s H-1B proposal could help outsourcers and hurt US-trained talent, Institute for Policy, September 24, 2025.
38. Jeremy Neufeld, Talent Recruitment Roulette: Replacing the H-1B Lottery: We could almost double the economic value of the H-1B program without changing the number of visas, Institute for Policy, January 17, 2025.
39. National Research Council, Building a Workforce for the Information Economy, The National Academies Press (Washington, DC).
40. For an examination of occupational shortages, see Ron Hira, “Is There Really a STEM Workforce Shortage,” Issues in Science and Technology, Col XXXVIII, No. 4, Summer 2022. For technology layoffs, see Layoffs.fyi, a website that tracks layoffs. For the state of the computer job market for recent graduates see, Natasha Singer, “Goodbye, $165,000 Tech Jobs. Student Coders Seek Work at Chipotle.” NY Times, August 10, 2025. For trends in computer science bachelor’s degrees, see National Center for Education Statistics, Table 322.10, Bachelor’s degrees conferred by postsecondary institutions, by field of study: Selected academic years, 1970-71 through 2021-2, 2023 Digest of Education Statistics, U.S. Department of Education. The vast majority, of computer science bachelor’s degrees, more than 90%, are earned by U.S. citizens and lawful permanent residents.
41. S.1233 – Keep STEM Talent Act of 2025, 119th Congress (2025-2026).
42. Natasha Singer, “Goodbye, $165,000 Tech Jobs. Student Coders Seek Work at Chipotle.” NY Times, August 10, 2025.
43. See for example; Eric Fan, Zachary Mider, Denise Lu, and Marie Patino, “How Thousands of Middlemen are Gaming the H-1B Program,” The Big Take, Bloomberg, July 31, 2024; Daniel Costa and Ron Hira, H-1B visas and prevailing wage levels: A majority of H-1B employers—including major U.S. tech firms—use the program to pay migrant workers well below market wages, Economic Policy Institute, May 4, 2020; Ron Hira, Congressional Testimony before the U.S. Senate Subcommittee on Immigration and the National Interest, hearing on “The Impact of High-Skilled Immigration on U.S. Workers,” Economic Policy Institute, February 25, 2016.
44. Eric Fan, Zachary Mider, Denise Lu, and Marie Patino, “How Thousands of Middlemen are Gaming the H-1B Program,” The Big Take, Bloomberg, July 31, 2024; Julia Preston, “Large Companies Game H-1B Visa Program, Costing the U.S. Jobs,” New York Times, November 10, 2015.
45. For more details about visa-ready H-1B abuse by outsourcing firms, see Ronil Hira’s testimony before the U.S. Senate Committee on the Budget, “Unlocking America’s Potential: How Immigration Fuels Economic Growth and Our Competitive Advantage,” September 13, 2023.
46. The White House, Restriction on Entry of Certain Nonimmigrant Workers, Presidential Proclamation, September 19, 2025.
47. Daniel Costa, Josh Bivens, Ben Zipperer, and Monique Morrissey, The U.S. benefits from immigration but policy reforms needed to maximize gains: Recommendations and a review of key issues to ensure fair wages and labor standards for all workers, Economic Policy Institute, October 4, 2024.
48. Jake Barnes, Janice Fine, Daniel J. Galvin, Jenn Round, Hana Shepherd, To Help U.S. Workers, We Need Labor Standards Enforcement, Not Mass Deportations, Data Brief, Workplace Justice Lab, Rutgers University, May 2025.
49. Rebecca Rainey, “DOGE’s $455 Million in Labor Savings Carry Costs for US Workers,” Bloomberg Law, April 2, 2025.
50. Adam Cancryn and Jennifer Scholtes, “Trump sends a scorched-earth budget plan. GOP lawmakers hate it already.” Politico, May 2, 2025.