Strengthening Wage Protections for the Temporary and Permanent Employment of Certain Aliens in the United States

ACTION:

Final rule.

SUMMARY:

In this final rule, the Department of Labor (the Department or DOL) adopts with changes an Interim Final Rule (IFR) that amended Employment and Training Administration (ETA) regulations governing the prevailing wages for employment opportunities that United States (U.S.) employers seek to fill with foreign workers on a permanent or temporary basis through certain employment-based immigrant visas or through H-1B, H-1B1, or E-3 nonimmigrant visas. Specifically, the IFR amended the Department's regulations governing permanent (PERM) labor certifications and Labor Condition Applications (LCAs) to incorporate changes to the computation of wage levels under the Department's four-tiered wage structure based on the Occupational Employment Statistics (OES) wage survey administered by the Bureau of Labor Statistics (BLS). The primary purpose of these changes is to update the computation of prevailing wage levels under the existing four-tier wage structure to better reflect the actual wages earned by U.S. workers similarly employed to foreign workers. This final rule will allow the Department to more effectively ensure the employment of immigrant and nonimmigrant workers admitted or otherwise provided status through the above-referenced programs does not adversely affect the wages and job opportunities of U.S. workers.

DATES:

This final rule is effective March 15, 2021.

FOR FURTHER INFORMATION CONTACT:

For further information, contact Brian D. Pasternak, Administrator, Office of Foreign Labor Certification, Employment and Training Administration, Department of Labor, 200 Constitution Avenue NW, Room N-5311, Washington, DC 20210, telephone: (202) 693-8200 (this is not a toll-free number). Individuals with hearing or speech impairments may access the telephone numbers above via TTY/TDD by calling the toll-free Federal Information Relay Service at 1 (877) 889-5627.

SUPPLEMENTARY INFORMATION:

I. Background

The Immigration and Nationality Act (INA or Act), as amended, assigns responsibilities to the Secretary of Labor (Secretary) relating to the entry and employment of certain categories of immigrants and nonimmigrants. This final rule concerns the calculation of the prevailing wage for job opportunities in the PERM, H-1B, H-1B1, and E-3 programs for which employers seek labor certification from the Secretary.

A. Permanent Labor Certifications

The INA prohibits the admission of certain employment-based immigrants unless the Secretary of Labor has determined and certified to the Secretary of State and the Attorney General that (1) there are not sufficient workers who are able, willing, qualified and available at the time of application for a visa and admission to the United States and at the place where the alien is to perform such skilled or unskilled labor, and (2) the employment of such alien will not adversely affect the wages and working conditions of workers in the United States similarly employed.

This “labor certification” requirement does not apply to all employment-based immigrants. The INA provides for five “preference” categories or immigrant visa classes, only two of which—the second and third preference employment categories (commonly called the EB-2 and EB-3 immigrant visa classifications)—require a labor certification. An employer seeking to sponsor a foreign worker for an immigrant visa under the EB-2 or EB-3 immigrant visa classifications generally must file a visa petition with the Department of Homeland Security (DHS) on the worker's behalf, which must include a labor certification from the Secretary of Labor. Further, the Department of State (DOS) may not issue a visa unless the Secretary of Labor has issued a labor certification in conformity with the relevant provisions of the INA. If the Secretary determines both that there are not sufficient able, willing, qualified, and available U.S. workers and that employment of the foreign worker will not adversely affect the wages and working conditions of similarly employed U.S. workers, the Secretary so certifies to DHS and DOS by issuing a permanent labor certification. If the Secretary cannot make one or both of the above findings, the application for permanent employment certification is denied.

Under the INA, the EB-2 classification applies to individuals who are “members of the professions holding advanced degrees or their equivalent or who because of their exceptional ability in the sciences, arts, or business, will substantially benefit prospectively the national economy, cultural or educational interests, or welfare of the United States.” United States Citizenship and Immigration Services (USCIS) regulations, in turn, define an “advanced degree” as any United States academic or professional degree or a foreign equivalent degree above that of baccalaureate. A United States baccalaureate degree or a foreign equivalent degree followed by at least five years of progressive experience in the specialty shall be considered the equivalent of a master's degree. If a doctoral degree customarily is required by the specialty, the alien must have a United States doctorate or a foreign equivalent degree. The regulation goes on to define “exceptional ability” as “a degree of expertise significantly above that ordinarily encountered in the sciences, arts, or business.”

The EB-3 program consists of three discrete classifications: “skilled workers,” defined as aliens who are “capable . . . of performing skilled labor (requiring at least two years training or experience), not of a temporary or seasonal nature, for which qualified workers are not available in the United States;” “professionals,” defined as aliens “who hold baccalaureate degrees and who are members of the professions;” and “other workers,” defined as aliens who are “capable . . . of performing unskilled labor, not of a temporary or seasonal nature, for which qualified workers are not available in the United States.”

B. Labor Condition Applications

The Secretary must certify an LCA filed by an U.S. employer before the employer may file a petition with DHS on behalf of a foreign worker for H-1B, H-1B1, or E-3 nonimmigrant classification. The LCA contains various attestations from the employer about the wages and working conditions that it will provide for the foreign worker. Most importantly, for the purposes of this final rule, the INA requires employers to pay H-1B workers the greater of “the actual wage level paid by the employer to all other individuals with similar experience and qualifications for the specific employment in question,” or the “the prevailing wage level for the occupational classification in the area of employment.”

The H-1B program allows U.S. employers to employ foreign workers temporarily in specialty occupations. “Specialty occupation” is defined as an occupation that requires the theoretical and practical application of a body of “highly specialized knowledge,” and a bachelor's or higher degree in the specific specialty, or its equivalent, as a minimum for entry into the occupation in the U.S. Similar to the H-1B visa classification, the H-1B1 and E-3 nonimmigrant visa classifications also allow U.S. employers to temporarily employ foreign workers in specialty occupations, except that these classifications specifically apply to the nationals of certain countries: The H-1B1 visa classification applies to foreign workers in specialty occupations from Chile and Singapore, and the E-3 visa classification applies to foreign workers in specialty occupations from Australia.

C. The Permanent Labor Certification Process

The Department's regulations at 20 CFR part 656 govern the labor certification process and set forth the responsibilities of employers who desire to employ, on a permanent basis, foreign nationals covered by the INA's labor certification requirement. The Department processes labor certification applications for employers seeking to sponsor foreign workers for permanent employment under the EB-2 and EB-3 immigrant visa preference categories. Aliens seeking admission or adjustment of status under the EB-2 or EB-3 preference categories are inadmissible “unless the Secretary of Labor has determined and certified . . . that—(I) there are not sufficient workers who are able, willing, qualified . . . and available at the time of application for a visa and admission to the United States and at the place where the alien is to perform such skilled or unskilled labor, and (II) the employment of such alien will not adversely affect the wages and working conditions of workers in the United States similarly employed.”

The Secretary makes this determination in the PERM programs by, among other things, requiring the foreign worker's sponsoring employer to recruit U.S. workers by offering a wage that equals or exceeds the prevailing wage and to assure that the employer will pay the foreign worker a wage equal to or exceeding the prevailing wage. Prior to filing a labor certification application, the employer must obtain a Prevailing Wage Determination (PWD) for its job opportunity from the Office of Foreign Labor Certification's (OFLC) National Prevailing Wage Center (NPWC). The standards and procedures governing the PWD process in connection with the permanent labor certification program are set forth in the Department's regulations at 20 CFR 656.40 and 656.41. If the job opportunity is covered by a collective bargaining agreement (CBA) that was negotiated at arms-length between a union and the employer, the wage rate set forth in the CBA agreement is considered the prevailing wage for labor certification purposes. In the absence of a prevailing wage rate derived from an applicable CBA, the employer may elect to use an applicable wage determination under the Davis-Bacon Act (DBA) or McNamara-O'Hara Service Contract Act (SCA), or provide a wage survey that complies with the Department's standards governing employer-provided wage data. In the absence of any of the above sources, the NPWC will use the BLS OES survey to determine the prevailing wage for the employer's job opportunity. After reviewing the employer's application, the NPWC will determine the prevailing wage and specify the validity period, which may be no less than 90 days and no more than one year from the determination date. Employers must either file the labor certification application or begin the recruitment process, required by the regulation, within the validity period of the PWD issued by the NPWC.

Once the U.S. employer has received a PWD, the process for obtaining a permanent labor certification generally begins with the U.S. employer filing an Application for Permanent Employment Certification, Form ETA-9089, with OFLC. As part of the standard application process, the employer must describe, among other things, the labor or services it needs performed; the wage it is offering to pay for such labor or services and the actual minimum requirements of the job opportunity; the geographic location(s) where the work is expected to be performed; and the efforts it made to recruit qualified and available U.S. workers. Additionally, the employer must attest to the conditions listed in its labor certification application, including that “[t]he offered wage equals or exceeds the prevailing wage determined pursuant to [20 CFR 656.40 and 656.41] and the wage the employer will pay to the alien to begin work will equal or exceed the prevailing wage that is applicable at the time the alien begins work or from the time the alien is admitted to take up the certified employment.”

Through the requisite test of the labor market, the employer also attests, at the time of filing the Form ETA-9089, that the job opportunity has been and is clearly open to any U.S. worker and that all U.S. workers who applied for the job opportunity were rejected for lawful, job-related reasons. OFLC performs a review of the Form ETA-9089 and may either grant or deny a permanent labor certification. Where OFLC grants a permanent labor certification, the employer must submit the certified Form ETA-9089 along with an Immigrant Petition for Alien Worker (Form I-140 petition) to DHS. A permanent labor certification is valid only for the job opportunity, employer, foreign worker, and area of intended employment named on the Form ETA-9089 and must be filed in support of a Form I-140 petition within 180 calendar days of the date on which OFLC granted the certification.

D. The Temporary Labor Condition Application Process

The Department's regulations at 20 CFR part 655, subpart H, govern the process for obtaining a certified LCA and set forth the responsibilities of employers who desire to temporarily employ foreign nationals in H-1B, H-1B1, and E-3 nonimmigrant classifications.

A prospective employer must attest on the LCA that (1) it is offering to and will pay the nonimmigrant, during the period of authorized employment, wages that are at least the actual wage level paid by the employer to all other employees with similar experience and qualifications for the specific employment in question, or the prevailing wage level for the occupational classification in the area of intended employment, whichever is greater (based on the best information available at the time of filing the attestation); (2) it will provide working conditions for the nonimmigrant worker that will not adversely affect working conditions for similarly employed U.S. workers; (3) there is no strike or lockout in the course of a labor dispute in the occupational classification at the worksite; and (4) it has provided notice of its filing of an LCA to its employee's bargaining representative for the occupational classification affected or, if there is no bargaining representative, it has provided notice to its employees in the affected occupational classification by posting the notice in a conspicuous location at the worksite or through other means such as electronic notification.

As relevant here, the prevailing wage must be determined as of the time of the filing of the LCA. In contrast to the permanent labor certification process, an employer is not required to obtain a PWD from the NPWC. However, like the permanent labor certification process, if there is an applicable CBA that was negotiated at arms-length between a union and the employer that contains a wage rate applicable to the occupation, the CBA must be used to determine the prevailing wage. In the absence of an applicable CBA, an employer may base the prevailing wage on one of several sources: A PWD from the NPWC; an independent authoritative source that satisfies the requirements in 20 CFR 655.731(b)(3)(iii)(B); or another legitimate source of wage data that satisfies the requirements in 20 CFR 655.731(b)(3)(iii)(C).

An employer may not file an LCA more than six months prior to the beginning date of the period of intended employment. 20 CFR 655.730. Unless the LCA is incomplete or obviously inaccurate, the Secretary must certify it within seven working days of its filing. Once an employer receives a certified LCA, it must file the Petition for Nonimmigrant Worker, Form I-129 (“Form I-129 Petition”) with DHS if seeking classification of the alien as an H-1B worker. Upon petition, DHS then determines, among other things, whether the employer's position qualifies as a specialty occupation and, if so, whether the nonimmigrant worker is qualified for the position.

II. Prevailing Wage Background

A. The Department's Prevailing Wage Determination Methodology

The Department has long relied on BLS OES data to establish prevailing wage levels. The OES is a comprehensive, statistically valid survey that, in many respects, is the best source of wage data available for satisfying the Department's purposes in setting wages in most immigrant and nonimmigrant programs. The OES wage survey is among the largest continuous statistical survey programs of the Federal Government. BLS produces the survey materials and selects the nonfarm establishments to be surveyed using the list of establishments maintained by State Workforce Agencies (SWAs) for unemployment insurance purposes. The OES collects data from over one million establishments. Salary levels based on geographic areas are available at the national and State levels and for certain territories in which statistical validity can be ascertained, including the District of Columbia, Guam, Puerto Rico, and the U.S. Virgin Islands. Salary information is also made available at the metropolitan and nonmetropolitan area levels within a State. Wages for the OES survey are straight-time, gross pay, exclusive of premium pay. Base rate, cost-of-living allowances, guaranteed pay, hazardous duty pay, incentive pay including commissions and production bonuses, tips, and on-call pay are included. These features are unique to the OES survey, which make it a valuable source for use in many of the Department's foreign labor programs.

The Department incorporated the wage component of the OES survey into its prevailing wage guidance in 1997. At the time, the Department divided OES wage data into two skill levels: A Level I wage for “beginning level employees” and a Level II wage for “fully competent employees.” Because the OES survey does not provide data about skill differentials within Standard Occupational Classification (SOC) codes, the Department established the entry and experienced skill levels mathematically. Specifically, under an Memorandum of Understanding (MOU), BLS computed a Level I wage calculated as the mean of the lowest paid one-third of workers in a given occupation (approximately the 17th percentile of the OES wage distribution) and a Level IV wage calculated as the mean wage of the highest paid upper two-thirds of workers (approximately the 67th percentile). This two-tier wage structure was based on the assumption that the mean wage of the lowest paid one-third of the workers surveyed in each occupation could provide a surrogate for the entry-level wage, but the Department did not previously conduct any meaningful economic analysis to test its validity, or otherwise explain how these levels were consistent with the INA's wage provisions.

In order to implement the INA's four-tier prevailing wage provision, the Department published comprehensive Prevailing Wage Determination Policy Guidance for Nonagricultural Immigration Programs (2005 Guidance), which expanded the two-tier OES wage level system to provide four “skill levels”: Level I “entry level,” Level II “qualified,” Level III “experienced,” and Level IV “fully competent.” The Department applied the formula in the INA to its two existing wage levels to set Levels I through IV, respectively, at approximately the 17th percentile, the 34th percentile, the 50th percentile, and the 67th percentile. In 2010, the Department centralized the prevailing wage determination process for nonagricultural labor certification programs within OFLC's NPWC. In preparation for this transition, the Department issued new Prevailing Wage Determination Policy Guidance for Nonagricultural Immigration Programs (2009 Guidance). This guidance currently governs OFLC's PWD process for the PERM, H-1B, H-1B1, and E-3 visa programs and will continue to govern OFLC's PWD process for these programs. No rulemaking to codify the old wage levels was ever undertaken, nor the public given an opportunity to comment on them.

When assigning a prevailing wage using OES data, the NPWC examines the nature of the job offer, the area of intended employment, and job duties for workers that are similarly employed. In particular, the NPWC uses the SOC taxonomy to classify the employer's job opportunity into an occupation by comparing the employer's job description, title, and requirements to occupational information provided in sources like the Department's Occupational Information Network (O*Net). Once the NPWC identifies the applicable SOC code, it determines the appropriate wage level for the job opportunity by comparing the employer's job description, title, and requirements to those normally required for the occupation, as reported in sources like O*Net. This determination involves a step-by-step process in which each job opportunity begins at Level I (entry level) and may progress to Level II (experienced), Level III (qualified), or Level IV (fully competent) based on the NPWC's comparison of the job opportunity to occupational requirements, including the education, training, experience, skills, knowledge, and tasks required in the occupation. After determining the prevailing wage level, the NPWC issues a PWD to the employer using the OES wage for that level in the occupation and area of intended employment.

B. The Interim Final Rule

On October 8, 2020, the Department published an Interim Final Rule (IFR) in the Federal Register, 85 FR 63872, revising the methodology the Department uses to determine prevailing wage levels for the H-1B, H-1B1, E-3, and PERM programs. As explained in the IFR, the Department concluded the existing wage levels were not consistent with the relevant statutory requirement that a government survey employed to determine the prevailing wage provide four wage levels commensurate with experience, education, and level of supervision. The Department also determined that the existing wage levels were artificially low and provided an opportunity for employers to hire and retain foreign workers at wages well below what their U.S. counterparts earn, creating an incentive to prefer foreign workers to U.S. workers, an incentive that is at odds with the statutory scheme and causes downward pressure on the wages of the domestic workforce. Therefore, the Department revised wage provisions at 20 CFR 655.731 and 656.40 to adjust the existing wage levels to ensure the wage levels reflect the wages paid to U.S. workers with similar experience, education, and responsibility to those possessed by similarly employed foreign workers.

In particular, the IFR amended paragraphs (a), (b)(2), and (b)(3) of 20 CFR 656.40, codifying the four-tier wage practice and revising the wage level computation methodology. A new § 656.40(b)(2)(i) specified the four new levels (Levels I through IV) to be applied. Paragraph (b)(2)(i)(A) explained the Level I wage would be calculated as the mean of the fifth decile of the wage distribution for the most specific occupation and geographic area available, rather than calculated as the mean of the bottom third of the OES wage distribution, as was the case prior to the IFR. Paragraph (b)(2)(i)(D) provided that the Level IV wage would be calculated as the mean of the upper decile of the wage distribution for the most specific occupation and geographic area available, rather than using the mean of the upper two-thirds of the distribution. As a result of these changes, the wage levels were increased, respectively, from approximately the 17th, 34th, 50th, and 67th percentiles to approximately the 45th, 62nd, 78th, and 95th percentiles. The IFR also made minor technical and clarifying amendments to sections 656.40 and 655.731, which the Department has adopted in this final rule with only a minor change to the location of one of the amended provisions, as explained further in section IV below.

The Administrative Procedure Act (APA), 5 U.S.C. 551 et seq., authorizes an agency to issue a rule without prior notice and opportunity to comment when the agency for good cause finds that those procedures are “impracticable, unnecessary, or contrary to the public interest.” The good cause exception for forgoing notice and comment rulemaking “excuses notice and comment in emergency situations, or where delay could result in serious harm.” The Department published the IFR with an immediate effective date, bypassing notice and comment due to exigent circumstances created by the coronavirus public health emergency that threatened immediate harm to the wages and job prospects of U.S. workers, as well as the need to avoid evasion by employers of the new wage rates. However, the Department requested public input on all aspects of the IFR during a post-promulgation 30-day public comment period and explained it would review and consider these comments before issuing a final rule. The public comment period ended on November 9, 2020, and resulted in receipt of more than two thousand comments. Most of the comments were not relevant and/or not substantive, but 148 relevant and substantive comments were received and are discussed further below.

C. Litigation

Four groups of plaintiffs separately challenged the Department's IFR. These groups of plaintiffs, which included academic institutions, businesses, and trade associations, claimed the Department lacked good cause to issue the IFR without undergoing notice and comment procedures under the APA and that the IFR was arbitrary and capricious and in violation of the INA. These plaintiffs further requested that the IFR be enjoined and the Department prevented from implementing it. In three of the four cases, the district court approved the parties' stipulation to convert plaintiffs' preliminary injunction motion to a motion for partial summary judgment on the notice and comment claim. In Chamber of Commerce, the district court issued a decision on December 1, 2020, granting plaintiffs' motion for partial summary judgment on their notice and comment claim and setting aside the Department's IFR. In Purdue University and Stellar IT (which were consolidated), the district court issued a decision on December 14, 2020, granting partial summary judgment to the plaintiffs on the basis that the Department lacked good cause to issue the IFR, and ordered the Department to re-issue prevailing wage determinations issued under the IFR on a mutually agreeable schedule. In the fourth case, ITServe Alliance, the district court issued a preliminary injunction on December 3, 2020, prohibiting the Department from enforcing the IFR against the plaintiffs in that case. In discussing plaintiffs' likelihood of success on the merits in that case, the court limited its analysis to plaintiffs' claim that the Department lacked good cause to forgo advance notice and comment. Following the district court's decisions in Chamber of Commerce and ITServe Alliance, OFLC took immediate action to comply with the courts' directives, including issuing a public announcement on its website on December 3, 2020, outlining the steps it was taking in response to the courts' orders.

Notwithstanding the district courts' orders to set aside the IFR on procedural grounds, the U.S. Supreme Court has acknowledged and affirmed the proposition that a procedurally flawed IFR does not taint a final rule relying upon an IFR as a proposed rule. The Department is satisfied that it meets the APA's objective requirements necessary for the promulgation of a final rule in this case. Specifically, the Department's IFR provided sufficient notice to the public by allowing for a 30 day comment period; “gave interested persons an opportunity to participate in the rule making through submission of written data, views or arguments”; the rule contained a “concise general statement of their basis and purpose”; and the rule will be published more than 30 days before it becomes effective. Accordingly, the Department maintains the legal authority to pursue this final rule based upon its compliance with the APA's procedural requirements satisfied in the IFR.

III. Discussion of Final Rule, Comments, and Responses

A. Overview

The IFR provided for the submission of public comments during a prescribed 30-day public comment period that closed on November 9, 2020. During this time, the Department received 2,340 comments. The Department received input from a broad range of commenters, including labor unions; employers; law firms; academic and research institutions; healthcare providers; public policy organizations; professional and trade associations; a federal agency; foreign workers, students, attorneys, and other individuals; and a significant number of anonymous commenters. Some commenters supported the new wage level computation methodology in the IFR generally or in concept as a necessary change to prevent abuse of the H-1B program, particularly its four-tier wage level system, by employers seeking to hire foreign workers at below market wages. However, the overwhelming majority of commenters opposed the new wage level computation methodology. Notably, however, commenters generally did not offer justifications or data to support the continued use of the old wage methodology.

Commenters opposed to the substantive changes in the IFR generally asserted that the revised wage levels do not correspond with wages paid to U.S. workers with similar qualifications or those employed in job opportunities with similar requirements, that the IFR wages do not reflect market wages as evidenced by comparisons to private wage surveys and wage data on various websites, and that the wage increases are arbitrary and unsustainable for most employers, especially given the immediate effective date of the IFR. Commenters expressed concern that the IFR would negatively impact the economy broadly by reducing labor demand, reducing American competitiveness in innovative industries, and encouraging outsourcing. A number of commenters asserted the IFR would disproportionately impact small businesses and start-ups; nonprofits; and academic, research, and healthcare institutions. Many commenters claimed that there is no need to raise wages to protect U.S. workers, asserting that foreign workers are not underpaid and employment of foreign workers creates, rather than reduces, employment opportunities for U.S. workers and benefits the economy broadly. Many commenters also expressed concern the IFR would harm currently employed foreign workers and their families, especially foreign workers with significant ties to the U.S. and for whom immigrant visa petitions have been filed but for whom visas are unavailable due to per country visa caps.

After careful and thorough consideration of the comments, the Department has adopted a number of modifications in this final rule to the wage methodology established by the IFR. In particular, the Department has adjusted the Level I wage and the Level IV wage downward to the 35th percentile and 90th percentile, respectively. The Department is also implementing in this rule a number of changes to how it uses data from BLS in the H-1B and PERM programs that will further reduce the incidence of inappropriately inflated wages identified by commenters. Finally, the Department is adopting a phase-in approach to how the new wage levels will be applied to give employers and workers time to adapt to the change. In combination, the Department believes these measures appropriately address commenters' concerns and will ensure that, going forward, the prevailing wage rates provided by the Department fully protect the wages and job opportunities of U.S. workers.

As the Department explained in the IFR, a primary purpose of the restrictions on immigration created by the INA, both numerical and otherwise, is “to preserve jobs for American workers.” Safeguards for American labor, and the Department's role in administering them, have been a foundational element of the statutory scheme since the INA was enacted in 1952. For the reasons set forth below, the Department has determined that the way it previously regulated the wages of certain immigrant and nonimmigrant workers in the H-1B, H-1B1, E-3, and PERM programs is inconsistent with the text of the INA. A substantial body of evidence examined by the Department, and discussed at length in the IFR, also suggests that the existing prevailing wage rates used by the Department in these foreign labor programs are causing adverse effects on the wages and job opportunities of U.S. workers and are therefore at odds with the purpose of the INA's labor safeguards. The current wage levels were also promulgated through guidance, without providing the public with any notice or an opportunity to comment, and without any meaningful economic justification. Accordingly, the Department is acting to adjust the wage levels to ensure they are codified and consistent with the factors the INA dictates must govern the calculation of foreign workers' wages. In so doing, the Department expects to reduce the dangers posed by the existing levels to U.S. workers' wages and job opportunities and thereby advance a primary purpose of the statute. While some commenters disagreed with the Department's conclusions about the effects of the old wage levels on U.S. workers, the Department continues to believe that the reasoning put forward in the IFR on this point is sound.

The modern H-1B program was created by the enactment of the Immigration Act of 1990 (IMMACT 90). Among other reforms, IMMACT 90 established “various labor protections for domestic workers” in the program. These protections were primarily designed “to prevent displacement of the American workforce” by foreign labor. In general, the purpose of the H-1B program is to “allow[ ] an employer to reach outside of the U.S. to fill a temporary position because of a special need, presumably one that cannot be easily fulfilled within the U.S.” Using a foreign worker as a substitute for a U.S. worker who is already working in or could work in a given job is therefore inconsistent with the broad aims of the program. Congress has recognized that repeatedly, both in enacting IMMACT 90 and in making subsequent changes to the H-1B program.

Wage requirements are central to the H-1B program's protections for U.S. workers. Under the INA, employers must pay H-1B workers the greater of “the actual wage level paid by the employer to all other individuals with similar experience and qualifications for the specific employment in question” or the “the prevailing wage level for the occupational classification in the area of employment.” By ensuring that H-1B workers are offered and paid wages that are no less than what U.S. workers similarly employed in the occupation are being paid, the wage requirements are meant to guard against both wage suppression and the replacement of U.S. workers by lower-cost foreign labor.

The OES prevailing wage levels that the Department uses in the H-1B program—as well as the related H-1B1 and E-3 “specialty occupation” programs for foreign workers from Chile, Singapore, and Australia—are the same as those it uses in its PERM program. Through the PERM program, the Department processes labor certification applications for employers seeking to sponsor foreign workers for permanent employment under the EB-2 and EB-3 immigrant visa preference categories. Aliens seeking admission or adjustment of status under the EB-2 or EB-3 preference categories are inadmissible “unless the Secretary of Labor has determined and certified . . . that—(I) there are not sufficient workers who are able, willing, qualified . . . and available at the time of application for a visa and admission to the United States and at the place where the alien is to perform such skilled or unskilled labor, and (II) the employment of such alien will not adversely affect the wages and working conditions of workers in the United States similarly employed.”

The Secretary makes this determination in the PERM program by, among other things, requiring the foreign worker's sponsoring employer to recruit U.S. workers by offering a wage that equals or exceeds the prevailing wage and to assure that the employer will pay the foreign worker a wage equal to or exceeding the prevailing wage. In this way, similar to its role in the H-1B program, the prevailing wage requirement in the PERM program furthers the statute's purpose of protecting the interests of, and preserving job opportunities for, American workers. Effectuating this purpose is the principle objective of the Department's regulatory scheme in the PERM program.

While the prevailing wage levels the Department sets in the H-1B, H-1B1, E-3, and PERM programs are meant to protect against the adverse effects the entry of immigrant and nonimmigrant workers can have on U.S. workers, they do not accomplish that goal—and have not for some time. For starters, the Department has never offered any explanation or economic justification for the way it currently calculates the prevailing wage levels it uses in these foreign labor programs. The INA requires that a government survey employed to determine the prevailing wage provide wage levels commensurate with experience, education, and level of supervision. However, it is clear that the Department's current wage levels are not sufficiently set in accordance with the relevant statutory factors. In setting the wage levels, the Department did not engage in an effort to tether them to the statutory factors, identify sources of wage data that would inform an analysis of how the levels should be calibrated so as to protect U.S. workers' wages and job opportunities, or otherwise articulate an analytical framework to guide and explain how the levels were established. It also set the levels outside the rulemaking process, instead promulgating them solely through a memorandum of understanding between departmental components.

Further, the Department's analysis of the likely effects of H-1B and PERM workers on U.S. workers' wages and job opportunities shows that the existing wage levels are not advancing the purposes of the INA's wage provisions. As explained below, under the existing wage levels, artificially low prevailing wages provide an opportunity for employers to hire and retain foreign workers at wages well below what their U.S. counterparts—meaning U.S. workers in the same labor market, performing similar jobs, and possessing similar levels of education, experience, and responsibility—make, creating an incentive—entirely at odds with the statutory scheme—to prefer foreign workers to U.S. workers, and causing downward pressure on the wages of the domestic workforce. The Department is therefore acting to adjust the existing wage levels to ensure the levels reflect the wages paid to U.S. workers with levels of experience, education, and responsibility comparable to those possessed by similarly employed foreign workers.

To accomplish this, the Department articulated an analytical framework in the IFR to govern how it adjusted the prevailing wage levels. In doing so, the Department considered, among other things, the statutory context in which the INA's prevailing wage provisions are found. In particular, because the prevailing wage levels are used primarily for high-skilled workers, most of whom are H-1B workers, the Department took into account the INA's definition of “specialty occupation,” which establishes the baseline minimum qualification requirements that foreign workers must possess to obtain an H-1B visa, and also looked to the qualification requirements for obtaining an EB-2 visa. From its review of these qualification requirements, the Department drew a number of conclusions about the least-skilled, or entry-level workers employed in the PERM and H-1B programs. Specifically, the Department determined that such workers often possess greater skills than many of the least qualified workers in the most common occupational classifications in which H-1B and PERM workers are found. For that reason, the Department concluded that the lower end of the wage distribution reported by the OES survey for those classifications should be discounted in setting an entry-level wage. Because wages for H-1B and PERM workers are, under the INA, to be based on the wages paid to U.S. workers with comparable education, experience, and responsibility, looking to the wage data of workers at the lowest points of the wage distributions for these occupations who likely would not be considered as working in a “specialty occupation” would therefore be inconsistent with the statute. Because the old wage methodology made such wage data a central element of the prevailing wage calculation, it did not, in the Department's judgment, comport with the INA.

The Department's review of the INA's qualification requirements for H-1B and EB-2 workers, in combination with an analysis of the demographic characteristics of workers in the H-1B program, led the Department to determine that, for purposes of identifying an entry-level wage, it should look to the wages paid to U.S. workers who possess a master's degree and limited work experience. Using such workers as wage comparators for entry-level H-1B and PERM workers, in the Department's judgment, is an appropriate way of determining what U.S. workers similarly employed and with comparable education and experience to such H-1B and PERM workers are paid. In analyzing wage data on such workers, the Department also determined that it was appropriate to focus its analysis on those occupations that account for one percent or more of all H-1B workers. As the Department acknowledged in the IFR, using a single wage structure across multiple programs, hundreds of different occupations, and for hundreds of thousands of different workers necessarily means that prevailing wage rates will not be perfectly tailored to every single job opportunity. While still giving due weight to other occupations in its analysis, the Department has determined that paying special attention to those occupations where foreign workers are most heavily concentrated, and where the risk to U.S. workers' wages and job opportunities from the employment of foreign labor is therefore most acute, is the optimal way of advancing the purpose of the INA's wage protections while accounting for the breadth of the programs and occupations covered by the four-tier structure. As discussed further below, while several commenters disagreed with various aspects of this analytical framework and the Department's interpretation of the INA, the Department, after considering those comments, continues to believe that its approach is appropriate.

Having determined how it would analyze the question of how to set prevailing wage levels, the Department proceeded to review data from various, credible government sources, specifically the surveys from the National Science Foundation (NSF) and the Current Population Survey (CPS), about the wages paid to master's degree holders with limited work experience employed in occupations that account for the vast majority of workers covered by the prevailing wage levels. Based on its analysis of this data, the Department concluded in the IFR that the range within the OES distribution where workers similarly employed and with levels of education and experience comparable to entry-level H-1B and PERM workers fall is between the 32nd and 49th percentiles of the distribution. The Department continues to believe that this conclusion is largely accurate, and that it is highly relevant to how it will set the entry-level wage in this final rule.

In the IFR, the Department relied on a number of qualitative considerations, including the relative strengths and weaknesses of the data it relied on to identify the entry-level wage range as well as the purpose of the INA's wage protections, to conclude that the entry-level wage should be placed higher up within the identified range at approximately the 45th percentile. Based on private wage data and other considerations provided by commenters, which are addressed below, the Department has reassessed this conclusion, and has now determined that the entry-level wage for the H-1B and PERM programs is more appropriate at the 35th percentile. In particular, data provided by commenters indicate that the lower end of the range may in fact provide a more accurate representation of what U.S. workers similarly employed to entry-level H-1B and PERM workers are paid. Concerns from commenters about how a potentially inflated entry-level wage would affect employers' ability to access the program, and how the IFR's reasoning was weighted too heavily to certain occupations and geographic areas, are also compelling reasons, in the Department's judgment, to favor a lower point in the range. Importantly, the Department believes that by staying within the range identified in the IFR, the entry-level wage it has selected will provide robust protection for U.S. workers.

The Department acknowledges commenters' reliance interests on the current wage methodology and understands that immediate changes to wage rates could cause some economic uncertainty for both employers and foreign workers. Thus, the Department is also adopting a series of transition provisions in this final rule to make it easier for employers and workers to adapt to the changed wage levels, thus avoiding disruption and striking a proper balance between stakeholders' reliance interests and the Department's obligation to comply with the INA and pursue a policy that is protective of U.S. workers. For many job opportunities, the new wage rates will phase in through two steps over a year and a half period. For job opportunities that will be filled by workers on track to become lawful permanent residents, and who therefore have greater reliance interests in the old wage methodology, the new wage rates will phase in through four steps over a three and a half year period. The Department also reduced the Level IV wage from approximately the 95th percentile to the 90th percentile, and made a number of other technical modifications to how it uses BLS data to produce prevailing wage rates. These changes, too, address commenters' concerns that wages under the IFR were inappropriately high.

B. Discussion

1. The Need for Rulemaking

Summary of Comments

The Department received a number of comments in support of the IFR, including one commenter that believed the IFR “makes important strides to bring wage requirements for the H-1B program closer to real prevailing wages in relevant industries.” These commenters agreed with the Department that the prior wage levels resulted in adverse effects on U.S. workers' wages and job opportunities. Some of these commenters noted that the Level I and II wages under the prior wage level methodology (approximately the 17th and 34th percentiles) were well below the median for the occupation and that 60 percent of H-1B positions were certified at one of these wage levels. One of these commenters expressed concern that the prior wage level methodology permitted H-1B employers to “engage in de facto wage arbitrage schemes.” A public policy organization noted that many employers “pay H-1B workers the lowest wages legally allowed, and outsource their H-1B employees to third-party firms.” The commenter asserted that employers opposed to the revised wage level methodology and increased wages claim “that employers will only hire H-1B workers if they are underpaid relative to similarly-situated U.S. workers,” which creates a wage “race to the bottom.” The commenter further stated that “other reliable sources of wage data” demonstrate that the wage results generated by the Department in the IFR are in fact too low. The commenter cited data from both the Department and NSF to draw the comparison and substantiate this claim, and it requested that the Department conduct a “systematic review” of major H-1B occupations to ensure that updates to the wage structure are in line with credible sources of salary data, such as the NSF's survey of recent college graduates. Another commenter believed the IFR would “prevent employers that seek specialized workers from being crowded out of the H-1B program by employers using the program to pay below market wages.” Some of these commenters believed the Level I wage should be set closer to the median for the occupation and one of the commenters stated that the Level I wage was the only wage level that mattered because the Department “has no adjudicative power over employer skill level claims.”

By contrast, the majority of comments received on the IFR expressed strong opposition to the rule and a number of commenters questioned whether adjustments to the prevailing wage level methodology are necessary. Many commenters believed there was no need to raise wages to protect U.S. workers, citing the Department's statement that many frequent H-1B program users pay wages above the required prevailing wage rates, as well as other external sources finding that foreign workers are paid as much or more than similarly employed U.S. workers and that foreign workers create jobs for U.S. workers or otherwise benefit U.S. workers and the economy broadly. Many commenters pointed to unemployment statistics and forecasted job growth in certain fields as evidence that the IFR changes are not necessary to protect U.S. workers. Three commenters stated that it is more expensive to hire foreign workers due to costs related to the visa process and that employers prefer to hire U.S. workers due to concern about the “instability of H-1B lottery systems.” Some commenters believed the regulatory requirement that H-1B employers must pay the highest of the actual or prevailing wage provides sufficient protection to U.S. workers because the employer must pay the actual wage in cases where the Department's PWD rate is lower. One commenter asserted the annual visa caps provide sufficient protection for U.S. workers and a second commenter asserted the recruitment requirements in the permanent labor certification regulations offer sufficient protection.

Several commenters claimed it was improper for the Department to cite higher actual wages paid by large H-1B employers as an indication that the prevailing wage levels were insufficient to protect U.S. workers. For example, an university commenter noted the Department's acknowledgment that many large “program users pay well in excess of the prevailing wage” and the commenter asserted this was an acknowledgment “that the issue it is trying to resolve . . . is non-existent.” This commenter stated that employers paying more than the prevailing wage might simply indicate these employers pay a higher actual wage “due to legitimate business factors.” Similarly, a public policy organization and a professional association stated that the fact that a group of H-1B employers pays more than the prevailing wage indicates only that some employers voluntarily increase wages for competitive reasons. Another commenter stated that pay differences are reflective of the “free market at work” and that “high profile tech companies . . . are in heavy competition . . . and have large enough profit margins” to pay higher wages. A group of associations stated that payment of higher wages by these employers may be due to geography and “intensity of the work” such that these employers must “pay a premium to attract both domestic talent and foreign-born talent . . .” By intensity of the work, the commenters referred to areas in which at least one percent of workers are employed in a particular occupation. The commenters stated that the OES “identifies for each SOC . . . [areas where] the number of employed individuals per each 1,000 employed persons in that particular occupation . . .” and that the Department should look to this as “a useful proxy for the intensity of activity in that particular occupation in a particular geography,” in addition to analyzing available LCA data to determine how often wages in excess of prevailing wages “are primarily for such high intensity jobs and locations.”

Many commenters asserted the Department failed to consider or “insufficiently weighted” a wide range of relevant and readily available studies and reports that indicate a revision to the wage level methodology is unnecessary. These commenters stated that the Department ignored ample evidence that H-1B workers are paid at least as much as their U.S. counterparts and that employment of H-1B workers may increase the wages earned by U.S. workers. A few commenters cited a GAO report finding H-1B workers earn the same or more than similar U.S. workers and an analysis by the website Glassdoor finding that across “10 cities and roughly 100 jobs” it examined, salaries for H-1B workers were “about 2.8 percent higher than comparable U.S. salaries . . . .” Similarly, several commenters cited a report published by the Partnership for a New American Economy, a research and advocacy organization dedicated to “mak[ing] the economic case for immigration,” finding that denials of H-1B petitions from 2007 to 2008 slowed job and wage growth for U.S. workers and that every one-percentage-point increase in the “foreign STEM share of a city's total employment . . . made possible by the H-1B visa program” increased wage growth by three to seven percentage points for U.S. workers. Other cited sources included:

• A Cato Institute report indicating roughly 80 percent of H-1B employers pay H-1B workers “above average market wages”;
• A working paper from the National Bureau of Economic Research finding that “complete elimination” of the H-1B program would have virtually no effect on the wages of “high-skilled Americans in year one and a slight reduction . . . by year three”;
• A National Foundation for American Policy (NFAP) report finding “on average, H-1B workers reduce overall unemployment and increase earnings growth within the fields they are employed by increasing firm productivity”;
• An NFAP report finding that each 1 percent increase in H-1B workers in science, engineering, technology, and mathematics (STEM) occupations “increased local wages of college educated Americans by 7-8 percent and non-college educated Americans by 3-4 percent”;
• A journal article concluding that “after controlling for human capital attributes, foreign I.T. professionals” earn more than their U.S. counterparts; and
• A National Survey of College Graduates comparative analysis finding that “controlling for socioeconomic and demographic characteristics, workers who hold a temporary work visa earn about thirty percent more than comparable” U.S. workers.

Commenters also cited a variety of studies and reports that conclude that the employment of foreign workers has little or no effect on employment rates for similarly employed U.S. workers. For example, a group comment cited a 2016 Journal of Economic Perspectives study on “Global Talent Flows” that the commenter said indicated “very little displacement of U.S.-born innovators and high-skilled professionals by high-skilled immigrants.” Another commenter stated “key fields such as software development and data science . . . are facing undeniable workforce supply shortages” and asserted this “undermin[ed] the argument that an influx in cheaper labor supply will result in lower possible earnings” for U.S. workers. In support, the commenter cited a Wall Street Journal article noting “tech job postings in the U.S. rose 32%” in the first half of 2019 and a 2018 BLS report projecting higher than average employment growth in high-tech services.

Some commenters also expressed concerns about the sources the Department did cite in the IFR in support of the need to revise wage levels. Citing an analysis of the IFR by labor economist and professor Dr. Madeline Zavodny, a trade association asserted the Department relied on “outdated, incorrect, or limited empirical data” and relied on sources that did not “include an analysis of the wages of H-1B workers in direct comparison with other workers having the same level of education, experience, or responsibility.” The commenter stated that the Associated Press analysis cited at footnote 122 provides “an incomplete picture” because it is not based on “actual workers in the U.S. who hold an H-1B visa” but instead is based on LCA data, which includes “applications that are denied (often because the wage is too low).” The commenter also stated that the analysis “does not control for any differences between applicants for an H-1B visa and U.S. workers, such as differences in age and education.” An anonymous commenter stated that the Associated Press article indicated that 58 percent of H-1B workers are paid more than their U.S. counterparts and asserted the article can only be used to support statements regarding wages paid to workers in computer occupations.

The trade association stated that the citations at footnote 121 in the IFR that the Department relied on to support its statement that H-1B IT workers earn roughly 25-33 percent less than U.S. workers failed to provide “a clear analysis of the wages of workers who hold an H-1B visa compared with other workers;” failed to include H-1B workers in the analysis; and failed to provide sufficient details of the wage analysis to determine the reason for the wage differentials. The anonymous commenter stated that the CRISIL Research citation in this footnote failed to cite evidence or provide data to support the statement that H-1B workers earn 25 percent less than U.S. workers and failed to provide a source for the claim that “local hires . . . cost 25-30% more.” The anonymous commenter stated that the third citation in this footnote is outdated, analyzing “only immigrant trends in the 1990s” and does not “specifically reference computer occupations.” The commenter also noted that the report recognizes that “the lower earnings of recent immigrants may reflect unobserved differences in the quality and type of education among immigrant cohorts” and the report “offers alternative factors that weigh into the wage trends of H-1B workers that [DOL] has not accounted for in this rule.”

An immigration law firm stated that the IFR misconstrued the CRISIL report, which the commenter asserted “actually shows that as a result of recent H-1B policy changes, it is harder to obtain H-1Bs for employees that are contracted to work at third-party worksites forcing U.S. employers to instead hire full-time employees to fill these roles” and “the increase in costs is attributed to the costs of full time employees” compared to the cost of “contract employees.” The commenter also asserted that the Department misconstrued Economic Policy Institute research when it claimed the research showed that only one of every two STEM graduates get a job in the field. The commenter stated that the researchers “found that half of students that do not enter the STEM industry found jobs in other industries.”

The anonymous commenter also asserted that the congressional testimony cited in this footnote provides no evidence to “establish the median wage as the appropriate compensation for any specific [H-1B] positions” and fails to consider that “that a Level 1 wage does not necessarily represent a position that requires less skill, but rather may have fewer experience requirements or supervisory duties.” The commenter also asserted that the journal article cited in this footnote is “outdated in its data” and “refers to computer occupations” so it “cannot be applied to any other occupational codes.”

Finally, a trade association noted that the Department cited findings by George Borjas regarding the impact of foreign workers on the wages of low-skill workers but failed to acknowledge Borjas's contribution to a 2016 National Academies of Sciences, Engineering, and Medicine (NASEM) literature review in which he stated “wage impacts from immigrants on U.S.-born college-educated workforce is minor (an increase for U.S. professionals of one-half of one percent in wage rates as a result of high-skilled immigration).” The commenter added that the NASEM review found that there is a “broad consensus with respect to high-skilled immigration that any impacts on U.S. wages by high-skilled, college-educated foreign-born professionals are close to negligible.”

Response to Comments

First, as the Department explained in the IFR, a primary and independently sufficient reason for reforming the manner in which it sets prevailing wage levels in the H-1B and PERM programs is that the old wage levels were never justified through an economic analysis, nor codified in rulemaking through notice and comment, and, on closer inspection, are in substantial tension with the statutory framework. Notably, commenters have also not provided data or analysis demonstrating that the wage rates under the old wage methodology produces wage rates commensurate with the wages paid to U.S. workers similarly employed and with comparable education, experience, and responsibility to H-1B and PERM workers, as required by statute. While some commenters urged the Department to preserve the old wage methodology, they provided no evidence for why that would be appropriate or consistent with the INA. Moreover, the Department notes that criticism of the way in which the wage levels are currently set is longstanding and exists across the political spectrum. Put simply, the old wage methodology is an outmoded method for calculating prevailing wage rates that is neither supported economic analysis, nor defended by commenters, and has never tied to the relevant statutory factors.

The Level I wage under the old methodology is set by calculating the mean of the bottom third of the OES wage distribution. That means the wages for many H-1B workers are set based on a calculation that takes into account wages paid to workers who, as explained in the IFR and below, almost certainly would not qualify to work in a “specialty occupation,” as defined by the INA. The Department has noted previously that “workers in occupations that require sophisticated skills and training receive higher wages based on those skills.” As a worker's education and skills increase, his wages are expected to increase as well. For that reason, it is likely that workers at the lowest end of an occupation's wage distribution generally have the lowest levels of education, experience, and responsibility in the occupation. In consequence, if the occupation by definition includes workers who do not have the level of specialized knowledge required of H-1B workers, as is the case with some of the most common occupations in which H-1B workers are employed, the very bottom of the wage distribution should be discounted in determining the appropriate point in the OES wage distribution at which to establish the entry-level wage under the four-tiered wage structure because workers at the bottom end are not similarly employed to H-1B workers. Yet the old wage structure made such workers a central component of that calculation. Similarly, the current Level IV wage is set by calculating the mean of the upper two-thirds of the wage distribution. That means that the wage level provided for the most experienced and highly educated H-1B workers is determined, in part, by taking into account a sizeable number of workers who do not even make more than the median wage of the occupation. Given the correlation between wages and skills, this calculation also would appear inconsistent with the statutory and regulatory framework. Common sense dictates that workers making less than the median wage of the occupation cannot be regarded as being similarly qualified to the most competent and experienced members of that occupation. That puts the old methodology in substantial tension with the governing statute and is in and of itself a sufficient reason for reassessing and revising the prior methodology in order to bring it more closely in line with the INA's wage provisions.

The Department also based its conclusion in the IFR that regulatory reform of H-1B and PERM prevailing wages was needed, in part, on a review of the academic literature on the subject, congressional testimony and media accounts of the practical consequences of the prior prevailing wage levels, and data on the actual wages that major users of the H-1B and PERM programs pay their foreign workers. As discussed at length in the preamble to the IFR, the Department considered numerous studies finding that H-1B workers are paid less than their U.S. counterparts. Other studies found this disparity to be especially true of H-1B employees working in computer science and information technology, fields in which two thirds of H-1B workers are employed. The Department's justification also took into account the fact that economic literature suggests that the introduction of low-cost foreign labor into a labor market suppresses wages in proportion to the number of foreign workers present in that labor market. Studies involving computer science workers confirm this general finding. Its review of this information led the Department to conclude that the old wage methodology resulted in adverse effects on both U.S. workers' wages as well as their job opportunities. After reviewing comments and the studies and information they provided, the Department continues to believe that, at least in some cases, the old prevailing wage methodology resulted in harm to U.S. workers and therefore should be revised.

The Department recognized, as did some commenters, the limitations of some of the wage studies it relied on in the IFR, noting that many of them compare H-1B and U.S. workers in the same occupation but do not directly compare workers in those occupations with the same levels of education, experience, and responsibility. However, in the IFR, the Department explained why these studies nonetheless allow for an instructive wage comparison: “[B]ecause H-1B workers are required to possess specialized knowledge and expertise that often exceeds the level of education and experience necessary to enter a given occupation generally, and greater skills are associated with higher earnings, the median H-1B workers should earn a wage that is at least the same, if not more, than the median wage paid to U.S. workers in the occupation. But a variety of studies show that the opposite is occurring.” Put another way, while the Department acknowledges that there is an inherent limitation in comparing median earnings of groups of workers, since doing so does not account for different levels of experience and education, the distortion in the data that results from such a limitation would be expected to show higher earnings for H-1B workers at the median given that a result of the INA's specialty occupation requirement for H-1B workers is that H-1B workers must possess more advanced education and experience than what is typically required to enter some of the most common occupations in which H-1B workers are employed. Yet the median earning of H-1B workers, according to these studies, are in fact skewed lower than the median U.S. worker in these occupations. Accordingly, the Department continues to believe this is a compelling data point demonstrating that H-1B workers in many cases make wages below those of similarly employed U.S. workers.

Further, the Department disagrees with commenters that other aspects of the methodology and reasoning relied on in the various studies that support the Department's position are flawed. These are, in many cases, studies from credible sources that are commonly cited in reporting and literature about the effects of the H-1B program on U.S. workers. Moreover, to the extent these studies focus on computer science and IT occupations, the Department believes that focus is appropriate. As explained at greater length below, the Department's analytic framework gives special attention to these occupations because they are where the largest concentration of H-1B and PERM workers are found, and therefore the places where the risks to U.S. workers that the Department is trying to guard against are most acute.

In addition, the Department considered testimony before the Senate Judiciary Committee as well as news reports about the displacement of U.S. workers by H-1B workers. As noted, some commenters criticized these sources as anecdotal and insufficient. But they were not the only sources on which the Department relied. The information from those sources supplemented the information the Department derived from studies and academic articles. Standing alone such information may (or may not) be insufficient to demonstrate systematic, adverse effects on U.S. workers, but, viewed in combination with other available evidence, it provides vital insight into the Department's understanding of the effects of the old wage methodology. The Department also views evidence about the real-world consequences of its wage methodology on U.S. workers, as shown in news reports, as important information that should not be ignored.

As detailed above, some commenters also claimed that the Department ignored or unfairly discounted studies showing that some H-1B workers earn more than U.S. workers. Far from ignoring or discounting such studies, the Department acknowledged their findings and addressed them in the IFR. While the Department did not discuss in the IFR every study of that kind that the commenters cite, it has reviewed the studies provided by commenters and notes that it did consider many sources with similar information, analysis, and conclusions to these studies. In addition, while some studies cited by commenters which were not directly addressed in the IFR offer additional analysis, they do not overwhelm the conclusions of other studies originally cited in the IFR. For example, reports that find that H-1B workers' wages exceed market wages often ignore that the prevailing wage level is fixed for the H-1B worker for three years, meaning that even if the H-1B worker is paid in excess of the market wage for an entry-level worker in year 1, this may not be the case in year 3 because the H-1B workers' wages should no longer be compared to entry-level workers. Other reports cited by critical commenters acknowledged that the research on employment of American workers in the presence of H-1B workers remains inconclusive or that the existing studies present mixed results on whether H-1B workers crowd out American workers. Some of these studies then focused on one segment of the American worker and H-1B market (e.g., recent college graduates) to obtain specific results which in many cases cannot be extrapolated to other workers cohorts. Others of these studies relied on data gathered only during recent economic recessions, which make it difficult to draw proper conclusions about the effect of H-1B workers on compensation and employment for competing workers under other (and more typical) economic conditions. The Department examined these studies concluding that some H-1B workers in some circumstances are better paid than U.S. workers, weighed them against other studies reaching the opposite conclusion, and, in its expert judgment, determined that there was reason to conclude that, at least in some instances, prevailing wage levels are set too low. An agency's choice of studies on which to rely is entitled to substantial deference. The Supreme Court has held that “[w]hen specialists express conflicting views, an agency must have discretion to rely on the reasonable opinions of its own qualified experts even if, as an original matter, a court might find contrary views more persuasive.” The studies cited by commenters rest on the same kinds of analyses and reach similar conclusions to those studies reviewed by the Department in development of the IFR. The Department has reviewed these studies and has concluded that they do not discredit, or even necessarily contradict, other sources of information that demonstrate that H-1B workers do, in some instances, adversely affect U.S. workers' wages and job opportunities, even if that is not true in all cases, as explained throughout. Accordingly, based on its review of these studies the Department continues to believe that some modification to the wage levels is necessary.

Contrary to the commenters' assertions, the Department considered studies showing that H-1B workers benefit U.S. workers. In the IFR, the Department acknowledged that in some instances the employment of H-1B workers fuels economic growth and job creation, as well as the fact that paying foreign workers at wages lower than U.S. workers may increase firms' profitability. Indeed, in the IFR the Department discussed studies that suggest the employment of H-1B workers has positive effects on the wages and job opportunities of U.S. workers and expressed a qualified agreement with them, specifically noting that “[w]hile the Department agrees that this is true in some instances, it is also clear that the current prevailing wage levels often result in adverse effects, and that adjustments to the wage levels are needed to ensure that the positive effects of the program will be enjoyed more widely.” In other words, the Department anticipates that bringing the wages of foreign workers in line with what similarly employed U.S. workers actually make will enhance the benefits resulting from the employment of such workers, which studies considered in the IFR as well offered by commenters show exist in some cases. The Department did not dispute in the IFR “that allowing firms to access skilled foreign workers can lead to overall increases in innovation and economic activity, which can, in turn, benefit U.S. workers,” but did conclude “H-1B workers' earnings data and other research indicate that, in many cases, the existing wage levels do not lead to these outcomes.” At no point in the IFR did the Department suggest that H-1B workers either always harm U.S. workers or always benefits U.S. workers and the firms that employ them. Rather, the Department concluded, and continues to conclude, that the positive benefits of the program, while real, are not as widespread as they might otherwise be, and that this is likely due to the fact that H-1B workers in some instances are paid wages below that paid to their U.S. counterparts.

One argument along these lines that the Department addressed in the IFR was made by the general counsel of a major user of the H-1B program in testimony before the Senate Judiciary Committee. In his testimony, he contended that H-1B workers raise the income of U.S. workers because they alleviate labor shortages, particularly in STEM and computer science. Importing workers to fill needs that would otherwise go unmet, he argued, allows companies to innovate and grow, creating more employment opportunities and higher-paying jobs for U.S. workers. The Department rejects the premise of the general counsel's argument that STEM jobs are going unfilled because there are no qualified American workers willing to take them, and therefore U.S. gross domestic product (GDP) would be smaller without importing foreign STEM workers. The Department notes that for every two students who graduate from a U.S. university with a STEM degree, only one obtains a STEM job. In the case of computer science occupations, another study cited by the Department challenges the notion that H-1B workers are filling needs unmet by U.S. workers. The study contains findings that foreign computer science workers have suppressed wages for U.S. computer science workers along with findings that “imply that for every 100 foreign [computer science] workers that enter the US, between 33 to 61 native [computer science] workers are crowded out from computer science to other college graduate occupations.” Further, while some commenters argued that the Department misconstrued the study showing that only half of U.S. STEM graduates go on to work in STEM fields on the grounds that many of these students find employment in other industries, the Department disagrees that the study is not relevant here. In fields where a graduate's degree signals certain skills to potential employers, such as computer science or many STEM fields, it is reasonable to assume that students who major in a particular field typically intend to find employment in that field. The fact that many of these particular students are able to find employment in other industries does not undercut the conclusion—indeed, it bolsters it—that at least some of their job opportunities in the fields for which they trained are limited by the presence of lower-paid foreign workers in some instances.

The Department also acknowledges commenters' point that in some circumstances H-1B workers contribute to innovation. Those contributions notwithstanding, “such outcomes are not the immediate objectives of the of the INA's wage protections.” Further, this rulemaking does not alter the number of H-1B workers permitted to work and it is unclear how the current wage levels promote greater innovation than the wages which will exist under this rule. The PERM program permits employers to hire aliens to work at permanent jobs where the Secretary of Labor has certified to the Secretary of State and the Secretary of Homeland Security that the employment of an alien seeking to enter the United States to perform skilled or unskilled labor “will not adversely affect the wages and working conditions of workers in the United States similarly employed.” In the case of H-1B workers, employers must file LCAs stating that the employer will offer wages that are, at a minimum, “the actual wage level paid by the employer to all other individuals with similar experience and qualifications for the specific employment in question,” or “the prevailing wage level for the occupational classification in the area of employment, whichever is greater.” In rulemaking, an agency is not required “to accord greater weight to aspects of a policy question than the agency's enabling statute itself assigns to those considerations.” In consequence, to the extent some comments and the studies cited therein criticized the Department's conclusion that the prevailing wage levels are set too low on the grounds that H-1B workers fuel innovation and economic growth, the Department affords them less weight. Such considerations are secondary to the Department's more immediate concern of fulfilling its statutory mandate to ensure that the presence of foreign workers does not adversely affect U.S. workers.

The Department also reemphasizes that while commenters preferred some studies and sources over others cited by the Department, they and their studies offered no affirmative argument in support of the old wage levels, nor did they explain how the prior wage levels reflect actual market wages. Rather, these commenters presented studies which the Department has already reviewed and which the Department does not believe align with the weight of the evidence which the Department continues to rely upon. The evidence amassed in the IFR provides a reasonable basis for increasing the wage rates, the Department stands by its determination that the old methodology did not adequately protect U.S. workers. The Department also notes that a number of commenters agreed with its conclusion that current wage levels often do not reflect prevailing wages and are set too low. For example, one commenter noted that, in some cases where H-1B workers are used to replace U.S. workers, “the H-1B workers have been hired with annual wages of around $30,000 to $40,000 less than the workers they have replaced.” These comments corroborate the Department's position that it weighted the conflicting evidence in a reasonable way and reached an appropriate conclusion that H-1B workers can and in many cases are used as low-cost alternatives to U.S. workers, and thereby undercut U.S. workers' wages and job opportunities.

The Department also acknowledges the comments it received (and studies cited therein) that argue that pointing to the higher actual wages that some employers pay H-1B and PERM workers to show that prevailing wage rates were too low is flawed reasoning because there may be other business factors beyond a worker's qualifications that explain why some employers pay a premium on the prevailing wage. The Department agrees that there may, in some instances, be legitimate business factors that explain why actual wages paid to H-1B workers would be higher than the prevailing wage rate. For example, a firm that faces a sudden increase in demand for its product relative to its competitors might be willing to pay premiums to both domestic and H-1B workers relative to its competitors. However, factors such as these are typically specific to a particular firm, employee, or geographic area, as some commenters acknowledged in their discussion of high-intensity occupation areas, and do not reflect the wages paid by the typical employer in a given labor market. In consequence, while the actual wages paid to H-1B workers might very well exceed the prevailing wage rate for legitimate reasons in some cases, such incidents should not be the norm across all employers, occupations, and locales. If the actual wage is consistently higher across the board than the prevailing wage rate, this suggests that the prevailing wage is not actually reflective of the market wage rate on offer in the labor market. As the data presented in the IFR shows, actual wages paid to H-1B workers not only exceed the prevailing wage rate, but do so consistently and substantially, on average, across many different employers. This suggests that legitimate business factors alone do not account for the extreme differences between the actual wages paid to H-1B workers and prevailing wage rates. Rather, it suggests that the prevailing wage rate is out of line with the market wage.

For similar reasons, the Department also rejects some commenters' contention (including as purportedly supported by the studies cited) that the fact that actual wages often exceeds the prevailing wage rate shows that there is no wage problem in the H-1B and PERM programs. One shortcoming such studies failed to acknowledge is that because the prevailing wage is in place for 3 years for H-1B workers, even if they are paid more than the prevailing wage in their first year, there is a distinct possibility that the prevailing wage will be low compared to the market for more experienced workers in the subsequent years. As the Department explained in the IFR, the INA takes a belt-and-suspenders approach to protecting U.S. workers' wages. Employers must pay the higher of the actual wage they pay to similarly employed workers or the prevailing wage rate set by the Department. Both rates generally should approximate the market wage for workers with similar qualifications and performing the same types of job duties in a given labor market as H-1B workers. It is therefore a reasonable assumption that, if both of the INA's wage safeguards were working properly, the wage rates they produce would, at least in many cases, be similar. Where the Department's otherwise applicable wage rate is significantly below the rates actually being paid by employers in a given labor market, it gives rise to an inference that the Department's current wage rates, based on statistical data and assumptions about the skill levels of U.S. workers, are not reflective of the types of wages that workers similarly employed to H-1B workers can and likely do command in the actual labor market. There is a mismatch between what the Department's prevailing wage structure says the relevant cohort of U.S. workers are or should be making and what employers are likely actually paying such workers, as demonstrated by the actual wage they are paying H-1B workers. Put another way, when many of the heaviest users of the H-1B program consistently pay wages well above the prevailing wage, it suggests that the prevailing wages are too low, and thus can be abused by other firms to replace U.S. workers with lower-wage foreign workers in cases where those firms do not have similarly employed workers on their jobsites whose actual wages would be used to set the wage for H-1B workers.

The Department also believes that looking to the pay practices of some of the most frequent users of the H-1B program is appropriate in determining whether the prevailing wage rates are set too low. Because the risk of harm to U.S. workers is most acute by employers in labor markets with heavy concentrations of H-1B workers, data on the actual wage rates at those employers and in those areas are entitled to special weight in the Department's analysis. Further, to the extent some commenters argue that looking at such firms unduly minimizes the Department's consideration of wage effects in rural areas or at smaller employers, the Department notes that, like its use of anecdotal evidence, the wage data it looked to from the heaviest users of the program is just one piece of various types of evidence on which it bases its conclusions about the effects of the old wage levels—no single piece of which is given dispositive weight. Rather, when considered in combination, this evidence provides a sound basis, in the Department's judgment, for concluding that the old wage methodology resulted in inappropriately low wages in a variety of circumstances.

The Department also disagrees that other safeguards in the INA are sufficient to protect U.S. workers and that updates to the prevailing wage levels are therefore unnecessary. Congress chose to enact multiple forms of protection for U.S. workers in these foreign labor programs. The Department must operationalize those protections entrusted to its administration as it sees best for the discharge of its legal responsibilities under the INA and its policy of more fully ensuring the protection of U.S. workers, including by updating the prevailing wage levels.

As explained in the IFR, the Department has determined that the conclusions it reached about adverse wage effects with respect to the H-1B program can also be extrapolated to the PERM program, about which the economic literature is far scanter. Critically, the PERM programs and the H-1B program are closely linked in both how they are regulated and used by employers. Unlike most nonimmigrant visas, H-1B visas are unusual in that they are “dual intent” visas, meaning under the INA, H-1B workers can enter the U.S. on a temporary status while also seeking to adjust status to that of lawful permanent residents. One of the most common pathways by which H-1B visa holders obtain lawful permanent resident status is through employment-based green cards, and in particular EB-2 and EB-3 visas. USCIS has estimated that over 80 percent of all H-1B visa holders who adjust to lawful permanent resident status do so through an employment-based green card. This is reflected in data on the PERM programs. In recent years, more than 80 percent of all individuals granted lawful permanent residence in the EB-2 and EB-3 classifications have been aliens adjusting status, meaning they were already present in the U.S. on some kind of nonimmigrant status. Given that the H-1B program is the largest temporary visa program in the U.S. and is one of the few that allows for dual intent, it is a reasonable assumption that the vast majority of the EB-2 and EB-3 adjustment-of-status cases are for H-1B workers. This is corroborated by the Department's own data, which shows that, in recent years, approximately 70 percent of all PERM labor certification applications filed with the Department have been for H-1B nonimmigrants.

Because of how many H-1B visa holders apply for EB-2 and EB-3 classifications, Congress has repeatedly amended the INA to account for the close connection between the programs. For example, while H-1B nonimmigrants are generally required to depart the U.S. after a maximum of six years of temporary employment, Congress has exempted from that requirement H-1B nonimmigrants who are beneficiaries of PERM labor certification applications with the Department, or who are beneficiaries of petitions for an employment-based immigrant visa with DHS that have been pending for longer than a year, if certain other requirements are met. Similarly, as noted above, Congress established the INA's prevailing wage requirements in section 212(p) with specific reference to the fact that they would apply in both the H-1B and PERM programs.

The various features of the statutory framework governing the programs, working in combination, have further tightened the relationship between them. In particular, because H-1B workers can have dual intent and, if they have a pending petition for an employment-based green card, can remain in the U.S. beyond the 6-year period of authorized stay limitation, many workers for whom an employer has filed a PERM labor certification application are already working for that same employer on an H-1B status. And because the method by which employment-based green cards are allocated can result in significant delays between when an alien is approved for a green card and when the green card is actually issued, the period during which a worker can, in some sense, have one foot in each program, is often protracted.

This system results in significant overlap in the principal uses of the H-1B and PERM programs. H-1B petitions approved in FY 2019, and the vast majority of individuals waiting for adjudication of EB-2- and EB-3-based adjustment of status applications, are concentrated in the same countries of origin. Relatedly, LCAs and applications for PERM labor certifications often are for job opportunities in the same occupations. Data from the Department's OFLC shows that of the ten most common occupations in which H-1B workers are employed, seven are also among the ten most common occupations in which PERM workers are employed. And PERM workers' wages are set based on the same methodology used for H-1B workers.

Given the evidence that these two programs are used similarly by employers, and employ in many instances the same or at least similarly situated foreign workers, the Department believes that it should treat the H-1B and PERM programs similarly. The upshot is that the H-1B and PERM programs are, in a variety of ways, inextricably conjoined. The rules governing the programs and how employers use them mean that, in many instances, workers in the PERM programs and workers in the H-1B program are often the exact same workers doing the same jobs in the same occupations for the same employers. And given the evidence of similarity, the Department can reasonably infer that the current wage levels under the four-tier structure—which result in inappropriately low wage rates in some instances for H-1B workers—also result in inappropriately low wage rates in some instances for the PERM programs. This is also borne out by the fact that, as noted in the IFR, the significant disparities between actual wages paid by heavy users of the programs and prevailing wage rates discussed above in connection with the H-1B program are also found in the PERM program.

2. Wage Level Methodology and Analytical Framework

Summary of Comments

Many commenters disagreed with the methodology and analytical framework the Department used to determine the appropriate prevailing wage rate, often asserting that the Department inappropriately relied on wage data from a limited pool of employers in the H-1B program and a pool of workers based on educational attainment limited to workers in information technology jobs, rather than basing the prevailing wage on market wages paid to workers in the applicable occupational classification based on the requirements of the occupation or employer's job opportunity. Several commenters also expressed concern about the chosen percentiles, asserting that an entry-level wage near the median for the occupation does not reflect real pay structures.

Some commenters asserted the Department inappropriately conflated the “actual” with the “prevailing” wage provisions in the INA and the Department's regulations at 20 CFR 655.731(a)(1) and (a)(2). A professional association stated it was improper to base prevailing wages on the “accomplishments, education or training of the employee” because that is the focus of the actual wage provision, whereas the prevailing wage is, “by regulation, based on the requirements for the position.” A university commenter noted that the prevailing wage is the wage paid to similarly employed workers, defined as “positions that have substantially comparable duties” in the occupation and area of employment and thus in prevailing wage determinations “the requirements of the position matters, not the skills that the individual worker brings to the table.” The commenter also asserted the IFR incorrectly states that the new methodology does not change the current wage determination process because the Department's 2009 PWD guidance indicates PWDs begin at entry level and “progress . . . only after considering the experience, education, and skill requirements of an employer's job description (opportunity).”

Related to these comments, many commenters believed it was improper for the Department to rely solely on wages paid to workers that possess a master's degree. A university commenter stated that the fact many H-1B workers possess a master's degree or higher is “attributed to the fact that USCIS favors beneficiaries with more advanced degrees.” Some commenters asserted that determining prevailing wage levels based only on wages paid to master's degree holders violates the INA because Congress did not include a master's degree requirement as a prerequisite for the employment-based visa programs. An association noted the statute defines “specialty occupation” as “an occupation requiring a bachelor's degree as the minimum qualification for entry.” Similarly, an immigration law firm believed that exclusion of wage data from workers possessing less than a master's degree is “baseless” because “attainment of a U.S. Bachelor's degree, or its equivalent, is sufficient for H-1B eligibility provided the petitioner can show a sufficient nexus between the degree earned and the offered position” and “the nexus of the degree specialty is a separate inquiry from prevailing wage requirements.” Noting that DHS regulations at 8 CFR 204.5(k)(2) “equate[ ] a master's degree to a bachelor's degree plus 5 years of progressively responsible work experience,” the commenter asked how the same Level I wage can represent both a position requiring a master's degree for entry and “entry level H-1B occupations that require a bachelor's degree in a specific specialty.”

Some commenters noted that a large number of occupations require at least a master's degree for entry and that it is improper for the Department to exclude the bottom third of wage data when determining the Level I prevailing wage in these occupations. For example, a university commenter stated that even if one accepts the prevailing wage was set too low for IT occupations “it is arbitrary to extrapolate from that very limited data set that the prevailing wage data set for other occupations is also lacking, especially for occupations where the normal educational requirement is an advanced degree.” Similarly, a professional association noted that at least 99 occupations require an advanced degree for entry according to DOL sources, including many that require a Ph.D., and that the bottom third of wages in these occupations “capture qualified and eligible H-1B individuals.” The commenter asserted the Department improperly excluded from consideration “one-third of the wages of individuals who are `similarly employed' ” and “essentially sets a minimum education level for entry as those with at least a master's degree in most professions.” One commenter from academia stated that many H-1B occupations that require a bachelor's degree are nonetheless specialized and thus the Department should consider all wage data for the occupation.

Several commenters also asserted that reliance on only wages paid to workers possessing a master's degree is particularly inappropriate for determining prevailing wages in the permanent labor certification context because many job opportunities in that program are in occupations that require no more than a bachelor's degree for entry. A group of associations asserted the Department ignored the fact that “about an equal number of individuals in H-1B status with advanced degrees and Bachelor's degrees are sponsored for green card status.” An immigration law firm stated the Department's reasoning focused centrally on wages paid to H-1B workers and asked the Department to explain how the “prior wage levels as applied in the PERM program negatively impact the wages of U.S. workers.” The commenter noted the PERM program differs from H-1B in relevant respects, including the labor market test requirement and the fact that employers file PERM petitions to fill “a future permanent position” that is “not necessarily the current position of the H-1B employee.” Noting the Department's acknowledgment that “not all SOC [occupations] qualify as a `specialty occupation,' ” this commenter asserted the IFR methodology “would arbitrarily raise salary requirements for occupations that are not used in the H-1B program but are used in the PERM program.” This commenter also noted that the Department acknowledged the new wage level methodology would create a “premium” on the wages of EB-3 workers and the commenter asserted the Department failed to cite authority to “require EB-3 petitioners to pay an additional fee, above what would be required to ensure the wages of U.S. workers are not negatively affected.”

Some commenters asserted that reliance on education alone when considering relevant wage data was inappropriate because many other factors can determine a worker's wage level. One commenter stated the Department provided no evidence that workers with a bachelor's degree “necessarily . . . make up a lower paid cohort of employees” and noted the Department's acknowledgment that “H-1B workers with master's degrees tend to be younger and less highly compensated than H-1B workers with bachelor's degrees.” The commenter noted that employers will accept equivalent credentials like experience and training and may base worker compensation on factors like “experience, special skills, history with the company or industry . . . [and] highly specialized knowledge.” Another commenter noted that someone with a bachelor's degree and 10 years of experience might be paid more for the same job opportunity than someone with a master's degree and 2 years of experience, whereas a bachelor's degree holder with 2 years of experience may be paid less. The prevailing wage in this case would be based on the requirements for the position, whereas the actual wage would be the wage paid to the worker employed in the position and may depend on the worker's education and experience.

A number of commenters asserted it was improper for the Department to rely only on wage data from workers in a limited set of information technology occupations as the relevant benchmark for determining the appropriate wage level. An anonymous commenter asserted that the Department's reasoning focused solely on “computer occupations” and the prevailing wage methodology based on that reasoning “can therefore only be applied to computer occupations.” A university commenter noted that many common occupations in the H-1B and PERM programs fall outside of this occupation set, including many occupations in the education sector, such as post-secondary teachers, several of which may require a Ph.D. for entry. The commenter added that even if one assumes wages are too low in the IT sector, “it is arbitrary to extrapolate from that very limited data set that the prevailing wage” is too low in other sectors.

Based on these concerns, some commenters urged the Department to reconsider its decision in the IFR to use a uniform wage structure across all occupations and programs. For example, a university commenter suggested the Department should apply the pre-IFR wage level methodology to occupations that normally require an advanced degree for entry, according to O*Net, rather than discounting the first one-third of occupational wage data for these occupations. One commenter suggested the Department should apply the revised wage level methodology to large IT employers and H-1B dependent employers, while applying the “PWD data from 07/01/2020-10/06/2020” to occupations in “medicine and health [070-079] and education [090-099].” Similarly, some commenters urged the Department to exempt specific positions in the medical field from revised wage methodology or exempt all ACWIA-eligible employers.

Many commenters also took issue with the reasoning behind setting the Level I wage for entry-level workers at approximately the 45th percentile. A public policy organization stated that placing entry level workers close to the median wage in the occupation “departs from the English language definition of median” and stated that, by definition, “[e]ntry level workers cannot be both at the bottom quarter of the wage scale and at almost the median of the wage scale.” A trade association stated that no employer sets compensation above the occupational median wage for all entry-level workers “completing graduate or professional degrees with little professional experience.” The commenter asserted the Department provided no evidence indicating a near-median wage is “the most reasonable and closest proxy” for the market wage paid to entry-level workers. A human resources professional association stated that it is “particularly important to reflect the lower and higher range” of an occupational wage distribution when using the SOC system because the SOC occupations are “hopelessly broad” and the commenter stated that SOC 11-9033 encompasses 126 distinct jobs in higher education.

Response to Comments

As noted, some commenters asserted that the Department misinterpreted the INA in the IFR, specifically disagreeing with the notion that the prevailing wage rate and the actual wage provided for by the INA should approximate one another, and similarly contending that the Department should not consider the accomplishments, education, or training of the employee as those are considerations associated with the actual wage requirement; rather, the Department should focus on the requirements for the position. This argument, however, misreads the statute, and also fails to understand a fundamental premise of the IFR. The Department is not ignoring its regulations or guidance on how prevailing wages rates are assigned; rather, the Department in this rulemaking is doing something different. It is making an assessment of how the four wage levels required by 8 U.S.C. 1182(p)(4) are to be established.

To begin with, as the IFR discussed in detail, the INA requires employers to pay H-1B workers the greater “of the actual wage level paid by the employer to all other individuals with similar experience and qualifications for the specific employment in question,” or the “prevailing wage level for the occupational classification in the area of employment.” The statute further provides that, when a government survey is used to establish the wage levels, “such survey shall provide at least 4 levels of wages commensurate with experience, education, and the level of supervision.” If an existing government survey produces only two levels, the statute provides a formula to calculate two intermediate levels. Thus, like the statute's actual wage clause, the prevailing wage requirement, when calculated based on a government survey, makes the qualifications possessed by workers, namely education, experience, and responsibility, an important part of the wage calculation.

Put slightly different, both clauses yield wage calculations that in similar fashions are designed to approximate the rate at which workers in the U.S. are being compensated, taking into account the area in which they work, the types of work they perform, and the qualifications they possess. The statute requires employers to pay the rate of whichever calculation yields the higher wage. In this way, the statutory scheme is meant to “protect U.S. workers' wages and eliminate any economic incentive or advantage in hiring temporary foreign workers.” If employers are required to pay H-1B workers approximately the same wage paid to U.S. workers doing the same type of work in the same geographic area and with similar levels of education, experience, and responsibility as the H-1B workers, employers will have significantly diminished incentives to prefer H-1B workers over U.S. workers, and U.S. workers' wages will not be suppressed by the presence of foreign workers in the relevant labor market.

The Department therefore disagrees with commenters' contention that the INA's actual wage clause and prevailing wage clause are not to be understood and operationalized in similar fashions. Moreover, the Department notes that, while commenters are correct that Department guidance and regulations discuss the “prevailing wage” as something that is assigned based on the requirements of a job opportunity, rather than the qualifications of the specific worker who will fill the position, the manner in which the “prevailing wage” for a specific job is assigned is different from the manner in which the Department establishes the four “prevailing wage levels” required by § 1182(p)(4). For one thing, a prevailing wage for a specific job opportunity is often assigned before the identity and actual qualifications of the worker who will fill the position are known. As a practical matter, it is therefore unavoidable that this would be done by reference to job requirements as opposed to the qualifications of an unknown worker. By contrast, the Department sets the four wage levels that are used to calculate specific prevailing wage rates by reviewing statistical data. The review of statistical data necessarily occurs at a more general level given that the four wage levels apply to broad swaths of workers and occupations and therefore relies on information from surveys, which often collect information about the skills possessed by particular workers rather than the job requirements of specific jobs. It is thus reasonable for the Department to consider the qualifications possessed by actual workers in operationalizing section 1182(p)(4).

In addition, the Department notes that it is a reasonable inference that, in many cases, the skills possessed by an actual worker will likely align with the qualification requirements of the job opportunity such worker fills. Looking to the skills possessed by actual workers thus should serve as a reasonable proxy in many cases for the requirements of the job opportunities in which they work. Moreover, to the extent the qualifications possessed by workers are different from the requirements of the jobs they fill, the Department believes that taking workers' actual skills and qualifications into account furthers the purpose of the statute. As explained throughout, the INA's wage provisions are designed to protect U.S. workers. In the labor market, workers compete with other workers based on the skills and qualifications those workers bring to the job—not based on what qualifications an employer lists in a job opening. Giving some weight to the actual characteristics of entry-level workers in the foreign labor programs thus takes into account important factors that determine how workers compete against one another over wages and job opportunities. Ignoring workers' actual qualifications in setting the wage levels would thus potentially weaken protections for U.S. workers insofar as it would mean the Department was leaving out of its analysis an important factor that influences employment outcomes.

Further, because, as noted, the actual wage clause and the prevailing wage clause of the INA are designed to achieve similar outcomes, serving as a form of belt-and-suspenders protection for U.S. workers, and given that the actual wage clause does take into account the specific qualifications possessed by actual workers, the Department believes it is reasonable to similarly take into account the actual qualifications of the workers when assessing survey data to set prevailing wage levels.

Finally, the Department also notes that, to the extent commenters suggest that the method by which the Department is setting the four wage levels pursuant to section 1182(p)(4) contradicts the previous method by which the Department set the wage levels, they are also mistaken. As noted, the Department has never previously set the wage levels through regulation. or has it ever explained its analysis or provided an economic justification for why the wage levels are set as they are. Rather, the old wage levels were set through a memorandum of understanding between DOL components, which offered no explanation for why the specific levels used were selected or how they comported with the statute. This rulemaking is therefore the first time the Department has undertaken to justify, and tether to the relevant statutory factors the manner in which the wage levels are established. There is no prior analytical framework to contradict because none was ever used. Again, the distinction between assigning a prevailing wage rate and setting prevailing wage levels pursuant to section 1182(p)(4) is key. While the Department has longstanding regulations on the former, this rulemaking is its first attempt to do the latter in a meaningful way.

Based, in part, on similar reasoning related to the actual demographics of workers in the H-1B program, the Department also concluded in the IFR, and continues to believe, that using master's degree holders with limited work experience as a proxy for entry-level workers in analyzing survey data to determine the entry-level wage for its H-1B and PERM programs is appropriate. In particular, in the IFR the Department examined the demographic characteristics of H-1B workers and concluded that many entry-level workers in the program are master's degree holders with limited work experience. In particular, a review of data from USCIS about the characteristics of individuals granted H-1B visas in fiscal years 2017, 2018, and 2019 indicates that H-1B workers with master's degrees tend to be younger and less highly compensated than H-1B workers with bachelor's degrees. On average, individuals with master's degrees in the program are approximately 30 years old, whereas bachelor's degree holders are, on average, 32 years old. This suggests that, while possessing a more advanced degree, master's degree holders in the program are likely to have less relevant work experience than their bachelor's degree counterparts. Relatedly, H-1B master's degree holders make, based on a simple average, $86,927, whereas bachelor's degree holders make on average $88,565. Given that differences in skills and experience often explain differences in wages, this gap in average earnings and age suggests that, while possessing a more advanced degree, master's degree holders in the H-1B program tend to be less skilled and experienced—and are therefore more likely to enter the program as entry-level workers—than are bachelor's degree holders.

This conclusion is further bolstered by the fact that master's degree holders have, in recent years, been the largest educational cohort within the program. In FY2019, for instance, 54 percent of the beneficiaries of approved H-1B petitions had a master's degree—whereas only 36 percent of beneficiaries had only a bachelor's degree. These facts, in combination with the age and earnings profiles of master's degree holders in the program, strongly suggest that a significant number of entry-level H-1B workers are individuals with a master's degree and very limited work experience. Because, as explained above, the Department has determined that the qualifications of actual workers are highly relevant to establishing prevailing wage levels pursuant to section 1182(p)(4), this analysis of the demographic characteristics of H-1B workers adds critical weight to the Department's conclusion to use master's degree holders as an analytical proxy for entry-level workers.

To further address commenters' concerns that master's degree holders with limited work experience are an inappropriate proxy for entry-level H-1B workers, the Department notes that, contrary to some commenters' contentions, this approach is consistent with the baseline qualification requirements in the INA for the H-1B program, as well as for EB-2 visas. For one thing, the statutory criteria for who can qualify as an EB-2 worker provide a clear, analytically useable definition of the minimum qualifications workers within that classification must possess. Even the least experienced individuals within the EB-2 classification are likely to have at least a master's degree or its equivalent. Possession of an advanced degree is thus a meaningful baseline with which to describe entry-level workers in the EB-2 classification.

As noted in the IFR, the baseline qualifications needed to obtain entry as an H-1B worker are different. An individual with a bachelor's degree in a specific specialty, or its equivalent, may qualify for an H-1B visa; a master's degree is not a prerequisite. However, the bachelor's degree or equivalent must be in a specific specialty. A generalized bachelor's degree is insufficient to satisfy the requirement that H-1B workers possess highly specialized knowledge. Further, the statute requires that the individual be working in a job that requires the application of “highly specialized knowledge.” Again, this means, contrary to some commenters' assertions, that for the H-1B program the possession of any kind of bachelor's degree is not the baseline qualification criterion for admission. Something more is needed. The ultimate inquiry rests also on whether the individual can and will be performing work requiring highly specialized knowledge.

As with aliens in the EB-2 classification, looking to the earnings of individuals with a master's degree provides an appropriate and analytically useable proxy for purposes of analyzing the wages of typical, entry-level workers within the H-1B program. For one thing, master's degree programs are, generally speaking, more specialized courses of study than bachelor's degree programs. Thus, while the fact that an individual possesses a bachelor's degree does not necessarily suggest one way or another whether the individual possesses the kind of specialized knowledge required of H-1B workers, the possession of a master's degree is significantly more likely to indicate some form of specialization. Although a master's degree alone does not automatically mean an individual will qualify for an H-1B visa, possession of a master's degree—something that is surveyed for in a variety of wage surveys—is thus a better proxy for specialized knowledge than is possession of a bachelor's degree for purposes of the Department's analysis. While possession of a bachelor's degree is also commonly surveyed for, mere possession of a bachelor's degree is not nearly as reliable an indicator that the degree holder possesses specialized knowledge.

Importantly, the Department is not claiming that all entry-level workers in the H-1B program possess a master's degree, or that possession of a bachelor's degree in a specific specialty such as would demonstrate specialized knowledge is in all cases the equivalent of having a master's degree. To reiterate, the Department is using master's degree holders with limited work experience as a proxy for entry-level workers purely for analytical purposes. As more fully explained below, because the OES survey does not capture data on workers' education and experience—the factors that the INA requires the Department to take into account in establishing wage levels—the Department sought in the IFR to identify where within the OES wage distribution the entry-level wage should fall by consulting other survey sources that do gather information on education and experience. Doing so necessarily requires the Department to identify an appropriate wage comparator or group of comparators for entry-level H-1B and PERM workers within those survey sources to ensure that the wage level for entry-level workers set based on that data reflects what workers with similar qualifications to entry-level H-1B and PERM workers are paid. For the reasons given above the Department, in its discretion, has determined that using master's degree holders as an analytical proxy for entry-level workers in these high-skilled programs is a reasonable method of assessing wage data for purposes of establishing the entry-level wage.

As noted, commenters also criticized the conclusion the Department reached about where to place the entry-level wage in the IFR based on its analysis of wage data about master's degree holders, arguing that placing the entry-level wage at approximately the 45th percentile is axiomatically in error given that entry-level workers do not, by definition, start out making more than almost half of all workers in an occupation. Although for the reasons given below the Department has decided to adjust the entry-level wage downward to the 35th percentile, the Department disagrees with commenters that setting the entry-level wage closer to the median of the OES distribution is inappropriate. As explained in the IFR, the interplay between the statutory framework governing the prevailing wage and the OES survey data demonstrate that, for the top H-1B and PERM occupations, workers at the lower end of the OES distribution in the most common H-1B occupations likely would not qualify as working in a “specialty occupation,” as that term is defined in the INA, and thus do not have education and experience comparable to even the least qualified H-1B worker—a contention generally not disputed by commenters—meaning their wage data must be discounted in setting wages for entry-level H-1B workers. In consequence, while a wage close to the median does not represent what all entry-level workers in a given occupation generally make, it is entirely reasonable that the wage for the vast run of entry-level workers covered by the four-tier wage structure, many of whom are required to possess more specialized skills, would fall closer to the median.

As explained above, the Department interprets the INA's wage provisions to require it to take into account the education, experience, and responsibility of workers in setting wage levels for the H-1B program. It is therefore necessary to identify what types of U.S. workers in a given occupation have comparable levels of education, experience, and responsibility to H-1B workers. The Department did so by looking to wage data about master's degree holders with limited work experience in occupations in which H-1B workers are commonly employed. While the INA makes clear that the prevailing wage levels must be set commensurate with education, experience, and level of supervision, it leaves assessment of those factors to the Department's discretion. How the Department exercises that discretion is informed by the legislative context in which the four-tier wage structure was enacted, which indicates that the wage levels are primarily designed for use in the Department's high-skilled and PERM foreign labor programs. Other provisions in the INA relating to the education and experience requirements of those programs—and in particular the statutory definition of “specialty occupation”—therefore serve as critical guides for how wage levels based on experience, education, and level of supervision should be formulated.

Under the INA, H-1B visas can, in most cases, only be granted to aliens entering the U.S. to perform services “in a specialty occupation.” The statute defines “specialty occupation” as an occupation that requires theoretical and practical application of a body of “highly specialized knowledge” and the “attainment of a bachelor's or higher degree in the specific specialty (or its equivalent) as a minimum for entry into the occupation in the United States.” An alien may be classified as an H-1B specialty occupation worker if the alien possesses “full state licensure to practice in the occupation, if such licensure is required to practice in the occupation,” “completion of [a bachelor's or higher degree in the specific specialty (or its equivalent)],” or “(i) experience in the specialty equivalent to the completion of such degree, and (ii) recognition of expertise in the specialty through progressively responsible positions relating to the specialty.” DHS regulations further clarify the requirements for establishing that the position is a specialty occupation and that the beneficiary of an H-1B petition must be qualified for a specialty occupation. The Department's regulations restate the statute's definition of specialty occupation essentially verbatim.

A few features of the definition bear emphasizing. First, the statute sets the attainment of a bachelor's degree in a specific specialty, or experience that would give an individual expertise equivalent to that associated with a bachelor's degree in the specific specialty, as the baseline, minimum requirement for an alien to qualify for the classification. Of even greater importance, having any bachelor's degree as a job requirement is not sufficient to qualify a job as a specialty occupation position—the bachelor's degree or equivalent experience required to perform the job must be “in the specific specialty.” In other words, the bachelor's degree required, or equivalent experience, must be specialized to the particular needs of the job, and impart a level of expertise greater than that associated with a general bachelor's degree, meaning a bachelor's degree not in some way tailored to a given field. These aspects of the definition play an important role in how the Department uses data from the BLS OES survey to set appropriate prevailing wage levels.

The OES survey categorizes workers into occupational groups defined by the SOC system, a federal statistical standard used by federal agencies to classify workers into occupational categories for the purpose of collecting, calculating, or disseminating data. An informative source on the duties and educational requirements of a wide variety of occupations, including those in the SOC system, is the Department's Occupational Outlook Handbook (OOH), which, among other things, details for various occupations the baseline qualifications needed to work in each occupation. A review of the OOH shows that only a portion of the workers covered by many of the occupational classifications used in the OES survey likely have levels of education and experience similar to those of H-1B workers in the same occupation. Some share of workers in these classifications likely do not have the education or experience qualifications necessary to be considered similarly employed to specialty occupation workers. Because the INA requires the prevailing wage levels for H-1B workers to be set based on the wages of U.S. workers with levels of experience and education similar to those of H-1B workers, the Department must take this into account when using OES data to determine prevailing wages.

For example, a common occupational classification in which H-1B nonimmigrants work is Computer Programmers. In some cases, the work of a computer programmer may involve writing basic computer code and testing it. The OOH's entry for Computer Programmers describes the educational requirements for the occupation as follows: “Most computer programmers have a bachelor's degree; however, some employers hire workers with an associate's degree.” In other words, while common, a bachelor's degree-level education, or its equivalent, is not a prerequisite for working in the occupation. USCIS and at least one court have reasoned from this that the mere fact that an individual is working as a Computer Programmer does not establish that the individual is working in a “specialty occupation.” Because a person without a specialized bachelor's degree can still be classified as a Computer Programmer, some portion of Computer Programmers captured by the OES survey are not similarly employed to H-1B workers because the baseline qualifications to enter the occupation do not match the statutory requirements.

The same is true for other occupational classifications in which H-1B workers are often employed. For example, the Medical and Health Services Manager occupation, as described by the OOH, does not in all cases require a bachelor's degree as a minimum requirement for entry. USCIS has therefore concluded that the fact that an individual works in that occupational classification does not necessarily mean that the individual is working in a “specialty occupation.” USCIS and its predecessor agency, the Immigration and Naturalization Service, have long emphasized that the term “specialty occupation” does not “include those occupations which [do] not require a bachelor's degree in the specific specialty.” In other words, if not all jobs in an occupational classification require a specialized bachelor's degree or equivalent experience, under the INA other evidence is needed to show that a worker will be performing duties in a specialty occupation beyond whether the job opportunity falls within a particular SOC classification.

A review of the OOH entries for the occupations in which H-1B nonimmigrants most commonly work demonstrates that most H-1B workers fall within SOC classifications that include some number of workers who would not qualify for employment in a specialty occupation. For instance, the OOH entries for Software Developers—an occupation accounting for over 40 percent of all certified LCAs —provides that such workers “usually have a bachelor's degree in computer science and strong computer programming skills.” For Computer Systems Analysts, which make up approximately 8.8 percent of all certified LCAs, “a bachelor's degree in a computer or information science field is common, although not always a requirement. Some firms hire analysts with business or liberal arts degrees who have skills in information technology or computer programming.” Similarly, the O*Net database, which surveys employers on the types of qualifications they seek in workers for various occupations, shows that, on average, over 13 percent of all jobs in the occupations that H-1B workers are most likely to work in do not require workers to have even a bachelor's degree. Moreover, the O*Net does not differentiate between jobs that require bachelor's degrees in specific specialties and job for which a general bachelor's degree will suffice. It is therefore a reasonable inference that the percentage of jobs in these occupations that would not qualify as specialty occupation positions for purposes of the INA is almost certainly even higher.

Simply put, the universe of workers surveyed by the OES for some of the most common occupational classifications in which H-1B workers are employed is larger than the pool of workers who can be said to have levels of education and experience comparable to those of even the least skilled H-1B workers performing work in a specialty occupation. Because the statutory scheme requires the Department to set the prevailing wage levels based on what workers similarly employed to foreign workers make, taking into account workers' qualifications and, as noted, the large majority of foreign workers are H-1B workers, it would be inappropriate to consider the wages of the least educated and experienced workers in these occupational classifications in setting the prevailing wage levels. To conclude otherwise would place the Department at odds with one of the purposes of the INA's wage protections: to ensure that foreign workers earn wages comparable to the wages of their U.S. counterparts.

As a result, it is entirely reasonable that the entry-level wage for H-1B workers would fall closer to the median of the OES distribution. The OES survey is not specifically designed to serve the Department's foreign labor programs. It does not survey for education and experience—the factors the INA requires the Department to consider in setting prevailing wage levels—which is why the Department looks to other survey sources, like the NSF and CPS, to make assessments about where within the OES distribution workers with particular education and experience levels are likely to fall. So too, as demonstrated by the above analysis of the OOH, its occupational classifications are not delineated so as to exclude workers who could not be regarded as working in a specialty occupation, meaning only a portion of the OES distribution for many occupations is actually relevant to how the Department sets wages for the H-1B program. As a result, the median of the OES distribution is not necessarily the median of the distribution of workers who have qualifications comparable to H-1B workers. The median of that distribution will likely in many cases fall above the median of the overall OES distribution since lower skilled, and therefore less highly compensated workers will be excluded.

On this last point, commenters also argued that the IFR's analysis improperly focused on only certain occupations, and that, for other occupations, most particularly those requiring an advanced degree, the above reasoning about how SOC classifications should be assessed in light of the statutory framework is inapposite. Relatedly, a number of commenters faulted the Department for focusing much of its analysis on the H-1B program, claiming the Department did not take adequate account of the array of occupations for which labor certification is sought in the PERM program. Despite these comments, for the reasons discussed above, the Department continues to believe that focusing its analysis on those programs and occupations that account for the largest share of workers covered by the four-tier wage structure is appropriate and consistent with the approach the Department has taken in setting wages in other foreign labor programs. Doing so is, in the Department's judgment, the most appropriate way to ensure U.S. workers are protected to the greatest extent possible in light of the fact that the Department's wage structure applies to a large and varied class of workers and occupations. Further, the Department acknowledges that PERM workers and advanced degree occupations are entitled to some weight in the Department's decision over how to set wage levels. As discussed at greater length below, taking into account these aspects of the issue addressed by this rule played an important part in the Department's decision to reduce the entry-level wage from the 45th percentile to the 35th percentile.

To explain its focus on H-1B workers, the Department notes that the H-1B program accounts, by order of magnitude, for the largest share of foreign workers covered by the Department's four-tier wage structure. Upwards of 80 percent of all workers admitted or otherwise authorized to work under the programs covered by the wage structure are H-1B workers. This, in combination with the fact that, as explained in an earlier section, the risk of adverse effects to U.S. workers posed by the presence of foreign workers is most acute where there are high concentrations of such workers, supports the Department's determination to pay special attention to the H-1B program in how it sets wages. Because the wage structure governs wages for hundreds of thousands of workers across five different foreign labor programs and hundreds of different occupations, no wage methodology will be perfectly tailored to the unique circumstances of every job opportunity. Advancing the INA's purpose of guarding against displacement and adverse wage effects against this statutory backdrop therefore means, in the Department's judgment, that particular weight should be given in the Department's analysis to those aspects of the problem this rule addresses where there is the greatest danger to U.S. workers' wages—hence the added focus on the H-1B program.

Relatedly, the Department notes that the H-1B program is linked closely to the PERM programs that are also covered by the Department's wage structure. For one thing, there is significant overlap in the types of occupations in which H-1B and PERM workers are employed. For example, the top ten most common H-1B occupations include seven of the ten most common PERM occupations. Through the third quarter of FY 2020, 80 percent of PERM cases were for jobs in Job Zones 4 and 5 —the most highly skilled job categories, which also account for 94 percent of all H-1B cases. Moreover, it is also clear that H-1B status often serves as a pathway to employment-based green card status for many foreign workers and that a very substantial majority of workers covered by PERM labor certification applications are already working in the U.S. as H-1B nonimmigrants. In FY 2019, 68.2 percent of all PERM applications were for aliens that at the time the applications were filed were already working in the U.S. on H-1B visas. For these reasons, giving particular attention to the H-1B program in determining how to adjust the wage levels is entirely consistent with also ensuring that how the wage levels are applied in the PERM programs is properly accounted for in the Department's analysis.

Similarly, the Department has concluded, in its discretion, that the Level I wage should be established based on the wages paid to workers in those occupations that make up a substantial majority of the applications filed in the H-1B, H-1B1, E-3, and PERM programs. This also ensures that the Department appropriately takes into account the size and breadth of the programs covered by the four-tier wage structure by giving special attention to those areas where the risk to U.S. workers' wages and job opportunities is most severe by virtue of having high concentrations of H-1B and PERM workers. Commenters are incorrect that the Department's decision to take this focus means it only looked at computer occupations. Rather, the Department looked at all occupations that account for one percent or more of the total H-1B population. While many of these occupations are computer-related, some are not. Further, while commenters are correct that there are as many as 99 occupations that require advanced degrees, the Department notes that this is out of a total of over 550 occupations covered by the OOH. Further, those 99 occupations account for an even smaller share of the actual workers who are employed under the Department's four-tier wage structure. While this does not mean that due attention should not be given to how the wage levels affect workers in advanced degree occupations, it does guide the relative weight these occupations are given in the Department's analysis.

Despite their disagreement with the methodology employed by the Department, commenters generally did not offer alternative ways to balance using a single wage structure across all five programs and hundreds of different occupations with varying skill requirements against the need to protect U.S. workers as fully as possible. Some commenters suggested as an alternative that different occupations or groups of occupations should be subject to a separate analysis and different wage structure. The Department has considered this option and believes that the utility of preserving a uniform wage structure across all programs and occupations outweighs any benefits that might be achieved by promulgating multiple, occupation or program-specific wage structures. The Department continues to believe that its method of doing so is the best available option as it is consistent with the approach the Department has taken in other foreign labor programs and focuses the Department's analysis on those areas where the risk to U.S. workers is greatest.

As for treating the PERM programs differently than the H-1B program, the Department notes that its analysis of highly skilled workers with advanced degrees and/or specialized knowledge—namely the EB-2 immigrant classification and the H-1B, E-3, and H-1B1 nonimmigrant programs—already takes into full account a large portion of the PERM program. With respect to the EB-3 classification, it is also noteworthy that many H-1B workers adjust status to that of lawful permanent residents through EB-3 classification, and the manner in which the programs operate means that, in many cases, foreign workers can, in some sense, have one foot in each program simultaneously for extended periods of time. Using different wage methodologies in the programs would therefore result in the incongruous possibility of a worker doing the same job for the same employer suddenly receiving a different wage upon adjusting status. Similarly, while having somewhat different eligibility criteria, the EB-2 and EB-3 classifications are not mutually exclusive: many workers that satisfy the eligibility criteria for one would also do so for the other. Applying the same wage methodology in both classifications is therefore important to ensure consistent treatment of similarly situated workers and prevent the creation of incentives for employers to prefer one classification over the other because different wage methodologies yield different wages. Thus, it is key in the Department's judgment that the EB-3 classification be treated the same as the EB-2 classification and H-1B program. More generally, continuing to employ the same wage structure across both the H-1B and PERM programs advances the Department's interest in administrative consistency and efficiency. Because there is significant overlap between the H-1B and PERM programs, they have long been regulated in connection with one another. Moreover, to the extent commenters assert that the IFR's wage levels resulted in inappropriately high wages for certain workers in advanced degree occupations, the Department notes that its decision to reduce the entry-level wage should, to some degree, ameliorate this concern.

For several reasons, the Department has also determined that occupation-specific wage structures are undesirable. For starters, calculating multiple different wage structures based on occupation would be a substantial and costly administrative undertaking for multiple components within the Department. There are over 800 different occupations in the SOC classification system used in the OES survey. The analysis needed to tailor different wage structures to each occupation would be an enormous undertaking, even assuming it were possible to conduct a meaningful, occupation-by-occupation analysis. Further, the burden on BLS to produce hundreds of different wage levels every year across various occupations would simply be unsustainable.

In addition, treating different occupations differently would create an opportunity and incentive, in some cases, for employers to misclassify workers in order to take advantage of lower wage rates. This is something that the Department already encounters by virtue of having different wage methodologies for different nonimmigrant programs that cover different types of jobs. Introducing the possibility of securing a different wage methodology within the H-1B and PERM programs would similarly allow employers ability to seek lower wages even if such wages are not the right wage for the job opportunity in question and result in adverse effects on U.S. workers. Again, this also means that, barring a compelling reason to introduce this kind of disuniformity into the H-1B and PERM programs, a single wage structure should be preserved. And the Department does not believe that there is such a compelling reason to disaggregate the wage methodology by occupation. While certain advanced degree occupations present somewhat different considerations in terms of how wage rates should be provided as compared to the top H-1B and PERM occupations the Department focused on in its analysis in the IFR, the Department reiterates that, as explained more fully below, the effects of the new wage methodology on advanced degree occupations have been given significant weight in the Department's analysis of where to set the entry-level wage. The Department therefore believes that adjusting the IFR's entry-level wage down to the 35th percentile—together with other features of the system, discussed below—adequately accounts for the interests of workers and employers in advanced degree occupations and will more consistently supply wage rates that are appropriate across a broader range of occupations. Moreover, other changes made in this final rule, including eliminating the use of the default wage of $208,000 per year for all four wage levels in cases where BLS cannot supply a Level IV wage (an issue that was of particular concern for commenters that discussed how the IFR affected employers of workers in advanced degree occupations), will also reduce the incidence of job opportunities requiring an advanced degree being assigned inflated wage rates.

Moreover, the Department notes that the use of a single wage structure has been its practice ever since it began using leveled wages in the H-1B and PERM programs. Twenty years of experience shows that using a single wage structure across all occupations is not unmanageable for employers. Indeed, given that the previous wage levels were selected with no analysis or explanation, the Department anticipates that its revised levels will in fact produce more appropriate outcomes in a larger number of cases across different occupations. For the first time the Department has undertaken a meaningful analysis of what wage levels will yield prevailing wage rates in the largest number of cases possible that are consistent with the wages paid to U.S. workers similarly employed and with comparable levels of education, experience, and responsibility to H-1B and PERM workers. In consequence, preserving a single wage structure should, if anything, be even more feasible and reasonable now than it was when the old wage levels were operative.

Further, the INA allows an employer to use the best available information at the time of filing an LCA in setting the wages in the H-1B program. If an employer does not believe the OES wage provided by the Department is the best available information at the time of filing, the employer may utilize an alternative prevailing wage survey provided by an independent authoritative source or another legitimate source of wage information. Such alternative sources of wage information are, in the Department's experience, widely available, and provide a backstop for employers, thereby reducing any need to create multiple, precisely tailored wage structures for different occupations.

The Department defines a prevailing wage survey published by independent authoritative source as “a prevailing wage survey for the occupation in the area of intended employment published . . . in a book, newspaper, periodical, loose-leaf service, newsletter, or other similar medium, within the 24-month period immediately preceding the filing of the employer's application.” The independent authoritative source should: (1) Reflect the average wage paid to workers similarly employed in the area of intended employment; (2) Reflect the median wage of workers similarly employed in the area of intended employment if the survey provides such a median and does not provide a weighted average wage of workers similarly employed in the area of intended employment; (3) be based upon recently collected data; and (4) represent the latest published prevailing wage finding by the authoritative source for the occupation in the area of intended employment.

In utilizing an independent authoritative source, the Department requires employers to follow the Department guidance, which explains the standards contained in the Department's regulations. Employers following the 2009 Guidance should ensure wage data collected is for similarly employed workers, meaning having substantially similar levels of skills. The survey should contain a representative sample of wages within the occupation that comports with recognized statistical standards and principals in producing prevailing wages. It is important to note that the nature of the employer, such as whether the employer is public or private, for profit or nonprofit, large or small, charitable, a religious institution, a job contractor, or a struggling or prosperous firm, do not bear in a significant way on the skills and knowledge levels required and should not limit the universe of employers surveyed. The relevant factors are the job, the geographic locality of the job, and the level of skill required to perform independently on the job. The Department provides a set of minimum survey standards in Appendix E of the 2009 Guidance, and encourages employers to reference these standards when seeking to use an independent authoritative source as the prevailing wage. Written documentation on the methodology used to conduct the survey and the validity of the methodology used in computing the occupational wage data covering the area of intended employment must be kept in the employer's data file and made available in the event of an investigation.

In addition, the Department allows employers to rely upon other legitimate sources of wage information if they do not have access to a published independent authoritative source. The only difference between a published independent authoritative source and another legitimate source of wage information is that the other legitimate source of wage information simply has to be “reasonable and consistent with recognized standards and principals in producing a prevailing wage” and does not need to be published. As with independent authoritative sources, the Department encourages employers to ensure the other legitimate source of wage information follows the Department's 2009 Guidance to ensure it is reasonable and consistent with recognized standards and principals in producing a prevailing wage.

The Department notes that since the IFR, employers have availed themselves of the ability to use independent authoritative sources and other legitimate sources of wage information at rates 274 percent greater than the same timeframe in 2019. In fact, since publication of the IFR, the Department has received 14,153 LCAs supported by an independent authoritative source or other legitimate source of wage information for 153 unique occupations, compared to 19,509 representing 216 unique occupations for the entirety of FY 2020. This increased use of private surveys is consistent with the Department's experience that alternative wage surveys are readily available across many different regions and industries.

This widespread use and availability of alternative age sources extends to advanced degree occupations. Since the IFR publication, employers filing 523 LCAs representing 950 positions for occupations requiring advanced degrees used an independent authoritative source or other legitimate source of wage information. Since the IFR, there have been 4,973 PWDs requested for occupations requiring an advanced degree using a survey as the wage source; this is 1,780 more PWDs relying on a survey as the wage source than for similar PWDs in FY 2020.

For the PERM program, too, employers are required to obtain a PWD from the Department; they have the option of providing an alternate wage source to the OES survey in this process as well. There are well-established standards of acceptance of alternative wage sources. In the weeks since the publication of the IFR, the Department has received more than 6,900 prevailing wage requests supported by private wage surveys in the PERM program, which is a 335% increase over the same timeframe in 2019. Again, this increase confirms that such sources of wage data are readily available for use in seeking a PWD not based on the OES survey if employers believe in anomalous cases that the OES survey does not produce an accurate wage. This obviates any need, in the Department's view, to create a complicated, administratively burdensome scheme of occupation-specific wage structures.

3. The IFR Wages and Market Wage Rates

Summary of Comments

The most common concern raised by commenters on this subject was that prevailing wages under the IFR's methodology do not reflect actual market rates and in many cases are unrealistically high such that they will require employers to lay off currently employed foreign workers, will make it “difficult if not impossible, to hire for highly specialized and hard-to-recruit for positions,” and will “frustrate equal pay principles for U.S. workers, and create an endless upward spiral of wage obligations that bear no relation to market dynamics.” A professional association asserted that the wage level methodology in the IFR produced “artificially high” prevailing wages and circumvented congressional intent by making it “virtually impossible for employers to use the H-1B visa program.” The commenter asserted the Department violated section 212(n) of the INA by “incorrectly setting the way data is leveled” and “prevent[ing] employers from obtaining” from the Department's Online Wage Library “a wage that is in fact the prevailing wage for the” occupation and Area of Intended Employment or a wage that represents “the best information available as of the time of filing the application.” An employer asserted the IFR would create spiraling wages because “the next collection of BLS data will be distorted by these new wage requirements, yielding new and even higher prevailing wage requirements, in a pattern that will repeat and multiply” over time. A public policy organization said the IFR would lead to inflated wages because employers must post at the worksite the H-1B worker's salary, which will compel employers to pay the increased IFR wage to similarly employed U.S. workers.

Commenters cited numerous general and specific examples of substantial wage increases for combinations of occupations and areas of employment that do not reflect, according to commenters, market wages. Several commenters cited an NFAP analysis that compared wages under the IFR to private survey wages and pre-IFR OES wages and found that for all occupations and geographic locations the new wages are “on average, 39% higher for Level 1 positions, 41% higher for Level 2, 43% higher for Level 3 and 45% higher for Level 4.” Examples included a 99.5 percent increase for Level I petroleum engineers and for electrical engineers, computer network architects, computer systems analysts, mechanical engineers, and database administrators at all wage levels. The most dramatic examples included a Level I wage increase of more than 206 percent for a computer and information systems manager in East Stroudsburg, Pennsylvania, and more than 177 percent for a pediatrician in Wichita, Kansas. Referencing an American Action Forum report, a trade association cited average IFR wage increases for several occupations, including “an 83 percent increase for Level 1 Computer and Information Systems Managers” and “a 44 percent increase for Level 2 Software Developers.” Several commenters asserted the required prevailing wages for some information technology occupations would exceed the salary cap implemented by some big tech employers, such as Level II and Level IV wages in Silicon Valley and Seattle that exceed a $160,000 salary maximum at Amazon.

Many commenters stated the wages produced under the IFR did not reflect data on prevailing wages found on websites like Payscale, Glassdoor, Indeed, or Levels.fyi. Examples cited include Level I software developer wages in Santa Clara and Level I engineer wages in Seattle that are lower than the 45th percentile, according to Levels.fyi, and a median salary for software developers in Cincinnati that is $20,000 per year lower than the entry level wage under the IFR, based on Payscale data. One commenter also stated that the Level IV IFR wage for electrical engineers in Seattle exceeded $168,820, the highest wage listed for the occupation in O*Net.

A few commenters expressed concern that the IFR wages are not consistent with prevailing wage determinations produced by private wage surveys. A public policy organization compared wages under the IFR to surveys conducted by Willis Towers Watson and found a divergence in wage determinations between the two, including IFR wages 63 percent higher for Level IV programmers in Chicago, three times higher for all levels of financial analysts in New York City, and 62 percent higher for Level I software developers in Los Angeles. This commenter noted that many private surveys use “precise methodologies and a wide range of data gathering to ensure that the surveys” are accurate and they are “used by employers for company-wide salary benchmarking.” Similarly, a trade association stated that private wage surveys “commonly collect compensation information reflecting education, experience, and responsibility,” and a professional association stated these surveys often “gather real market data for what companies are paying employees at different levels,” in contrast to the OES, which gathers “general data without regard to experience levels.”

In addition to arguing that the IFR's wage rates were too high, a number of commenters highlighted what can be described as second and third order consequences of prevailing wage rates being out-of-step with market wages. For instance, comments primarily from academic and research institutions and related organizations and individuals expressed concern that if wages are untethered from market rates, particularly for post-doctoral research positions, clinical faculty, administrative positions, and teaching assistants, and the prevailing wage requirement would be untenable for institutions reliant on grant funding, especially those reliant on government funding. As a result, commenters believed the IFR would produce a shortage of qualified faculty and diminish the quality of education students receive; reduce already declining foreign student enrollment and tuition revenue; and derail critical research projects in science, healthcare, and technology.

Most of these commenters asserted wages under the IFR often are significantly higher than prevailing wages in the higher education or research sectors, and several commenters cited specific examples, like a Level I wage increase for post-doctoral researchers that would raise the wage higher than the salary of many experienced tenure track faculty. Several commenters asserted the increased wages would be especially burdensome for employers reliant on grant funding that may be subject to statutory or other limits on the funding amounts and the ways the employer can expend the funds. For example, a university stated that federal research organizations lack adequate funding to pay the IFR wages for work on research projects funded by federal awards and will need to reduce the size of those project groups or attempt to avoid employing H-1B workers on any of those projects. Other commenters noted more specifically that grants like those awarded by the National Institute of Health (NIH) are subject to rules limiting the amount that can be used for “administrative costs, including salaries,” and one commenter stated that the IFR prevailing wage for biological scientists would exceed the NIH salary cap by as much as 79 percent in some areas.

Commenters expressed concern the wage increases would diminish the quality of education universities provide by making it difficult or impossible to retain or hire qualified faculty, researchers, and workers in other jobs like administrative positions. A leading teaching and medical research hospital stated that the inability to retain researchers at the IFR wage levels would jeopardize critical research projects and the jobs in which U.S. workers are employed in “assistant, tech and coordinator roles.” Commenters also believed the wage increases would reduce post-graduation career opportunities significantly for international students and would reduce already declining foreign student enrollment, which in turn would contribute to a shortage of skilled labor in higher education and research and in the United States broadly. For example, some commenters asserted the IFR would reduce the number of available and qualified graduate teaching assistants, tutors, post-doctoral researchers, and similar workers because international students constitute a substantial portion of this labor force. An employer expressed concern about the impact of the IFR on the STEM and engineering labor force, noting that foreign graduates account for more than 70 percent of workers possessing a master's degree or Ph.D. in electronics engineering or related fields, according to a referenced 2018 National Center for Education Statistics Integrated Postsecondary Education Data System survey. Several of the commenters also stated the enrollment decline would reduce not only tuition revenue but also tax revenue and consumer spending.

Citing budget constraints and the importance of its work, a research organization reliant on NIH grant funding urged the Department to provide an exemption from the wage rule for ACWIA-eligible employers, which would encompass institutions of higher education and related or affiliated nonprofit entities, as well as nonprofit and governmental research organizations. The commenter added that the Department should continue to work to “update the ACWIA wage library.”

Comments primarily from healthcare providers and academic institutions expressed concerns similar to concerns of higher education commenters. The commenters asserted the new wage rates would exceed market rates, particularly for physicians subject to a $208,000 wage in many areas and for resident physicians. Two commenters asserted university clinical programs and medical research programs did not have adequate funds to pay the increased wages and asserted this would set back important “biomedical research during a pandemic” and curtail their ability “to care for and treat those afflicted.” A professional association stated that resident physicians are physicians in training and asserted that use of the OES to determine the prevailing wage for these job opportunities would produce wages higher than the actual prevailing wage for residents.

Most of these commenters asserted the increased wages would lead to a shortage of healthcare workers, including bilingual workers and mental health professionals and would reduce the quality of and access to healthcare and the quality of care available. Several commenters expressed concern this would have a particularly significant impact on providers in rural areas that have difficulty recruiting, cannot afford to pay the same wages as employers in larger areas, and often rely on foreign workers allocated to underserved areas through the Conrad-30 waiver program. One commenter also asserted the increased wages under the IFR “may cause elimination of the Conrad-30 waiver program” altogether.

Several commenters expressed concern the IFR would adversely impact small employers, start-ups, and nonprofits in particular because many these employers cannot afford competitive base wages due to limited resources and instead compete based on intangibles or use incentives like stock options. One commenter asserted that “incremental compliance costs” for small employers would be as much as three percent of revenue in 2020-21 and that these employers would effectively “be shut out of the H-1B visa program for new workers.” Some commenters asserted these employers are more likely to rely on DOL issued wages than private wage surveys, either due to inability to afford the survey or because they operate in small or nonmetropolitan areas and “private wage surveys are based on metropolitan area wages and do not cover many small market areas or less commonly utilized occupations because of data limitations.”

Many commenters expressed concern that the IFR would require employers to pay foreign workers more than the wage paid to U.S. workers or foreign workers hired prior to the IFR effective date and this would require employers to increase wages across the board due to the potential for worker resentment or decreased morale or because federal and state laws prohibiting discrimination require equal pay. For example, a professional association expressed concern that the IFR would require employers to pay the IFR wage to similarly employed workers to avoid potential pay equity claims under federal and state laws prohibiting discrimination, including Title VII of the Civil Rights Act of 1964 and a New York state law requiring equal pay for “equal or substantially similar work.” Similarly, some higher education commenters were concerned that they would need to pay the IFR wage to a broad range of U.S. workers due to “pay equity demands” or an “actual wage analysis” requiring payment of the higher wage to “all comparable workers.” Several commenters expressed general concern that the IFR would produce entry-level wages higher than wages paid to mid-career professionals or even the managers or supervisors of those workers.

By contrast, a number of commenters suggested that IFR's entry-level wage was set too low, that the entry-level wage should be placed no lower than the median of the OES distribution, and that some place even higher up within the distribution may be appropriate. A public policy organization asserted that wages “close to and above the median . . . will ensure H-1B workers are not being sought out simply because employers can save on labor costs.” A second public policy organization expressed concern that the pre-IFR wage level methodology that set rates below the median in the occupation “failed to require that firms pay market wages to H-1B workers.” A third public policy organization supported the increased wages under the IFR but expressed concern that setting the Level I wage “just below the local median wage” would “permit employers to pay H-1B workers at below market wage rates.” Similarly, a labor union and a commenter from academia supported the Department's decision to increase the Level I wage closer to the median, which the labor union asserted “is reflective of the minimum market rate that should be paid to an H-1B worker in order to safeguard U.S. wage standards and ensure that migrant workers in H-1B status are compensated fairly.” Another public policy organization and an academic commenter suggested the Department should increase the Level I wage to the 75th percentile and require that all H-1B job opportunities be certified “at a wage that is no lower than the national median wage for the occupation.”

Other suggestions about how to set the wage levels included one from an anonymous commenter, who urged the Department to set the prevailing wage at the highest prevailing wage in the country for the occupation, such as requiring all employers to pay the prevailing wage for physicians in New York City if that is the highest wage among all areas in the country. The commenter believed this would “equalize the cost to [all] employers” and would incentivize employers to recruit in other regions of the United States before hiring foreign workers. Another anonymous commenter suggested the Department should set the wage levels at the average of the IFR and pre-IFR levels, stating this would result in wage levels at the 31st, 48th or 50th, 64th or 66th, and 81st or 83rd percentiles for Levels I through IV, respectively.

Response to Comments

At the outset, the Department notes that commenters generally did not offer data or economic justifications purporting to show that the old wage level methodology produced wages across many different occupations and geographic areas that reflect the wages paid to U.S. workers similarly employed to H-1B and PERM workers. Further, as explained above, the Department has reasonably concluded that the old wage methodology, in many instances, is a source of harm to U.S. workers' wages and job opportunities. This fact, on its own, in the Department's view, gives rise to a clear inference that the old wage levels were not set in a manner that yielded prevailing wage rates on par with market wages. Whatever merits some commenters might see in the old methodology, it is clear it did not advance the purpose of the INA's wage provisions to protect U.S. workers. Of equal importance, and a reason independently sufficient for concluding that adjustments to the old wage methodology are needed, is the fact that the old methodology, as noted previously, is in tension with the governing statute.

The need for this rulemaking clear, the question then turns to how the wage levels should be adjusted. Notably, a number of commenters agreed with the foundational premise of the IFR that the Department should set prevailing wage levels based on an assessment of what workers with similar levels of education and experience to the foreign workers covered by the four-tier wage structure are paid. As one commenter said, “DOL reasonably claims that a well-functioning system for prevailing wages determinations would find that the wages that need to be paid for foreign national workers subject to these requirements `generally should approximate the going wage for workers with similar qualifications and performing the same types of job duties in a given labor market.' ” This commenter, and others, therefore did not disagree with the aim of the IFR, but rather simply claimed that the Department had overshot the mark and adjusted the wage levels so high that they do not reflect actual market wages.

The Department agrees with these commenters, and the reasoning in the IFR, that prevailing wage rates produced by the four-tier wage structure should approximate actual market wages to the greatest extent possible. The Department also takes seriously commenters' concerns that the IFR's wage levels may yield prevailing wage rates that do not meet that goal. It has therefore taken into account data and analysis provided by commenters to supplement and inform the analysis used in the IFR. Based on this reassessment of the conclusions it reached in the IFR, the Department has determined that it is appropriate to reduce the entry-level wage from the mean of the fifth decile, or the 45th percentile, to the 35th percentile. Doing so will, in the Department's expert judgment, and based on a review of the relevant data sources, including those provided by commenters, result in entry-level prevailing wage rates that approximate the wages paid to U.S. workers similarly employed to H-1B and PERM workers.

While the Department believes that data and analysis provided by commenters warrants a reassessment of the IFR's wage levels, the Department, as discussed in detail above, has determined that the analytical framework relied on in the IFR remains the appropriate lens through which to understand how the levels should be set.

While the INA provides the relevant factors and general framework by which the wage levels are to be set, it leaves the precise manner in which this is accomplished, including the types of data and evidence to be used and how such data and evidence are weighed, to the Department's discretion and expert judgment. In exercising that discretion, the Department's decision on how to adjust the wage levels is informed by the statute's purpose of protecting the wages and job opportunities of U.S. workers. This means the Department has focused its analysis on those areas where the risk to U.S. workers is most acute, taken into account how the foreign labor programs are actually used by employers, and, where appropriate, resolved doubts in favor of refining the wage calculations so as to eliminate to the greatest extent reasonably possible adverse effects on U.S. workers caused by the employment of foreign workers, while also ensuring that the program is still accessible to employers.

As explained in the IFR, to determine the wages typically made by individuals having comparable levels of education, experience, and responsibility to the prototypical entry-level H-1B and EB-2 workers and working in the most common H-1B and PERM occupations, the Department consulted a variety of data sources, most importantly wage data on individuals with master's degrees or higher and limited years of work experience—the type of worker the Department determined to be an appropriate wage comparator for entry-level H-1B and EB-2 workers—from the 2016, 2017, and 2018 CPS conducted by the U.S. Census Bureau, and data on the salaries of recent graduates of master's degree programs in STEM occupations garnered from surveys conducted by the NSF in 2015 and 2017. Both of these surveys represent the highest standards of data collection and analysis performed by the federal government. Both surveys have large sample sizes that have been methodically collected and are consistently used not just across the federal government for purposes of analysis and policymaking but by academia and the broader public as well.

In the case of the CPS survey, the Department used a wage prediction model to identify the wages an individual with a master's degree or higher and little-to-no work experience (based on age) would be expected to make and matched the predicted wage with the corresponding point on the OES wage distribution. Using the NSF surveys, the Department calculated the average wage of individuals who recently graduated from STEM master's degree programs and matched the average wage against the corresponding point on the OES distribution.

These analyses located three points within the OES wage distribution at which the wages of U.S. workers with similar levels of education and experience to the prototypical entry-level workers in specialty occupations and the EB-2 program are likely to fall. In particular, the 2015 NSF survey data indicate that workers in some of the most common H-1B and PERM occupations with a master's degree and little-to-no relevant work experience are likely to make wages at or near the 49th percentile of the OES distribution.¹⁷¹ The 2017 NSF survey suggests that these workers are likely to make wages at or near the 46th percentile of the OES distribution. On the low end, the CPS data suggest that such individuals make wages at or near the 32nd percentile.

The Department thus identified a range within the OES data wherein fall the wages of workers who, while being relatively junior within their occupations, clearly possess the kinds of specialized education and/or experience that the vast majority of foreign workers covered by the Department's wage structure are, at a minimum, required to have. Put another way, through an assessment of the experience and education generally possessed by some of the least skilled and least experienced H-1B and EB-2 workers—workers who are likely entry-level workers within their respective programs—the Department determined what U.S. workers with similar levels of education and experience are likely paid. Accordingly, it is appropriate for the wages paid to such U.S. workers to govern the entry-level prevailing wage paid under the Department's wage structure.

In the IFR, the Department explained that translating the identified range into an entry-level wage for the Department's use in the H-1B and PERM programs could be accomplished in a number of ways. One option would be to simply calculate the average wage of all workers that fall within the range, meaning those workers whose reported wage falls between the 32nd and 49th percentiles, which would place the entry-level wage at approximately just above the 40th percentile. An alternative would be to identify a subset of wages within the range—either on the lower end or the higher end of the range—and calculate the average wage paid to workers within such subset. Because of the greater suitability of the NSF data for the Department's purposes, likely distortions in the wage data of both surveys caused by the presence of lower-paid foreign workers in the relevant labor markets, and the purposes of the INA's wage protections, the Department determined in the IFR that the most appropriate course was to set the entry-level wage by calculating the average of a subset of the data located at the higher end of the identified wage range. This resulted in the entry-level wage being placed at approximately the 45th percentile. Notably, commenters did not dispute these three qualitative considerations the Department offered for why it favored the higher end of the range.

The Department therefore continues to believe that the reasoning that led it to set the entry-level wage at the higher end of the identified range remains relevant to its decision in this rule. For one thing, as between the two data sources and the manner in which they were analyzed, the NSF data are better tailored to the Department's purposes in identifying an entry-level wage for the H-1B program. The NSF surveys provide data on the wages of individuals with degrees directly relevant to the specialized occupations in which they are working, namely degrees in STEM fields. By contrast, the CPS data only show whether a person does or does not have a master's degree and does not identify what field the master's degree or the individual's undergraduate course of study was in. It is therefore likely that some of the wage data relied on in generating the CPS estimate were based on the earnings of individuals who possess degrees not directly related to the occupation in which they work. Given that the CPS data used only accounted for persons with little-to-no experience, such individuals would therefore be unlikely to have the qualifications needed to work in a “specialty occupation,” as that term is defined in the INA. Having neither a specialized degree nor experience, and therefore lacking in specialized skills or expertise, at least with respect to the occupations in which they work, such individuals would not qualify as similarly employed to even the least skilled H-1B workers and are thus not appropriate comparators for identifying an entry-level wage in the H-1B program. Because of these workers' relative lack of skill and expertise, they are likely to command lower wages, and thus decrease the predicted wage below what would be an appropriate entry-level wage for the Department's foreign labor programs.

Relatedly, the Department's method for approximating experience in the CPS data is also not as closely tailored to the goal of determining what U.S. workers similarly employed to the prototypical entry-level H-1B and EB-2 workers are paid as is the NSF data. The CPS analysis relied on potential experience as a proxy for actual experience, which was calculated using a standard formula of subtracting from individuals' ages their years of education and six, based on the common assumption that most individuals start their education at the age of six. While a standard measure for potential experience, this method of approximation is imprecise because it shows each individual of the same age and education level as having the same level of work experience. In reality, such individuals may vary significantly in their levels of experience.

For starters, the approximation does not take into account the possibility of a worker temporarily exiting the workforce, and would count the time spent outside the workforce as work experience. It also does not account for gaps between when a person received his or her bachelor's degree and when he or she enrolled in a master's degree program. In such cases, the work experience captured by the proxy of potential experience may thus not be directly relevant to the work a person performs after he or she graduates from a master's degree program since in some cases the work experience in question was likely acquired before the individual enrolled in a master's degree program. In consequence, the sample used in the CPS analysis almost certainly includes some individuals who have no relevant experience in the specialized occupations in which they are working, which likely decreases the wage estimate calculated using the CPS data and makes it a less precise and reliable estimation of the wages of U.S. workers with similar levels of education and experience to the prototypical, entry-level H-1B and EB-2 workers. In other words, the CPS data allows for only a rough approximation of experience—a key factor the Department must take into account in adjusting the prevailing wage levels. This, in combination with the fact that some workers contained within the CPS dataset likely also lack specialized education relevant to the occupations in which they work, means that CPS data is, in some degree, distorted by wage earners who should be discounted in identifying the appropriate entry-level wage because they likely possess neither the type of specialized experience nor the education in their field that is comparable to that possessed by entry-level H-1B and EB-2 workers.

The NSF survey data, by contrast, are uniquely suited to the Department's purposes. The NSF surveys in 2015 and 2017 capture wage data about exactly the sort of workers the Department has determined serve as the appropriate comparators for entry-level H-1B and EB-2 workers. They surveyed individuals with master's degrees in STEM fields who are working in STEM occupations, including some of the most common H-1B and PERM occupations, and who are approximately three years or less out of their master's degree programs. In other words, the NSF surveys report wage data for individuals with specialized knowledge and expertise working in the occupations in which H-1B and PERM workers are most often employed and who are relatively junior within their respective occupations. The NSF data therefore provide a more accurate wage profile of workers similarly employed to entry-level H-1B and EB-2 workers. While both data sources are useful in helping determine a wage range for entry-level H-1B and PERM workers, of the two, the NSF surveys provide information more relevant to the Department's assessment of what is the appropriate entry-level wage. Therefore, the Department's analysis relies more on the NSF surveys. This weighs in favor of placing the entry-level wage higher up in the identified wage range given that is where the NSF survey results fall.

Beyond the relative weight of each data source, the Department also takes into account in identifying the appropriate entry-level wage the fact that both sources are likely distorted to some degree by the presence, in both the surveyed population and the labor market as a whole, of the very foreign workers the Department has determined are, in some instances, paid wages below the market rate. As noted above, various studies and data demonstrate that some H-1B workers are paid wages substantially below the wages paid to their U.S. counterparts, and that this has a suppressive effect on the wages of U.S. workers. Further, these adverse effects are most likely to occur and be severe in occupations with higher concentrations of foreign workers. It is therefore relevant to how the Department weighs the data that many of the occupations examined in the analyses of the NSF and CPS datasets have very high concentrations of H-1B workers. H-1B nonimmigrants make up about 10 percent of the total IT labor force in the U.S. In certain fields, including software developers, applications (22 percent); statisticians (22 percent); computer occupations, all other (18 percent); and computer systems analysts (12 percent), H-1B workers likely make up an even higher percentage of the overall workforce.

From this, the Department draws two conclusions. First, the respondents reporting wages in the CPS and NSF surveys are likely in some cases H-1B or PERM workers, given that both surveys contain responses from both U.S. citizens and noncitizens and the surveyed occupations have high concentrations of such foreign workers. The reported wages are thus in some instances likely not the market wage paid to U.S. workers similarly employed to H-1B and PERM workers, but rather the wages of the foreign workers themselves, which, as discussed previously, will be likely lower than the wages of U.S. workers in some cases. Second, even the reported wages of respondents who are not H-1B and PERM workers are likely not perfectly accurate reflections of what the market rate would be absent wage suppression given that high concentrations of lower-paid foreign workers likely decrease the overall average wage paid in the relevant labor market, as detailed above.

The need to account for these distortions also weighs in favor of setting the entry-level wage at the higher end of the identified wage range. To discount this consideration would mean that, far from ensuring that the adjusted wage levels guard against adverse effects on U.S. workers caused by the presence and availability of lower-cost foreign labor, the Department would, to some degree, be basing its regulations on a preexisting distortion caused by the old, flawed wage methodology.

Finally, the purpose of the relevant INA authorities, particularly the prevailing wage requirement, also weighs in favor of adjusting the entry-level wage higher up within the identified wage range. As emphasized throughout, the guiding purpose of the INA's prevailing wage requirements is to “protect U.S. workers' wages and eliminate any economic incentive or advantage in hiring temporary foreign workers.” Giving due weight to the purpose of the statutory scheme suggests, in the Department's judgment, that uncertainties should, to some extent, be resolved so as to eliminate the risk of adverse effects on U.S. workers' wages and job opportunities. That also countenances in favor of placing the entry-level wage at the higher end of the wage range.

However, in response to the IFR commenters provided the Department with additional data and considerations, which have led the Department to modify the wage levels established in the IFR. As noted, the principal concern commenters expressed about the IFR was that the wages it produces are significantly higher than the actual market wages employers pay their workers. To substantiate this criticism, various commenters offered wage figures from private and public wage surveys, and, in some instances, reported what specific employers pay their workers. The wage data from commenters analyzed by the Department generally dealt with wages paid to what commenters represented to be starting or entry-level positions.

To allow for a meaningful comparison with the wage figures used in the IFR, the Department selected a cross section of the wage data provided by commenters and used the same mode of analysis it used in the IFR to match those figures with percentiles in the OES. In particular, it compared annual wage data offered for specific jobs in specific metropolitan areas with OES data for the occupation in which the job falls in the same metropolitan area. OES data provides annual wage data for the 10th, 25th, 50th, 75th, and 90th percentiles for occupations at national, state, and metropolitan area levels. Using these data, the Department interpolated annual wages data provided by commenters at each of the missing percentiles between the 10th and the 25th, the 25th and the 50th, the 50th and the 75th, and the 75th and the 90th percentiles. This allowed the Department to approximate the specific percentile at which the wages offered by employers fall.

In general, the Department found that the annual wage data for specific jobs in specific metropolitan areas offered by commenters were clustered around percentiles in the 30s. Some annual wage data offered by commenters fell in lower percentiles, and a few fell higher in the distribution.

A number of commenters cited annual wage data based on salary offers for L3 software developers with no relevant work experience from major employers that are significant users of H-1B workers in the Seattle-Tacoma-Bellevue, WA Metropolitan Statistical Area (MSA) and San Jose-Sunnyvale-Santa Clara, CA. These offers ranged between the 25th percentile and 42nd percentile of the OES distribution. Excluding the lowest offer and the highest offer, most offers were clustered between the 32nd percentile and 41st percentile.

One commenter cited annual wage data from Glassdoor for entry-level tax managers at public accounting firms in the New York-Newark-Jersey City, NY-NJ-PA MSA. The Department found that the annual wage was between 33rd percentile and the 34th percentile. Another commenter offered Indeed and Payscale annual wage data for accountants in the Dallas-Fort Worth-Arlington, TX MSA. Using the higher annual wages from the two surveys, annual wages were between the 19th percentile and the 20th percentile.

One comment cited Glassdoor, Payscale, and ZipRecruiter data for minimum and maximum annual wages for statisticians in the New York-Newark-Jersey City, NY-NJ-PA MSA. Review of this data showed the minimum annual wages were less than the 10th percentile. Another comment cited Glassdoor average annual wage data for financial analysts with no experience in in the Dallas-Fort Worth-Arlington, TX MSA, which showed that the average annual wages were between the 31st percentile and 32nd percentile.

A commenter cited annual wages offered by a major university in the Bloomington, IN MSA. Because of data limitations in the OES, the Department could only compare the annual wages for the computer system analyst position provided by the commenter. The Department found that the annual wages for this position were between the 68th percentile and 69th percentile.

A commenter cited the annual wages of an assistant professor of clinical pediatrics/physician surgeon at a major university in the Chicago-Naperville-Elgin, IL-IN-WI MSA. The Department found the annual wages were between the 44th percentile and the 45th percentile.

One commenter cited the annual wages of four employees of a major university in the Salt Lake City, UT MSA: (1) A computer and information research scientist, (2) a database architect, (3) a foreign language instructor, and (4) a pediatric endocrinologist. The Department found that these annual wages were (1) between the 36th percentile and the 37th percentile, (2) between the 32 percentile and the 33rd percentile, (3) between the 12th percentile and 13th percentile, and (4) between the 34th percentile and the 35th percentile, respectively.

Another commenter cited Glassdoor annual wage data for a structural engineer with four to six years of experience in the Boston-Cambridge-Nashua, MA-NH MSA. The Department found that the annual wages were between the 24th percentile and the 25th percentile.

One commenter cited Willis Tower Watson private wage survey data for eight jobs in different metropolitan area that compare with Level 1 and Level 4 OES. The Department focused on the Level 1 data and found the following:

In sum, most of the wage data offered by commenters was for salaries paid by employers to entry-level workers in positions typically filled by H-1B workers. While there are outliers, most of these wage observations fell between the 30th and 40th percentiles of the OES distribution. Importantly, wage data about entry-level software developers employed by some of the largest users of the H-1B program fell between the 32nd and 41st percentiles. This is noteworthy given that such data may allow for the closest comparison to the IFR's data of all the private wage data submitted by commenters. This is because, as noted above, the IFR's analysis also focused on software developers and other occupations in the IT sector to account for the fact that such occupations comprise the largest share of the relevant programs.

It is also notable, in the Department's judgment, that, while the wage data submitted by commenters tends to be lower on the OES distribution than the IFR's 45th percentile entry-level wage, it still generally falls within the wage range between the 32nd and 49th percentiles identified by the IFR as the portion of the OES distribution where U.S. workers similarly employed to entry-level H-1B workers are likely to be found. From this, the Department draws two conclusions. First, the IFR's determination that wages paid to workers similarly employed to entry-level H-1B and PERM workers likely fall in this range seems to be largely accurate. While there are outliers in the wage data provided by commenters that fall both well above and well below the range, the data from commenters does not give the Department reason to abandon its conclusion in the IFR that some point within that range will serve as the appropriate entry-level wage.

Second, while consistent with the IFR's wage range, the commenters' data suggests, contrary to the IFR's reasoning, that the lower half, as opposed to the upper half of the range, would be a more appropriate place to set the entry-level wage. While the IFR offered a variety of reasons for why the NSF data, which falls at the higher end of the range, were likely better suited as compared to the CPS data for informing the Department's decision about where to set the entry-level wage, and the Department still views those considerations as relevant, the commenters' data suggests otherwise. As noted, the CPS data suggest that a point closer to the 32nd percentile would be the appropriate place to set the entry-level wage, which many data from commenters would seem to confirm.

As was the case in the IFR, the Department does not evaluate the data from either the government sources it analyzed or the private wage data submitted by commenters in a vacuum. Various qualitative considerations, including key points raised by commenters, shape the Department's assessment of what conclusions to derive from this data.

First, DOL regulations and guidance establish quality standards for the use of private wage sources in setting prevailing wage rates. Some of the private wage sources provided by commenters—particularly the comments that offer a single example of a wage paid by one employer in one geographic area—would almost certainly not satisfy these standards if an employer sought to use them to establish a wage rate for its H-1B workers. These data are therefore arguably entitled to less weight than the data relied on in the IFR. Similarly, even as to the private wage survey sources offered by commenters that may satisfy DOL's standards, the NSF and CPS data are, in the Department's judgment, of higher quality. These are highly credible government surveys administered by agencies with extensive experience in gathering wage data. This too suggests that the data provided by commenters is entitled to less weight in the Department's analysis than the data used in the IFR.

Similarly, as explained above, the analysis used in the IFR controlled for characteristics relevant to setting a wage rate under the INA's framework. Because the Department is seeking to set an appropriate wage primarily for workers in specialty occupations—not for workers generally—the IFR took, among other things, the INA's minimum qualification requirements for working in a specialty occupation into account in deciding what data to use. It is at best unclear whether some of the surveys offered by commenters are also limited to workers who could be described as working in a specialty occupation, and therefore similarly employed to H-1B workers. For example, while data from one commenter suggest that an entry-level computer programmer working in the Chicago area makes wages that fall between the 32nd and 33rd percentiles of the OES distribution, computer programmers will likely not in all cases be properly regarded as working in a specialty occupation. For example, in some cases, the job of a computer programmer may involve writing basic computer code and testing it. As explained previously, because a person without a specialized bachelor's degree can still be classified as a Computer Programmer, some portion of Computer Programmers captured by the OES survey are not similarly employed to H-1B workers because the baseline qualifications to enter the occupation do not match the statutory requirements. It is therefore possible that the computer programmer described as an entry-level worker by the commenter may not in fact have the same level of qualifications as an entry-level H-1B computer programmer. In such cases, the wage data provided by commenters, being based on the wages paid to workers who lack the specialized knowledge required of H-1B workers, is likely below the level that would be an appropriate entry-level wage for the Department's foreign labor programs. This, in turn, suggests that the data provided by commenters are entitled to less weight than the IFR's analysis, which controlled for the INA's specialty occupation requirement, and may also explain some of the extreme outliers at the lower end of the OES distribution found among commenters' data.

Relatedly, some of the commenters' private wage surveys report the bare minimum wage paid to workers in the occupation as the entry-level wage. Given that entry-level workers typically fall within a range of the wage data, as opposed to falling only at the very low end of the distribution, some of the private wage data arguably does not represent what would count as a reasonable entry-level wage, even if some portion of entry-level workers do in fact make wages at the bottom end of the distribution. Indeed, as the Department explained above, the purpose of the INA's wage provisions to protect U.S. workers suggests that uncertainty over how to read available wage data should be resolved in favor of placing the entry-level wage higher up within the distribution to eliminate as much as possible risks to U.S. workers from the employment of foreign labor. Yet these private wage sources do just the opposite, offering what is the absolute bare minimum wage that an entry-level worker might be expected to make. This too likely accounts for some of the outliers in the commenters' data that fall below the IFR's identified wage range, and suggests a wage higher up within the range should be selected.

On the other side of the equation, and in addition to the data they provided, commenters have provided the Department with various considerations that pull in the direction of favoring the lower end of the IFR's wage range. As explained previously, commenters detailed various second and third order consequences that would result if prevailing wage rates do not approximate actual market wages. These consequences include limiting healthcare providers', universities', and small businesses' ability to use the H-1B program, which would, in turn, disrupt research and impede access to healthcare, particularly in rural areas. Commenters also expressed concerns about the effect overly inflated prevailing wages would have on their ability to comply with pay equity laws. The Department takes these concerns seriously, and has determined that they weigh in favor of placing the entry-level wage at the lower end of the range identified by the IFR.

To begin with, the Department notes that many if not all of these problems are eliminated if prevailing wages rates are set in line with actual market wages. Each of these issues arises principally because, according to commenters, the IFR's wages do not approximate market wages. Setting an appropriate entry-level wage based on available data and other relevant considerations is thus the appropriate way to address these concerns.

As explained previously, the Department continues to believe that the range identified by the IFR accurately reflects the portion of the OES distribution where workers with levels of education, experience, and responsibility similar to the vast run of entry-level H-1B and PERM workers likely fall—something that commenters' wage data largely confirms. However, as the Department has also acknowledged, there is some level of indeterminacy about the exact point in that range at which placing the entry-level wage will yield optimal outcomes in the largest number of cases given that different data sources point toward somewhat different conclusions. In the IFR, the Department reasoned that the purpose of the INA's wage provisions to protect U.S. workers warranted resolving such indeterminacy in favor of placing the wage higher up within the range. However, the Department also recognizes that a purpose of the H-1B program more generally is to ensure that employers can access needed high-skilled labor to supplement their workforces. Given that prevailing wage rates that are substantially above actual market wages can impede employers' access to the program, and cause various problematic, secondary consequences, the importance of avoiding such outcomes weighs in favor of resolving indeterminacy in favor of the lower end of the identified range. While the INA's wage provisions must be implemented in a way that fully protects U.S. workers' wages, raising wages to such a degree that the program becomes unusable for many employers defeats the entire reason Congress created the program. Placing the entry-level wage at a lower point within the range is one way to ensure that does not occur.

Relatedly, because the four-tier wage structure covers hundreds of thousands of workers employed across hundreds of different occupations by a wide variety of different employers, there is some level of variability as between different workers and what would constitute an appropriate entry-level wage for each of them. As explained above, in establishing the identified range, the Department focused its analysis on those occupations that account for the largest number of workers covered by the four-tier wage structure. The Department continues to believe this is appropriate given that occupations with large numbers of foreign workers are where U.S. workers are most at risk of experiencing adverse wage effects due to competition from foreign labor. However, the Department also acknowledges that some occupations, such as physicians, that account for a smaller share of H-1B and PERM workers and are therefore given less weight in how the Department identified the entry-level wage range, may have entry-level market wages that are somewhat lower within the OES distribution than the top H-1B occupations. This is because, as commenters explained, occupations like physicians typically require all workers in them to possess an advanced degree, meaning that, while in the top H-1B occupations the INA's specialty occupation requirement will generally mean that wages paid to H-1B workers should be placed higher up within the OES distribution, that is less true of advanced degree occupations. Workers in such occupations with qualifications similar to the least skilled H-1B worker might be found closer to the lower end of the OES distribution.

In consequence, while the analysis used to identify the entry-level wage range largely focused on top H-1B occupations, the decision of where within that range the entry-level wage should be set should give additional weight to occupations that account for a smaller number of workers within the program, particularly the advanced degree occupations about which commenters raised concerns. This suggests that the lower end of the entry-level range would be a more appropriate point to place the first wage level. Indeed, the Department notes that data from at least one commenter about the starting salary of a pediatric endocrinologist—which falls between the 34th and the 35th percentiles of the OES distribution—suggest that the lower end of the range may yield an appropriate entry-level wage for some positions in advanced degree occupations. Further, as discussed previously, some commenters suggested that the bottom third of the distribution for advanced degree occupations consists of entry-level workers similarly employed to H-1B workers. If commenters are correct, that means that the lowest points within the entry-level range identified by the Department does in fact cover the highest paid entry-level workers in such occupations.

Accounting for small businesses and rural employers that use the H-1B and PERM programs in selecting a point within the entry-level range identified by the Department also weighs in favor of the lower part of the range. As commenters note, large employers are able in some cases pay higher wages than small businesses. Further, wages in metropolitan areas may be higher to the extent that these are high-intensity occupational areas. The Department notes that some of these differences are already accounted for by other aspects of the regulatory framework governing prevailing wage rates. In particular, the Department issues wages based not only on the occupation a worker is in, but also on the geographic area in which the worker is employed. Thus, for example, while the wage data described above from large tech companies fall between the 32nd and 41st percentiles of the wage data gathered for the metropolitan areas in which those firms operate, such data fall well above the 60th percentile of the national OES wage distribution. By taking geographic area into account in analyzing what the appropriate entry-level wage is, the Department has thus, to some degree, already accounted for the differences between employers about which some commenters expressed concern. However, the Department also recognizes that higher wages may still be less manageable for small businesses and rural employers, which suggests that the lower part of the entry-level range would be appropriate.

Moreover, the Department acknowledges that placing the entry-level wage at any place within the identified range—even the lowest point—will result in significant wage increases for employers that may, in some cases, be difficult to adapt to given how long the old wage methodology has been in place. As detailed at greater length below, the Department is addressing this concern by phasing in the new wage rates over a period of time. However, the Department also believes that, even with a phased-in approach, the ability of employers to adapt to a significant change is relevant to the decision of where to set the entry-level wage. Insofar as a smaller increase—albeit one that is still substantial—will be more manageable for employers, the Department considers that also to be a reason to favor the lower end of the range.

On balance, the Department has determined that the factors pointing to the lower end of the identified range carry greater weight than the reasoning relied on in the IFR to select the higher end of the range. Accounting for advanced degree occupations, employers' ability to access the program and adapt to the change effected by this rule, and private wage data are all compelling considerations put forward by commenters that, in the Department's judgment, warrant a reassessment of its decision in the IFR. Thus, while in the IFR the Department chose to set the entry-level wage at approximately the 45th percentile, which fell at approximately the midpoint of the upper half of the entry-level range, the Department is now adjusting the level downward to approximately the midpoint of the lower half of the range, which is the 35th percentile.

Importantly, setting the wage at the 35th percentile will, in the Department's view, still provide the full protection to U.S. workers contemplated by the INA. The 35th percentile falls within the range identified in the IFR as the portion of the OES distribution where workers with qualifications comparable to entry-level H-1B and PERM workers are likely to fall. The manner in which the Department identified that range, as recounted above, relied on a variety of considerations, including the INA's specialty occupation requirement and how that interplays with the OES data, to ensure that the interests of U.S. workers are fully and properly accounted for in how the wage levels are set. As a result, while lower than the level set in the IFR, the 35th percentile will still achieve the purpose of the INA's wage provisions. While a point higher up within the range may also be reasonable, and the Department may reassess how to set the entry-level wage as it gains experience administering the entry-level at the 35th percentile, the Department believes that the 35th percentile strikes the right balance between fully protecting workers' wages and job opportunities while also preserving employers' ability to access the program.

By favoring the lower end of the range, the Department is confident the second and third order consequences identified by commenters as a product of prevailing wage rates that are inflated above actual market wages will be reduced if not eliminated by the downward adjustment in the entry-level wage. The Department notes that the downward adjustment is substantial. To compare the effects of the final rule on prevailing wages with the effects on prevailing wages produced by the IFR, the Department calculated the prevailing wages for two common occupations for H-1B workers (web develops and electrical engineers) in five metropolitan area (Atlanta-Sandy Springs-Roswell, GA; Austin-Round Rock, TX; Chicago-Naperville-Elgin, IL-IN-WI; San Jose-Sunnyvale-Santa Clara, CA; and Seattle-Tacoma-Bellevue, WA) under the IFR and the final rule. The Department then analyzed the differences. Comparing the prevailing wages under the final rule and interim final rule, the Department found that the prevailing wages are significantly lower under the final rule for both occupations in all five metropolitan areas at all four levels expect for the prevailing wage for level 4 web developers in Seattle, which is $7,322 or 3.8% higher (see table below).

Further, the Department notes that many of commenters' concerns are also addressed by other measures the Department is taking in this final rule. For example, commenters' complaints about overly inflated wages for physicians, particularly in rural areas, focused in many cases on the fact that the IFR resulted in a default wage of $208,000 a year for all four levels in a number of different locations. As detailed more fully below, the Department is eliminating the influence of outliers on the upper level wage, reducing the upper level wage, and providing a default rule for cases where BLS is unable to calculate the upper level wage to ensure that the Department provides leveled wages wherever possible. These measures will further alleviate complications healthcare providers and other employers in rural areas encountered under the IFR.

The Department disagrees with comments that suggested that the median of the OES distribution should be the absolute minimum for the entry-level wage, and that some point even higher up in the distribution might be appropriate. The purpose of having a four-tier wage structure is to provide gradually increasing wages as workers skill levels increase. The entry-level wage should therefore be set not based on what the median wage is of all workers, but rather based on an assessment of what other entry-level workers with qualifications comparable to H-1B and PERM workers possess. As detailed at length above, the Department's review of the relevant data and other considerations indicates that a point below the median is the right place to set the entry-level wage.

The Department also rejects other alternatives suggested by commenters. For example, the recommendation that the Department set wages by averaging the IFR's wage levels with the old wage levels is flawed because, as noted, the old wage levels were selected arbitrarily, and therefore should not be a significant factor in how the Department determines the new wage levels, except insofar as the Department takes into account employers' and workers' reliance interests in the prior methodology. The Department also disagrees with the commenter that suggested that the prevailing wage should be the highest prevailing wage in the nation for any given occupation. Doing so would ignore the importance variations in labor markets by geographic area have long played in how prevailing wage rates are provided, as well as the statutory requirement that prevailing wage rates be based in part on geographic area.

4. Reliance Interests

Summary of Comments

Many commenters expressed concern about the IFR's negative impact on current H-1B visa holders in the United States, especially those with families and strong ties in the United States and those with pending or approved I-140 Immigrant Petitions for Alien Workers or pending I-485 Applications to Register Permanent Residence or Adjust Status (“green cards”). Several commenters discussed the impact on foreign workers who had expected to continue working in the United States and for some, obtain lawful permanent status through their employer. Commenters expressed concern that employers would terminate H-1B visa holders and “potential green-card recipients” would have to leave the country. An individual commenter asserted that the IFR would inhibit job opportunities for international graduates of U.S. universities, regardless of their capabilities, and contended that the new wage levels would disincentivize legal immigration. Similarly, another individual commenter described the rule as “eliminating legal immigration paths” and warned that it will cause foreign workers who have contributed greatly to the U.S. tax base to go out of status.

Commenters stated that since some employers will not be able to afford the wage increases and will terminate foreign workers, the IFR would have devastating effects on the lives of foreign workers with families, property, and ties to a community. One lobbying organization stated the IFR would mean that “many talented foreign nationals [would be] forced to leave the U.S. because these new wage requirements make it impractical to continue employing them in our country.” Based on polls of their membership conducted by some of the signatories, a group of professional associations and advocacy organizations asserted that as many as 70 percent of H-1B workers who are making progress toward obtaining a green card, and in many cases have “developed permanent ties to the United States” through home ownership or U.S.-born children, may have to abandon the process. The commenter also stated that the IFR “understates or ignores altogether the reliance interests” of the nearly 600,000 H-1B workers currently employed in the United States. This professional association warned that H-1B workers whose status is threatened by the IFR will need to leave the country abruptly, impacting not only the workers, but also their spouses and children, and it expressed concern about COVID-19 complicating further their ability to relocate. The commenter maintained that suddenly changing the longstanding rules that have before now allowed workers to buy homes, raise children, and otherwise create ties to the United States over time is unfair and unreasonable. Meanwhile, an attorney claimed that the IFR's economic and social impacts will be acute for Indian nationals in particular because they often face long delays while waiting for green cards, which the commenter said results in many purchasing houses and having children here. One public policy organization that supported the IFR opposed immediate implementation, asserting the abrupt wage increase would put currently employed workers in a “precarious position” and “may cause churn if employers [are] unwilling to pay real market wages [and] decline to renew their workers' H-1B visas or initiate petitions for permanence.”

One commenter also expressed concern that an employer may violate DOL regulations at § 655.731(a) if it pays the IFR wage to workers hired after the IFR effective date, but continues to pay a current H-1B worker the lower wage issued prior to the IFR, because the employer will be paying less than the actual wage to the first employee. The commenter suggested that this would result in additional disruption to employers' operations as the new wage levels would result in increases in the wages owed to new H-1B workers, but would result in immediate changes to the wages owed to workers already employed.

Many commenters expressed concern that immediate implementation of the IFR dramatically increased prevailing wages too abruptly, jeopardizing operations by disrupting long-term budget and other planning, interfering with contractual obligations, and preventing employers from adapting to the wage increases by adjusting operations and hiring and training new workers.

A professional association expressed concern that the immediate implementation of the IFR would increase costs for “human resources and compensation staff to bring their companies into compliance with the rule” and asserted the Department failed to consider staffing changes that may be necessary for employers that cannot “support” wages at levels produced by the IFR methodology. A trade association expressed concern that the IFR may cause “material disruption” to employers' “operations or delivery models . . . because of long-term contractual commitments . . .” and that these employers may be “forced to operate at a loss” because they are unable to re-negotiate contracts entered into prior to the IFR effective date. Another trade association stated that the IFR forced employers to put “talent acquisition and workforce development decisions on hold” and required them to “reconsider work schedules, cost increases, and performance metrics that impact their entire workforce.” The commenter expressed concern that immediate implementation of the IFR created operational disruptions because employers relied on the published July 2020 OES wages “to create plans, develop strategies and hiring, and consider talent retention and immigration programs.” A third trade association asserted the IFR may cause long-term damage to employers and disrupt U.S. worker hiring processes because employers “plan and budget their hiring months and often years in advance.”

A higher education policy organization noted that colleges and universities have planned budgets and salaries and signed employment contracts in reliance on wages produced by the Department's wage surveys and expressed concern the IFR would require these employers to “re-visit all of those plans, in the midst of a pandemic and in the middle of an academic year.” The commenter also stated the Department's wage rules are complex and that universities have “spent years developing the methodology according to DOL requirements” and “invested significant resources over the years to train international offices on DOL prevailing wage methodology.” A higher education professional association noted hiring cycles at academic institutions “often run over a year” and that employers have already made offers to foreign workers “based on the ability to sponsor H-1B status and/or green cards.” A university submitted a similar comment and expressed concern that immediate implementation of the IFR would require the employer to renegotiate employment offers “in some cases . . . just days before the expiration of the beneficiary's current status.”

Several commenters, including some that expressed support for the IFR, urged the Department to provide for a transition period or to phase in the wages over time to permit employers to adjust to the wage increases. A commenter from academia suggested the Department should phase in the new wage levels over no more than a two-year period, which the commenter believed would be sufficient time for employers to adjust to the new wage levels while also preventing employer “exploitation of artificially low wage rules.” A public policy organization that supported a phase-in period also suggested the Department should work with DHS to “create positive incentives for employers who match the new wage requirements for their existing workforce.” A public policy organization also suggested the Department should apply the revised wage level methodology only to “new workers in . . . temporary work visa programs with [LCAs] submitted after the IFR took effect” to avoid “discourag[ing] renewals and petitions for lawful permanent residence by employers unwilling to pay market wage rates.” The commenter stated this would protect workers who were “contracted under one set of rules and expectations” by avoiding an unreasonable change to those terms and conditions of employment.

Response to Comments

While the Department believes that adjusting the wage levels in the IFR to a level that more closely approximates the actual wage typically paid to U.S. workers similarly employed to H-1B workers will address many if not most of the concerns raised by commenters about the impact of the new wage methodology, it also recognizes that implementing such an immediate and significant change may cause disruption to employers' and foreign workers' reliance interests in the old methodology. While such reliance interests are difficult to quantify, the Department has sought to account for these interests and ensure that the new wage levels are implemented in a way that appropriately balances the need to protect U.S. workers with the Department's obligation to consider reliance interests engendered by its prior methodology, the Department has decided to adopt a series of measures to ease the transition to the new wage structure.

In particular, the Department is including in the final rule a delayed implementation period under which adjustments to the new wage levels will not begin until July 1, 2021. Further, once adjustments begin, they will be made in a phased approach, with most job opportunities not becoming subject to the full increase to the new levels until July 1, 2022. For workers who are on track to receive lawful permanent resident (LPR) status, as indicated by their being the beneficiaries of approved employment-based green card petitions, or otherwise eligible to extend their H-1B status beyond the six-year limit, the Department has determined that a more gradual phase-in occurring in four steps that results in job opportunities filled by such workers being placed at the new wage levels beginning on July 1, 2024, is appropriate. Finally, to the extent that employers' actual wage obligations under the INA may result in more immediate changes to the wages they must pay workers who have already received work authorization on a previously approved LCA, the Department will take this into account in exercising the discretion afforded it by the INA when enforcing such obligations.

In effecting an adjustment to the wage levels previously used to set the prevailing wage in the H-1B and PERM programs, the Department is obligated to consider whether “its prior policy has engendered serious reliance interests.” In the IFR, the Department recognized that the old wage levels “have been in place for over 20 years, and that many employers likely have longstanding practices of paying their foreign workers at the rates produced by the current levels.” The Department further acknowledged that making significant adjustments to the wage levels “may result in some employers modifying their use of the H-1B and PERM programs,” and “will also likely result in higher personnel costs for some employers.” Despite these considerations, the Department concluded that “to the extent employers have reliance interests in the existing levels . . . setting the wage levels in a manner that is consistent with the text of the INA and that advances the statute's purpose of protecting U.S. workers outweighs such interests and justifies such increased costs.”

As explained above, the Department continues to believe that the old wage levels are the source, in many cases, of serious, adverse effects on U.S. workers' wages and job opportunities. Adjusting the levels to bring them in line with the wages paid to U.S. workers with levels of education, experience, and responsibility comparable to H-1B workers—and thereby reducing the danger posed to U.S. workers by the employment of foreign workers—remains the principal aim of this rulemaking. Ensuring that the Department's wage structure is set in accordance with the relevant statutory factors is also necessarily a controlling objective in the Department's assessment of how best to reform the prevailing wage levels. The old levels have never been justified by economic analysis, and, as detailed above, are in tension with the statutory scheme insofar as they are based, in many instances, on data about the earnings of workers who cannot be regarded as similarly employed to workers in specialty occupations. Effecting a significant adjustment to the wage levels, and doing so as expeditiously as is reasonably possible, is therefore of paramount importance in the Department's judgment.

That said, concerns raised by commenters about disruptions to business operations, fairness to foreign workers, and the feasibility of adapting to significant changes to the wage levels in a short period of time are also entitled to weight in how the Department implements adjustments to the levels. The old levels were set far too low, which means that the adjustment necessary to bring them in line with what similarly employed U.S. workers make, and therefore be consistent with the statutory scheme, is substantial. The Department notes that shifting the entry-level wage from approximately the 45th percentile provided for by the IFR to the 35th percentile means the adjustment employers will have to make to accommodate themselves to the new levels is less dramatic. But it is still significant. Indeed, approximately 60 percent of all LCAs in recent years have been for job opportunities at the first and second wage levels, which are at roughly the 17th and 34th percentiles of the OES distribution. Setting the lowest wage level at the 35th percentile thus means that the prevailing wage for all H-1B workers going forward will likely be higher—and in many cases substantially so—than the prevailing wage for as much as 60 percent of the current H-1B population. The Level III and Level IV wages will also now be, in many cases, higher than the highest wage required under the old Level IV wage. Considerations brought to the Department's attention by commenters about the effects of an adjustment of this magnitude have provided the Department with greater insight into how to implement such a substantial change.

For that reason, the Department has reassessed how it balanced in the IFR reliance interests in the old wage levels with the need to adjust the wage levels. To begin with, the Department reiterates that setting wages so as to protect U.S. workers is the central purpose of the INA's wage requirements. To the extent commenters suggest that business practices have evolved around and been shaped by the old wage levels, and that the old levels, or something close to them, should therefore be maintained indefinitely or for extended periods of time to prevent disruption to employers' operations, the Department disagrees. The fact that some employers have long benefited from inappropriately low wage rates cannot justify the continued perpetuation of the harms to U.S. workers that result from foreign workers earning wages that do not reflect what similarly employed U.S. workers are paid.

However, in light of the comments it received, the Department has determined that a wage increase that is both dramatic and immediate is also undesirable, and indeed may be counterproductive to the aims of this rule. For one thing, as some commenters noted, immediate disruptions to business operations, such as might lead to the termination of contracts or the shuttering of offices, may in fact threaten U.S. workers with job losses or reductions in work. Adopting a rule that eliminates workers' jobs in order to protect their wages advances neither the interests of workers nor the purposes of the INA.

Similarly, the Department acknowledges that, while the aim of the INA's wage requirements is to protect U.S. workers, one purpose of the H-1B program more generally is to ensure that employers can access needed high-skilled labor to supplement their workforces. Although permitting employers to access temporary foreign labor must be accomplished in a way that works no harm on the wages and job opportunities of U.S. workers, it is also important to ensure that reforms to the prevailing wage do not unnecessarily limit employers' use of the program. Helping employers bring the wages they pay their H-1B workers in line with the requirements of the INA while avoiding the kind of abrupt change that might make it unreasonably difficult for employers to adapt is therefore consistent with the broader goals of the H-1B program.

For those reasons, the Department has determined that a gradual transition to the new wage levels is needed to account for employers' reliance interests on the prior system while still ensuring that U.S. workers' wages and job opportunities are fully protected. Such an approach is a reasonable method of effecting a regulatory change that results in increased costs on regulated entities. Modifying the existing system over a period of time, even where the prior system is inconsistent with the governing statute, can assist affected parties in “reorder[ing] their affairs.” The Department's decision to implement the new wage rates through a transition rather than through an immediate adjustment is also consistent with the notice the IFR gave to the public of the intended policy change.

Modifying the prevailing wage levels through a delayed or graduated transition matches how Congress and other agencies have instituted similar changes to employers' wage obligations in other contexts. For example, all increases in the Federal minimum wage that Congress enacted over the last 60 years were phased in over two or more years. Only two of the ten minimum wage adjustments since the enactment of the Fair Labor Standards Act have been made fully effective immediately. The three most recent amendments to the Federal minimum wage were implemented over two or three year periods. In so doing, Congress has sought to minimize any loss of jobs or other economic disruptions that an immediate, one-step increase in the Federal minimum wage might cause to labor markets. Changes to minimum wage laws at the state level are also often made through incremental adjustments. Similarly, the Department has employed comparable transition provisions when implementing wage changes in other foreign labor programs.

Similarly, the Wage and Hour Division has typically implemented changes to employers' obligations to provide overtime pay through delayed effective periods. The most recent change to overtime rules was made effective more than 90 days after the final rule was published—more time than is required by either the Administrative Procedure Act or the Congressional Review Act. The 2016 overtime rule (later enjoined) was made effective more than five months after publication. The 2004 overtime rule was made effective 120 days after publication. In both cases, the Department determined that a delayed effective date would “provide employers ample time to make any changes necessary to ensure compliance with the final regulations.”

The Department notes that changes to the minimum wage or overtime obligations are different in important respects from the adjustments the Department is making to the prevailing wage levels. For one thing, changes to the minimum wage and overtime requirements are often made in light of gradual changes in economic conditions that make it necessary to reassess a prior policy determination. By contrast, in undertaking a change to the prevailing wage levels, the Department is giving meaningful consideration to what employers' wage obligations should be based on available economic data for the first time since the Department began using a multi-level wage structure in its foreign labor programs. Similarly, while Congress's decision to adjust the minimum wage is driven entirely by competing policy considerations, the Department's discretion to adjust the wage levels is to some degree confined by the INA. As explained above, the Department is adjusting the manner in which it sets prevailing wage rates not only because the existing wage levels are the source, in some cases, of harm to U.S. workers' wages and job opportunities, but also because they are inconsistent with the governing statute. In consequence, the reform the Department is undertaking in this rulemaking is long overdue and of greater significance than similar kinds of changes to employers' wage obligations in other contexts. Finally, as explained further below, the adjustments to the prevailing wage levels will not have the same kind of immediate impact on employers' wage obligations with respect to all workers currently on their payroll as changes to the minimum wage do. Employers will be able to pay H-1B workers currently employed in many cases at the current wage levels for the duration of the validity period of their current LCAs. Increases in the wage levels will generally have an immediate impact only on new workers or where the employer seeks to renew a current worker for a new period of employment. In consequence, immediate changes to the wage levels are likely to be less disruptive than immediate increases in the minimum wage. In combination, these considerations weigh in favor of keeping the transition period to the new wage levels of short duration, even if that means employers will still be required to adapt quickly to a significant increase in the wage levels.

The Department has therefore decided to implement the adjustments to the prevailing wage levels through a combination of a delayed effective period and multi-step adjustments occurring over approximately a year and half period. The first adjustment to employers' prevailing wage obligation will not occur until July 1, 2021. This delay from publication of this rule until the first wage increase will give employers time to plan for the adjustment. Adjusting the wage levels on July 1st is also consistent with historical practice at the Department, which has typically published the new annual wage rates for the H-1B and PERM programs each year at the beginning of July. Employers are thus accustomed to modifications being made at that time of the year.

On July 1st, the entry-level wage will increase from roughly the 17th percentile to 90 percent of the 35th percentile wage, as provided by BLS—a point approximately halfway between the current Level I wage and the 35th percentile, which, as explained above, is the point in the OES distribution that the Department has determined is appropriate for setting entry-level wage rates. Similarly, at the same time the Level IV wage will increase from roughly the 67th percentile to 90 percent of the 90th percentile wage. The following year, on July 1, 2022, the wage levels will again increase, and be placed at the 35th percentile for the entry-level wage and the 90th percentile for the uppermost level, at which point the transition to the new wage structure will be complete.

The Department determined the appropriate step up in wages by analyzing national wage data for the top ten occupations in which H-1B workers are employed. In particular, the Department averaged the wages estimated to fall at various percentile in the OES distribution using linear interpolation, and weighted that average by the share of H-1B workers in each occupation relative to the total number of H-1B workers in the top ten occupations. In so doing, the Department relied on the same basic methodology it used to determine the appropriate entry-level wage. As explained elsewhere, the Department's interest in maintaining a single, uniform wage methodology for the H-1B and PERM programs means that the wage provided will not be perfectly tailored to every job opportunity or geographic location. Providing wages that are closely tailored to the unique circumstances of as many job opportunities as possible while still using a single wage structure necessarily means that the Department must focus on nationwide data and those occupations that account for the largest share of the affected programs.

An analysis of national data for the top ten H-1B occupations indicates that 90 percent of the average wage at the 35th percentile falls approximately at the midpoint between wages at the 17th percentile—a rough proxy for the wages yielded by the old wage methodology—and wages at the 35th percentile. Similarly, 90 percent of the average wage at the 90th percentile is approximately the midpoint between wages at the 67th percentile—a rough proxy for the Level IV wages yielded by the old methodology—and the 90th percentile. Requiring employers to pay wages that are 90 percent of the 35th percentile for entry-level workers and 90 percent for Level IV workers in the first stage of the two-step implantation of this rule will thus ensure an even and gradual adjustment over the period of time the Department has determined is appropriate to allow employers to adapt to the new wage rates.

The Department recognizes that, even under this incremental approach, wage rates will still increase significantly in a relatively short period. An analysis of wage rates based on current OES data suggests that an increase in the entry-level wage from roughly the 17th percentile to 90 percent of the 35th percentile may equate in many cases to a real dollar increase of approximately 14 percent in the annual wages employers will be required to pay their foreign workers. However, for the reasons given above, the Department believes that a transition consisting of both a delayed effective period and a gradual increase to the new wage levels occurring over a year and a half period is the appropriate way to balance the need to ensure U.S. workers are not harmed by the presence of foreign workers in the labor market while giving employers time to adapt to the new wage system. Further delay in adjusting to the new levels would, absent some other compelling consideration, entail too great a risk to U.S. workers' wages and job opportunities, in the Department's judgment.

Beyond employers' general reliance on the old wage levels, the Department notes that some employers also have reliance interests in a specific worker or group of workers currently working who were hired on the understanding that they would be employed at wages based on the prior prevailing wage methodology. Immediate changes to the wages employers are required to pay could change the expectations employers had about the cost of employing such workers when they invested in sponsoring them for a visa. Such concern would only pertain to visa workers who have already been approved and who are already working. It is unlikely that this kind of immediate change to employers' wage obligations to current workers will occur, however, to the extent it does the Department possesses some enforcement discretion to mitigate against any such potential impact on visa workers hired under the prior prevailing wage methodology.

As some commenters noted, there is a possibility that employers' wage obligations as to current workers will be immediately affected by significant adjustments to the prevailing wage levels, even though the Department has already approved LCAs for these workers, which contain prevailing wage rates that will remain valid for the duration of the LCA's validity period. This may occur through operation of employers' actual wage obligation under the INA and the Department's regulations, which is to say their obligation to pay the higher of the actual wage or the prevailing wage to their H-1B workers. As the Department's regulations note, “employers are cautioned that the actual wage component to the required wage may, as a practical matter, eliminate any wage-payment differentiation among H-1B employees based on different prevailing wage rates stated in applicable LCAs.” While new prevailing wage rates based on this rule's revised methodology will not immediately change the prevailing wage for H-1B workers with already-approved LCAs, the arrival of new H-1B workers at the same worksite that is subject to a higher prevailing wage under the new methodology could potentially modify employers' actual wage obligations with respect to current H-1B workers and result in the employer having to pay a higher wage.

While acknowledging this issue, the Department believes, as a practical matter, it is unlikely that the introduction of new H-1B workers at a worksite will result in immediate and significant increases in the wages an employer is required to pay current H-1B workers who have already been approved to work at prevailing wage rates based on the prior wage methodology. First, the Department's Wage and Hour Division has never brought a case in which an employer was deemed to have violated its actual wage obligations as a result of a different H-1B worker being paid a higher prevailing wage rate. This is so for a few reasons. For instance, for the wage paid to a new H-1B worker to be relevant to the employer's actual wage obligation to a current worker, the new worker would not only have to be stationed at the same specific worksite, but also possessed of similar qualifications and experience as the current worker and be performing the same set of duties and responsibilities. Thus, the wages paid to many new H-1B workers will likely simply not be relevant to employers' actual wage obligations to current workers.

Second, the actual wage “reflects the application of an employer's actual pay system.” Employers are therefore permitted to establish the actual wage they pay H-1B workers by taking into account “Experience, qualifications, education, job responsibility and function, specialized knowledge, and other legitimate business factors.” In consequence, even as between H-1B workers with similar qualifications and experience performing the same duties and responsibilities, an employer may have other legitimate reasons for paying these workers different wages. The fact that one worker has a significantly higher prevailing wage rate will, in many cases, be only one of many relevant factors governing the employers' actual wage obligation.

In those instances where the employer has not documented and cannot reconstruct its actual wage system, the Department may base the actual wage on averaging the wages paid to all similarly employed workers. In those instances, the introduction of a new H-1B worker at the worksite will not necessarily cause the actual wage owed to current H-1B workers to immediately increase to whatever the new workers' prevailing wage rate is. Rather, a more modest increase may be required based on an average of what the new worker is being paid as compared to what similarly employed current workers are making.

Finally, although the Department does not believe that employers' actual wage obligations to current H-1B workers are likely to change immediately as a result of adjustments to the prevailing wage levels, the Wage and Hour Division will, where appropriate, take the above factors into consideration in enforcement actions. In some cases, the Department has discretion over whether to launch an investigation into potential violations of the INA's wage requirements. Similarly, even in those cases where the Department is obligated by statute to initiate an investigation and make a determination as to whether a violation has occurred, the assessment of civil money penalties, where such penalties are applicable at all, is sufficiently flexible to take all of the facts and circumstances into account.

In the unlikely event that violations of this kind arise the Department will evaluate them on a case-by-case basis, and, in choosing whether to bring an enforcement action or impose civil monetary penalties, the cause of the violation will be taken into account.

Once a currently employed worker's LCA expires, the employer will, except as explained below, be required to pay the worker a prevailing wage rate based on the new methodology if the employer seeks a new labor certification. As noted above, some commenters suggested that this will result in certain employers being unable to renew their workers for a new period of employment as it will be too costly to do so, and that this will be disruptive to business operations. While this may be the case in some instances, the Department emphasizes that H-1B visas provide only temporary work authorization. Neither employers nor guest workers on H-1B visas can claim a permanent interest in a temporary employment relationship. Further, requiring employers to file new LCAs periodically to continue employing H-1B workers gives teeth to the INA's wage protections by ensuring that the prevailing wage an employer must pay is not based on out-of-date information. Allowing all current H-1B workers to continue working at the prevailing wage rates below the level the Department has determined is appropriate after the LCAs associated with their positions have expired and their employers have filed new LCAs would undermine the Department's determination that significant adjustments are needed to the wage levels to adequately protect U.S. workers.

In consequence, when an employer files a new LCA as part of the process of renewing an H-1B worker for a new period of employment, the Department has concluded that it is appropriate that the new prevailing wage rates should, except as noted below, apply. To the extent employers may have had expectations that current workers could be renewed at rates based on the old wage levels, such expectations are naturally circumscribed by the fact that H-1B visas are inherently temporary in nature and there is no legal guarantee that work authorizations will be renewed on the terms that they were previously granted. Further, any such expectations are, in the Department's view, outweighed by the need to guard against adverse effects on U.S. workers' wages and job opportunities.

Beyond concerns about being able to renew current H-1B workers generally, some commenters also noted that employers' and guest workers' reliance interests in the old wage methodology are particularly weighty in cases where the employer has sponsored the H-1B worker for LPR status. As one commenter noted, H-1B workers who are on the path to obtaining LPR status “often have purchased a home, developed permanent ties to the United States, or made a decision to have children here, counting on obtaining Lawful Permanent Resident status.” That commenter also suggested that an immediate and abrupt change in the wage rates could mean that “65%-70% of all individuals being sponsored for green card status through a Permanent Employment Certification may be unable to continue in the process” as their employers will be unable to pay the increased wage rates. Relatedly, employers of such workers have undertaken additional investments in the workers beyond what would ordinarily be expended on sponsoring an H-1B worker as part of the permanent labor certification process.

The Department agrees with commenters that H-1B workers who are on the path to becoming employment-based lawful permanent residents present unique considerations for how the Department transitions current H-1B workers to wage rates produced by the new wage methodology. These individuals, in many cases, have spent extended periods of time in the United States, during which they have developed greater connections to this country than the typical temporary visa holder. What's more, they have done so under a legal regime established by Congress that permits and, indeed, encourages them to develop strong ties to the United States. In other words, not only have these individuals built lives in the United States in reliance on the prior wage methodology, which set the terms of their employment, but their expectation of being able to remain in the country indefinitely has been fostered by congressional enactments specifically designed to treat this group of individuals differently than other H-1B visa holders. For that reason, the Department has concluded that accelerated, significant increases in the wages employers owe these workers, insofar as it may result in large numbers of these workers losing their current employment, and therefore potentially being required to depart the country, would work a unique hardship and unfairness on both the workers themselves as well as the employers that have made greater investments in retaining these workers. In consequence, the Department has determined that a more gradual transition to the new wage rates for these workers is appropriate.

As the Department noted in the IFR, unlike most nonimmigrant visas, H-1B visas are unusual in that they are “dual intent” visas, meaning under the INA H-1B workers can enter the U.S. on a temporary status while also seeking to adjust status to that of lawful permanent residents. One of the most common pathways by which H-1B visa holders obtain lawful permanent resident status is through employment-based green cards, and in particular EB-2 and EB-3 visas. USCIS has estimated that over 80 percent of all H-1B visa holders who adjust to lawful permanent resident status do so through an employment-based green card. This is reflected in data on the PERM programs. In recent years, more than 80 percent of all individuals granted lawful permanent residence in the EB-2 and EB-3 classifications have been aliens adjusting status, meaning they were already present in the U.S. on some kind of nonimmigrant status. Given that the H-1B program is the largest temporary visa program in the U.S. and is one of the few that allows for dual intent, it is a reasonable assumption that the vast majority of the EB-2 and EB-3 adjustment of status cases are for H-1B workers. This is corroborated by the Department's own data, which shows that, in recent years, approximately 70 percent of all PERM labor certification applications filed with the Department have been for H-1B nonimmigrants.

Because of how many H-1B visa holders apply for EB-2 and EB-3 classifications, Congress has repeatedly adapted the INA to account for the close connection between the programs. For example, while H-1B nonimmigrants are generally required to depart the U.S. after a maximum of six years of temporary employment, Congress has created an exception that allows H-1B nonimmigrants for whom PERM labor certification applications have been filed with the Department or petitions for employment-based immigrant visas have been filed with DHS that have been pending for longer than a year to be exempt from the six year period of authorized admission limitation if certain requirements are met. In such cases, the workers are able to renew their H-1B status in one-year increments indefinitely until the process by which they can obtain lawful permanent resident status is resolved. Similarly, aliens who are the beneficiaries of an approved petition for an EB-1, EB-2, or EB-3 green card and who are eligible to be granted LPR status but for application of the per country limitations are permitted to extend their stay beyond the usual six year limit in three year increments.

Congress created these exceptions to the temporary limits of H-1B status in recognition of the fact that the method by which employment-based green cards are allocated—namely through the operation of caps on the number of visas that can be allocated to nationals of a given country in any given year—can result in significant delays between when an alien is approved for a green card and when the green card is actually issued. Put another way, the system for allocating employment-based green cards often results in protracted periods during which a worker can, in some sense, have one foot in the temporary H-1B program and another in the PERM program as they progress to LPR status. These workers, while not yet possessed of LPR status, have made substantial, formal steps toward acquiring such status, and, in so doing, acquired more permanent ties to the United States than does the typical temporary worker. Congress recognized as much and singled out this group for a special accommodation that allows their temporary status to continue indefinitely. In so doing, Congress further increased the degree to which such workers can reasonably expect to be permitted eventually to remain in the country on a permanent basis.

Congress's creation of exceptions to the six-year limit on H-1B status was also undertaken in recognition of the fact that requiring workers on track to receive LPR status to leave the United States after six years before they receive a green card would be disruptive to the employers of such workers. As noted above, employers that have sponsored H-1B workers for an employment-based green card have undertaken investments in retaining such workers beyond what would ordinarily be required to continue renewing such workers' H-1B status. Similarly, in many cases these workers will likely have been with their employer for longer than the typical H-1B worker, meaning the employer may have developed a greater reliance on the services of these particular workers. Absent these workers being able to extend their stays indefinitely, they “would otherwise be forced to return home at the conclusion of their allotted time in H-1B status, disrupting projects and American workers.” As a result, Congress chose to allow “these individuals to remain in H-1B status until they are able to receive an immigrant visa and adjust their status within the United States, thus limiting the disruption to American businesses.”

In sum, H-1B workers whose employers have taken substantial, formal steps toward obtaining an employment-based green card are uniquely situated as compared to other H-1B visa holders subject to the Department's prevailing wage methodology such that applying a sudden and significant change in wages would work a special hardship to such workers and their employers to the extent it might result in some workers losing their H-1B status. Not only have many of these workers spent extended periods of time in the United States, and begun building lives here, but they have done so with a guarantee from Congress that they legally may remain here beyond the six year limit that usually applies to H-1B visa holders until their application for LPR status is resolved. And because such workers are seeking employment-based green cards, their employers in many cases also have substantial reliance interests on such workers' continued presence in the country beyond what would normally be the case for other H-1B workers. The special status of workers who are the beneficiaries of an approved employment-based green card petition, or who are otherwise eligible to extend their status beyond the six-year limit, has also been recognized by the Department of Homeland Security in a separate rulemaking that singled this group out for unique treatment for many of the same reasons outlined above.

Consequently, as suggested by some commenters, the Department is adopting a phase-in approach to how it applies the new wage methodology to job opportunities that will be filled by workers who are on track to obtaining employment-based green cards. While, for the reasons given above, the Department believes that a two-step transition is appropriate with respect to new H-1B workers and many other workers for whom their employer seeks renewed status, the Department has concluded that the unique circumstances of workers who are on track to receive LPR status warrant a longer transition period. These workers and their employers have more substantial expectations of their being able to remain employed in the United States that have been engendered by congressionally created exceptions to the six year limit on H-1B status.

The Department is also cognizant of its obligation to ensure that U.S. workers' wage and job opportunities are protected. That consideration, as elaborated previously, means that any transition to the new wage structure should be kept as short as reasonably possible while still accommodating the reliance interests identified by commenters. The Department believes that a delayed implementation period followed by a four-step adjustment occurring over a three and a half year period for job opportunities filled by workers on track to receive LPR status appropriately balances these competing considerations.

By making the phase-in nearly twice as long for these workers, and stretching it out over a period of more than three years, the Department has taken into account the fact that most LCAs are approved for a three year period, meaning that all employers seeking to renew the status of H-1B workers on track to receive LPR status will be able to do so at least once at wage levels below the new levels set by this rule and that in many cases will be closer to the prevailing wage rates that would have obtained if the prior methodology had been left in place. This allows for a more gradual transition than would be achieved if these job opportunities were subject to the two-step phase-in occurring over a year and a half. Gradually increasing the wage rates that will be available for these job opportunities over a period of time also takes into account the need to protect U.S. workers by not allowing the current, inappropriately low wage levels to remain in place beyond the initial, delayed effective period, as well as the fact that wage increases that occur further out in time from the date this rule is published will be more manageable for both employers and workers to plan for. Moreover, the Department notes that, because employers have undertaken significant investments in the long-term employment of these workers, a longer transition period is also unnecessary insofar as such employers can be expected to have an incentive to undertake the additional expenditures needed to retain the workers at the new prevailing wage levels by the time the transition is complete.

The Department recognizes that many H-1B workers on track to receive LPR status will still be on H-1B status and have their green card petitions pending at the time the transition to the new wage rates is complete. Workers in the green card backlog as of October 2020 may not be able to obtain an employment-based green card for a decade or more. However, in the Department's judgment, delaying full implementation of the new wage rates for what amounts to a significant share of the current H-1B population until all workers on track to receive LPR status have had their green card petitions resolved would result in far too lengthy of a delay that would result in ongoing harm to U.S. workers' wages and job opportunities. A three and a half year, graduated transition gives these workers adequate time to adjust to the new wage rates, whether by allowing their employers sufficient time to adapt or, in some cases, allowing such workers additional time to find a new employer that is able to pay the higher wage rates.

Using the same methodology and data it used to set the wage rate at the intermediate step of the two-step transition, the Department has concluded that the wage rates for the three and a half year transition will be 85 percent of the wage rates produced by the 35th and 90th percentiles beginning in July, 2021; 90 percent of such wage rates beginning in July, 2022; and 95 percent of such rates beginning in July, 2023. For the reasons given with respect to the year and a half transition, these rates allow for a gradual, even adjustment to the wage levels the Department has determined are appropriate. Beginning in July 2024, the wage rates provided for any job opportunity filled by an alien on track to receive LPR status will be the same as the wage rates provided for all H-1B job opportunities.

Finally, the Department has decided that the job opportunities that should be eligible for these special transition wage rates are those that will be filled by any H-1B workers who, as of October 8, 2020, were the beneficiaries of approved employment-based green card petitions, or who were otherwise eligible to extend their temporary status beyond the six year limit under the American Competiveness in the 21st Century Act. October 8th is the date the Department published the IFR and thereby gave notice to employers and workers that it would be increasing wage rates. It thus provides a clear, administrable delineation of the class of workers who can benefit from the three and a half year transition period, and takes into account the fact that workers whose expectation of being able to remain in the country indefinitely became settled before such notice was provided have the most compelling reliance interests in the prior wage methodology.

5. Wage Data and Sources

a. OES

Summary of Comments

Some commenters expressed concern about the Department's exclusive reliance on the OES to determine prevailing wages. Citing an NFAP policy brief, a public policy organization commented the “fundamental problem” with prevailing wage determinations is that the “process requires statistical precision that simply is not available” because “no government survey [ ] collects data within occupations with detailed wage levels, much less a survey that seeks to assemble data to calculate wage levels based on experience, education or level of supervision.” The commenter further stated that the OES produces “two average wage figures, neither of which is based on the collection of data connecting compensation to education, experience or supervision.” The commenter expressed concern that this method is less reliable than “asking employers directly what they pay employees at different levels of education, experience, or supervision” and that “a government agency can adjust the formula in a way that makes the required wages far higher than the market rate.” An employer expressed concern the OES “does not measure workers' skills or duties or “reflect what workers in the survey are paid” and instead “simply records [the] set of DOL-established pay bands” within which a worker can be classified.

Several commenters also expressed concern that the OES fails to consider total compensation, including stock options and bonuses, for example, resulting in an underestimation of the total earnings of U.S. and foreign workers. An individual commenter noted that many workers, particularly those in information technology occupations, earn much more than their base salary when accounting for total compensation and asserted that the IFR unfairly advantages “companies with a cash-heavy pay structure” and harms small start-ups that are more likely to compete by providing “equity and stock options.” A trade association asserted the IFR ignores an “important evolution” in the compensation of professionals “whereby many employers add to annual salaries with variable compensation tied to productivity, performance, or other specific goals” and may “incentivize employers to abandon variable compensation schemes altogether, in order to use available resources in an attempt to meet the new required wages.” Citing a Society for Human Resources Management article stating “85% of employers use variable pay. . .” an employer asserted that consideration of fixed pay exclusively is outdated because an increasingly important component of compensation packages is variable pay, including “incentive plans, bonuses, profit-sharing plans, performance-sharing plans, and equity.”

Many commenters expressed concern that the Department would issue a prevailing wage of “exactly $100 an hour, or $208,000 a year, for any occupation and geographic area” for which the Department lacks sufficient OES wage data to determine a prevailing wage for each wage level. Many commenters cited a finding by a public policy organization that this $208,000 wage requirement would apply to at least 18,000 combinations of occupations and geographic locations. A university stated that assigning a “default wage rate of $100” per hour “for each of the four wage levels . . . artificially inflates the wage data for each of the wage levels for affected occupations.” A trade association expressed concern that OES wage data is “skewed toward employers in large metro areas” and that the failure to collect sufficient wage data would result in many non-metropolitan employers receiving a “default” prevailing wage of $208,000 under the IFR. A professional association believed the lack of BLS data and resulting “default” wage of $208,000 was due to the Department's decision to use data for a limited “pool of workers who use the H-1B . . . and PERM programs,” rather than using a “prevailing wage data pool [ ] based on all wage data within the occupation, regardless of the number of years of education, experience, and level of responsibility.” A second professional association asserted assignment of a $208,000 wage in this context violates the INA, 8 U.S.C. 1182(p)(4), because the Department provides only one wage level, despite the four levels of wages required by Congress, and that it is contrary to a 1990 Congressional directive that BLS must “make determinations on prevailing wages” and make this information “readily available to employers and workers.” Many of these commenters provided examples of prevailing wages far exceeding the market wage, such as a prevailing wage of $208,000 for an entry-level software developer in California, despite a private wage survey determination that the prevailing wage is approximately $70,600 per year.

A public policy organization and an academic commenter that supported the IFR wage increases urged the Department to clarify an employer's wage obligation in these cases, expressing concern that the policy created confusion that threatens necessary wage reform efforts. Specifically, one of the commenters requested clarification of whether the employer must pay the $208,000 salary, must “use an alternative method to the OFLC-generated OES wage rates in these cases,” or may choose either option.

Response to Comments

The Department received many comments regarding the prevalence of the use of the OES footnote wage to set prevailing wage rates under the IFR's wage levels. This issue arises when BLS cannot provide a wage estimate for a Level IV wage. BLS is unable, at times, to produce a wage estimate when the survey results at the upper end of the wage distribution exceed the highest wage interval BLS uses, which is $100 an hour or $208,800 annually. In such cases, BLS reports a default wage, or footnote wage, of $208,000 for the Level IV wage to OFLC as that is the highest wage value available. Currently, BLS collects actual wage data from employers and then converts the actual wage data into wage intervals, which range from under $9.25 an hour to $100.00 an hour and over. In situations when BLS reports a footnote wage for the Level IV wage to the Department, the Department's standard practice has been to note that leveled prevailing wages for an occupation and/or geographic area was unavailable and only to provide the OES footnote wage for all four levels.

Under the Department's proposal in the IFR, the mean of the upper decile produced an OES footnote wage for more than 18,000 occupations, up from roughly 6,000 occupations under the old prevailing wage methodology. The higher prevalence of the use of the footnote wage under the IFR's methodology resulted in the default wage of $208,000 per year being used for a number of occupations where its use was likely not appropriate, as some commenters noted. The Department has therefore determined that it a change to its standard practice of not providing leveled wages in these situations is warranted.

Upon the effective date of this final rule, when BLS is able to report a Level I wage, the Department will utilize the OES footnote only as the Level IV wage estimate in cases where the 90th percentile wage value exceeds the highest wage interval value used by BLS. This change will allow the Department to provide leveled wages even where the footnote wage must be used for the Level IV wage and ensure that entry-level wages are not improperly inflated. In making this change, the Department expects there will be far fewer instances of the Department being unable to provide leveled wages than was the case under the IFR, or even the old wage methodology.

This change to how the Department handles situations where the footnote wage is used for the Level IV wage will ensure that leveled wages and an entry-level wage appropriately set at the 35th percentile will be provided wherever possible. This change will largely eliminate those incidents commenters expressed concern about, such as in healthcare occupations, where even an entry-level wage under the IFR was set at $208,000 per year, and is thereby inflated well above both the previous entry-level wage as well as what the Department has determined is an appropriate entry-level wage. Like its decision to move the entry-level wage to the 35th percentile, this change will ensure that prevailing wage rates more accurately reflect actual market wages and are more manageable for employers. Further, as discussed in more detail below, the changes the Department is making to how it calculates the Level IV wage—namely by using the 90th percentile as the Level IV wage instead of the mean of the upper decile—will eliminate the influence of extreme outlier at the upper end of the distribution, thereby reducing the reported Level IV rate to a level that is not inflated by anomalous data, and thus potentially reducing the frequency with which the footnote wage is used even for Level IV wage.

The Department acknowledges that there will continue to be instances, as there are currently, where BLS will report to OFLC an OES footnote wage for all levels in an occupation because the survey results received by BLS at and above the 35th percentile are all in the wage interval of $100.00 an hour and over. This will occur in a few very highly compensated occupations. Importantly, in such cases the use of the footnote wage will actually result in a lower prevailing wage rate than would otherwise be the case if actual wage data were available because BLS only reports up to the maximum interval of $100.00 an hour and in these situations the actual wages are at or over $100.00 an hour. Put another way, the use of the footnote wage in these cases, unlike its use under the IFR, will not result in wages that are inflated beyond what the actual market wage would be if actual wage data were available. Until BLS moves away from collecting all wage data in intervals this will continue to occur. But the Department believes that as BLS expands its collection of actual wage data this issue will cease to occur even in those few very highly compensated occupations. The Department anticipates that this change to its standard procedures will allow the Department to report leveled wages in more occupations and/or geographic areas than has historically been the case.

Relatedly, many commenters expressed concern that because the Department raised the Level IV wage to the mean of the upper decile, it caused more physician occupations, in particular, to default to the OES footnote wage of $100.00 an hour, or $208,000 annually at an especially high rate. As discussed above, the Department's changes to its standard procedures to use the OES footnote wage only as the Level IV wage estimate when a Level I wage is also reported from BLS will allow the Department to report leveled wages in these instances, thus reducing, if not altogether eliminating this concern.

Similarly, many commenters suggested that the failure of the Department to provide leveled wages would disproportionately harm employers outside of large urban areas and cause rural communities to lose access to healthcare. Many of these commenters suggested that under the IFR the Department is unable to provide leveled wage estimates for physicians and researchers in rural areas who would therefore be provided the OES footnote that is significantly higher than what some of those employees' supervisors are paid, which would be unsustainable and potentially result, among other things, in undermining the Conrad-30 program in certain areas. However, as previously stated, the Department has reviewed the commenters concerns and determined it is appropriate to make changes to the standard procedures of not providing leveled wage estimates in these situations. Instead, upon the effective date of this Final Rule the Department will use the OES footnote wage only as the Level IV wage estimate, allowing the Department to provide leveled wage estimates, except in those cases where the wage at the 35th percentile is also above the highest OES wage interval value. This will reduce if not eliminate the incidents of inappropriately high wages being provided for these specific occupations and areas.

The Department also acknowledges commenters' concerns with flaws in the OES collection of wage data from employers that result from BLS collecting data in 12 wage intervals as opposed to reporting actual wages. Though the OES survey does collect most wage data in wage intervals, BLS does collect actual wage data from employers in some instances and is exploring the ability to collect and report actual wage data from employers on a more consistent basis. As BLS phases in the collection of actual wage data from employers, wage estimates reported to the Department will become even more accurate and all instances of the OES footnote wage being used to set prevailing wage rates, which is a product of the current practice of using wage intervals, should cease. Further, even if BLS ultimately does not convert all wage data collection from employers to actual wages, this methodology of using wage intervals has been in place since the inception of the OES survey and has in most cases produced accurate wage estimates at the levels defined by the Department. Given the low incidence of the footnote wage being used; the modifications made by the Department to how it provides default wages that both further reduce the use of the footnote wage and eliminate its use in cases where it would result in an inappropriately inflated wage; and the other strengths of the OES data discussed below, the Department continues to believe that the OES survey serves as the best possible source of wage data for use in various foreign labor programs and that its reliance on wage intervals does not warrant the Department abandoning its longstanding practice of using the OES.

As noted above, the Department received several more general comments regarding the suitability of the BLS OES data for setting wages in the foreign labor certification programs. Some of the comments cited the fact that the OES data uses broad occupational classifications that encompass a wide range of different positions, some of which only fall at the lower end of the pay scale. Others commented that the OES data does not survey for education and experience, making it a poor fit for use in setting H-1B wage levels.

As the Department stated in the IFR, the Department reviewed the statutory framework of the INA and its interplay with the BLS OES survey data that the Department uses to calculate prevailing wages. This review demonstrated that, while the OES survey is the best source of wage data available for use in the Department's foreign labor certification programs, it is not specifically designed for such programs, and therefore does not account for the requirement that workers in the H-1B program possess highly specialized knowledge in how it gathers data about U.S. workers' wages. This fact necessarily shapes how the Department integrates the OES survey into its foreign labor programs.

The Department has long relied on OES data to establish prevailing wage levels. That is because it is a comprehensive, statistically valid survey that is the best source of wage data available for satisfying the Department's purposes in setting wages in most immigrant and nonimmigrant programs. As the Department has previously noted, the OES wage survey is among the largest continuous statistical survey programs of the federal government. BLS produces the survey materials and selects the nonfarm establishments to be surveyed using the list of establishments maintained by State Workforce Agencies (SWAs) for unemployment insurance purposes. The OES collects data from over one million establishments. Salary levels based on geographic areas are available at the national and State levels and for certain territories in which statistical validity can be ascertained, including the District of Columbia, Guam, Puerto Rico, and the U.S. Virgin Islands. Salary information is also made available at the metropolitan and nonmetropolitan area levels within a State. Wages for the OES survey are straight-time, gross pay, exclusive of premium pay. Base rate, cost-of-living allowances, guaranteed pay, hazardous duty pay, incentive pay including commissions and production bonuses, tips, and on-call pay are included. The features described above are unique to the OES survey, which is a comprehensive, statistically valid, and useable wage reference. The OES survey's quality and characteristics have made it, and continue to make it, a useful tool for setting prevailing wage levels in the Department's foreign labor programs. There are no consistently and readily available alternative surveys or sources of wage data that would provide DOL with wage information at the same level of granularity needed to properly administer the H-1B and PERM programs. For these reasons, the Department continues to believe that the OES survey is the best possible source of wage data for use in various foreign labor programs.

The Department also notes that the OES survey is what is currently used to set prevailing wage rates in the H-1B and PERM programs. As a result, even if the modifications to the prevailing wage levels in this final rule were not adopted, the OES would continue to be the source used to produce prevailing wage rates by the Department. As explained, the Department believes that continuing to use the OES is the best way to advance the policy aims of the INA's wage protections. However, even if reconsideration of the Department's use of the OES were warranted, the Department believes that the more immediate goal of correcting how the wage levels are set is the appropriate focus of this rule.

However, as noted, the OES survey is not specifically designed to serve these programs. For one thing, “the OES survey captures no information about differences within the [occupational] groupings based on skills, training, experience or responsibility levels of the workers whose wages are being reported” —the factors the INA requires the Department to rely on in setting prevailing wage levels. Relatedly, “there are factors in addition to skill level that can account for OES wage variation for the same occupation and location.” Further, the geographic areas used by BLS to calculate local wages do not always match up exactly with the “area of employment” for which wage rates are set, as that term is defined by the INA for purposes of the H-1B program. So while the OES survey is the best available source of wage data for the Department's purposes, it is not a perfect tool for providing wages in the H-1B, H-1B1, E-3, and PERM programs—a fact that the Department must take into consideration in how it uses the OES data.

The Department also acknowledged in the IFR that the universe of workers surveyed by the OES for some of the most common occupational classifications in which H-1B workers are employed is larger than the pool of workers who can be said to have levels of education and experience comparable to those of even the least skilled H-1B workers performing work in a specialty occupation. Commenters are therefore correct that BLS's occupational classifications are not delineated with the H-1B and PERM programs in mind. But, as explained in the IFR, the Department took steps to account for this potential mismatch. In particular, because the statutory scheme requires the Department to set the prevailing wage levels based on what workers similarly employed to foreign workers make, taking into account workers' qualifications and, as noted, the large majority of foreign workers are H-1B workers, the Department determined it would be inappropriate to consider the wages of the least educated and experienced workers in these common H-1B occupational classifications in setting the prevailing wage levels.

To address the fact that the OES survey does not itself contain information about experience and education, the Department sought to determine the wages typically earned by individuals having comparable levels of education, experience, and responsibility to the prototypical entry-level H-1B and EB-2 workers working in the most common H-1B and PERM occupations by looking to other credible government surveys that do gather such information and comparing their data to the OES data. In particular, the Department consulted a variety of data sources, most importantly wage data on individuals with master's degrees or higher and limited years of work experience from the 2016, 2017, and 2018 CPS conducted by the U.S. Census Bureau, and data on the salaries of recent graduates of master's degree programs in STEM occupations garnered from surveys conducted by the NSF in 2015 and 2017. Both of these surveys represent the highest standards of data collection and analysis performed by the federal government. Both surveys have large sample sizes that have been methodically collected and are consistently used not just across the federal government for purposes of analysis and policymaking, but by academia and the broader public as well. Comparing their data to OES wage distributions thus allowed the Department to take into account education and experience in determining how to use OES data. Further, though the CPS and NSF surveys provide a good approximation of where U.S. workers with similar skills to entry-level H-1B and EB-2 workers, fall within the OES distribution; they are not conducted on a regular basis with enough granularity as the OES survey to produce wage estimates at the occupational and geographic levels, nor are the produced frequently enough to provide the up to date wage data necessary to ensure accurate prevailing wages. They thus are useful for assessing how the OES data should be used in the Department's foreign labor programs, but could not be used as a substitute for the OES, which, as noted above, has unique attributes that make it, in the Department's judgment, the best possible source of wage data even though it does not survey for education and experience. The Department is therefore confident that its use of the OES continues to be appropriate in the H-1B and PERM programs, and that the IFR's methodology properly accounted for the fact that the OES does not survey for education and experience.

As noted, some commenters suggested that the BLS OES survey is flawed because it is a voluntary survey and some smaller or more rural employers are less likely to respond to the survey, which in turn means, according to commenters, that such employers will be given inappropriately high wages because they will be grouped in with establishments in metropolitan statistical areas with higher labor costs due to a lack of survey responses. The Department recognizes that the BLS OES survey is voluntary. However, BLS sends the OES survey to over 1 million establishments and those establishments are encouraged to respond to the survey. The survey is recognized as a statistically valid, comprehensive source of wages nationwide. As the Department has discussed, the OES survey is not the perfect tool for setting wages in the foreign labor certification programs, but it is the largest and best single source of wage data available for setting wages across hundreds of occupational classifications in hundreds of geographical areas. The Department endeavors to produce as many statistically valid wage estimates as possible and therefore will move to the next geographic area until it can report a statistically valid wage. While it may be the case that in some instances wage rates provided for areas of the country with fewer establishments responding to the survey will result in those areas being grouped in with adjacent regions, the Department believes, as elaborated on previously, that the value in having a single, uniform survey that produces consistent and reliable results for its foreign labor programs outweighs any benefits that might result from using different sources of wage data for specific areas of employment. Moreover, the fact that the Department permits employers to use alternative sources of wage data to set prevailing wage rates gives employers some recourse if they believe, in certain instances, that the OES prevailing wage rate is not accurate.

Some commenters suggested that the Department should use a separate survey for certain occupations, such as physicians, because there are better surveys for those specific occupations. The Department declines to make this change. As explained throughout, the Department has determined that the OES survey is the largest and best available survey to rely upon for setting wages in the foreign labor certification programs. The Department understands the shortfalls that a survey the size of the OES survey has, and, as discussed above, has taken various steps to account for the fact that the OES survey is not specifically designed for use in the Department's foreign labor programs. For administrative uniformity the Department believes that providing one set of data, from a government conducted survey, has more benefits than using on potentially less reliable surveys conducted by private organizations that could be discontinued or have changes to their methodology made without the Department's input. Further, as noted previously, employers already have a method for utilizing a survey other than the BLS OES survey. If employers believe there are better surveys for their occupations than the BLS OES survey, they may rely upon those surveys, either through the Prevailing Wage Determination process or listing a valid wage survey as the source of the prevailing wage when submitting an LCA in the FLAG system. Indeed, the Department notes that the AAMC survey itself is often used by employers as the source of the prevailing wage on their LCAs and PWD applications.

6. The Upper and Intermediate Wage Levels

Summary of Comments

Several commenters expressed concern that use of the mean of the top decile of the OES distribution to approximate the prevailing wage for Level IV workers produces a Level IV wage above the 95th percentile due to outlier wages at the top of the distribution and that this, in turn, skews the intermediate wage levels because they are “set by statute by interpolating the data for levels” I and IV. Some commenters cited a Cato Institute finding that “extreme outliers” in the data used to determine the level IV wage resulted, in some cases, in Level II and III wage determinations “up to 26 percent higher than predicted in” the IFR. A university commenter and an anonymous commenter stated that this methodology resulted in situations where the Level II wage increases to the 78th percentile and the Level III wage increases to the 90th percentile. An employer stated that the IFR methodology would produce clearly inaccurate prevailing wages in industries with bi-modal salary distributions. An individual commenter stated that the 95th percentile represents workers “nearing the end of their career, with decades of experience.”

Similarly, a few commenters expressed concern about specific errors or discrepancies in prevailing wages produced by the IFR at the intermediate levels. An individual commenter asserted that of “437,593 Area Code-SOC Code combinations” there are prevailing wage “discrepancies in 228,836.” As an example, the commenter noted that the Level II wage for SOC 15-2031 in “[a]rea code 37980” based on what the Department estimated would be at the 62nd percentile is higher than the pre-IFR Level IV wage, which the Department estimated to be at the 67th percentile. Similarly, a trade association stated that its members reported that the Level II 62nd percentile wage is higher in many cases than the pre-IFR Level IV 67th percentile wage. In these cases, commenters noted that the wage increases effected by the IFR appeared to be even greater than the Department anticipated or intended. By contrast, two commenters asserted that prevailing wages published in the Department's Online Wage Library clearly were too low in some cases, citing examples like a level I wage of $22,000 for Electrical Engineers in College Station, Texas, much lower than entry-level wages indicated in a NSF survey.

Response to Comments

To begin, the Department agrees with commenters that setting the top wage at the mean of the upper decile skews the wages of the intermediate wage levels by including, sometimes extreme, outliers. For the reasons given below, the Department continues to believe that the Level IV wage should be placed at the uppermost end of the OES distribution. However, to avoid the statistical issues that resulted in overly inflated wages at both the upper and intermediate wage levels under the IFR, the Department has adjusted the manner in which BLS will provide data for the Level IV wage.

As the Department explained in the IFR, the highest wage level should be commensurate with the wages paid to the most highly compensated workers in any given occupation because such workers are also generally the workers with the most advanced skills and competence in the occupation, and therefore the type of workers who are similarly employed to the most highly qualified H-1B and PERM workers. Again, it is generally the case that, as a worker's education and experience increase, so too do his wages. Further, while the INA places baseline, minimum skills-based qualifications on who can obtain an H-1B or EB-2 visa, it does not place any limit on how highly skilled a worker can be within these programs. Thus, while the Department necessarily discounted the lower end of the OES wage distribution in determining the entry-level wage, full consideration must be given to the uppermost portion of the distribution in adjusting the Level IV wage.

H-1B workers can be, and at least in some cases already are among the most highly paid, and therefore likely among the most highly skilled workers within their respective occupations. This is demonstrated by a review of the highest salaries paid to H-1B workers in the most common occupations in which H-1B workers are employed. In Fiscal Year (FY) 2019, for example, the most highly compensated H-1B nonimmigrants employed as Computer Systems Analysts commanded annual wages as high as $450,000. That figure was $357,006 for H-1B workers in other Computer Occupations. The wages of workers at the 90th percentile of the OES distribution for these occupations, by contrast, are significantly lower. Computer Systems Analysts at the 90th percentile in the OES distribution make approximately $142,220. That figure is $144,820 for workers in other computer occupations. In other words, H-1B workers in some instances make wages far in excess of those earned by 90 percent of all U.S. workers in the same occupation. Indeed, a review of the wages of the top five percent highest earners among H-1B nonimmigrants, and therefore the earners likely to have the highest levels of education, experience, and responsibility, in the 16 occupational classifications that account for one percent or more of all approved H-1B petitions in FY2019 shows that such workers make wages that are, on average, at least 20 percent higher than those made by workers at the 90th percentile in the OES wage distribution.

Further demonstrating that H-1B workers can be and sometimes are among the most skilled and competent workers in their occupations, an examination of the top end of the wage distribution within the H-1B program shows that, for H-1B nonimmigrants with graduate and bachelor's degrees, the association between education and income level begins to break down to some extent. Among the most highly compensated H-1B workers, the higher the income level, the more likely the foreign worker beneficiary only has a bachelor's degree. This strongly suggests that individuals at the fourth wage level truly possess the most advanced skills and competence—the only remaining parameters that can reasonably account for significant wage differentials—within their occupations, as additional years of education are largely irrelevant in explaining wages among top earners. The U.S. workers who are similarly employed to the most highly qualified H-1B workers are, therefore, also likely to be among the most highly skilled, and, therefore, the most highly compensated workers within the OES wage distribution.

The high levels of pay that the most skilled H-1B workers can command is also shown by the fact that, due to their advanced skills, diversified knowledge, and competence, workers placed at the fourth wage level are likely to be far more productive than their less experienced and educated peers. Whereas experience itself generally increases on a linear basis, as a function of age and time spent in an occupation, productivity and an individual's supervisory responsibilities, as a function of experience and skills, do not. For example, the nature of senior management or supervisory roles, in particular, means workers who serve as productivity multipliers are more likely to fill such positions, which in turn translates to higher wages. Perhaps even more relevant to the Department's assessment of the wages paid to H-1B workers is the nature of the work these individuals do, which is highly specialized and typically occurs in computer or engineering-related fields. In such occupations, experience and abilities can result in exponentially divergent levels of productivity, which in turn means that workers with the most advanced skills and competence can command wages far above what other workers in those occupations do.

All of these considerations strongly indicate that U.S. workers similarly employed to the H-1B and PERM workers with the most advanced skills and competence are themselves among the most highly skilled workers in any given occupation, and therefore the most highly compensated. Thus, because the INA requires wages for H-1B and PERM workers to be set based on the wages paid to similarly employed U.S. workers, taking into account education, experience, and responsibility, and the Level IV wage is used for job opportunities filled by the most highly skilled workers, the Level IV wage should, in the Department's judgment be placed at the uppermost end of the OES distribution.

Importantly, commenters by and large did not dispute the Department's conclusion that H-1B workers in some cases are among the most skilled and educated workers in an occupation, and therefore should be compensated at rates that reflect what the most skilled and educated U.S. workers in those occupations make. Rather, as noted, commenters' primary concern was with the statistical methodology the Department used to calculate the Level IV wage. Because the Department agrees with commenters that the methodology contained certain unforeseen flaws, it has decided to take a new approach in the final rule that, while still resulting in wage rates that reflect what some of the most highly skilled, and therefore the most highly compensated individuals in a given occupation, make will eliminate the influence of outliers on prevailing wage rates that result in anomalous and overly inflated rates at both the upper and intermediate wage levels. In consequence, the Department has determined that the Level IV should be calculated as the 90th percentile of the OES distribution, as opposed to the mean of the upper decile used in the IFR. This change will reduce significantly, if not eliminate, the influence of outliers on wage rates because outlier data at the very upper end of the distribution will no longer be a significant factor in how the Level IV wage is calculated.

In particular, as commenters noted, the extremely high wages paid to a few “superstar” outliers in an occupation in a geographic area may raise the mean of the upper decile of workers in that occupation and geographic area far above the median of the upper decile, which is the 95th percentile. Thus, using the mean of the upper decile to calculate Level IV wages and derive Level II and III wages may boost Level II, III, and IV wages higher than the Department anticipated or intended in the IFR. Changing to the 90th percentile to calculate the Level IV wages and derive Level II and III wages means the Level IV wages will more accurately reflect the wages paid to workers with levels of education, experience, and responsibility comparable to the typical U.S. worker at the high end of the distribution, rather than workers with abnormally high levels of compensation even for that part of the distribution. For example, a “superstar” senior software designer (OES code 15-1256) that makes over $750,000 per year working in San Jose, California in 2019 would affect the mean of the top decile, but would not affect the 90th percentile wage figure of software engineers in San Jose, California, which was $207,200 in 2019, according to OES statistics. Thus, using the mean of the top decile to calculate Level IV wages and derive Levels II and III wages allows the presence of a few “superstar” outliers in an occupation in a geographic area to inflate Level II, III, and IV wages for an occupation in a geographic area.

In addition, there are other considerations weighing against using the mean of the upper decile to calculate Level IV wages and derive Levels II and III wages. The extremely high wages that employers pay to “superstar” outliers in an occupation in a geographic area of course do not necessarily mean that employers also pay high wages to other workers in the same occupation in the same geographic area. Thus, using the mean of the top decile to calculate Level IV wages and derive Levels II and III wages not only inflate Level II, III and IV wages so that they do not accurately reflect the overall wage distribution for an occupation in a geographic area, but also introduces the potential for significant unpredictability in wages from year to year that is not based on any systemic change to the labor market. Consider the same “superstar” senior software designer that makes over $750,000 per year working in San Jose, California in 2019 and suppose his employer agreed to let him work remotely in 2020, and he moved to Salt Lake City, Utah. That decision would affect the mean of the top decile, reducing it in San Jose and increasing it in Salt Lake City, but would not affect the respective 90th percentiles of $207,200 in San Jose and $157,290 in Salt Lake City. Changing the work location for one “superstar” outlier would not affect the distribution of wages for 80 percent of software developers earning between the 10th and 90th percentiles in either San Jose or Salt Lake City. Software developers would still make more on average at the every level in San Jose than in Salt Lake City. Moreover, because the OES survey does not necessarily capture the same workers year-over-year, the unpredictability in wages that can result from the presence and then absence of an outlier in the wage data can occur even if that same worker has not changed locations. The weakening of the linkage between supply and demand factors affecting wages for most workers in an occupation and the Level II, III, and IV wages was not the Department's intention in the IFR, and is not consistent with the INA's wage provisions. Using the 90th percentile instead to calculate the Level IV wages and derive Level II and III wages for an occupation in a geographic area eliminates the distortions and minimizes the excessive and unintended variability in Levels II, III, and IV wages arising from the inclusion of a few “superstar” outliers in the mean of top decile.

Finally, the Department has decided to use a percentile calculation instead of a mean calculation because the Department can produce such data more efficiently. In addition, experience with the IFR's methodology has demonstrated that taking the mean of a small portion of the OES distribution, such as of a decile, can in some cases result in exceedingly small sample sizes being used to produce the wage figure, which make the figure produced potentially less reliable.

Based on its review of the comments received, the Department also believes that a percentile calculation will be easier for employers, workers, and the public to understand than a mean calculation. As noted above, some commenters challenged the wage figures provided under the IFR as being incorrect because some wages the Department estimated as falling at the 62nd percentile wage were significantly higher than what the Department had described as the 67th percentile wage under the old methodology. While, for the reasons given above, it is likely that this occurred in some cases due to the presence of outliers in the data used to calculate the Level IV wage, there is also another explanation. Specifically, describing the wage figures produced under the old methodology and the IFR's methodology as percentiles was, as explained in the IFR, simply a shorthand way of describing a rough approximation of what a mean calculation yields. For example, under the old methodology, the Level IV wage was provided as the mean of the upper two-thirds of the OES distribution, meaning the average of the wage data falling between the 33rd and 100th percentiles. The midpoint of that portion of the distribution is the 67th percentile, but its mean will not necessarily be the 67th percentile. Put more simply, the average of a set of numbers does not always fall at the median of those numbers. As a result, discussing two different means calculated based on different portions of the distributions by describing them as percentiles gives a false sense of comparability, as demonstrated by some of the discrepancies raised by commenters.

To avoid confusion about how it describes the wages it provides going forward, the Department will speak more clearly about the kinds of data it is providing and will consequently report the wage based on a percentile calculation. This means that the Department will no longer take the average of portion of the wage distribution, but instead will provide a wage that falls at a particular predetermined point within the distribution.

As to the precise values of the intermediate levels, the Department notes that it will continue to calculate the two intermediate wage levels in accordance with 8 U.S.C. 1182(p)(4), which provides that, in establishing a four-tier wage structure, “[w]here an existing government survey has only 2 levels, 2 intermediate levels may be created by dividing by 3, the difference between the 2 levels offered, adding the quotient thus obtained to the first level and subtracting that quotient from the second level.” The BLS OES survey is, as provided in the statute, an existing survey that has long provided two wage levels for Department's use in setting the prevailing wage rates.

The Department will apply the statutory formula as follows: the difference between the two levels provided by the OES survey data is 55 percentiles. Dividing this by three yields a quotient of 18.33. This quotient, added to the value of the Level I wage at the 35th percentile, yields a Level II wage at approximately the 53rd percentile. When subtracted from the value of the Level IV wage at the 90th percentile, the quotient yields a Level III wage at approximately the 72nd percentile of the OES distribution.

Finally, while eliminating the influence of outliers on how the upper level wage is calculated and moving to percentile calculations will reduce unpredictability in the data, prevent the inflation of wages beyond the levels the Department has determined appropriate, and make the wage structure easier to understand for the public, it is possible that there will continue to be anomalies as the Department moves from a mean-based to a percentile-based methodology. However, the Department does not expect these will be common.

7. Other Suggested Alternatives and Additional Comments

One public policy organization suggested the Department should require use of a government survey to determine prevailing wages, stating the INA does not require the Department to permit use of other sources and expressing concern that employers “have routinely relied on LCA prevailing wage sources that do not fit the `independent authoritative source' or `another legitimate source of wage information.”

The Department believes that allowing employers the flexibility of choosing to use an independent authoritative source or another legitimate source of wage data provides a backstop for cases in which OES data on an occupation in a given region is insufficient or the OES data provides an anomalous result. This flexibility serves the goal of ensuring that the wage requirement actually reflects the market wage for the job.

Another public policy organization stated it is unclear how independent authoritative and other non-OES sources “compare to OFLC-generated OES prevailing wage” and urged the Department to conduct a study comparing OES-based wages and wages produced by private surveys and non-OES sources “to identify whether there are any systematic biases” in non-OES sources.

The quality of independent wage surveys is an important subject to which OFLC pays attention and will continue to pay attention. Although private surveys are conducted independently of the Department, the Department in its regulations and guidance has set standards that private surveys must attain. As discussed above, the regulations restrict independent authoritative sources to publications within 24 months of the application and require them to use recent and valid data. Independent sources must be “reasonable and consistent with recognized standards and principals in producing a prevailing wage.”

Guidance that the Department issued in 2009 requires that wage data collected by an independent authoritative source is for similarly employed workers, meaning workers having substantially similar levels of skills. The survey should contain a representative sample of wages within the occupation that comports with recognized statistical standards and principles in producing prevailing wages. The Department provides a set of minimum survey standards in Appendix E of the 2009 Guidance and encourages employers to reference these standards when seeking to use an independent authoritative source as the prevailing wage. Written documentation on the methodology used to conduct the survey and the validity of the methodology used in computing the occupational wage data covering the area of intended employment must be kept in the employer's data file and made available in the event of an investigation. Two commenters suggested the Department should combine data collected by the OES survey with “certain data from private, independently published compensation surveys” to produce prevailing wages that would more accurately reflect skill, education, and experience levels than wages determined using OES pay band data alone. One of these commenters suggested BLS could “layer” the private survey data “over the OES data” and asserted this would not be difficult because H-1B workers are heavily concentrated in IT occupations that are included in private surveys, though the commenter acknowledged private surveys are not available for all occupations and localities. Other general suggestions included applying a higher wage to “tech companies” or applying a higher wage as “the number of visas grow for an employer.”

The Department does not believe that combining or layering data from studies that may not be measuring quite the same occupations in the same regions would yield more accurate results. OES data is comprehensive and reliable. As the commenter acknowledged, private survey data is not available for some occupations and localities. An advantage of the OES survey is that it allows uniformity in the Department's methodology. That advantage would be lost if the Department adopted the commenters' proposal. The system the Department has adopted allows for cases where private survey data may be more accurate. As discussed, using other authoritative or legitimate sources is an option available to employers.

Various commenters asserted increased wages under the IFR methodology would have negative macroeconomic impacts, including: Brain drain and loss of American competitiveness in a global economy, stifling innovation in areas like artificial intelligence and manufacturing 4.0; increased prices for or elimination of products and services; elimination or increased outsourcing of jobs and a general reduction in labor demand; and reduced revenues, including local, State, and Federal tax revenue and reduced consumer spending from foreign workers and students. Many commenters also expressed concern that the higher IFR wages would result in increased outsourcing of jobs, rather than increased opportunities for U.S. workers. One of these commenters noted that U.S. employers can hire workers through foreign affiliates and cited a Wharton School of Business study finding H-1B restrictions “caused foreign affiliate employment increases at the intensive and extensive margins.”

The Department does not anticipate that the harms the commenters envisage will be the consequences of more accurately calculating prevailing wages of H-1B and PERM workers. Some of the consequences are possible, but in setting wage requirements, Congress accepted that there would be costs resulting from its chosen means of protecting U.S. workers. The Department has not been assigned the function of reconsidering Congress's decision. Rather, the Department's obligation under the INA is to match as closely as possible workers' pay with their occupations and qualifications.

Two public policy organizations believed the Department must address employer misclassification of job opportunities by reviewing “the qualifications of individual workers before DHS petitions are approved to ensure that wage levels match up with age, education, and experience” to ensure the employer is paying an accurate prevailing wage. One of these commenters asserted some employer petitions contain the same prevailing wage for different job opportunities, such as listing the same wage for a software engineer and a senior software engineer.

These comments propose actions that may be undertaken by DHS but not by the Department. The Department cannot review DHS petitions before DHS approves them.

Some commenters suggested new definitions of the terms `employer' and `employment,' enhanced regulation of foreign labor recruiters, a ban of staffing companies from the H-1B program, and enhanced wage protections in the H-2A program. Other commenters expressed concerns related to DHS regulations and recent rulemaking either unrelated or not directly related to this rulemaking, including a DHS IFR regarding specialty occupation determinations.

These comments express concerns or provide suggestions that exceed the scope of this rulemaking. Accordingly, they need not be addressed in this preamble.

IV. Amendments to the Computation of Prevailing Wage Levels Created by the Final Rule

In light of the foregoing, this final rule amends the Department's regulations at part 20, sections 656.40 and 655.731 to reflect the wage level computations the Department will use to determine prevailing wages in the H-1B, H-1B1, E-3, EB-2, and EB-3 classifications. These amendments are in accordance with the President's Executive Order (E.O.) 13788, “Buy American and Hire American,” which instructed the Department to “propose new rules and issue new guidance, to supersede or revise previous rules and guidance if appropriate, to protect the interests of United States workers in the administration of our immigration system.” Additionally, the Department has determined that the existing prevailing wage levels were artificially low and provided an opportunity for employers to hire and retain foreign workers at wages well below what their U.S. counterparts earn, creating an incentive to prefer foreign workers to U.S. workers, an incentive that is at odds with the statutory scheme and causes downward pressure on the wages of the domestic workforce. Therefore, the amendments discussed below revising the wage provisions at 20 CFR 655.731 and 656.40 will ensure the prevailing wage levels reflect the wages paid to U.S. workers with similar experience, education, and responsibility to those possessed by similarly employed foreign workers.

1. Prevailing Wage Levels Based on the OES in the Permanent Labor Certification Program (20 CFR 656.40)

The IFR amended this section to codify the practice of using four prevailing wage levels and to specify the manner in which the wages levels are calculated. Additionally, the IFR incorporated minor technical amendments to clarify the prevailing wage process and to codify the Department's practice of having the OFLC Administrator announce, via a notice of implementation, annual updates to OES wage data. After a careful review of the comments and as discussed above, this final rule adopts a revised wage level computation methodology and other clarifying and technical amendments to § 656.40.

Paragraph (b)(2)(ii)(A) describes the computation of the Level I Wage following implementation of transition wage rates specified under paragraph (b)(2)(iii). This first wage level—calculated as the mean of the fifth decile of the OES wage distribution under the IFR—will now be calculated as the 35th percentile of the wage distribution for the most specific occupation and geographic area available. Roughly speaking, this means that the Level I Wage will be adjusted downward from the approximate 45th percentile under the IFR to the exact 35th percentile of the relevant OES wage distribution in this final rule.

Next, paragraph (b)(2)(ii)(D) provides that the Level IV Wage—calculated as the mean of the upper decile of the OES wage distribution—will now be calculated as the exact 90th percentile of the wage distribution for the most specific occupation and geographic area available. This means the Level IV Wage will decrease approximately from the 95th percentile under the IFR to exactly the 90th percentile of the relevant OES wage distribution. Further, where the Department is unable to compute a Level IV Wage for an occupation and geographic area due to wage values exceeding the uppermost interval of the OES wage interval methodology, the Level IV Wage will be the highest of: (1) The current hourly wage rate applicable to the highest OES wage interval for the specific occupation and geographic area (also known as the footnote wage), or (2) the mean of the wages of all workers for the most specific occupation and geographic area available.

For the two intermediate levels, II and III, the Department will continue to rely on the mathematical formula Congress provided in the INA. Thus, new paragraph (b)(2)(ii)(B) states that the Level II Wage shall be determined by first dividing the difference between Levels I and IV by three and then adding the quotient to the computed value for Level I. The Level III Wage is defined in new paragraph (b)(2)(ii)(C) as a level determined by first dividing the difference between Levels I and IV by three and then subtracting the quotient from the computed value for Level IV. This yields second and third wage levels at approximately the 53rd and 72nd percentiles, respectively, under this final rule as compared to the computations under the IFR, which placed Level II Wage at approximately the 62nd percentile and Level III Wage at approximately the 78th percentile.

Section 656.40(b)(2)(ii) in the IFR explained that the OFLC Administrator will publish the prevailing wage rates at least once in each calendar year, on a date to be determined by the Administrator, codifying the Department's current practice of announcing updates to OES wage data via a notice of implementation, rather than publishing multiple prevailing wage rates in the Federal Register. The Department has adopted the language of the provision without change, but has made a minor technical change moving the provision to paragraph (b)(2)(iv) in order to accommodate revisions to the wage level computation provisions in this final rule.

The Department is adopting without change revisions to § 656.40(b)(2) that provide greater precision in the language used by changing the term “DOL” to “BLS” when describing which entity administers the OES survey and eliminate redundancy by deleting the language “except as provided in (b)(3) of this section.” Because the Department is now specifying within the regulation exactly how the prevailing wage levels are calculated, the revised text also removes the existing reference to how the levels are calculated—namely the reference to the “arithmetic mean”—and will instead read: “If the job opportunity is not covered by a CBA, the prevailing wage for labor certification purposes shall be based on the wages of workers similarly employed using the wage component of the OES survey, in accordance with subparagraph (b)(2)(i), unless the employer provides an acceptable survey under paragraphs (b)(3) and (g) of this section or elects to utilize a wage permitted under paragraph (b)(4) of this section.” The Department also is adopting without change the revisions to paragraph (a) that remove an out-of-date reference to the role of the SWAs in the prevailing wage determination process and an unnecessary reference to “arithmetic mean” that is specified in other paragraphs.

2. Amending the Wage Requirement for LCAs in the H-1B, H-1B1, and E-3 Visa Classifications (20 CFR 655.731)

The IFR made minor technical amendments to this section to remove out-of-date references, clarify use of the BLS's OES survey and other permissible wage sources to determine prevailing wages, and specify that these determinations will be made in a manner consistent with the amended section 656.40(b)(2). After a careful review of the comments and as discussed above, this final rule adopts, without change, these clarifying and technical amendments to § 656.731.

This final rule adopts amendments to paragraph (a)(2)(ii)(A) that removes an out-of-date reference to SWAs' role in the prevailing wage determination process to reflect current practice and to provide for operational flexibilities in the future with respect to where PWD requests are processed. Non-agricultural PWD requests are no longer processed by SWAs; since 2010 they have solely been processed by the Department at a National Processing Center (NPC). PWD requests are primarily adjudicated by the NPWC, located in Washington, DC, but through interoperability, they may be processed by any NPC. The regulatory text is amended to reflect current DOL practice and to provide maximum flexibility for DOL to ensure PWDs are issued in a timely manner.

The Department also adopts without change revised language in § 655.731 that more clearly explains the Department will use BLS's OES survey to determine the prevailing wages under this paragraph, as well as an additional sentence that specifies these determinations will be made in a manner consistent with amended § 656.40(b)(2). The revised language in paragraphs (a)(2)(ii), (a)(2)(ii)(A), and (a)(2)(ii)(A)(2) also includes technical and clarifying revisions regarding other permissible wage sources (i.e., applicable wage determinations under the Davis-Bacon Act or McNamara-O'Hara Service Contract Act), as well as other independent authoritative or legitimate sources of wage data in accordance with paragraph (a)(2)(ii)(B) or (C).

This final rule adopts without change language that removed the reference to “arithmetic mean” in paragraph (a)(2)(ii) and now states “. . . the prevailing wage shall be based on the wages of workers similarly employed as determined by the OES survey in accordance with 20 CFR. 656.40(b)(2)(i) . . .” The revisions also correct an error referencing “H-2B nonimmigrant(s)” by changing the reference to “H-1B nonimmigrant(s)” in paragraph (a)(2)(ii)(A)(2). The revisions further provide that an NPC will continue to determine whether a job is covered by a collective bargaining agreement that was negotiated at arms-length, but in the event the occupation is not covered by such agreement, an NPC will determine the wages of workers similarly employed using the wage component of the BLS OES, unless the employer provides an acceptable wage survey. An NPC will determine the prevailing wage in accordance with sections 212(n) and 212(t) of the INA and in a manner consistent with the newly revised 20 CFR 656.40(b)(2).

3. Transition Wage Rates for Implementing Changes Created by the Final Rule

As stated in the IFR, the Department applied the new regulations to applications for prevailing wage determination pending with the NPWC as of the effective date of the regulation; applications for prevailing wage determinations filed with the NPWC on or after the effective date of the regulation; and LCAs filed with the Department on or after the effective date of the regulation where the OES survey data is the prevailing wage source, and where the employer did not obtain the PWD from the NPWC prior to the effective date of the regulation. However, the Department received a number of comments expressing concerns that immediate implementation of the revised wage levels may have a significant negative impact on the economy, and that a phased implementation of the revised wage levels is appropriate to allow employers to adjust to the new computation methodology and plan payroll, budget, and contractual obligations accordingly.

To address these concerns and support an orderly and seamless transition between the rules, the Department is adding paragraph (b)(2)(iii) to this section to provide a phased implementation period to the new prevailing wage levels. A short transition period also allows the Department to implement necessary changes to program operations, OES wage databases, and technology systems, and to provide training and technical assistance to the NPC, employers, and other stakeholders in order to familiarize them with changes required by this final rule. The wage level computations contained in this section will only apply to applications for prevailing wage determination pending with the NPWC on or during the effective date(s) of each transition period; applications for prevailing wage determinations filed with the NPWC on or during the effective date(s) of each transition period; and LCAs filed with the Department on or during the effective date(s) of each transition period where the OES survey data is the prevailing wage source, and where the employer did not obtain the PWD from the NPWC prior to the effective date(s) of each transition period.

Accordingly, paragraph (b)(2)(iii)(A) describes the computations of the wage levels for the period beginning on the effective date of this final rule through June 30, 2021. The Level I Wage will continue to be calculated as the mean of the lower one-third of the wage distribution for the most specific occupation and geographic area available, which roughly approximates the 17th percentile of the wage distribution. The Level IV Wage will continue to be calculated as the mean of the upper two-thirds of the wage distribution for the most specific occupation and geographic area available, which roughly approximates the 67th percentile of the wage distribution. For the two intermediate levels, II and III, the Department will continue to rely on the mathematical formula Congress provided in the INA.

Paragraph (b)(2)(iii)(B) describes the computations of the wage levels for the period beginning on July 1, 2021, through June 30, 2022. The Level I Wage will be set as either (1) 90 percent of the wage value calculated at the 35th percentile of the wage distribution under paragraph (b)(2)(ii)(A), or (2) the mean of the lower one-third of the wage distribution under paragraph (b)(2)(iii)(A)(1), whichever is highest. The Level IV Wage will be set as either (1) 90 percent of the wage value calculated at the 90th percentile of the wage distribution under paragraph (b)(2)(ii)(D), or (2) the mean of the upper two-thirds of the wage distribution under paragraph (b)(2)(iii)(A)(2), whichever is highest. For the two intermediate levels, II and III, the Department will continue to rely on the mathematical formula Congress provided in the INA based on the wage levels derived under this paragraph.

Paragraph (b)(2)(iii)(C) describes transition wage rates that will apply only to LCAs and, as applicable, applications for prevailing wage determinations submitted by employers seeking to employ a H-1B nonimmigrant worker in job opportunity where such H-1B nonimmigrant worker was, as of October 8, 2020, the beneficiary of an approved I-140 Petition or eligible for an extension of his or her H-1B visa status under AC21, and eligible to be granted immigrant status but for application of the per country visa limitations or remains eligible for an extension of his or her H-1B visa status at the time the LCA is filed.

Where these requirements pertaining to job opportunities for which LCAs are filed are met, paragraph (b)(2)(iii)(C)(1) describes the computations of the wage levels for the period beginning on July 1, 2021, through June 30, 2022. The Level I Wage will be set as either (1) 85 percent of the wage value calculated at the 35th percentile of the wage distribution under paragraph (b)(2)(ii)(A), or (2) the mean of the lower one-third of the wage distribution under paragraph (b)(2)(iii)(A)(1), whichever is highest. The Level IV Wage will be set as either (1) 85 percent of the wage value calculated at the 90th percentile of the wage distribution under paragraph (b)(2)(ii)(D), or (2) the mean of the upper two-thirds of the wage distribution under paragraph (b)(2)(iii)(A)(2), whichever is highest. For the two intermediate levels, II and III, the Department will continue to rely on the mathematical formula Congress provided in the INA based on the wage levels derived under this paragraph.

Paragraph (b)(2)(iii)(C)(2) describes the computations of the wage levels for the period beginning on July 1, 2022, through June 30, 2023. The Level I Wage will be set as either (1) 90 percent of the wage value calculated at the 35th percentile of the wage distribution under paragraph (b)(2)(ii)(A), or (2) the wage value provided from the calculation specified under paragraph (b)(2)(iii)(C)(1)(i), whichever is highest. The Level IV Wage will be set as either (1) 90 percent of the wage value calculated at the 90th percentile of the wage distribution under paragraph (b)(2)(ii)(D), or (2) the wage value provided from the calculation specified under paragraph (b)(2)(iii)(C)(1)(ii), whichever is highest. For the two intermediate levels, II and III, the Department will continue to rely on the mathematical formula Congress provided in the INA based on the wage levels derived under this paragraph.

Paragraph (b)(2)(iii)(C)(3) describes the computations of the wage levels for the period beginning on July 1, 2023, through June 30, 2024. The Level I Wage will be set as either (1) 95 percent of the wage value calculated at the 35th percentile of the wage distribution under paragraph (b)(2)(ii)(A), or (2) the wage value provided from the calculation specified under paragraph (b)(2)(iii)(C)(2)(i), whichever is highest. The Level IV Wage will be set as either (1) 95 percent of the wage value calculated at the 90th percentile of the wage distribution under paragraph (b)(2)(ii)(D), or (2) the wage value provided from the calculation specified under paragraph (b)(2)(iii)(C)(2)(ii), whichever is highest. For the two intermediate levels, II and III, the Department will continue to rely on the mathematical formula Congress provided in the INA based on the wage levels derived under this paragraph.

Following this transition period and beginning on July 1, 2024, paragraph (b)(2)(iii)(C)(4) requires that all prevailing wage calculations for job opportunities for which LCAs are filed shall be provided by the OFLC Administrator as specified under paragraph (b)(2)(ii) of this section. Where the Department is unable to compute a Level IV Wage under paragraph (b)(2)(iii) for an occupation and geographic area due to wage values exceeding the uppermost interval of the OES wage interval methodology, paragraph (b)(2)(iii)(D) specifies that the OFLC Administrator shall determine the Level IV Wage as the highest of: (1) The current hourly wage rate applicable to the highest OES wage interval for the specific occupation and geographic area, or (2) the mean of the wages of all workers for the most specific occupation and geographic area available.

V. Statutory and Regulatory Requirements

A. Executive Orders 12866 (Regulatory Planning and Review), Executive Order 13563 (Improving Regulation and Regulatory Review), and Executive Order 13771 (Reducing Regulation and Controlling Regulatory Costs)

Under E.O. 12866, the OMB's Office of Information and Regulatory Affairs (OIRA) determines whether a regulatory action is significant and, therefore, subject to the requirements of the E.O. and review by OMB. 58 FR 51735. Section 3(f) of E.O. 12866 defines a “significant regulatory action” as an action that is likely to result in a rule that: (1) Has an annual effect on the economy of $100 million or more, or adversely affects in a material way a sector of the economy, productivity, competition, jobs, the environment, public health or safety, or State, local, or tribal governments or communities (also referred to as economically significant); (2) creates serious inconsistency or otherwise interferes with an action taken or planned by another agency; (3) materially alters the budgetary impacts of entitlement grants, user fees, or loan programs, or the rights and obligations of recipients thereof; or (4) raises novel legal or policy issues arising out of legal mandates, the President's priorities, or the principles set forth in the E.O. Id. Pursuant to E.O. 12866, OIRA has determined that this is an economically significant regulatory action. However, OIRA has waived review of this regulation under E.O. 12866, section 6(a)(3)(A). Pursuant to the Congressional Review Act (5 U.S.C. 801 et seq.), OIRA has designated that this rule is a “major rule,” as defined by 5 U.S.C. 804(2).

E.O. 13563 directs agencies to propose or adopt a regulation only upon a reasoned determination that its benefits justify its costs; the regulation is tailored to impose the least burden on society, consistent with achieving the regulatory objectives; and in choosing among alternative regulatory approaches, the agency has selected those approaches that maximize net benefits. E.O. 13563 recognizes that some benefits are difficult to quantify and provides that, where appropriate and permitted by law, agencies may consider and qualitatively discuss values that are difficult or impossible to quantify, including equity, human dignity, fairness, and distributive impacts.

Outline of the Analysis

Section III.B.1 describes the need for the final rule, and section III.B.2 describes the process used to estimate the costs of the rule and the general inputs used to reach these estimates, such as wages and number of affected entities. Section III.B.3 explains how the provisions of the final rule will result in costs and transfer payments and presents the calculations the Department used to reach the cost and transfer payment estimates. In addition, this section describes the qualitative transfer payments and benefits of the changes contained in this final rule. Section III.B.4 summarizes the estimated first-year and 10-year total and annualized costs, perpetuated costs, and transfer payments of the final rule. Finally, section III.B.5 describes the regulatory alternatives that were considered during the development of the final rule.

Summary of the Analysis

The Department expects that the final rule will result in costs and transfer payments. As shown in Exhibit 1, the final rule will have an annualized cost of $2.90 million and a total 10-year cost of $20.34 million at a discount rate of 7 percent in 2019 dollars. The final rule will result in annualized transfer payments of $14.97 billion and total 10-year transfer payments of $105.16 billion at a discount rate of 7 percent in 2019 dollars. When the Department uses a perpetual time horizon to allow for cost comparisons under E.O. 13771, the annualized cost of this final rule is $1.86 million at a discount rate of 7 percent in 2016 dollars.

The total cost associated with the final rule includes only rule familiarization. The rule is not expected to result in any cost savings. Transfer payments are the result of changes to the computation of prevailing wage rates for employment opportunities that U.S. employers seek to fill with foreign workers on a temporary basis through H-1B, H-1B1, and E-3 nonimmigrant visas. See the costs and transfer payments subsections of section III.B.3 (Subject-by-Subject Analysis) below for a detailed explanation.

The Department was unable to quantify some transfer payments and benefits of the final rule. The Department describes them qualitatively in section III.B.3 (Subject-by-Subject Analysis).

1. Need for Regulation

The Department has determined that this rulemaking is needed to update the computation of prevailing wage levels under the existing four-tier wage structure to better reflect the actual wages earned by U.S. workers similarly employed to foreign workers, eliminate economic incentive or advantage in hiring foreign workers on a permanent or temporary basis in the United States, and further the goals of E.O. 13788, Buy American and Hire American. See 82 FR 18837. The “Hire American” directive of the E.O. articulates the executive branch policy to rigorously enforce and administer the laws governing entry of nonimmigrant workers into the United States in order to create higher wages and employment rates for U.S. workers and to protect their economic interests. Id. sec. 2(b). It directs Federal agencies, including the Department, to propose new rules and issue new guidance to prevent fraud and abuse in nonimmigrant visa programs, thereby protecting U.S. workers. Id. sec. 5.

The Department is therefore amending its regulations at Sections 656.40 and 655.731 to update the methodology it will use to determine prevailing wages using wage data from the BLS OES survey for job opportunities in the H-1B, H-1B1, E-3, and permanent labor certification programs. The reports discussed and analyses provided in the preamble above explain how application of the current wage methodology for the four-tier OES wage structure fails to produce prevailing wages at a level consistent with the actual wages earned by U.S. workers similarly employed to foreign workers and, therefore, has a suppressive effect on the wages of U.S. workers similarly employed. The Department has a statutory mandate to protect the wages and working conditions of U.S. workers similarly employed from adverse effects caused by the employment of foreign workers in the United States on a permanent or temporary basis.

2. Analysis Considerations

The Department estimated the costs and transfer payments of the final rule relative to the baseline (the regulations governing permanent labor certifications at 20 CFR part 656 and labor condition applications at 20 CFR part 655, subpart H).

In accordance with the regulatory analysis guidance articulated in OMB's Circular A-4 and consistent with the Department's practices in previous rulemakings, this regulatory analysis focuses on the likely consequences of the final rule (i.e., costs and transfer payments that accrue to entities affected). The analysis covers 10 years (from 2021 through 2030) to ensure it captures major costs and transfer payments that accrue over time. The Department expresses all quantifiable impacts in 2019 dollars and uses discount rates of 3 and 7 percent, pursuant to Circular A-4.

Exhibit 2 presents the number of entities affected by the final rule. The number of affected entities is calculated using OFLC performance data from fiscal years (FY) 2018, 2019, and 2020. The Department uses them throughout this analysis to estimate the costs and transfer payments of the final rule.

Estimated Number of Workers and Change in Hours

The Department presents the estimated average number of foreign worker applicants and the change in burden hours required for rule familiarization in section III.B.3 (Subject-by-Subject Analysis).

Compensation Rates

In section III.B.3 (Subject-by-Subject Analysis), the Department presents the costs, including labor, associated with implementation of the provisions contained in this final rule. Exhibit 3 presents the hourly compensation rates for the occupational categories expected to experience a change in the number of hours necessary to comply with the final rule. The Department used the BLS mean hourly wage rate for private sector human resources specialists. Wage rates were adjusted to reflect total compensation, which includes non-wage factors such as overhead and fringe benefits (e.g., health and retirement benefits). We used an overhead rate of 17 percent and a fringe benefits rate based on the ratio of average total compensation to average wages and salaries in 2019. For the private sector employees, we used a fringe benefits rate of 42 percent.

The Department received one comment on the adjustment of wage rates to reflect total compensation. One commenter said the Department had underestimated the cost of the program because fringe and overhead were included in calculations of costs and transfers. In response to the commenter's concern, the wage transfer calculations in the IFR and the final rule do not include overhead or fringe benefits; they are raw wages. Overhead and fringe benefits were only applied to staffing wages in the cost section. The commenter's calculation of fringe and overhead application was incorrect when suggesting how they were applied. The 17 percent overhead rate is not applied after calculating the fringe rate; instead, the fringe rate and the overhead rates are applied simultaneously to wages as shown in Exhibit 3.

The fringe wage rate is based on Employer Costs for Employee Compensation data which includes paid leave; supplemental pay (i.e., overtime and premium, shift differentials, and nonproduction bonuses); insurance (i.e., life, health, short-term disability, and long-term disability); retirement and savings; and legally required benefits (i.e., Social Security, Medicare, federal unemployment insurance, state unemployment insurance, and workers' compensation). As wages increase the costs associated with paid leave, retirement savings, and supplemental pay will also increase.

The Department used the hourly compensation rates presented in Exhibit 3 to estimate the labor costs.

3. Subject-by-Subject Analysis

The Department's analysis below covers the estimated costs and transfer payments of the final rule. In accordance with Circular A-4, the Department considers transfer payments as payments from one group to another that do not affect total resources available to society. The regulatory impact analysis focuses on the costs and transfer payments that can be attributed exclusively to the new requirements in the final rule.

Costs

The following section describes the costs of the final rule.

Rule Familiarization

When the final rule takes effect, existing employers of foreign workers with H-1B, H-1B1, E-3 visas, and those employers sponsoring foreign workers for permanent employment, will need to familiarize themselves with the new regulations. Consequently, this imposes a one-time cost for existing employers in the temporary and permanent visa programs in the first year. Each year, there are new employers that participate in the temporary and permanent visa programs. Therefore, in each year subsequent to the first year, new employers will need to familiarize themselves with the new regulations.

To estimate the first-year cost of rule familiarization, the Department calculated the average (83,312) number of unique employers requesting H-1B certifications and PERM certifications. The average number of unique H-1B and PERM employers (83,312) was multiplied by the estimated amount of time required to review the rule (1 hour). This number was then multiplied by the hourly, fully loaded compensation rate of Human Resources Specialists ($51.93 per hour). This calculation results in an initial cost of $4.33 million in the first year after the final rule takes effect. Each year after the first year the same calculation is done for the average number of new unique employers requesting H-1B and PERM certifications in FY 2019 and FY 2020 (47,339). This calculation results in a continuing annual undiscounted cost of $2.46 million in years 2-10 of the analysis. The one-time and continuing cost yields a total average annual undiscounted cost of $2.65 million. The annualized cost over the 10-year period is $2.75 million and $2.90 million at discount rates of 3 and 7 percent, respectively.

Transfer Payments

Quantifiable Transfer Payments

This section discusses the quantifiable transfer payments related to changes to the computation of the prevailing wage levels.

As discussed in the preamble, the Department determined that current wage level methodology results in prevailing wage rates for temporary and permanent workers that are far below what their U.S. counterparts are likely paid, which has a suppressive effect on the wages of similarly employed U.S. workers. While allowing employers to access higher-skilled H-1B workers to fill specialized positions can help U.S. workers' job opportunities in some instances, the benefits of this policy diminish or disappear when the prevailing wage levels do not accurately reflect the wages paid to similarly employed workers in the U.S. labor market. The distortions resulting from a poor calculation of the prevailing wage allow some firms to replace qualified U.S. workers with lower-cost foreign workers.

Under this final rule, the Department will compute the Level I Wage for PERM labor certifications and LCAs as the 35th percentile of the OES wage distribution for the most specific occupation and geographic area available, rather than the mean of the fifth decile used in the IFR. Roughly speaking, this means that the first wage level will be decreased from the 45th percentile to the 35th percentile. The Department will compute the Level IV Wage as the 90th percentile of the OES wage distribution for the most specific occupation and geographic area available, rather than the arithmetic mean of the upper decile used in the IFR. This means the fourth wage level will decrease approximately from the 95th percentile to the 90th percentile.

Consistent with the formula provided in the INA, the Level II Wage will be calculated by dividing by three, the difference between Levels I and IV, and adding the quotient to the computed value for Level I. The Level III Wage will be calculated by dividing by three the difference between Levels I and IV, and subtracting the quotient from the computed value for Level IV. This yields a Level II Wage at approximately the 53rd percentile and a Level III Wage at approximately the 72nd percentile, as compared to the current computation, which places Level II at approximately the 34th percentile and Level III at approximately the 50th percentile.

This final rule also provides for a transition period from the current wage methodology to the wage methodology contained in this final rule to give foreign workers and their employers time to adapt to the new wage rates. For most job opportunities, the transition will occur in two steps, following a short delayed implementation period, and conclude on July 1, 2022. For job opportunities that will be filled by workers who are the beneficiary of an approved Immigrant Petition for Alien Worker, or successor form, or is eligible for an extension of his or her H-1B status under sections 106(a) and (b) of the American Competitiveness in the Twenty-first Century Act of 2000 (AC21), Public Law 106-313, as amended by the 21st Century Department of Justice Appropriations Authorization Act, Public Law 107-273 (2002), the transition will occur in four steps, following a short delayed implementation period, and conclude on July 1, 2024.

For the two-step transition the current wage levels will be in effect from January 1, 2021 through June 30, 2021. From July 1, 2021 through June 30, 2022, the prevailing wage will be 90 percent of the final wage level. From July 1, 2022 and onward the prevailing wage will be the final wage levels. For the three and a half year transition the current wage levels will be in effect from January 1, 2021 through June 30, 2021. From July 1, 2021 through June 30, 2022 the prevailing wage will be 85 percent of the final wage levels; from July 1, 2022 through June 30, 2023 the prevailing wage will be 90 percent of the final wage levels; from July 1, 2023 through June 30 2024 the prevailing wage will be 95 percent of the final wage levels; and from July 1, 2024 onwards the prevailing wage will be the final wage levels.

Finally, the Department is revising § 655.731 to explain that it will use the BLS's OES survey wage data to establish the prevailing wages in the H-1B, H-1B1, and E-3 visa classifications. The Department added a sentence to explain that these determinations will be made by the OFLC NPC in a manner consistent with § 656.40(b)(2).

The Department calculated the impact on wages that will occur from implementation of the prevailing wage computation changes contained in the final rule. It is expected that the increase in prevailing wages under the final rule will incentivize some employers to employ U.S. workers instead of foreign workers from the H-1B program, but nonetheless, the Department still expects that the same number of H-1B visas will be granted under the annual caps. For many years, the Department has observed that the number of petitions exceeds the numerical cap, as the annual H-1B cap was reached within the first five business days each year from FY 2014 through FY 2020, and higher prevailing wage levels do not necessarily mean that demand for temporary foreign labor will fall below the available supply of visas. Under existing prevailing wage levels, which the Department has shown are too low and do not accurately reflect the wages paid to similarly employed U.S. workers, demand for temporary foreign labor far exceeds the statutory limits on supply. Usually prices rise in a market when demand exceeds supply. However, given the statutory framework of the H-1B system, along with the lower wages for comparable work in many other countries and the non-pecuniary benefits of participating the H-1B program, prices for temporary foreign labor under the H-1B program have stayed too low to depress overall employer demand.

Under the final rule, wage transfers will still occur in cases where U.S. workers are employed instead of H-1B workers; therefore, no adjustments to the wage estimates are necessary due to this effect. However, it is possible that prevailing wage increases will induce some employers to train and provide more working hours to incumbent workers, resulting in no increase in employment but an increase in earnings. It is also possible that prevailing wage increases will induce some employers to not hire a worker at all (either a U.S. worker or a worker from the H-1B program that is subject to the annual cap or not subject to the annual cap), resulting in a decrease in employment of guest workers. However, given that participation in temporary labor certification programs is voluntary, and there exists an alternative labor market of U.S. workers who are not being prevented from accepting work offered at potentially lower market-based wages, there is some reason to doubt whether an increase in prevailing wages will lead to an efficiency loss from decreased labor demand. Due to data limitations on the expected change in labor demand and supply of U.S. workers, the Department cannot accurately measure the efficiency gains or losses to the U.S. labor market created by the new prevailing wage system. The Department discusses this potential impact qualitatively; the Department invited comment on how to estimate changes to efficiency from the new prevailing wage levels, but did not receive any such comment.

The Department received two comments suggesting that the transfers of the rule were underestimated.

One commenter suggests that the analysis in the IFR underestimates the transfer payments of the IFR. They cite a 2020 Cato Institute study that found the wage increases, using interpolated wages from the publicly available BLS OES dataset resulted in underestimates of the wage impacts of the IFR. In addition, they suggested that the use of the 90th percentile as a proxy for the 95th percentile significantly underestimated wages.

In response to the commenter's concern, the IFR estimate of wages was based on BLS OES data publicly available at the time of publication. Therefore, the estimated wage impacts in the IFR were conservative, particularly for workers with wages set at the 95th percentile where wage impacts were calculated based on the publicly available 90th percentile. In this final rule the Department revises its wage tier methodology, including setting the Level IV percentile at the 90th percentile. The change in methodology will result in wage tiers that are set at percentiles that are lower than those presented in the IFR and that will be phased in over a period of 2 years for applicants that are new to the H-1B program, and three and a half years for applicants on track for lawful permanent residency (LPR).

Another commentator suggested the transfers were underestimated and they calculated that the IFR was based on wage increases of $4,825 to $9,651 per worker based on Exhibit 5 and Exhibit 6 of the IFR.

In response to the commenter's concern, Exhibit 5 and Exhibit 6 of the IFR contained illustrative wage data for a particular SOC-code and area in BLS OES and do not reflect the average impact of the IFR. They instead serve the purpose of illustrating the Department's wage impact calculations. Wage increases vary by SOC code and geographic area and therefore can be higher than these examples. The analysis for the IFR estimated that workers facing a wage increase (i.e., those that were offered less under the baseline than required by the IFR) had an average increase of $27,000.

Under this final rule the Department revises the wage level percentiles of the IFR with some modifications to account for the two-step and three and a half year transition periods that are new to the final rule. Therefore, the final rule wage impact estimation follows four main steps: Step 1—simulate wage impacts with the revised percentiles for each transition wage level using historical certification data and adjust wage impacts for USCIS approval rates. Step 2—project 10-year series wage impacts incorporating the transition schedule. Step 3—during the transition period adjust the population of workers eligible for the two-step transition versus the three and a half year transition. Step 4—Estimate total transfers by combining adjusted two-step and three and a half year transition total wage impacts. This methodology is described in more detail below.

Step 1—simulate wage impacts with the revised percentiles for each transition wage level and adjusted based on USCIS approval rates. For each H-1B certification in FY 2018, FY 2019, and FY 2020, the Department used the difference between the estimated prevailing wage level under the final rule and the wage offered under the current baseline to establish the wage impact of the prevailing wage computation changes in each calendar year of the certification's employment period. Under the H-1B visa classification, employment periods for certifications can last for up to three years in length and generally begin up to six months after a certification is issued by the Department. Therefore, a given fiscal year can have wage impacts that start in that calendar year and last up to three years, or wage impacts that could start in the following calendar year and have an end-date up to four calendar years past the fiscal year. For example, an employment start date in March of 2019 may be associated with an H-1B application certified by the Department during FY 2018 and, if that certified application contains a three-year employment period, the wage impacts on the employer will extend through March of 2022. This final rule does not retroactively impact certified wages, so there will be new H-1B applications certified by the Department during FY 2020 that may extend well into the analysis period. Therefore, the first year of the rule will only impact new certifications, in the second year new and continuing certifications from year 1 will be impacted, and in the third year and beyond both new and continuing certifications from years 1 and 2 will be impacted.

To account for this pattern of wage impacts, we classify certifications into three length cohorts and calculate annual wage impacts for each length cohort based on FY 2018 through FY 2020 data. The length cohorts are: Certifications lasting less than 1 year, certifications lasting 1-2 years, and certifications lasting 2-3 years. For each length cohort we calculate wage impacts for their first calendar year (“new”), their second calendar year (“ongoing”), and third or more calendar year (“ongoing +”)

H-1B, H-1B1, or E-3 applications certified by the Department do not necessarily result in employer wage obligations. After obtaining a certification, employers applying under the H-1B and H-1B1 programs, and in certain situations, the E-3 program must then submit a Form I-129, Petition for a Nonimmigrant Worker for approval by U.S. Citizenship and Immigration Services (USCIS). USCIS may approve or deny the H-1B visa petition. USCIS approval data represents approvals of petitions based on both certifications issued by the Department that used OES data for the prevailing wage, or certifications that were based on other approved sources to determine the prevailing wage (e.g., Collective Bargaining Agreements, employer-provided surveys). Exhibit 4 summarizes FY 2018 and FY 2019 data on H-1B, H-1B1, and E-3 certifications with their prevailing wage based on the OES survey, adjusted USCIS approvals, and approval rate. To account for approval rates that may differ by geographic location and whether a certification is new or continuing, we adjust each certification's wage impact by the approval rate of the State of intended employment for the employer's certification and whether it is a new or continuing application.

To estimate the wage impacts of new percentiles contained in this final rule, the Department used publicly available BLS OES data that reports the 10th, 25th, 50th, 75th, and 90th percentile wages by SOC code and metropolitan or non-metropolitan area. In order to estimate wages for the new final rule levels of 35th, 53rd, 72nd, and 90th percentiles, the Department linearly interpolated between relevant percentiles for reported wages at each SOC code and geographic area combination. For each certification from FY 2018 through FY 2020 the new wage was estimated for the final rule wage levels as well as all transition periods (i.e., 90 percent for the two-step transition; 85 percent, 90 percent, and 95 percent for the three and a half year transition).

An illustrative example of calculations used to calculate wage impacts under the final rule is provided in Exhibit 5 and Exhibit 6 below. In Exhibit 5, to calculate projected wage impacts under the final rule, the Department first multiplied the number of certified workers by the number of hours worked in each calendar year (2,080 hours) and the new prevailing wage for the level the workers were certified at for their particular SOC and the geographic area combination. The examples in Exhibit 5 set forth how the Department calculated the final rule wage impact for an individual case of each length cohort.

After the total wages for the final rule was determined, the Department calculated the baseline wage. The baseline wage is always equal to or greater than the baseline prevailing wage because some certifications offer a wage higher than the prevailing wage. The methodology is the same as that used to estimate the projected wages under the final rule: Number of certified workers is multiplied by the number of hours worked in each calendar year (based on 2,080 hours in a full year) of certified employment and the actual offered wage for the certified workers (Exhibit 6 provides an example of the calculation of the baseline wages for the same case as in Exhibit 5).

Once the baseline offered wage was obtained, the Department estimated the wage impact of the final rule prevailing wage levels by subtracting the baseline offered wage for each calendar year from the final rule prevailing wage. The total wage impact was then multiplied by the average USCIS petition beneficiary approval rate for the State of intended employment. Here, the Department presents the wage impacts for the examples in Exhibits 5 and 6, above. For the length cohort less than 1 year, the impact in 2018 was $149,632 (($2,868,437 − $2,413,146) * 0.33) and $238,303 in 2019 (($4,568,251 − $3,843,158) * 0.33). For the length cohort of 1-2 years, the impact in 2019 was $1,528,388 (($7,230,496 − $4,116,300) * 0.49), and in 2020 was $1,528,388 (($7,230,496 − $4,116,300) * 0.49). The example for length cohort 2-3 years had wage impacts in 2018, 2019, and 2020. In the 2018 the wage impact was $103,580 (($1,445,030 − $1,113,014) * 0.31), $408,042 in 2019 (($5,692,544 − $4,384,600) * 0.31), and $2,947,905 in 2020 (($5,692,544 − $4,384,600) * 0.31).

Existing prevailing wage data from the Foreign Labor Certification (FLC) Data Center, accessible at http://www.flcdatacenter.com,, contains wage data for each SOC code and geographic area combination that are not readily available in the public OES data used to estimate new prevailing wage levels. For example, when an OES wage is not releasable for a geographic area, the prevailing wage available through the FLC Data Center may be computed by BLS for the geographic area plus its contiguous areas. Additionally, in publicly available OES data, some percentiles are missing for certain combinations of SOC codes and geographic areas. These two factors result in a small number of certifications having no match with a new prevailing wage level. To estimate wage impacts for workers associated with these certifications, the average wage impact per worker, for the given cohort and fiscal year the certification is associated with, is calculated and then applied to an adjusted number of workers associated with the certification that does not match. It is unlikely that all unmatched certifications will have a wage impact so the calculated wage impact per worker is applied to 85 percent of workers associated with unmatched certifications. This produces a series of estimated wage impacts for workers that are not matched with new prevailing wages in the public OES data for each calendar year for which they have employment. These imputed wage impacts are then added to the calculated wage impact to produce a final total wage impact for each length cohort and percentile group in each calendar year.

Exhibit 7 summarizes the wage impacts of each length cohort for all percentile groups involved in the two wage transitions based on FY 2018 through FY 2020 certification data. The result of this analysis is an annual average wage impact for each length cohort and percentile group that is used in following steps to construct projected 10-year wage impacts. In Exhibit 7 some calendar years do not have values because the cohort, based on FY 2018 through FY 2020 data, does not have a full year of data for those years. For example, calendar year 2021 does have new entries from FY 2020 data but it is not a complete year of data as FY 2021 would also have new entries, and therefore it is not included.

Step 3—project 10-year series of wage impacts incorporating transition schedule. To project 10-year wage transfers the average annual values from Exhibit 7 are used to construct a 10-year series that incorporates the transition schedule and change in worker population eligible for the two-step transition or three and a half year transition. Based on data provided by USCIS there are approximately 266,500 workers in backlog for a Green Card that are on continuing H-1B visas and are therefore eligible for the three and a half year transition. On average from FY 2018 to FY 2020 316,845 workers were approved annually by USCIS. Therefore, approximately 84 percent of applications are currently eligible for the three and a half year transition and the remaining 16 percent will use the two-step transition. Over time USCIS estimates that 30,000 workers would be processed through the backlog every year resulting in a declining population of workers eligible in each subsequent year for wages under the three and a half year transition. The Department assumes that the total population of applicants will not change, therefore the percent of applicants applying to the H-1B visa program for two-step transition wages (or the final wage level after the transition) will grow over time and the population of workers eligible for wages under the three and a half year transition will decline. A summary of this population transition as well as the wage transition for each group is presented in Exhibit 8.

To illustrate the application of the wage and population transitions to the average annual wages provided above in Exhibit 7 we describe an example of this calculation for new applications in 2021. Exhibit 9, below, provides an example calculation for new applicants in 2021 under the two-step transition wage (90 percent of final wage levels).

Average annual wage impacts for each length cohort represent a full year of wage impacts, however the wage transition does not begin until July 1, 2021. Therefore, the proportion of working days in July 1, 2021 through December 31, 2021 (50.6%) is used to adjust each length cohort's average annual wage impact. A second adjustment is made to account for the population transition (16% of the total applicant population faces wages under the two-step transition in 2021). Ongoing wages from new applications in 2021 occur in 2022 and 2023. Therefore, the estimates of ongoing wages from Exhibit 7 are included in 2022 and 2023 and also adjusted by 2021 transition and population adjustments (because these ongoing wages are associated with the 2021 new applicants).

This process was repeated for each year of 2021-2024 to account for each new year of applicants (i.e., in 2022, under the two-step transition, half of applicants have impacts at 90 percent of final wage levels and half at the final wage levels). In addition, the population of applicants under the two-step transition increases from 16 percent in 2021 to 25 percent in 2022. From 2025 onwards all new applicants are subject to the final wage levels.

Step 4—estimate total transfer payments. The Department determined the total impact of the final rule by summing wage impacts from new applicants in each year and ongoing wage impacts from new applicants in prior years. The results of this is presented below in Exhibit 10.

The changes in prevailing wage rates constitute a transfer payment from employers to employees. The Department estimates the total transfer over the 10-year period is $130.83 billion and $105.16 billion at discount rates of 3 and 7 percent, respectively. The annualized transfer over the 10-year period is $15.34 billion and $14.97 billion at discount rates of 3 and 7 percent, respectively.

With the increases in prevailing wage levels under the final rule, some employers may decide not to hire a U.S. worker or a foreign worker on a temporary or permanent basis. The prevailing wage increase may mitigate labor arbitrage and induce some employers to train and provide more working hours to incumbent workers, resulting in no increase in employment. The Department is unable to quantify the extent to which these two factors will occur and therefore discusses them qualitatively.

The labor economics literature has a significant volume of research on the impact of wages on demand for labor. Of interest in the context of the H-1B program is the long-run own-wage elasticity of labor demand that describes how firms demand labor in response to marginal changes in wages. There is significant heterogeneity in estimates of labor demand elasticities that can depend on industry, skill-level, region, and more. A commonly cited value of average long-run own-wage elasticity of labor demand is −0.3. This would mean that a one percent increase in wage would reduce demand for labor by 0.3 percent. The average annual increase in wage transfers is anan 18.8 percent increase in wage payments, which would imply a potential reduction in labor demand by 5.64 percent (18.8 * .3). It is likely that U.S. employers will pay higher wages to H-1B workers or replace them with U.S. workers to the extent that is possible. However, we can approximate that, if U.S. employers were limited in the ability to pay higher wages and did reduce demand for workers in these roles, it would reduce the transfer payment by approximately 5.64 percent. The annual average undiscounted wage transfer estimate of $15.57 billion would therefore be reduced to $14.69 billion.

Non-Quantifiable Transfer Payments

This section discusses the non-quantifiable transfer payments related to changes to the computation of the prevailing wage levels. Specifically, the Department did not quantify transfer payments associated with new certifications under the Permanent Labor Certification Program because they are expected to be de minimis.

The PERM programs have a large proportion of certifications issued annually to foreign beneficiaries that are working in the U.S. at the time of certification and would have changes to wages under the final rule prevailing wage. Prior to the PERM certification, these beneficiaries are typically working under H-1B, H-1B1, and E-3 temporary visas and wage transfers for these PERM certifications are therefore already factored into our wage transfer calculations for H-1B, H-1B1, and E-3 temporary visas. Below, Exhibit 11 illustrates the percentage of PERM certifications that are on H-1B, H-1B1, or E-3 temporary visas, the percent that are not on a temporary visa and/or are not currently in the U.S. and would therefore enter on an EB-2 or EB-3 visa, and all other visa classes.

Approximately 10 percent of PERM certifications are issued annually by OFLC to foreign beneficiaries who do not currently reside in the U.S. and would enter on immigrant visas in the EB-2 or EB-3 preference category. Employment-based immigrant visa availability and corresponding wait times change regularly for different preference categories and countries. Foreign workers from countries with significant visa demand consistently experience delays, at times over a decade. Therefore, employers would not have wage obligations until, at the earliest, the very end of the 10-year analysis period, and the number of relevant certifications is a relatively small percent of all PERM certifications; the Department therefore has not included associated wage transfers in the analysis.

Benefits Discussion

This section discusses the non-quantifiable benefits related to changes to the computation of the prevailing wage levels.

The Department's increase in the prevailing wages for the four wage levels is expected to result in multiple benefits that the Department is unable to quantify but discusses qualitatively. One benefit of the final rule's increase in prevailing wages is the economic incentive to increase employee retention, training, and productivity which will increase benefits to both employers and U.S. workers. The increase in prevailing wages is expected to induce employers—particularly those using the permanent and temporary visa programs—to fill critical skill shortages, to minimize labor costs by implementing retention initiatives to reduce employee turnover, and/or to increase the number of work hours offered to similarly employed U.S. workers. Furthermore, for employers in the technology and health care sectors, this could mean using higher wages to attract and hire the industry's most productive U.S. workers and to provide them with the most advanced equipment and technologies to perform their work in the most efficient manner.

This high-wage, high-skill approach to minimizing labor costs is commonly referred to as the “efficiency wage” theory in labor economics—a well-established strategy that allows companies employing high-wage workers to minimize labor costs and effectively compete with companies employing low-wage workers. The efficiency wage theory supports the idea that increasing wages can lead to increased labor productivity because workers feel more motivated to work at higher wage levels. Where these jobs offer wages that are significantly higher than the wages and working conditions of alternative jobs, workers will have a greater incentive to be loyal to the company, impress their supervisors with the quality of their work, and exert an effort that involves no shirking. Thus, if employers increase wages, some, or even all, of the higher wage costs can be recouped through increased staff retention, lower costs of supervision, and higher labor productivity.

Strengthening prevailing wages will also help promote and protect jobs for American workers. By ensuring that the employment of any foreign worker is commensurate with the wages paid to similarly employed U.S. workers, the Department will be protecting the types of white-collar, middle-class jobs that are critical to ensuring the economic viability of communities throughout the country.

There is some evidence that the existing prevailing wage levels offer opportunities to use lower-cost alternatives to U.S. workers doing similar jobs by offering at the two wage levels below the median wage. For example, in FY 2019, 60 percent of H-1B workers were placed at either the first or second wage level, meaning a substantial majority of workers in the program could be paid wages well below the median wage for their occupational classification. By setting the Level I wage level at the 35th percentile, employers using the H-1B and PERM programs will have less of an incentive to replace U.S. workers doing similar jobs at lower wage rates when there are available U.S. workers. This will increase earnings and standards of living for U.S. workers. It also will level the playing field by reducing incentives to replace similarly employed U.S. workers with a low-cost foreign alternative.

In addition, because workers with greater skills tend to be more productive, and as a result can command higher wages, raising the prevailing wage levels will lead to the limited number of H-1B visas going to higher-skilled foreign workers, which will likely increase the spillover economic benefits associated with high-skilled immigration.

Finally, ensuring that skilled occupations are not performed at below-market wage rates by foreign workers will provide greater incentives for firms to expand education and job training programs. These programs can attract and develop the skills of a younger generation of U.S. workers to enter occupations that currently rely on elevated levels of foreign workers.

4. Summary of the Analysis

Exhibit 12 below summarizes the costs and transfer payments of the final rule. The Department estimates the annualized cost of the final rule at $2.90 million and the annualized transfer payments (from H-1B, H-1B1, and E-3 employers to workers) at $14.97 billion, at a discount rate of 7 percent. The Department did not estimate any cost savings. For the purpose of E.O. 13771, the annualized cost, when perpetuated, is $1.86 million at a discount rate of 7 percent in 2016 dollars.

5. Regulatory Alternatives

The Department considered two alternatives to the chosen approach of establishing the prevailing wage for Levels I through IV, respectively, at approximately 35th percentile, the 45nd percentile, the 72nd percentile, and the 90th percentile with a transition period.

First, the Department considered an alternative that would modify the number of wage tiers from four levels to three levels. Under this alternative, prevailing wages would be set for Levels I through III at the 35th, 72nd, and 90th percentile, respectively. Modifying the number of wage tiers to three levels would allow for more manageable wage assignments that would be easier for employers and employees to understand due to decreased complexity to matching wage tiers with position experience. A three-tiered prevailing wage structure would maintain the minimum entry-level and fully competent experience levels and simplify the intermediate level of experience by combining the current qualified and experienced distinctions. The Department prefers the chosen methodology over this alternative because the chosen four-tiered prevailing wage structure is likely to produce more accurate prevailing wages than a three-tiered structure due to the ability to have two intermediate wage levels. In addition, creating a three-tiered prevailing wage structure would require a statutory change.

The Department considered a second alternative that would modify the geographic levels for assigning prevailing wages for the SOC code within the current four-tiered prevailing wage structure, which ranges from local MSA or BOS areas to national, to a two-tiered geographic area structure containing only statewide or national area estimates. By assigning prevailing wages at a statewide or, where statewide averages cannot be reported by the BLS, national geographic area, this second alternative would again simplify the prevailing wage determination process by reducing the number of distinct wage computations reported by the BLS and provide employers with greater certainty regarding their wage obligations, especially where the job opportunity requires work to be performed in a number of different worksite locations within a state or regional area. This process would also reduce variability in prevailing wages within a state for the same occupations across time, making prevailing wages more consistent and uniform. However, this method would not account for wage variability that may occur within states and that can account for within-state differences in labor market dynamics, industry competitiveness, or cost of living.

The Department prefers the chosen methodology because it preserves important differences in county and regional level prevailing wages and better aligns with the statutory requirement that the prevailing wage be the wage paid in the area of employment.

The Department received one comment on the regulatory alternatives considered in the IFR. One commenter representing 23 organizations suggested that the Department consider an alternative where data from private sector compensation surveys is layered on top of BLS OES data to provide more accurate prevailing wage data for certain occupations and localities where private sector compensation surveys may have coverage.

Supplementing BLS OES data from private sector compensation surveys may result in an increased ability to quantitatively connect education, experience, or employee responsibility with wages for certain occupations and localities. However, this introduction of fidelity in certain locales and not others could lead to inconsistent treatment of wages in the same occupation in different geographic areas depending on whether prevailing wages are based on BLS OES or the private sector compensation survey. In addition, such an approach would reduce transparency of prevailing wages by introducing additional complexity in the wage determination as well as non-public data sources.

B. Regulatory Flexibility Act

The Regulatory Flexibility Act of 1980 (RFA), 5 U.S.C. 601 et seq., as amended by the Small Business Regulatory Enforcement Fairness Act of 1996, Public Law 104-121 (March 29, 1996), hereafter jointly referred to as the RFA, requires that an agency prepare an initial regulatory flexibility analysis (IRFA) when proposing, and a final regulatory flexibility analysis (FRFA) when issuing, regulations that will have a significant economic impact on a substantial number of small entities. The agency is also required to respond to public comment. The Chief Counsel for Advocacy of the Small Business Administration submitted public comment on the Initial Regulatory Flexibility Analysis (IRFA) which is addressed below.

The Department believes that this final rule will have a significant economic impact on a substantial number of small entities and therefore the Department publishes this FRFA.

1. Objectives of and Legal Basis for the Final Rule

The Department has determined that new rulemaking is needed to better protect the wages and job opportunities of U.S. workers, minimize incentives to hire foreign workers over U.S. workers on a permanent or temporary basis in the United States under the H-1B, H-1B1, and E-3 visa programs and the PERM program, and further the goals of Executive Order 13788, Buy American and Hire American. Accordingly, this final rule revises the computation of wage levels under the Department's four-tiered wage structure based on the OES wage survey administered by the BLS to ensure that wages paid to immigrant and nonimmigrant workers are commensurate with the wages of U.S. workers with comparable levels of education, experience, and levels of supervision in the occupation and area of employment.

The Department is amending its regulations at Sections 656.40 and 655.731 to reflect the methodology the Department will use to determine prevailing wages based on the BLS's OES survey for job opportunities in the H-1B and PERM programs. The revised methodology will establish the prevailing wage for Levels I through IV, respectively, at approximately the 35th percentile, the 53rd percentile, the 72nd percentile, and the 90th percentile. In addition, the final rule allows for a transition period by setting an interim year of wages at 90 percent of the above wage levels for new H-1B visas, and a three and a half year transition period of 85 percent, 90 percent, 95 percent of the above wage levels for workers on track for lawful permanent residency (LPR).

The INA assigns responsibilities to the Secretary relating to the entry and employment of certain categories of employment-based immigrants and nonimmigrants. This rule relates to the labor certifications that the Secretary issues for certain employment-based immigrants and to the LCAs that the Secretary certifies in connection with the temporary employment of foreign workers under the H-1B, H-1B1, and E-3 visa classifications. The Department has a statutory mandate to protect the wages and working conditions of similarly employed U.S. workers from adverse effects caused by the employment of foreign workers in the U.S. on a permanent or temporary basis.

2. The Agency's Response to Public Comments

The Department did not receive public comment on the IRFA.

3. Response to Comments by the Chief Counsel for Advocacy of the Small Business Administration

The Department received a comment on the IRFA by the Chief Counsel for Advocacy of the Small Business Administration that suggested the Department underestimated the economic impacts of the IFR, and therefore underestimated the significant impacts on small entities. The comment suggested that the IFR underestimated impacts based on IFR RIA Exhibits 5 and 6 which indicated a wage increase of $4,825 to $9,651 per worker and the comment provided examples from the Department's online wage library showing examples of higher wage increases.

IFR RIA Exhibit 5 and Exhibit 6 contain illustrative wage data for a particular SOC-code and area in BLS OES and do not reflect the average impact of the IFR. They instead serve the purpose of illustrating the Department's wage impact calculations. Wage increases vary by SOC code and geographic area and therefore can be higher than these examples. The analysis for the IFR estimated that workers facing a wage increase (i.e., those that were offered less under the baseline than required by the IFR) had an average increase of approximately $27,000.

Under the final rule the Department revises its wage tier estimates so that wages will be transitioned over a period of two to three and a half years reducing impacts in some years. In addition, the final wage levels (after transition) will be set at lower percentiles than the IFR resulting in reduced wage obligations from the IFR, therefore reducing impacts on small businesses. Finally, Department wage estimates are based on DOL H-1B disclosure data. However, USCIS does not approve all certifications contained in the disclosure data. As a result, the estimated wage obligations for some small entities may be overestimated, and the overall number of impacted small entities at all levels of impact may be overestimated.

4. Description of the Number of Small Entities to Which the Final Rule Will Apply

i. Definition of Small Entity

The RFA defines a “small entity” as a (1) small not-for-profit organization, (2) small governmental jurisdiction, or (3) small business. The Department used the entity size standards defined by SBA, in effect as of August 19, 2019, to classify entities as small. SBA establishes separate standards for individual 6-digit NAICS industry codes, and standard cutoffs are typically based on either the average number of employees, or the average annual receipts. For example, small businesses are generally defined as having fewer than 500, 1,000, or 1,250 employees in manufacturing industries and less than $7.5 million in average annual receipts for nonmanufacturing industries. However, some exceptions do exist, the most notable being that depository institutions (including credit unions, commercial banks, and non-commercial banks) are classified by total assets (small defined as less than $550 million in assets). Small governmental jurisdictions are another noteworthy exception. They are defined as the governments of cities, counties, towns, townships, villages, school districts, or special districts with populations of less than 50,000 people.

ii. Number of Small Entities

The Department collected employment and annual revenue data from the business information provider Data Axle and merged those data into the H-1B, H-1B1, and E-3 visa program disclosure data (H-1B disclosure data) for FY 2019. This process allowed the Department to identify the number and type of small entities using the H-1B program and their annual revenues. A single employer can apply for H-1B workers multiple times; therefore, unique employers were identified. The Department was able to obtain data matches for 34,203 unique H-1B employers. Next, the Department used the SBA size standards to classify 26,354 of these employers (or 77.1 percent) as small. These unique small employers had an average of 75 employees and average annual revenue of approximately $18.61 million. Of these unique employers, 22,430 of them had revenue data available from Data Axle. The Department's analysis of the impact of this final rule on small entities is based on the number of small unique employers (22,430 with revenue data).

To provide clarity on the types of industries impacted by this regulation, Exhibit 13 shows the number of unique H-1B small entity employers with certifications in FY 2019 within the top 10 most prevalent industries at the 6-digit and 4-digit NAICS code level. Depending on when their employment period starts and the length of the employment period (up to 3 years), small entities with certifications in FY 2019 can have wage obligations in calendar years 2018 through 2023.

iii. Projected Impacts to Affected Small Entities

The Department has considered the incremental costs for small entities from the baseline (the regulations governing permanent labor certifications at 20 CFR part 656 and labor condition applications at 20 CFR part 655, subpart H) to this final rule. We estimated the cost of (a) the time to read and review the final rule and (b) wage costs. These estimates are consistent with those presented in the E.O. 12866 section.

The Department estimates that small entities using the H-1B program, 22,430 unique employers would incur a one-time cost of $51.93 to familiarize themselves with the rule.

In addition to the total first-year cost above, each small entity using the H-1B program may have an increase in annual wage costs due to the revisions to the wage structure if they currently offer a wage lower than the final rule's prevailing wage levels. For each small entity, we calculated the likely annual wage cost as the sum of the total final rule wage minus the total baseline wage for each small entity identified from the H-1B disclosure data in FY 2019. We added this change in the wage costs to the total first-year costs to measure the total impact of the final rule on the small entity. Small entities with certifications in FY 2019 can have wage obligations in calendar years 2018 through 2023, depending on when their employment period starts and the length of the employment period (up to 3 years). Because USCIS does not approve all certifications, the estimated wage obligations for some small entities may be overestimated. The Department is unable to determine which small entities had certifications approved or not approved by USCIS and therefore estimates the total wage obligation with no adjustment for USCIS approval rates. As a result, estimates of the total cost to small entities are likely to be inflated. The Department sought public comments on how to best estimate which small entities had certifications approved by USCIS but did not receive any comments that discussed a method for estimating certification approval by USCIS. Exhibit 14 presents the number of small entities with a wage impact in each year, as well as the average wage impact per small entity in each year.

The Department determined the proportion of each small entity's total revenue affected by the costs of the final rule to determine if the final rule would have a significant and substantial impact on small entities. The cost impacts included estimated first-year costs and the wage costs introduced by the final rule. Wage costs are based on the final wage levels as these represent the largest annual impacts a small entity would face (as opposed to wage impacts during the transition to the final wage levels). The Department used a total cost estimate of 3 percent of revenue as the threshold for a significant individual impact and set a total of 15 percent of small entities incurring a significant impact as the threshold for a substantial impact on small entities.

The Department has used a threshold of three percent of revenues in prior rulemakings for the definition of significant economic impact. This threshold is also consistent with that sometimes used by other agencies. The Department also maintains that 15 percent of small entities experiencing a significant impact represents an appropriate threshold to determine whether the rule has a substantial impact on small entities generally. The Department has used the same threshold in prior rulemakings for the definition of substantial number of small entities.

Of the 22,430 unique small employers with revenue data, up to 13 percent of employers would have more than 3 percent of their total revenue affected in 2019, up to 22 percent in 2020 and 2021, and up to 16 percent in 2022. Exhibit 15 provides a breakdown of small employers by the proportion of revenue affected by the costs of the final rule.

5. Projected Reporting, Recordkeeping, and Other Compliance Requirements of the Final Rule

The final rule does not have any reporting, recordkeeping, or other compliance requirements impacting small entities.

6. Steps the Agency Has Taken To Minimize the Significant Economic Impact on Small Entities

The RFA directs agencies to assess the effects that various regulatory alternatives would have on small entities and to consider ways to minimize those effects. Accordingly, the Department considered two regulatory alternatives to the chosen approach of establishing the prevailing wage for Levels I through IV, respectively, at approximately the 35th percentile, the 53rd percentile, the 72nd percentile, and the 90th percentile with a transition period.

First, the Department considered an alternative that would modify the number of wage tiers from four levels to three levels. Under this alternative, the Department attempted to set the prevailing wages for Levels I through III, respectively, at the 35th, 72nd, and 90th percentile. Modifying the number of wage tiers to three levels would allow for more manageable wage assignments that would be easier for small entities and their employees to understand due to decreased complexity to matching wage tiers with position experience. The Department decided not to pursue this alternative because the chosen four-tiered wage methodology is likely to be more accurate than the three-tiered wage level because it has two intermediate wage levels. In addition, creating a three-tiered wage level would require a statutory change. Although the Department recognizes that legal limitations prevent this alternative from being actionable, the Department nonetheless presents it as a regulatory alternative in accord with OMB guidance.

The Department considered a second alternative that attempted to modify the geographic levels for assigning prevailing wages for the occupation from the current four-tiered structure, which ranges from local MSA or BOS areas to national, to a two-tiered structure containing statewide or national levels. By assigning prevailing wages at a statewide or national level (depending on whether statewide averages can be reported by BLS), this second alternative attempted to simplify the prevailing wage determination process by reducing the number of distinct wage computations reported by the BLS. It would also provide small entities with greater certainty regarding their wage obligations, especially where the job opportunity requires work to be performed in a number of different worksite locations within a State or regional area. The Department decided not to pursue this alternative because the chosen methodology preserves important differences in county and regional level prevailing wages, and because it would require a statutory change.

C. Unfunded Mandates Reform Act

The Unfunded Mandates Reform Act of 1995 (UMRA) is intended, among other things, to curb the practice of imposing unfunded Federal mandates on State, local, and tribal governments. Title II of UMRA requires each Federal agency to prepare a written statement assessing the effects of any Federal mandate in a proposed or final agency rule that may result in a $100 million or more expenditure (adjusted annually for inflation) in any one year by State, local, and tribal governments, in the aggregate, or by the private sector. The inflation-adjusted value equivalent of $100 million in 1995 adjusted for inflation to 2019 levels by the Consumer Price Index for All Urban Consumers (CPI-U) is approximately $168 million based on the Consumer Price Index for All Urban Consumers.

While this final rule may result in the expenditure of more than $100 million by the private sector annually, the rulemaking is not a “Federal mandate” as defined for UMRA purposes. The cost of obtaining prevailing wages, preparing labor condition and certification applications (including all required evidence) and the payment of wages by employers is, to the extent it could be termed an enforceable duty, one that arises from participation in a voluntary Federal program, applying for immigration status in the United States. This final rule does not contain such a mandate. The requirements of Title II of UMRA, therefore, do not apply, and DOL has not prepared a statement under UMRA. Therefore, no actions were deemed necessary under the provisions of the UMRA.

D. Congressional Review Act

The Office of Information and Regulatory Affairs, of the Office of Management and Budget, has determined that this final rule is a major rule as defined by 5 U.S.C. 804, also known as the “Congressional Review Act,” as enacted in section 251 of the Small Business Regulatory Enforcement Fairness Act of 1996, Public Law 104-121, 110 Stat. 847, 868, et seq.

E. Executive Order 13132 (Federalism)

This final rule would not have substantial direct effects on the states, on the relationship between the national government and the states, or on the distribution of power and responsibilities among the various levels of government. Therefore, in accordance with section 6 of Executive Order 13132, it is determined that this final rule does not have sufficient federalism implications to warrant the preparation of a federalism summary impact statement.

F. Executive Order 12988 (Civil Justice Reform)

This final rule meets the applicable standards set forth in sections 3(a) and 3(b)(2) of Executive Order 12988.

G. Regulatory Flexibility Executive Order 13175 (Consultation and Coordination With Indian Tribal Governments)

This final rule does not have “tribal implications” because it does not have substantial direct effects on one or more Indian tribes, on the relationship between the Federal Government and Indian tribes, or on the distribution of power and responsibilities between the Federal Government and Indian tribes. Accordingly, E.O. 13175, Consultation and Coordination with Indian Tribal Governments, requires no further agency action or analysis.

H. Paperwork Reduction Act

The Paperwork Reduction Act of 1995 (PRA), 44 U.S.C. 3501, et seq., and its attendant regulations, 5 CFR part 1320, require the Department to consider the agency's need for its information collections and their practical utility, the impact of paperwork and other information collection burdens imposed on the public, and how to minimize those burdens. This final rule does not require a collection of information subject to approval by OMB under the PRA, or affect any existing collections of information.

List of Subjects

20 CFR Part 655

• Administrative practice and procedure
• Australia
• Chile
• Employment
• Employment and training
• Immigration
• Labor
• Migrant labor
• Wages

20 CFR Part 656

• Administrative practice and procedure
• Employment
• Foreign workers
• Labor
• Wages

DEPARTMENT OF LABOR

Accordingly, for the reasons stated in the preamble, the Department of Labor amends parts 655 and 656 of Chapter V, Title 20, Code of Federal Regulations, as follows:

PART 655—TEMPORARY EMPLOYMENT OF FOREIGN WORKERS IN THE UNITED STATES

1. The authority citation for part 655 is revised to read as follows:

Authority: Section 655.0 issued under 8 U.S.C. 1101(a)(15)(E)(iii), 1101(a)(15)(H)(i) and (ii), 8 U.S.C. 1103(a)(6), 1182(m), (n), (p), and (t), 1184(c), (g), and (j), 1188, and 1288(c) and (d); sec. 3(c)(1), Pub. L. 101-238, 103 Stat. 2099, 2102 (8 U.S.C. 1182 note); sec. 221(a), Pub. L. 101-649, 104 Stat. 4978, 5027 (8 U.S.C. 1184 note); sec. 303(a)(8), Pub. L. 102-232, 105 Stat. 1733, 1748 (8 U.S.C. 1101 note); sec. 323(c), Pub. L. 103-206, 107 Stat. 2428; sec. 412(e), Pub. L. 105-277, 112 Stat. 2681 (8 U.S.C. 1182 note); sec. 2(d), Pub. L. 106-95, 113 Stat. 1312, 1316 (8 U.S.C. 1182 note); 29 U.S.C. 49k; Pub. L. 107-296, 116 Stat. 2135, as amended; Pub. L. 109-423, 120 Stat. 2900; 8 CFR 214.2(h)(4)(i); 8 CFR 214.2(h)(6)(iii); and sec. 6, Pub. L. 115-218, 132 Stat. 1547 (48 U.S.C. 1806).

Subpart A issued under 8 CFR 214.2(h).

Subpart B issued under 8 U.S.C. 1101(a)(15)(H)(ii)(a), 1184(c), and 1188; and 8 CFR 214.2(h).

Subpart E issued under 48 U.S.C. 1806.

Subparts F and G issued under 8 U.S.C. 1288(c) and (d); sec. 323(c), Public Law 103-206, 107 Stat. 2428; and 28 U.S.C. 2461 note, Public Law 114-74 at section 701.

Subparts H and I issued under 8 U.S.C. 1101(a)(15)(H)(i)(b) and (b)(1), 1182(n), (p), and (t), and 1184(g) and (j); sec. 303(a)(8), Public Law 102-232, 105 Stat. 1733, 1748 (8 U.S.C. 1101 note); sec. 412(e), Public Law 105-277, 112 Stat. 2681; 8 CFR 214.2(h); and 28 U.S.C. 2461 note, Public Law 114-74 at section 701.

Subparts L and M issued under 8 U.S.C. 1101(a)(15)(H)(i)(c) and 1182(m); sec. 2(d), Public Law 106-95, 113 Stat. 1312, 1316 (8 U.S.C. 1182 note); Public Law 109-423, 120 Stat. 2900; and 8 CFR 214.2(h).

2. Amend § 655.731 by revising paragraphs (a)(2)(ii) introductory text, (a)(2)(ii)(A) introductory text, and (a)(2)(ii)(A)( 2) to read as follows:

PART 656—LABOR CERTIFICATION PROCESS FOR PERMANENT EMPLOYMENT OF ALIENS IN THE UNITED STATES

3. The authority citation for part 656 is revised to read as follows:

Authority: 8 U.S.C. 1182(a)(5)(A), 1182(p); sec.122, Pub. L. 101-649, 109 Stat. 4978; and Title IV, Pub. L. 105-277, 112 Stat. 2681.

4. Amend § 656.40 by revising paragraphs (a) and (b)(2) and (3) to read as follows: