AAU, Associations Express Concerns on Prevailing Wage NPRM

AAU joined the College and University Professional Association for Human Resources (CUPA-HR) and 14 other higher education associations in sending comments in response to DOL’s proposed rule on prevailing wage levels under the H-1B and related programs.

Dear Administrator Pasternak,

On behalf of the College and University Professional Association for Human Resources (CUPA-HR) and undersigned organizations, we write in response to the Notice of Proposed Rulemaking (NPRM) published March 27, 2026 in the Federal Register by the Department of Labor (Department), entitled Improving Wage Protections for the Temporary and Permanent Employment of Certain Foreign Nationals in the United States.¹

CUPA-HR serves as the voice of human resources in higher education, representing more than 25,000 human resources professionals and other higher-education leaders at more than 1,725 colleges and universities across the country, including 88 percent of all U.S. doctoral institutions, 66 percent of master’s institutions, 45 percent of bachelor’s institutions, and nearly 600 two-year and specialized institutions. The institutions represented by this comment include major research universities, academic medical centers, regional public universities, liberal arts colleges, and community colleges that rely on highly skilled international faculty, researchers, instructors, clinicians, and other professionals to support institutional teaching, research, and patient-care missions.

CUPA-HR and the undersigned organizations submit this comment to express substantial concern regarding the proposed rule’s effects on institutions of higher education and affiliated research enterprises. Higher-education employers use the H-1B, H-1B1, E-3, and PERM (permanent labor certification) programs principally to recruit and retain faculty, postdoctoral researchers, research scientists, instructors, physicians, and other specialized personnel in fields where advanced graduate training is the norm and where institutions compete in national and international academic labor markets. The proposed changes would significantly affect higher education’s ability to recruit and retain this workforce and would disrupt academic hiring, research operations, clinical programs, and grant-funded positions across the sector.

CUPA-HR and the undersigned organizations support the Department’s interest in ensuring that the H-1B and related programs are administered in a manner that protects U.S. workers, prevents misuse of prevailing-wage requirements, and maintains the wage safeguards Congress established. We are concerned, however, that the proposed rule rests on generalized labor-market assumptions that do not reflect how the H-1B and PERM programs operate in higher education. The Department’s analysis is built principally on labor-market dynamics outside higher education and does not reflect the structure of academic hiring, research employment, clinical appointments, and faculty compensation at colleges and universities. The rule’s underlying premise—that existing prevailing-wage levels systematically permit the underpayment of H-1B workers—is not borne out in higher education, where CUPA-HR survey data indicate that H-1B faculty are paid at parity with or above similarly situated non-H-1B faculty at the same institutions.²

The American Competitiveness and Workforce Improvement Act (“ACWIA”), at 8 U.S.C. § 1182(p)(1), already directs the Department to account for distinctions in the higher-education labor market by requiring that the prevailing wage for employees of institutions of higher education consider only employees at such institutions and related nonprofit research organizations. The H-1B cap exemption at 8 U.S.C. § 1184(g)(5)(A) reflects the same statutory distinction.³The proposed rule does not separately analyze this labor market or adequately account for the ways in which higher education hiring, compensation structures, and workforce composition differ from the sectors principally discussed in the rulemaking record.

We therefore respectfully request that the Department: (1) exempt institutions of higher education from the proposed percentile changes pending sector-specific analysis calibrated to the academic labor market that is published in the Federal Register with a notice and comment period; (2) alternatively, delay application of the revised percentile structure to higher-education-filed Labor Condition Applications (LCA) and PERM applications for at least two years to accommodate higher education’s reliance interests and transition burdens; and (3) retain without modification the existing authority for employer-provided alternative wage surveys under 20 CFR 656.40(b)(3) and (g).

CUPA-HR and allied higher-education organizations previously submitted sector-specific comments regarding the Department’s 2020 Interim Final Rule⁴ and 2021 Request for Information, and we supplement and update that record here.⁵

I. Background.

The proposed rule would revise the four-level methodology used to determine prevailing wages under the H-1B, H-1B1, E-3, and PERM visa programs by shifting each wage level upward across the Occupational Employment and Wage Statistics (“OEWS”) wage distribution. Under the proposal, Level I would increase from approximately the 17th to the 34th percentile, Level II from the 34th to the 52nd percentile, Level III from the 50th to the 70th percentile, and Level IV from the 67th to the 88th percentile. The same percentile schedule would apply uniformly across all four programs.

The current four-tier wage structure has governed academic hiring continuously since 2005. In that year, the Department issued the Prevailing Wage Determination Policy Guidance for Nonagricultural Immigration Programs, implementing INA § 212(p)(4)’s requirement that prevailing-wage surveys provide “at least 4 levels of wages commensurate with experience, education, and the level of supervision.” Under the Department’s longstanding methodology, the Level I wage corresponded approximately to the 17th percentile of the OEWS distribution, Level II to the 34th percentile, Level III to the 50th percentile, and Level IV to the 67th percentile. The Department reissued the guidance in updated form in November 2009 in connection with the centralization of prevailing-wage processing within the Office of Foreign Labor Certification’s National Prevailing Wage Center.⁶

On October 8, 2020, the Department issued an Interim Final Rule (“IFR”) that substantially revised the prevailing-wage framework, raising Level I to approximately the 45th percentile of the OEWS distribution and Level IV to approximately the 95th percentile, effective immediately and without prior notice and comment.⁷The IFR created substantial operational and budgeting challenges for colleges and universities that had already made hiring, retention, and grant-funded staffing commitments under the existing wage framework. CUPA-HR and eighteen other higher-education associations submitted a detailed comment on the IFR on November 9, 2020, expressing concern regarding both the methodology and the rule’s abrupt implementation. Three federal district courts subsequently set aside or enjoined the IFR on procedural grounds, and higher-education institutions, including Purdue University, were among the lead challengers.⁸

On January 14, 2021, the Department promulgated a Final Rule that retained the IFR’s overall approach while modifying the percentile levels, setting Level I at the 35th percentile and Level IV at the 90th percentile.⁹The Department subsequently delayed implementation of the Final Rule twice, in February and May 2021. In the May delay, the Department acknowledged that “delaying the implementation of the Final Rule is likely to have an impact on the wages paid to workers” and further recognized the need to review concerns raised regarding the rule’s methodology and implementation.¹⁰On April 2, 2021, the Department issued a Request for Information (“RFI”) seeking input on alternative data sources and methodologies for determining prevailing wage levels. On June 1, 2021, CUPA-HR and five other higher-education associations filed a second detailed comment in response. On June 23, 2021, on the Department’s unopposed motion, the Northern District of California vacated and remanded the January 2021 Final Rule.¹¹

The Department engaged in no further rulemaking on these methodologies until Presidential Proclamation 10973 directed the Secretary of Labor in September 2025 to initiate rulemaking to revise H-1B prevailing wage levels.¹²The Department published the present proposed rule on March 27, 2026.

The Department offers two principal reasons for the proposal. First, the Department concludes that the OEWS occupational distributions on which prevailing wages are based capture some workers whose education and experience fall below the H-1B “specialty occupation” threshold and the analogous statutory qualifications for the other programs to which the four-tier structure applies. The Department therefore concludes that the existing 17/34/50/67-percentile schedule fails to satisfy INA § 212(p)(4)’s requirement that wage levels be “commensurate with experience, education, and the level of supervision.” Second, the Department identifies an aggregate adverse effect on U.S. workers. It observes that 63 percent of certified LCAs in FY 2024 were assigned to Levels I or II and that certified LCA wages averaged approximately $10,191 below the OEWS average for similarly classified occupations. The Department concludes that the current methodology permits employers to access lower-cost labor in ways that adversely affect U.S. workers’ wages. Although the Department’s analysis focuses principally on the H-1B program, it applies the same percentile schedule uniformly across the H-1B, H-1B1, E-3, and PERM programs.¹³

The proposed rule applies prospectively to Applications for Prevailing Wage Determinations (“PWD”) pending with the National Prevailing Wage Center as of the effective date and to LCAs and PERM applications filed on or after that date; it does not apply to previously approved PWDs, LCAs, or PERM certifications. The Department designates the proposed rule economically significant.

II. The NPRM does not separately analyze higher education or the higher-education workforce.

Although the Department proposes sweeping changes to prevailing wage levels across the H-1B, H-1B1, E-3, and PERM programs, the NPRM does not meaningfully examine how those changes would affect colleges and universities, academic medical centers, or affiliated nonprofit research institutions. “Higher education,” “university,” “college,” and “research institution” do not appear as sector or employer categories in the NPRM’s analysis, and “ACWIA” appears only once, in connection with an alternative the Department is not proposing.¹⁴The closest the NPRM comes to engaging the sector is a single footnote addressing the distinction between cap-subject and cap-exempt H-1B employment, in which the Department states that it has “no reason to assume that the distribution of cap-exempt employees is different from the distribution of non-cap-exempt LCAs.”¹⁵The Department’s proposed methodology was calibrated principally against the H-1B program as a whole and not against the distinct characteristics of higher-education employment.

That omission is significant because Congress itself has long recognized that higher education occupies a distinct position within the H-1B framework. ACWIA, at 8 U.S.C. § 1182(p)(1), requires that prevailing wages for employees of institutions of higher education take into account only employees at such institutions and related nonprofit research organizations. The H-1B cap exemption at 8 U.S.C. § 1184(g)(5)(A) reflects the same distinction by separately exempting institutions of higher education and affiliated nonprofit entities from the numerical cap. These statutory distinctions reflect Congress’s recognition that higher education institutions compete in distinct labor markets and operate under different economic and operational constraints than many private-sector employers.¹⁶

Yet the NPRM largely analyzes the labor market through generalized references to wage suppression, labor substitution, and employer incentives within private-sector labor markets without separately evaluating whether those assumptions apply similarly within higher education. The Department’s discussion focuses heavily on high-volume H-1B usage in technology and related industries and repeatedly frames the proposal as a response to concerns regarding labor substitution and the use of lower prevailing wage levels to obtain (acknowledging that the ACWIA-covered methodology must be run separately “as required by statute”). lower-paid labor.¹⁷The NPRM does not separately evaluate whether those assumptions hold in higher-education employment settings.

The NPRM likewise does not analyze whether OEWS-based percentile adjustments calibrated to broader occupational groupings accurately reflect academic labor markets. Many university positions involve highly specialized combinations of teaching, research, clinical practice, and grant administration responsibilities that are not readily comparable to positions in the commercial sector grouped within the same occupational classifications. Nor does the NPRM separately assess how prevailing wage increases would affect research-intensive and clinically affiliated academic employers.

The Department’s small-entity analysis has the same blind spot. The Initial Regulatory Flexibility Analysis (IRFA) names Custom Computer Programming Services (29 percent), Computer Systems Design Services, Engineering Services, Offices of Lawyers, and Accounting Services as the small-entity NAICS categories most affected by the proposed rule.¹⁸Higher-education NAICS classifications — including 611210 (junior colleges) and 611310 (colleges, universities, and professional schools) — do not appear. Small private colleges and small two-year institutions meet the Small Business Administration size standards for those NAICS codes and use the H-1B and PERM programs directly affected by the proposed rule.

The scale of the proposed rule compounds the gap in the Department’s analysis. The Department designates the rule as economically significant and projects ten-year transfer payments of up to $46.09 billion.¹⁹CUPA-HR’s analysis of the Department’s own disclosure data identifies substantial wage floor exposure for the higher education sector under the proposed methodology, discussed further in Section III. The Department’s Executive Order 12866 analysis quantifies the proposed wage floor increases in isolation. It does not account for the additional $100,000 payment that Presidential Proclamation 10973 imposes on certain H-1B specialty occupation petitions for workers entering the United States.²⁰For institutions with limited alternatives to the H-1B classification, the combined burden of those costs determines whether hiring a prospective employee remains feasible. An analysis that quantifies one cost while omitting the other does not fully capture the proposed rule’s effect on higher education employers.

The Department acknowledges the 2021 Request for Information and the comments submitted in response, including those submitted by the undersigned higher education organizations, but does not separately address the specific concerns and data presented in those submissions. CUPA-HR submitted sector-specific analysis to the Department in its November 2020 comment on the Interim Final Rule and its June 2021 comment on the Request for Information. We provide that analysis again here, updated and expanded with five additional years of disclosure data and findings from CUPA-HR’s 2025-26 annual survey cycle, all of which provide additional evidence regarding the operation of the H-1B program in higher education.²¹

III. The higher-education H-1B workforce differs materially from the labor market described in the NPRM.

Although higher education represents a relatively small share of total H-1B filings, the sector is disproportionately concentrated in Wage Level I, the wage level most affected by the proposed rule. Between FY2020 and FY2025, 59.9 percent of certified higher-education H-1B LCAs were assigned to Wage Level I, compared with 13.6 percent in all other industries — a 4.4-fold higher concentration.²²That concentration reflects the structure of academic employment and the concentration of highly credentialed early-career faculty and researchers at the beginning of their academic careers.

Certified-LCA data and CUPA-HR incumbent workforce data show patterns that differ sharply from the aggregate H-1B population on which the Department calibrated the proposed methodology.

DOL’s proposed methodology uses the certified LCA population to construct a “Benchmark Value” and then selects new wage level percentiles designed to align average prevailing wages with that benchmark. Higher education’s LCA distribution differs sharply from the aggregate distribution on which the Department calibrated the proposed percentiles. This distribution in higher education has remained stable between FY2020 and FY2024, the Level I share in higher education remained between 58.1 percent and 61.1 percent.²³

In higher education, Level I concentration reflects the structure of academic career ladders rather than the labor-substitution dynamics emphasized in the NPRM. Many postsecondary teaching and research occupations require doctoral or equivalent terminal credentials at the entry level yet still classify as entry-rung academic appointments. The same Level I concentration therefore carries a different labor-market meaning in higher education than in the broader H-1B market.

Table III-1. Wage Level Composition, Higher Education vs. All Other Industries

Certified H-1B LCAs, FY2020 – FY2025

Source: CUPA-HR analysis of DOL OFLC H-1B LCA Disclosure Data, FY2020 Q1 – FY2025 Q4. Methodology at Appendix A.

The divergence is not limited to wage-level distribution. It extends to the occupations that make up the Level I population itself. At Wage Level I, the top ten Standard Occupational Classification (“SOC”) codes in higher education and the top ten SOCs outside higher education share almost no common occupations. In higher education, the leading Level I occupations are medical scientists, biochemists, and postsecondary teaching faculty across engineering, business, computer science, mathematics, and health specialties. Outside of higher education, the leading occupations are software developers, accountants, and mechanical, civil, and electrical engineers.

The divergence is sharpest in the postsecondary teaching occupations. Multiple postsecondary teaching occupations recorded either zero or only negligible non-higher education Level I filings across six fiscal years, including political science, communications, sociology, geography, anthropology, environmental science, social work, and criminal justice.²⁴Other teaching SOCs appear only marginally outside of higher education. Engineering Teachers (HE 4,237; non-HE 7), Business Teachers (3,790 vs. 2), Computer Science Teachers (3,477 vs. 8), and Economics Teachers (1,555 vs. 1) are effectively absent from DOL’s non-higher education record.

These occupations drive higher education’s Level I filing volume, yet they are largely absent from the non-higher education workforce underlying the Department’s analysis. The problem is therefore not simply that higher education’s occupational composition differs from the aggregate LCA population, it is that the Benchmark Value and resulting percentile calibrations are derived principally from occupations that do not describe higher-education employment.²⁵Because many of the occupations driving higher-education Level I usage are minimally represented or absent outside higher education, the occupational mix underlying the Department’s Benchmark Value does not accurately describe the higher education workforce affected by the proposed rule.

Table III-2. Top 10 Occupations at Wage Level I, Higher Education vs. All Other Industries

Certified H-1B LCAs at Wage Level I, FY2020 – FY2025

Source: CUPA-HR analysis of DOL OFLC H-1B LCA Disclosure Data, FY2020 Q1 – FY2025 Q4. Methodology at Appendix A.

Wage level distribution and occupational composition describe the structure of higher education’s Level I filings. Offered wages describe how those filings operate in practice under the H-1B wage requirements. Across higher education Level I filings, the offered wage exceeds the prevailing wage by a median of 24.9 percent. In all other industries, the corresponding median is 6.4 percent. The NPRM interprets large offered-vs-prevailing-wage gaps as evidence that prevailing wages are set below the market value of comparable

U.S. workers. Higher education presents a materially different factual pattern: the sector combines the highest concentration of Level I usage with substantially larger offered wage premiums. Many Wage Level I positions in higher education nonetheless require doctoral or other terminal credentials and compete in labor markets for highly credentialed early-career faculty and researchers. The higher-education record therefore does not support the NPRM’s inference that Level I offered-vs-prevailing-wage gaps necessarily indicate systemic underpayment relative to market wages.

Table III-3. Offered Wage Premium Over Prevailing Wage at Level I

Median values, certified H-1B LCAs at Wage Level I, FY2020 – FY2025

Source: CUPA-HR analysis of DOL OFLC H-1B LCA Disclosure Data, FY2020 Q1 – FY2025 Q4. The offered-vs-prevailing-wage gap is computed at the case level as (offered wage − prevailing wage) ÷ prevailing wage; the table reports the median of the per-case gap values within each population.

Methodology at Appendix A.

Wage level distribution, occupational composition, and the Level I offered-vs-prevailing-wage gap describe higher education’s H-1B filings in the certified-LCA record. The next pattern appears in CUPA-HR incumbent workforce data and operates at the level of the workers themselves. One of the largest H-1B-incumbent populations in higher education’s Professionals workforce is the academic research staff hierarchy — Research Assistant, Research Scholar, Research Associate, Senior Research Scholar, and Principal Research Scholar — which includes a substantial portion of the postdoctoral research workforce at U.S. colleges and universities.

These positions are concentrated in funded research environments and are typically filled by early-career researchers holding doctoral degrees. Postdoctoral researchers conduct a substantial portion of the bench-level research supporting the U.S. academic research enterprise.²⁶The National Center for Science and Engineering Statistics estimates approximately 66,000 postdoctoral researchers nationally.²⁷CUPA-HR’s 2024–25 survey cycle reported 6,470 Research Scholar positions, 17 percent of which were held by H-1B workers, implying roughly 11,000 H-1B postdoctoral researchers at U.S. colleges and universities.²⁸

These positions are directly affected by the proposed increase to Level I prevailing-wage floors in research occupations, yet they are not the type of lower-credentialed labor pool on which the NPRM’s adverse effects rationale is principally based. Table III-4 reports the corresponding position-level pay ratios from the 2025–26 survey cycle.

Table III-4. H-1B Pay Ratios, Academic Research-Staff Positions

CUPA-HR incumbent survey data, 2025-26

Source: CUPA-HR Professionals in Higher Education Survey, 2025-26. The five positions form the academic research-staff hierarchy under SOC 19-0000 in CUPA-HR‘s coding. Together they account for 1,272 H-1B incumbents — one of the largest single H-1B-incumbent populations in CUPA-HR’s higher-education Professionals universe. Pay ratio = median salary of H-1B incumbents ÷ median salary of non-H-1B incumbents in the same position. Pooled ratio is the H-1B-incumbent-weighted mean of position-level ratios.

Across the five positions, H-1B and non-H-1B incumbents in comparable positions are paid at parity or near parity. At the three positions most commonly associated with early-career and postdoctoral appointments— Research Scholar (1.030), Research Associate (1.031), and Research Assistant (1.050) — H-1B incumbents earn at or slightly above non-H-1B peers, with median years in position of 2.0 or fewer for both groups. The two senior positions sit modestly below parity at 0.923 and 0.897; these are also the two positions where H-1B incumbents have substantially fewer years in position than non-H-1B peers (2.0 vs. 3.0 and 2.0 vs. 4.0, respectively). The academic research-staff classification is one of the largest H-1B-incumbent populations in higher education and includes a substantial portion of the postdoctoral research population at U.S. colleges and universities. The higher-education workforce patterns reflected in the certified-LCA and incumbent-workforce records also produce materially different exposure to the proposed prevailing-wage floors themselves.

Table III-5. Estimated Aggregate Wage Floor Impact of the Proposed Rule on Higher Education

Certified filings at Wage Level I, FY2020 – FY2025

Source: CUPA-HR analysis of DOL OFLC LCA and PERM Disclosure Data, FY2020 Q1 – FY2025 Q4, and OFLC ACWIA Wage Data Tables (EDC_Export), Wage Years 2020-21 through 2024-25. Methodology at Appendix A.

Across the six fiscal years of available disclosure data, 80,808 certified higher-education filings at Wage Level I would have been subject to the proposed new floor. The proposed methodology would have increased the required wage in 44,377 of those filings — 54.9 percent of all higher-education Wage Level I filings during the study period. The median shortfall among affected filings was approximately $9,865 for LCAs and $11,439 for PERM filings. Aggregated across the period, the estimated employer wage-floor shortfall totals approximately $489.7 million.

Four empirical findings emerge from the higher education record. First, higher education’s H-1B population is concentrated at Wage Level I to a degree materially different from the aggregate LCA population on which the Department based its methodology. Second, the occupational composition of higher education’s Level I filings differs sharply from the non-higher-education population, including many postsecondary teaching and research occupations that appear minimally or not at all outside higher education. Third, higher education’s Level I filings show substantially larger offered-wage premiums than corresponding filings in other industries. Fourth, the proposed methodology would increase the required wage in more than half of higher education Wage Level I filings during the study period, despite the sector’s demonstrated parity and credential patterns. Taken together, these patterns undermine the NPRM’s assumption that aggregate H-1B labor-market dynamics can be generalized to higher education without separate sector-specific analysis.

IV. The Department’s empirical reasoning does not fit the higher education record.

The Department’s empirical reasoning rests on three premises about how the H-1B program operates in practice. First, the Department assumes that H-1B workers are systematically underpaid relative to comparable U.S. workers. Second, it treats the gap between Level I prevailing wages and offered wages as evidence that the Level I floor is too low. Third, it reasons that the lower portion of the OEWS distribution includes workers who lack the educational qualifications typically associated with H-1B employment, requiring an upward adjustment to the Level I percentile. These premises do not reflect the higher education record. This Section addresses each premise and explains why those premises do not align with the higher education labor market.

IV.A H-1B and non-H-1B faculty in higher education are paid at parity.

The Department’s wage-suppression premise is testable. Under INA § 212(n)(1)(A), the H-1B offered wage must equal at least the higher of the actual wage paid to comparably situated workers at the employer or the prevailing wage in the area. The Department’s prevailing-wage framework approximates that comparison through OEWS, a synthetic distribution of U.S.-worker wages within the same SOC and metropolitan statistical area, with Wage Level I anchored at a percentile of that distribution. CUPA-HR’s four annual Higher Education Surveys permit a more direct comparison. They collect incumbent-level salary data from more than 1,000 participating non-profit colleges and universities and permit a direct comparison between the salaries of H-1B incumbents and non-H-1B incumbents in the same faculty categories at the same institutions.²⁹

The CUPA-HR comparison differs from the Department’s OEWS-based framework in several respects. The CUPA-HR surveys compare incumbent salaries within faculty categories and disciplines across participating institutions rather than within OEWS occupation-and-geography groupings. But the CUPA-HR comparison is closer to the statutory actual-wage inquiry in one critical respect: the non-H-1B reference group is the actual same-employer peer workforce identified by the statute, not a synthetic distribution drawn from the broader labor market. It is also the only direct H-1B-versus-non-H-1B compensation comparison in the rulemaking record. If the NPRM’s wage-suppression premise accurately described higher education, the CUPA-HR data would be expected to show systematic H-1B pay disparities within the same institutional faculty populations.

CUPA-HR’s 2025-26 survey cycle directly identifies 12,026 H-1B faculty and full-time staff across more than 1,000 participating institutions. Of those, 8,868 are faculty: 6,008 in tenure-track ranks, 2,295 in non-tenure-track teaching positions, and 565 in non-tenure research positions. H-1B faculty are 4.0 percent of all faculty whose H-1B status was reported by the institution — 4.4 percent of tenure-track faculty, 2.9 percent of non-tenure-track faculty, and 7.6 percent of non-tenure research faculty.³⁰The H-1B share in faculty ranks is materially higher than the cap-subject H-1B share in the broader labor market, a distinction the Department’s footnote 162 set aside without analysis.³¹

Across that population, H-1B faculty are paid, on average, slightly more than their non-H-1B peers in the same faculty categories at the same institutions. Table IV-1 reports the comparison for the 2025-26 cycle. The overall H-1B-to-non-H-1B pay ratio across all faculty is 1.040. Three faculty categories produce ratios above 1.0 — Tenure-Track Assistant Professor at 1.087, Non-Tenure-Track Faculty at 1.179, and Tenure-Track Faculty (all ranks combined) at 1.010. Two additional categories sit effectively at parity — Tenure-Track Professor at 0.978 and Tenure-Track Associate Professor at 0.994. The remaining below-parity figure, Non-Tenure Research Faculty at 0.866, reflects differences in career-stage composition discussed below.

Table IV-1. H-1B Pay Ratios, Higher-Education Faculty, 2025-26

CUPA-HR Faculty in Higher Education Survey, 2025-26

Source: CUPA-HR Faculty in Higher Education Survey, 2025-26. Pay ratio = median salary of H-1B incumbents in the faculty category ÷ median salary of non-H-1B incumbents in the same faculty category at the same set of participating institutions.

Two of the three below-parity ratios in Table IV-1 — 0.978 at the Professor rank and 0.994 at the Associate Professor rank — are consistent with differences in career-stage composition between the H-1B and non-H-1B populations within those ranks. H-1B Professors have a median 5 years in position, compared with 8 years for non-H-1B peers; H-1B Associate Professors have a median 3 years, compared with 5. The absolute median salaries are within $2,550 (Professor) and $610 (Associate Professor) of one another, on populations of 55,567 and 44,698 non-H-1B incumbents respectively. The ratios are not consistent with a pattern of systematic wage suppression. They are what the data would be expected to show when an H-1B population concentrated at earlier career stages is compared against a non-H-1B population with longer tenure within rank.

The remaining below-parity figure, Non-Tenure Research Faculty at 0.866, reflects the same pattern in stronger form. H-1B Non-Tenure Research Faculty have a median 2 years in position, compared with 4 years for non-H-1B incumbents. Section III and Table III-4 document the same pattern at the position level for the academic research-staff hierarchy under SOC 19-0000. The early-career rungs run at or above parity (Research Scholar 1.030, Research Associate 1.031, Research Assistant 1.050), while the senior rungs sit below parity (Senior Research Scholar 0.923, Principal Research Scholar 0.897), consistent with the same career-stage concentration.³²

The parity pattern holds across the disciplines that account for the largest H-1B faculty populations in higher education. Table IV-2 reports H-1B-to-non-H-1B pay ratios for the eight disciplines that account for the largest H-1B faculty populations across the two faculty categories where the headline pattern is most direct — Tenure-Track and Non-Tenure-Track. Together, these disciplines contain the majority of H-1B faculty in those two categories.³³Within those disciplines, tenure-track ratios cluster at or above parity (Business 1.042, Mathematics 1.093, Physical Sciences 1.000, Biological Sciences 1.039, Health Professions 1.146, Social Sciences 1.072, Computer Science 0.993, Engineering 0.983). Non-tenure-track ratios run higher (Health Professions 1.207, Business 1.203, Biological Sciences 1.311, Computer Science 1.062, Mathematics 1.053, Social Sciences 1.117, Physical Sciences 0.991, Engineering 0.955).

These discipline-level data include faculty in fields central to the U.S. technical and scientific workforce pipeline, including Engineering, Computer Science, Mathematics, and Physical Sciences. Those faculty educate the students who enter the labor markets the Department identifies as significant in the NPRM’s broader discussion of the H-1B program. Yet the higher-education compensation data do not show systematic underpayment of H-1B faculty relative to non-H-1B peers in those fields.

Table IV-2. H-1B Pay Ratios by Discipline, Higher-Education Faculty (Tenure-Track and Non-Tenure-Track), Top Eight Disciplines by H-1B Faculty Volume, 2025-26 CUPA-HR Faculty in Higher Education Survey, 2025-26

Source: CUPA-HR Faculty in Higher Education Survey, 2025-26. Pay ratio = median salary of H-1B incumbents in the discipline-and-category cell ÷ median salary of non-H-1B incumbents in the same cell at the same set of participating institutions.

Where comparison data exist across CUPA-HR cycles, the parity pattern remains stable. In 2020, the Tenure-Track Professor pay ratio was 1.140; in 2025-26, 0.978. The Associate Professor ratio moved from 1.070 to 0.994 over the same period, while the Assistant Professor ratio remained consistently above parity (1.100 in 2020; 1.087 in 2025-26). The downward movement at the senior ranks is consistent with the population-growth-driven career-stage composition described above: between 2020 and 2025-26, the H-1B Professor population increased from 314 to 582, and the H-1B Associate Professor population from 728 to 1,043.

Section III, ¶4, documents that the median Level I offered wage in higher education exceeds the prevailing wage by 24.9 percent, approximately four times the gap observed elsewhere in the H-1B economy. Those same higher-education employers pay H-1B faculty 4.0 percent above their non-H-1B peers in the same positions at the same institutions. The two findings answer related questions in the same direction: higher education does not use the Level I floor to suppress H-1B wages relative to the broader labor market, and higher education does not pay H-1B faculty less than the U.S.-worker faculty employed alongside them.

CUPA-HR submitted the same incumbent-level pay-ratio comparison to the Department in 2021, on prior cycles of the same four surveys. The 2025-26 cycle confirms the pattern on more recent data. On this record, the central premise of the proposed methodology — that H-1B workers are systematically underpaid relative to similarly employed U.S. workers — has no support in higher education.

IV.B The $10,191 Gap Is What a Two-Prong Wage Requirement Produces, Not Evidence the Floor Is Too Low.

The H-1B wage requirement has two prongs that measure two different things. The prevailing wage is calculated for the position — the duties, occupation, and area of intended employment — while the actual wage is calculated from what the employer pays comparable employees with similar experience and qualifications.³⁴The position-based prong asks what the position requires; the actual-wage prong asks what the employer pays the people performing it. H-1B beneficiaries frequently hold qualifications above the minimum requirements of the position, and the actual-wage prong captures that qualifications premium while the prevailing wage remains tied to the position floor. A gap between offered and prevailing wages is therefore consistent with the structure of the statutory framework when employers pay for qualifications above minimum position requirements.

The Department’s Step 1 inference treats the gap as a defect in the wage floor. In Step 1, the Department points to a $10,191 average difference between offered and prevailing wages on certified H-1B LCAs filed in FY2020 through FY2025. The Department reads that gap as evidence that the prevailing wage is set “below the market value of comparable U.S. workers.”³⁵That reading does not account for the structure of the wage requirement. The H-1B employer must pay the higher of the actual and prevailing wages.³⁶Where the employer’s comparably situated workers are paid above the prevailing wage, the actual wage prong binds, and the offered wage rises to meet it. A non-negative gap between offered and prevailing wages is therefore built into the statutory framework at any prevailing-wage floor. A positive average gap on the certified-LCA universe is what we see whenever the actual-wage prong binds above the prevailing wage. The $10,191 figure is consistent with the design of the requirement and with ordinary market behavior at and above a working floor.

The NPRM does not clearly explain how to distinguish between ordinary operation of the dual wage framework and evidence that the prevailing-wage floor itself is improperly calibrated. Even under the proposed methodology, employers would still frequently pay above the prevailing wage because the actual-wage requirement would continue to bind where institutions compete for highly qualified workers. The offered-vs-prevailing-wage gap may also reflect limitations in the geographic specificity of the ACWIA wage table, an issue discussed further in Section V.A.

Higher education’s record shows this architecture working. At Wage Level I, where higher-education filings concentrate, the median offered wage exceeds the prevailing wage by 24.9 percent, compared with 6.4 percent in all other industries.³⁷Pay parity to non-H-1B peers in the same positions at the same institutions holds across the tenure-track ranks, with

H-1B Assistant Professors paid 8.7 percent above their non-H-1B peers. The educational composition of the work that drives higher-education Level I filings differs fundamentally from the occupations on which the Department’s OEWS-coverage rationale is based. The existing methodology, applied to higher education’s credentialed early-career workforce at the wage levels where the sector’s filings concentrate, produces wage outcomes consistent with the statutory framework. In higher education, the gap the Department treats as evidence of an inadequate floor is instead evidence that the prevailing-wage and actual-wage requirements are functioning as designed.

IV.C In Higher Education, “Entry Level” Reflects Career Stage, Not Lower Educational Qualification.

The Department’s rationale for raising the Level I percentile rests on a claimed mismatch between the educational qualifications associated with H–1B employment and the workers included in the OEWS wage distribution. OEWS measures wages across all workers within an occupational classification regardless of whether those workers possess the educational qualifications typically required for H–1B employment. The Department therefore reasons that the lower portion of the OEWS distribution includes workers who would not qualify for H–1B employment and that anchoring Level I at the 17th percentile understates the market wage for specialty-occupation workers.³⁸ The NPRM illustrates this concern principally through computer-related occupations, including Software Developers, Computer Systems Engineers/Architects, and Registered Nurses, where the Department concluded that portions of the OEWS distribution include workers whose educational qualifications fall below the H–1B specialty-occupation threshold.³⁹

In higher education, however, the occupations driving Level I H–1B usage operate differently. Across the FY2020 through FY2025 LCA data, 82.2 percent of certified higher-education Level I filings — 64,179 of 78,083 cases — fall within postsecondary teaching occupations and life, physical, and social-scientist occupations, occupational families in which the standard entry qualification already meets or exceeds the H–1B statutory minimum. Postsecondary teaching occupations account for 42.4 percent of higher-education Level I filings, while life, physical, and social-scientist occupations account for another 39.8 percent.⁴⁰An assistant professor in mathematics, business, computer science, or engineering is ordinarily hired with a doctoral degree. Medical scientists and biochemists, likewise, typically hold doctoral degrees or enter the position through multi-year postdoctoral appointments that themselves ordinarily require doctoral training. In these occupations, “entry level” describes career stage rather than educational attainment.

The Department’s own Level I framework recognizes that entry-level positions may still require advanced educational qualifications. Under 20 CFR 656.40(d)(1), a position is classified at Level I where the position’s requirements place it at the entry end of the occupational range. In higher education’s dominant Level I occupations, however, the entry end of the occupation is not a less-educated workforce. It is the beginning stage of a doctoral or research-intensive professional career. An assistant professor or early-career research scientist may hold the terminal qualification of the field while still occupying the entry rung of the academic career structure associated with that qualification. Raising the Level I percentile from the 17th to the 34th percentile therefore does not correct a mismatch between H–1B educational qualifications and the lower portion of the OEWS distribution in those occupations. Instead, it raises prevailing wages for the same highly educated workforce that OEWS has historically priced under the existing framework.

The NPRM asks whether the proposed percentiles “appropriately approximate” the education and experience expected in the occupation.⁴¹In higher education’s dominant Level I SOCs, the existing Level I framework already prices a workforce that meets or exceeds the educational qualifications associated with H–1B employment. The proposed percentile increase therefore does not correct a mismatch between H–1B qualifications and workers without the educational qualifications typically associated with H–1B employment. Instead, it applies a percentile adjustment derived principally from materially different occupational labor markets to doctoral and research-trained academic workforces that the NPRM does not separately analyze.

The same mismatch problem appears geographically as well. For ACWIA-covered higher-education employers, the Department’s prevailing-wage framework increasingly substitutes national fallback estimates for area-specific labor-market wages.

V. The proposed methodology compounds existing limitations in the higher-education prevailing-wage framework and underscores the importance of retaining alternative wage surveys.

The NPRM focuses principally on percentile selection within the OEWS framework. For ACWIA-covered higher-education employers, however, the operation of the prevailing-wage system also depends on the geographic specificity of the underlying wage data and the continued availability of alternative wage surveys where OEWS classifications or area estimates do not accurately reflect higher-education labor markets. As discussed below, the Department’s own published wage tables show that the ACWIA-adjusted OEWS framework already relies heavily on national fallback estimates rather than area-specific wages, and the Department’s stated reasons for retaining alternative surveys apply with particular force to higher education.

V.A Geographic Specificity in the ACWIA-Adjusted OEWS Wage Table

The NPRM’s methodological issues are not limited to percentile selection. For ACWIA-covered higher-education employers, the prevailing-wage framework also relies heavily on geographic fallback mechanisms that substitute national estimates for local labor-market wages.

ACWIA at 8 U.S.C. § 1182(p)(1) directs that prevailing wages for institutions of higher education and related nonprofit research organizations take into account wages paid at such institutions in the area of intended employment. The Department implements that requirement through a separate ACWIA-adjusted OEWS wage table — the EDC (Education) table at the Foreign Labor Certification Data Center — alongside the all-industries ALC table.

Each prevailing-wage estimate in the EDC and ALC tables corresponds to a particular occupation in a particular geographic area. The Department classifies those estimates by “Geo Level”: Geo Level 1 reflects the actual MSA or Balance-of-State area; Geo Level 2 combines the area with contiguous areas; Geo Level 3 substitutes statewide data; and Geo Level 4 substitutes a national average where no releasable state-level estimate exists.

For Wage Year 2024-25, the EDC table is dominated by Geo Level 4 national fallbacks, while the ALC table remains dominated by local-area estimates. The full distribution, including five years of trend data, is reported in Table V-1.⁴²

Table V-1. Geographic Specificity of DOL-Published Prevailing Wages, Higher Education (EDC/ACWIA) vs. All Industries (ALC)

Panel A: Share of SOC × area combinations at each Geo Level, Wage Year 2024-25

Panel B: EDC and ALC Geo Level 1 (local) and Geo Level 4 (national) shares, Wage Years 2020-21 through 2024-25

Panel C: Distribution of ACWIA-covered SOCs by share of area entries at Geo Level 4, Wage Year 2024-25

Source: CUPA-HR analysis of DOL OFLC EDC_Export.csv and ALC_Export.csv wage tables, Wage Years 2020-21 through 2024-25. Geo Level field as defined in the Department’s Technical Release Notes, July 2024 Wage Year. Methodology at Appendix A.

The deterioration over time is concentrated in the EDC table. The EDC Level 4 share has risen from 57.8 percent in WY2020-21 to 69.0 percent in WY2024-25. The ALC Level 4 share has remained roughly stable at approximately 11 percent across the same five years. The all-industries wage table is essentially unchanged in its geographic-specificity profile; the higher-education-specific wage table is becoming less area-specific.

The Department acknowledges the geographic-fallback issue briefly at 91 Fed. Reg. 15486 in the course of the Transfer Payments calculation. The Department observes that an FLC Data Center wage may, when not releasable for the area, be computed “for the geographic area plus its contiguous areas,” and characterizes the resulting fallback as affecting “a small number of certifications.” The NPRM discusses the contiguous-area fallback (Geo Level 2), but does not separately disclose the Geo Level distribution, distinguish EDC from ALC, or address the higher-education-specific magnitude reflected in the Department’s own published wage tables.⁴³

The cell-level distribution in Panels A and B understates how concentrated the geographic-fallback problem is at the occupation level. Panel C reports the distribution of ACWIA-covered SOCs by the share of their geographic entries that fall to Geo Level 4. For 144 of the 606 ACWIA-covered SOCs (23.8 percent), every published area entry is at Geo Level 4 — meaning that the EDC table publishes a single national wage for the occupation regardless of where the work is performed. For 263 SOCs (43.4 percent, inclusive of the 144), more than 90 percent of geographic entries fall at Geo Level 4. On the Department’s own data, nearly one in four ACWIA-covered occupations are represented exclusively by national-average wages, and nearly half are nationalized for more than 90 percent of geographic entries. The wage source the ACWIA statute requires higher-education employers to use is becoming less area-specific over time at both the cell level and the occupation level.

The proposed rule does not address the geographic non-locality already present in the EDC table; it raises the percentile floor applied to that table. Level I moves from the 17th to the 34th percentile of the OEWS distribution, with corresponding increases at Levels II, III, and IV. For the 69 percent of occupation-and-area entries in the EDC table already represented by national-average wages — and for the 144 ACWIA-covered SOCs for which every geographic entry is national — the proposal produces a higher national average rather than a more geographically specific wage.

Whether the percentile shift moves additional occupation-and-area entries into Geo Level 4 — by raising the threshold at which BLS suppresses area-level estimates — is a separate empirical question the rulemaking record does not address. We do not assert that effect; we note only that the Department’s analysis at 91 Fed. Reg. 15486 does not address it either.

Where the EDC table cannot produce an area-specific wage, the alternative-survey mechanism retained in Alternative 2 provides the principal means by which higher-education employers can supply one consistent with the statute’s area-of-intended-employment requirement.

V.B Retention of Alternative Wage Surveys Under Alternative 2

CUPA-HR supports the Department’s decision to retain the option for employers to use private wage surveys, evaluated as Alternative 2 in the regulatory alternatives analysis.⁴⁴ The Department explains that alternative surveys remain important because government data sources may not capture niche labor markets or specialized positions, eliminating the option could disrupt longstanding compliance practices, private surveys may better reflect market wages in specialized industries or geographic areas, and the existing framework permits the Department to evaluate surveys against established methodological standards. Those rationales apply with particular force to higher education.

The existing framework for alternative wage surveys is set out in 20 CFR 656.40(b)(3) and (g) and the Department’s 2009 Prevailing Wage Determination Policy Guidance. Under that framework, qualifying surveys produce an arithmetic-mean wage for the position in the area without reliance on the OEWS percentile structure.

The Department’s concern regarding niche labor markets and unique job requirements applies directly to higher education. The OEWS classification system aggregates many substantively distinct higher-education positions within single SOC codes. SOC 11-9033 (Education Administrators, Postsecondary), for example, encompasses 150 distinct higher-education positions in CUPA-HR’s 2025-26 cycle, ranging from admissions officers through directors of academic programs to deans, provosts, and presidents — substantively different roles with materially different compensation structures aggregated under a single OEWS estimate. CUPA-HR raised the same SOC-aggregation concern in its November 2020 IFR comment, where the comparable figure for SOC 11-9033 was 126 positions.

The aggregation issue also operates differently across SOCs. In some administrator classifications, the issue is the breadth of positions grouped within a single SOC code. In others, the issue is aggregation at the major-SOC level, where higher-education research positions classified within SOC 19-0000 (Life, Physical, and Social Science Occupations) are not separately identifiable within OEWS. Where these aggregation issues occur, the alternative-survey route provides a mechanism for producing an area-specific wage at the position level.

The Department’s discussion of longstanding compliance practices also applies directly to higher education. CUPA-HR has conducted incumbent-level salary surveys for more than fifty years for staff positions and more than thirty years for faculty positions. CUPA-HR surveys have been used as alternative wage sources under the 2009 guidance and the framework at 20 CFR 656.40 throughout that period.

The Department’s recognition that private surveys may better reflect wages in specialized industries or geographic areas likewise reflects the structure of higher-education compensation. ACWIA itself recognizes higher education as a specialized labor market by directing that prevailing wages be based only on wages paid at institutions of higher education and related nonprofit research organizations in the area of intended employment. The Geo Level distribution discussed in Section V.A reflects the same concern in practice. CUPA-HR’s four 2025-26 Higher Education Surveys cover more than 1,000 nonprofit institutions, with response rates generally exceeding 35 percent, and produce arithmetic-mean wages for positions in the area consistent with 20 CFR 656.40(g) and the Department’s 2009 guidance.⁴⁵

CUPA-HR and the undersigned organizations supported retention of the alternative-survey framework in comments submitted on the 2020 Interim Final Rule and the 2021 Request for Information and continue to support retention here. The 20 percent average premium the Department reports for survey-based wages over OEWS wages at 91 Fed. Reg. 15492 is calculated from five SOC codes within the 15-xxxx computer-related occupational family. Whether survey premiums for higher-education-specific positions resemble figures derived from those occupations is a separate question that the NPRM does not address.⁴⁶

The existing framework for evaluating alternative wage surveys in 20 CFR 656.40(b)(3) and (g) and the Department’s 2009 guidance provides a workable framework for evaluating alternative wage surveys and should be retained.

VI. Reliance interests in the existing methodology are particularly significant in higher education.

Higher-education institutions have structured academic hiring, research staffing, clinical programs, and immigration sponsorship practices around a prevailing-wage framework that has remained materially stable for approximately two decades. Those structures operate on multi-year timelines that often extend well beyond the validity period of an individual LCA or H-1B approval. As a result, institutions routinely make hiring, compensation, grant, and sponsorship commitments years in advance under the wage framework currently in effect.

The Department acknowledges that regulated entities have developed reliance interests in the existing methodology but concludes prospective-only implementation provides sufficient transition latitude.⁴⁷The higher-education operational and workforce structures discussed below are not materially mitigated by prospective-only implementation and are not separately addressed in the NPRM’s balancing analysis.

VI.A Academic Hiring Cycles and Institutional Budgeting Operate on Multi-Year Planning Horizons

Faculty searches in most disciplines begin in the fall, proceed through the winter, and conclude with spring hiring decisions for positions that begin the following academic year. Postdoctoral and research staff recruitment frequently operates on the same cycle. As a result, institutions recruiting for academic-year 2026–27 positions have, in many cases, already conducted searches, extended offers, received candidate acceptance, and committed institutional funding under the prevailing-wage framework currently in effect. Compensation commitments, immigration sponsorship decisions, and institutional budgets were established against the existing four-level wage methodology. CUPA-HR raised the same concern in its 2020 comment on the Department’s Interim Final Rule, observing that academic hiring cycles often extend more than a year and that institutions and employees make consequential employment and sponsorship decisions in reliance on existing wage requirements.⁴⁸

Academic medical centers operate under similar planning constraints, with the additional consideration that clinical staffing obligations continue year-round and cannot readily pause for mid-cycle compensation changes. Faculty physicians, clinical-laboratory personnel, and health-sciences staff employed in H-1B status support ongoing patient care, research, and teaching functions funded through combinations of clinical revenue, sponsored research, and public appropriations that are generally fixed on annual budgeting cycles. The Department also does not address the limited availability of replacement labor in several affected clinical occupations. The Association of American Medical Colleges projects a national physician shortage of up to approximately 86,000 physicians by 2036. Against these existing workforce constraints, academic medical centers may face limited practical flexibility to absorb abrupt increases in prevailing wage obligations through substitution of alternative labor sources.⁴⁹

Higher-education revenue streams — including tuition revenue, state appropriations, sponsored-research funding, clinical revenue, and endowment distributions — are generally established through annual or multi-year budgeting processes and may not adjust contemporaneously with substantial changes in wage-floor obligations. In many institutional compensation systems, increases in base salary also produce corresponding increases in employer retirement contributions, payroll-tax obligations, and benefit costs that are calculated as proportions of salary. The operational cost of a prevailing-wage increase therefore may extend beyond the direct offered-wage increase itself. CUPA-HR’s analysis identifies affected higher-education Wage Level I filings with a median prevailing-wage shortfall of approximately $9,865 per filing per year under the proposed methodology. Because H-1B employment commonly operates through multi-year appointments and extensions, the cumulative compensation impact for a single sponsored research or faculty position may extend substantially beyond the initial filing year.

The interaction between increased prevailing-wage floors and the INA’s actual-wage requirement may also create broader compensation effects within institutional comparison groups. INA § 212(n)(1)(A) requires employers to pay H-1B workers at least the higher of the prevailing wage or the actual wage paid to similarly situated employees. Where revised prevailing-wage floors materially exceed existing compensation levels within institutional comparison groups, institutions may need to reevaluate compensation relationships among similarly situated employees for compliance and equity purposes. As a practical matter, compensation adjustments affecting H-1B incumbents may therefore extend beyond the directly sponsored employee population.

The NPRM’s Initial Regulatory Flexibility Analysis principally estimates forward-looking costs associated with future LCA filings. It does not separately analyze several additional categories of operational burden identified here, including positions already recruited and budgeted under the existing framework, downstream total-compensation effects associated with salary-linked benefits and payroll obligations, and broader compensation effects that may arise through interaction with the INA actual-wage requirement.⁵⁰

VI.B Multi-Year Research and Faculty Pipelines Are Not Fully Addressed by Prospective-Only Implementation

Higher education H-1B employment frequently operates through multi-year research and faculty-development pipelines. Doctoral graduates commonly enter postdoctoral or research-staff appointments before advancing into longer-term academic or tenure-track positions. These pathways often extend across many years and involve sequential institutional commitments to compensation, sponsorship, training, and research continuity.

The higher education H-1B population identified in Sections III and IV is concentrated within research and early-career academic positions where employees are more likely to require future H-1B extensions while remaining within the same institutional pipeline. Section III, Table III-4, for example, identifies 1,272 H-1B incumbents within the academic research-staff hierarchy under SOC 19-0000, with H-1B representation concentrated at earlier career stages. Section IV.A identifies a similar pattern among non-tenure-track research faculty.

The Department concludes that prospective-only implementation provides sufficient accommodation for reliance interests because the revised methodology would apply only to future filings rather than previously approved determinations. In higher education, however, employment and sponsorship commitments routinely extend beyond the validity period of an initial LCA or H-1B approval. H-1B status is commonly granted in three-year increments, while postdoctoral appointments, research positions, and tenure-track timelines frequently extend beyond that period.

As a result, institutions that recruited and sponsored employees under the existing wage framework may later be required to seek extensions under materially different prevailing-wage requirements while the employee remains in the same academic or research pipeline. A research scholar in the third year of a four-year appointment or an assistant professor midway through a tenure-track period may therefore face extension-related wage obligations that were not reasonably foreseeable when the original institutional commitment was made. The concentration of higher-education H-1B filings at Level I, discussed in Section III, further magnifies the significance of those transition effects within early-career academic populations.

Federally sponsored research grants, including grants administered through the National Institutes of Health and the National Science Foundation, are generally awarded against fixed project budgets that incorporate personnel costs at the compensation levels in effect at the time of award. Budget-adjustment authority during the grant period is often limited. Where revised prevailing-wage obligations materially increase personnel costs for in-position employees, institutions may need to redirect existing grant funds from other research activities, absorb the additional expense through institutional funding, or reduce the scope of supported research activity.

The same reliance interests extend beyond current H-1B employees to the broader international-student and research-training ecosystem on which U.S. higher education depends. U.S. colleges and universities hosted more than 1.17 million international students in academic year 2024–25, and international students made up approximately six percent of U.S. higher-education enrollment.⁵¹Post-graduation employment pathways are part of that ecosystem: international students increasingly consider opportunities for U.S.-based professional experience when deciding where to study.⁵²Abrupt increases in prevailing-wage requirements for entry-level specialty occupations may therefore affect not only current research and faculty pipelines, but also the attractiveness of U.S. higher education to prospective international students and scholars.

Although the Department retained the existing alternative-survey framework under Alternative 2, that decision does not fully address the reliance and transition concerns discussed above. The disruptions identified in this Section affect positions, employees, grants, and institutional commitments already structured under the existing prevailing-wage framework, including circumstances in which institutions may not separately obtain individualized survey determinations. On the record presently before the Department, those reliance interests support the sector-specific relief requests set out at the beginning of this comment: either exemption of higher education pending development of sector-appropriate analysis and methodology, or, in the alternative, delayed implementation for higher-education-filed LCAs and related applications.

VII. Conclusion.

The proposed rule substantially recalibrates the prevailing-wage structure governing the H-1B, H-1B1, E-3, and PERM programs based principally on labor-market assumptions and empirical patterns drawn from sectors outside higher education. The record developed in this comment demonstrates that higher education operates within a materially different labor market characterized by distinct occupational composition, career-stage employment structures, compensation systems, research staffing models, and statutory treatment under the INA. The Department’s NPRM does not separately analyze that labor market or adequately address the sector-specific evidence presented here and in CUPA-HR’s prior comments.

We therefore urge the Department to exempt institutions of higher education from the proposed rule pending development of sector-specific analysis and methodology appropriate to the academic labor market that is published in the Federal Register with a notice and comment period. Congress has already recognized higher education as a distinct labor market within the H-1B framework through both ACWIA’s prevailing-wage provisions at 8 U.S.C. § 1182(p)(1) and the cap exemption at 8 U.S.C. § 1184(g)(5)(A). Until the Department develops methodology that adequately reflects this sector, the proposed percentile changes should not be applied to higher-education employers.

If the Department declines that request, we urge the Department, at minimum, to delay application of the revised percentile structure to higher-education-filed LCAs and PERM applications for at least two years. In considering alternatives to the proposed rule, the Department separately considered delaying implementation for small businesses and nonprofits for two years in light of transition burdens and increased labor costs. The Department declined to adopt that approach because the proposal would apply prospectively only to new LCAs and PERM applications and because, in the Department’s view, delay would permit continued use of prevailing wage standards set below appropriate market levels.⁵³ The higher-education record developed in this comment does not reflect the same compensation patterns identified elsewhere in the NPRM. CUPA-HR survey data indicate that H-1B faculty and research employees in higher education are generally compensated at parity with similarly situated non-H-1B employees, and the record overall reflects labor-market dynamics materially different from those discussed in the Department’s broader analysis. At the same time, the record demonstrates substantial reliance interests associated with academic hiring cycles, multi-year research and faculty-development pipelines, federally funded grants, clinical staffing obligations, and long-term institutional workforce planning. A delayed implementation period would allow institutions to manage those transition burdens in an orderly manner while adapting to materially increased prevailing wage obligations.

In any final rule, we ask the Department to preserve the existing authority under 20 CFR 656.40(b)(3) and (g) permitting employer-provided alternative wage surveys.⁵⁴ The alternative-survey framework remains the practical mechanism by which higher-education-specific compensation data, calibrated to the position-specific level at which academic labor markets operate, remains available to higher-education employers making prevailing-wage determinations. We support the Department’s decision not to adopt Alternative 2, which would have eliminated the use of private wage surveys.

We appreciate the Department’s consideration of this comment and welcome the opportunity to discuss any aspect of the record submitted here.

Sincerely,

Basil Thomson, CUPA-HR Government Relations 

On behalf of:

American Association of Community Colleges
American Association of State Colleges and Universities
American Association of Veterinary Medical Colleges
American Council on Education
American Council of Learned Societies
Association of American Universities
Association of Governing Boards of Universities and Colleges (AGB)
Association of Public and Land-grant Universities
College and University Professional Association for Human Resources
Council for Advancement and Support of Education
Council for Christian Colleges & Universities
Council of Graduate Schools
EDUCAUSE
NAFSA: Association of International Educators
National Association of College and University Business Officers
National Association of Independent Colleges and Universities

1 91 Federal Register 15454 (Mar. 27, 2026), pp. 15454-15501.

2 CUPA-HR’s research team conducts four annual Higher Education Surveys — covering Faculty, Administrators, Professionals, and Staff — collecting incumbent-level salary, position, and employment-status data from more than 1,000 participating non-profit U.S. colleges and universities. Current cycle coverage is approximately 270,000 faculty and 500,000 administrators, professionals, and staff. Survey Methodology at https://www.cupahr.org/research-data/about-our-surveys/.

3 91 Fed. Reg. 15490 (Department’s acknowledgment that ACWIA-covered prevailing-wage calculations are “required by statute”).

4 Comment of CUPA-HR et al. on Department of Labor Interim Final Rule, Strengthening Wage Protections for the Temporary and Permanent Employment of Certain Aliens in the United States, DOL Docket No. ETA-2020-0006 (Nov. 9, 2020), available at https://www.regulations.gov/comment/ETA-2020-0006-2134.

5 Comment of CUPA-HR et al. on Department of Labor Request for Information, Determining Prevailing Wage Levels, Docket No. ETA-2021-0003 (June 1, 2021), available at https://www.regulations.gov/comment/ETA-2021-0003-0076.

6 INA § 212(p)(4), codified at 8 U.S.C. § 1182(p)(4); U.S. Department of Labor, Employment and Training Administration, Prevailing Wage Determination Policy Guidance, Nonagricultural Immigration Programs (rev. Nov. 2009).

7 85 Fed. Reg. 63872 (Oct. 8, 2020).

8 Chamber of Commerce v. DHS, No. 4:20-cv-07331 (N.D. Cal. Dec. 1, 2020); Purdue University v. Scalia, No. 1:20-cv-03006 (D.D.C. Dec. 14, 2020); ITServe Alliance v. Scalia, No. 1:20-cv-14604 (D.N.J. Dec. 3, 2020).

9 86 Fed. Reg. 3608 (Jan. 14, 2021).

10 86 Fed. Reg. 26164 (May 14, 2021).

11 Chamber of Commerce v. DHS, No. 4:20-cv-07331 (N.D. Cal. June 23, 2021); 86 Fed. Reg. 70683 (Dec. 13,

2021).

12 Proclamation 10973, 90 Fed. Reg. 45717 (Sept. 19, 2025).

13 91 Fed. Reg. 15462–66 (qualifications-mismatch reasoning); 91 Fed. Reg. 15467 (63 percent of FY 2024 LCAs assigned to Levels I or II; certified LCA wages averaging approximately $10,191 below OEWS averages); 91 Fed. Reg. 15462 (“approximately 80 percent” of program workers are H-1B, supporting the Department’s analytical focus on the H-1B program and uniform application across H-1B1, E-3, and PERM).

14 Term-search results from the Federal Register text, 91 Fed. Reg. 15454–15501. “ACWIA” appears once (Alternative 1, 91 Fed. Reg. 15490); “non-profit” appears once (rejected delayed-implementation alternative, 91 Fed. Reg. 15499); “higher education”, “university”, “college”, and “research institution” do not appear as sector or employer categories.

15 91 Fed. Reg. 15471 n. 162.

16 8 U.S.C. § 1182(p)(1); 20 U.S.C. § 1001(a) (defining “institution of higher education”); 91 Fed. Reg. 15490

17 91 Fed. Reg. 15462–67 (discussing the Department’s rationale for revising prevailing wage levels, including concerns regarding wage suppression, labor substitution, outsourcing-firm practices, and employer incentives in the H-1B labor market).

18 91 Fed. Reg. 15495 (Initial Regulatory Flexibility Analysis, Exhibit 10).

19 91 Fed. Reg. 15481.

20 The cost-benefit analysis at 91 Fed. Reg. 15479–88 quantifies the rule familiarization cost and the wage-floor transfer payments; it does not aggregate the $100,000 fee against those figures.

21 CUPA-HR 2025-26 data were collected in CUPA-HR’s Higher Education Annual Surveys with an effective date of November 1, 2025. Analyses include only non-profit institutions of higher education. Survey Methodology at https://www.cupahr.org/research-data/about-our-surveys/.

22 CUPA-HR analysis of U.S. Department of Labor, Office of Foreign Labor Certification, LCA Disclosure Data, FY2020 Q1 – FY2025 Q4. Higher education = NAICS 611210 + 611310. Certified cases with annual prevailing wages. See Attachment A, Table 1.

23 Across the five fiscal years for which complete year-end data are available (FY2020 through FY2024), the Level I share of certified HE H-1B LCAs ranges from 58.1% (FY2022) to 61.1% (FY2024), consistent with the six-year aggregate reported in Table III-1. The pattern predates the current rulemaking and is not an artifact of any single year.

24 Ten postsecondary teaching SOCs recorded zero non-higher-education certified Level I H-1B filings between FY2020 and FY2025: Political Science Teachers, Postsecondary (25-1065; 704 HE filings); Communications Teachers, Postsecondary (25-1122; 697); Area, Ethnic, and Cultural Studies Teachers, Postsecondary (25-1062; 424); Recreation and Fitness Studies Teachers, Postsecondary (25-1193; 313); Sociology Teachers, Postsecondary (25-1067; 284); Environmental Science Teachers, Postsecondary (25-1053; 248); Geography Teachers, Postsecondary (25-1064; 212); Anthropology and Archeology Teachers, Postsecondary (25-1061; 199); Social Work Teachers, Postsecondary (25-1113; 153); and Criminal Justice and Law Enforcement Teachers, Postsecondary (25-1111; 119).

25 91 Fed. Reg. 15471 (defining the “Benchmark Value” as the average OEWS mean across the occupational composition of the LCA population as a whole). 

26 The research-staff hierarchy corresponds to credentialed positions under SOC 19-0000 (Life, Physical, and Social Science Occupations) in CUPA-HR’s coding. The functional description of postdoctoral roles in this paragraph reflects standard NIH and NSF descriptions of postdoctoral training appointments.

27 National Center for Science and Engineering Statistics, Survey of Graduate Students and Postdoctorates in Science and Engineering (NCSES 25-316), available at https://ncses.nsf.gov/pubs/nsf25316.

28 College and University Professional Association for Human Resources (CUPA-HR), Data on H-1B Status for Faculty and Professionals (Sept. 29, 2025), https://www.cupahr.org/resource/data-on-h-1b-status-for-faculty-and-professionals/

29 CUPA-HR Faculty in Higher Education Survey, 2025-26.

30 The 4.0 percent figure represents the H-1B share of all surveyed faculty for whom the participating institution reported H-1B status.

31 91 Fed. Reg. 15470 n. 162 (the Department “has no reason to assume that the distribution of cap-exempt employees is different from the distribution of non-cap-exempt LCAs,” inviting comment “if any member of the public has data that indicate a significant discrepancy”).

32 The research-staff hierarchy houses 1,272 H-1B incumbents, the largest single H-1B-incumbent population in CUPA-HR’s higher-education Professionals universe.

33 Counts drawn from the discipline-level cells for Tenure-Track and Non-Tenure-Track only. Across the two categories combined, the eight disciplines listed in Table IV-2 contain 5,671 of the 8,276 H-1B faculty whose discipline was reported in those two categories (68.5 percent).

34 20 CFR 655.731(a)(1) (defining the actual wage by reference to workers “with similar experience and qualifications for the specific employment in question”); 20 CFR 655.731(a)(2)(i) (defining the prevailing wage by reference to workers “similarly employed” in the “occupational classification in the area of intended employment”).

35 91 Fed. Reg. 15471

36 8 U.S.C. § 1182(n)(1)(A); 20 CFR 655.731(a).

37 The offered-vs-prevailing-wage gap is computed at the case level as (offered wage − prevailing wage) ÷ prevailing wage; the figure is the median of the per-case gap values across certified H-1B LCAs at Wage Level I, FY2020 Q1 – FY2025 Q4. Methodology at Appendix A.

38 91 Fed. Reg. 15464–65 (arguing that “only a portion of the workers covered by many of the occupational classifications used in the OEWS survey likely have levels of education and experience similar to those of H–1B workers” and that “it would be inappropriate to consider the wages of the least educated and experienced workers” in setting the Level I floor).

3991 Fed. Reg. 15464–65 (Software Developers as 32 percent of all certified LCAs).

40 CUPA-HR analysis of OFLC H–1B LCA Disclosure Data, FY2020 Q1 – FY2025 Q4.

41 91 Fed. Reg. 15479 (the Department “is particularly interested” in comment on, among other items, whether Levels I and IV “appropriately approximate” the education and experience requirements of the occupation).

42 CUPA-HR analysis of DOL OFLC Wage Data Tables, EDC_Export.csv (ACWIA) and ALC_Export.csv (All Industries), Wage Years 2020-21 through 2024-25. Geo Level field as defined in the Department’s Technical Release Notes, July 2024 Wage Year. Methodology at Appendix A.

43 91 Fed. Reg. 15486 (in the Transfer Payments calculation, observing that “when a 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,” and characterizing the resulting fallback as affecting “a small number of certifications that cannot be matched to a new prevailing wage level”).

44 91 Fed. Reg. 15491–92, 15498 (the Department’s evaluation of Alternative 2 — eliminating the option for employers to use private wage surveys — and the Department’s retention decision).

45 Comparison of Department of Labor OEWS Wage Guidelines and CUPA-HR Survey Procedures (Nov. 20, 2025) at https://www.cupahr.org/wp-content/uploads/surveys/CUPA-HR-Surveys-Per-DOL-OES-Guidelines.pdf.

46 91 Fed. Reg. 15492 n. 211 (“The 20% figure represents the average percentage difference between Survey-based and OES (Occupational Employment Statistics) prevailing wages across five key SOC codes (15-1121, 15-1132, 15-1199, 15-1252, 15-1299) in FY 2024 H-1B LCA data.

47 91 Fed. Reg. 15478–79; 91 Fed. Reg. 15478.

48 CUPA-HR Nov. 2020 IFR Comment, at 3.

49 Association of American Medical Colleges, The Complexities of Physician Supply and Demand: Projections From 2021 to 2036 (Mar. 2024)

50 91 Fed. Reg. 15479–84 (Initial Regulatory Flexibility Analysis).

51 Institute of International Education, Open Doors 2025 Report on International Educational Exchange (2025), reporting 1,177,766 international students enrolled at U.S. institutions during academic year 2024–25, representing approximately 6 percent of total U.S. higher-education enrollment, available at https://www.iie.org/research-initiatives/open-doors/

52 See NAFSA: Association of International Educators, summarizing 2025 IDP survey data reporting that opportunities for post-graduation employment and professional experience are significant factors in international student enrollment decisions, available at https://www.nafsa.org/OPT

53 91 Fed. Reg. 15499

54 91 Fed. Reg. 15489 (Alternative 2).

Download the PDF