Comment on NPRM Removing Deference to Prior USCIS Determinations of Eligibility

Note: footnotes are provided in the attached document

November 13, 2026

U.S. Citizenship and Immigration Services
Office of Policy and Strategy
Regulatory Coordination Division
5900 Capital Gateway Drive
Camp Springs, MD 20746

Re: Comment on "Removing Deference to Prior USCIS Determinations of Eligibility," 91 Fed. Reg. 65119 (Oct. 14, 2026), DHS Docket No. USCIS-2026- 0529, RIN 1615-AD26

Dear Officer:

The U.S. Chamber of Commerce (“Chamber”) respectfully submits the following comment to the Department of Homeland Security's (DHS) notice of proposed rulemaking (NPRM) to rescind 8 C.F.R. § 214.1(c)(5), the regulation authorizing USCIS's deference policy for extension petitions filed on Form I-129.1. The current framework has enabled American employers to make long-term investments, expand operations, and create American jobs by providing predictable rules for hiring and retaining high-skilled talent. Preserving that stability is essential to maintaining a strong talent pipeline, supporting economic growth, and ensuring that U.S. businesses can compete for the world's best talent.

Discussion
I. The Deference Standard Has a Twenty-Year Pedigree
The “deference” standard originates from an April 2004 USCIS memorandum instructing immigration officers to defer to a prior approval involving the same parties and facts absent material error, a material change in circumstances, or new material information. The policy was eventually codified by regulation in 2024.2 The NPRM would eliminate that instruction.3 Deference has thus governed the great majority of extension adjudications for two decades, with only a three-year interruption whose consequences are documented below.

That codification followed full notice-and-comment rulemaking in which DHS concluded that deference "does not equate to a lack of USCIS review" and that petitioners "continue to have the burden to present all required and relevant evidence."

II. The NPRM's Legal Rationale Does Not Withstand Scrutiny
The NPRM describes deference as being "in tension" with the Immigration and Nationality Act (“INA”) § 291 and caselaw.5 However, the 2024 rule DHS now seeks to rescind cited the identical authority and found deference fully consistent with it. INA § 291 does not require an officer to begin an extension review with a blank slate and does not prohibit referring to prior, unrebutted factual findings alongside new evidence. The deference standard never relieved petitioners of the burden of proof but requires immigration officers not to disregard a verified record absent a reason to doubt it. It is no different from the reasonable evidentiary presumptions the Supreme Court has sanctioned where consistent with the statutory scheme. The NPRM itself concedes uncertainty, stating that "[w]hether or not the deference regulation is consonant with the statute, DHS now believes that it has the practical effect of shifting the burden of proof."

The cited statutory authorities, INA §§ 103(a), 214(a)(1), (c)(1), 235(d)(3), 287(b) and Homeland Security Act §§ 402, 451,10establish only that DHS has broad rulemaking and fact-finding discretion; none requires, or even suggests, that deference to an unchanged factual finding is unlawful. DHS may exercise its discretion differently than it did in 2024, but only through reasoned decision-making supported by the record.

III. DHS's Factual Justifications Are Contradicted by the Agency's Published Data
A. The NPRM would drive up denials and RFEs, but not fraud detection. 

Published USCIS data show that after the October 2017 memorandum ended the deference standard, the H-1B denial rate for continuing employment rose 140%, from 5% in FY 2017 to 12% in FY 2018 and FY 2019, falling back to 2% in FY 2021 once deference was reinstated.11 RFE rates moved in lockstep: the share of H-1B completions receiving an RFE rose from 21.4% in FY 2017 to 38% in FY 2018 and 40.2% in FY 2019, a 78–88% increase in the two fiscal years after rescission.12 No other material change in adjudication standards, statutory requirements, or applicant composition occurred during these periods that would otherwise explain this rise and subsequent fall in denial and RFE rates; the only intervening variable was the deference policy itself.

Rather than address these data, DHS relies on an internal difference-in-differences analysis claiming denial rates "did not statistically differ" across the two periods and that RFE rates rose only 2.9 percentage points relative to a comparison group, contradicting the agency's own published statistics, a conflict DHS has not reconciled.

B. The site-visit fraud statistics DHS cites do not show what DHS claims. 

The NPRM separately points to higher combined non-compliance and fraud rates at H-1B (22.39% vs. 16.77%) and L-1A (24.46% vs. 18.37%) worksites during the years with the deference standard in effect (FY 2022–2025) compared to the no-deference years (FY 2018–2020). DHS itself acknowledges that "other factors besides or in addition to deference may have led to this higher percentage," and offers no analysis isolating deference as the cause rather than, for example, changes in site-visit targeting, post-pandemic labor-market shifts, or sponsorship volume. A bare correlation between two multi-year periods affected by numerous confounding variables cannot justify rescinding a codified rule.

IV. The Proposal Undervalues Reliance Interests and Reverses DHS's 2024
Reasoning Without New Evidence: The NPRM claims employers and workers "do not have a legitimate expectation that an erroneous or unsupported prior approval would continue to result in future approvals," but this mischaracterizes the interest at stake: reliance centers on correct approvals, the overwhelming majority of extensions, involving no material change in facts, not erroneous ones. Long-tenured workers who have built careers and families around stable employment are not perpetuating fraud; they seek the predictability the 2021 policy and 2024 rule were designed to provide, an interest DHS's own 2024 record recognized as legitimate. The NPRM dismisses this in a single paragraph, without quantifying the cost to workers, families, or employer workforce planning. 

Caselaw requires an agency to give "a reasoned explanation . . . for disregarding facts and circumstances that underlay or were engendered by the prior policy." DHS offers only that it has "reconsidered," without new data, a changed statutory landscape, or an intervening judicial decision—an inadequate basis for reversing a rule adopted through full notice-and-comment rulemaking barely two years earlier.

Conclusion
The deference standard DHS proposes to eliminate has governed extension adjudications for nearly two decades, was adopted and twice reaffirmed, and is consistent with the applicable burden of proof. DHS's own historical data show that removing deference sharply raises denial and RFE rates, a conflict with its internal analysis that DHS has not reconciled—a failure to consider relevant evidence that renders the rulemaking arbitrary and capricious. The fraud statistics DHS offers are merely correlational and do not establish causation. The U.S. Chamber urges DHS to withdraw the proposed rescission, or at minimum adopt the discretionary, case-by-case alternative it considered and rejected, rather than eliminating deference entirely.
 

Respectfully submitted,

Patrick Shen
Vice President, Immigration Policy
U.S. Chamber of Commerce

About the author

Patrick Shen

Patrick Shen is vice president for Immigration Policy at the U.S. Chamber of Commerce.

Read more