By Cyrus D. Mehta and Manjeeta Chowdhary

A federal district court in Nebraska has reopened a consequential debate over how U.S. Citizenship and Immigration Services adjudicates EB-1A extraordinary ability petitions. In Agarwal v. Miller, Chief District Judge Brian C. Buescher dismissed a challenge to USCIS’s “final merits determination,” declining to follow the earlier decision in Mukherji v. Miller, which had invalidated that second step of the agency’s analysis.

 

The result is striking. Two judges in the same district have reached opposing conclusions about an adjudicatory framework that is not expressly set out in the Immigration and Nationality Act or its implementing regulations. The disagreement matters not only to EB-1A petitioners, but also to the broader question of how much interpretive latitude agencies retain after Loper Bright Enterprises v. Raimondo.

The EB-1A Framework and the Contested Second Step

Congress reserved the EB-1A classification for individuals of extraordinary ability in the sciences, arts, education, business, or athletics whose ability has been demonstrated by sustained national or international acclaim and whose achievements have been recognized through extensive documentation. See INA § 203(b)(1)(A).  The regulations at 8 C.F.R. § 204.5(h)(3)(i)-(x) permit a petitioner to establish eligibility through a major, internationally recognized award or by satisfying at least three of ten enumerated evidentiary criteria.

Following the Ninth Circuit’s decision in Kazarian v. USCIS, the agency developed a two-step framework. At step one, the adjudicator decides whether the evidence meets at least three regulatory criteria. At step two—the “final merits determination”—the adjudicator evaluates the record as a whole to decide whether the beneficiary possesses the required high level of expertise and sustained acclaim.

The controversy arises because neither the statutory text nor the regulation expressly labels or describes this second step. In practice, petitioners can satisfy three, four, or even five criteria and still be denied because USCIS concludes that the evidence, viewed in its totality, does not show that the beneficiary is among the small percentage at the very top of the field.

Mukherji’s Challenge to the Final Merits Determination

In Mukherji v. Miller, another judge in the District of Nebraska set aside an EB-1A denial after USCIS acknowledged that the petitioner satisfied five regulatory criteria. The court concluded that the agency’s two-step framework had not been validly adopted through notice-and-comment rulemaking, that USCIS had not adequately explained the policy change, and that the legality of the final merits determination presented a question of law for independent judicial judgment after Loper Bright.

USCIS initially appealed but later withdrew the appeal. That left Mukherji intact for the individual petitioner without creating binding appellate precedent. It also left unresolved whether other courts would accept its reasoning.Agarwal Takes the Opposite Path

Agarwal arose from a materially similar challenge. The plaintiff attacked USCIS’s denial of an EB-1A petition and the agency’s reliance on the Kazarian two-step policy. Judge Buescher rejected the challenge and dismissed the complaint with prejudice:

“In these circumstances, even though Agarwal prefers the post-Loper Bright Enterprises decision in Mukherji v. Miller, No. 4:24CV3170, 2026 WL 895708 (D. Neb. Jan. 28, 2026), the Court declines to follow Mukherji because it is unsound and unconvincing.”

The court relying on the Fifth Circuit decision in Amin v. Mayorkas treated the final merits inquiry not as an unlawful extra-regulatory hurdle, but as a permissible means of deciding whether the petitioner ultimately satisfied the statutory requirement of extraordinary ability demonstrated by sustained acclaim. On this view, meeting three regulatory criteria is necessary when there is no qualifying one-time achievement, but it is not automatically sufficient. The enumerated criteria identify types of evidence; they do not displace the statute’s ultimate eligibility requirements.

According to Amin, which was decided before Loper Bright,  the statute does not merely require a petitioner to accumulate three categories of evidence. It requires extraordinary ability, sustained national or international acclaim, recognition through extensive documentation, an intention to continue working in the area of expertise, and prospective substantial benefit to the United States. A court can therefore reason that USCIS must evaluate whether the evidence proves those ultimate statutory elements. But this does not fully answer the objection to the agency’s final merits framework. There is a meaningful difference between applying statutory elements and imposing a separate, binding adjudicatory rule with elastic standards untethered to the regulatory text. The central question is not whether USCIS may consider the quality and probative value of evidence. Of course it may. The harder question is whether the agency may use an administratively created second step to discount evidence that has already satisfied the regulation, particularly when that step has become the principal basis for denial.

Corner Post and Agarwal’s Six-Year Limitations Analysis

Agarwal also raises an important threshold issue under 28 U.S.C. § 2401(a), the default six-year statute of limitations for civil actions against the United States. USCIS issued the Kazarian analysis in 2010, and the government argued that any challenge to it became time-barred in 2016. Judge Buescher did not accept that position. Consistent with the Supreme Court’s decision in Corner Post, Inc. v. Board of Governors of the Federal Reserve System, Agarwal treated the limitations period as running from the agency’s denial of the petitioner’s EB-1 petition.

In Corner Post, the Supreme Court held that an Administrative Procedure Act claim does not accrue under § 2401(a) until the plaintiff is injured by final agency action. A right of action accrues when the plaintiff has a “complete and present cause of action”—that is, when the plaintiff can file suit and obtain relief. Thus, the six-year clock does not necessarily begin when an agency first issues a rule or policy. It begins when that agency action causes injury to the particular plaintiff.

Applied to Agarwal, the relevant injury was USCIS’s final denial of the EB-1 petition using the Kazarian analysis, not the agency’s issuance of that analysis in 2010. Before the denial, Agarwal could not show the concrete injury for which relief was sought. Because the action was brought within six years of the final denial, the claim was not barred under § 2401(a), even though more than six years had passed since 2010.

This aspect of Agarwal is therefore favorable to EB-1 petitioners. The court rejected the government’s attempt to turn § 2401(a) into a statute of repose that would categorically insulate the Kazarian analysis from challenge after 2016. Under Corner Post, the provision is a statute of limitations measured from accrual of the particular plaintiff’s right of action. The distinction is especially compelling in immigration adjudication. A prospective EB-1 petitioner may not have filed a petition, received a final decision, or even possessed a claim when USCIS adopted the Kazarian analysis in 2010. Starting and ending the limitations period in 2010 and 2016 would allow the claim to expire before the policy was ever applied to that petitioner. Agarwal correctly recognized that the six-year period begins with the final denial that first injures the petitioner.

Why Loper Bright Still Matters

Agarwal should not be read as restoring broad judicial deference to USCIS. Loper Bright requires courts to exercise independent judgment on questions of statutory meaning. An agency’s interpretation may persuade, but it does not control merely because the statute is ambiguous.

The post-Loper Bright inquiry should therefore focus on whether the INA and the regulation, properly interpreted, authorize the precise work that USCIS asks the final merits determination to perform. If the second step simply assesses whether the evidence proves sustained acclaim and top-of-field status, it may be characterized as application of the statutory standard. If it adds new requirements—such as indefinite preeminence, uninterrupted recognition, or evidentiary expectations not found in the statute or regulation—it remains vulnerable to challenge.

This distinction is critical. Judicial review cannot stop just because the USCIS has developed a  “final merits determination” without notice and comment.  Courts must examine how USCIS actually used the framework in the case before them. An agency may not evade meaningful review by describing an extra-textual requirement as a holistic assessment.

Kisor v. Wilkie Does Not Rescue the Final Merits Determination

Kisor v. Wilkie is also relevant because it sharply limits deference when an agency claims to be interpreting its own regulation. Under Kisor, a court must first exhaust the traditional tools of construction and find the regulation genuinely ambiguous. Even then, the agency’s interpretation must be reasonable, authoritative, within its substantive expertise, and the product of its fair and considered judgment rather than a convenient litigating position or an unfair surprise.

The final merits determination does not comfortably fit that framework. USCIS is not interpreting an ambiguous term in the EB-1 evidentiary regulations when it adds a second evaluative stage after determining that the prescribed criteria have been satisfied. The EB-1A regulation states that evidence of a qualifying one-time achievement or at least three of ten listed criteria is required; the EB-1B regulation identifies at least two of six. Final merits does not resolve the meaning of a word or phrase within those criteria. It overlays an additional decisional structure on the regulations.

That distinction matters. Kisor concerns how an agency may interpret its own genuinely ambiguous regulation; it does not authorize an agency to supplement an unambiguous regulatory scheme with a new substantive test. Calling final merits an “interpretation” cannot substitute for identifying the precise regulatory ambiguity being construed. If USCIS cannot point to genuinely ambiguous text that the second step interprets, there is no basis for Kisor deference.

Even if some regulatory language were genuinely ambiguous, a court would still have to decide independently whether the final merits framework falls within the bounds of a reasonable interpretation and reflects fair and considered judgment. A framework that permits adjudicators to demand indefinite preeminence, uninterrupted acclaim, or other proof not specified in the regulation risks creating precisely the unfair surprise that Kisor cautions against.

Loper Bright and Kisor therefore address complementary problems. Loper Bright requires independent judicial judgment when an agency interprets a statute. Kisor confines any deference when an agency interprets its own regulation. Here, USCIS’s more fundamental difficulty is that final merits appears not to interpret the EB-1 regulation at all, but to add a second substantive hurdle that the regulatory text does not contain.

Exceeding the Minimum Criteria Must Count for Something

The final merits analysis should also account for how far a petitioner has surpassed the regulatory threshold. For EB-1A, the regulation generally requires three out of ten criteria; for EB-1B outstanding professors and researchers, it generally requires two of six. If USCIS agrees that an EB-1A petitioner has satisfied more than three criteria, or that an EB-1B petitioner has satisfied more than two, those findings should bolster rather than disappear from the agency’s ultimate assessment.

This is consistent with the concern addressed in Amin v. Mayorkas. Although satisfying the minimum number of criteria does not mechanically compel approval, the criteria cannot be treated as a meaningless counting exercise. Each additional satisfied criterion adds probative weight to the record. USCIS must rationally explain why evidence sufficient to meet multiple regulatory categories nevertheless fails to establish extraordinary ability or outstanding achievement when considered collectively.

A final merits determination becomes vulnerable when it acknowledges that a petitioner has substantially exceeded the threshold but then repeats generalized conclusions about lacking sustained acclaim or top-of-field standing. The agency must grapple with its own favorable findings. Otherwise, step two risks becoming an undefined veto that nullifies step one rather than a reasoned evaluation of the same evidence.

Scripps Remains a Useful Path

Scripps College v. Jaddou offers a more nuanced way to understand—and constrain—the final merits determination. Scripps College sought EB-1B classification for art historian Julia Lum as an outstanding professor or researcher. USCIS agreed that Dr. Lum satisfied three of the six regulatory criteria: judging the work of others, making original scholarly contributions, and authoring scholarly books or articles. The regulation required only two. USCIS nevertheless denied the petition at final merits.

The District of Nebraska did not hold that satisfying the threshold criteria automatically compelled approval, nor did it invalidate final merits review wholesale. Instead, it applied ordinary APA review and asked whether USCIS had engaged in reasoned decision-making. The court concluded that the denial was arbitrary and capricious, granted summary judgment to Scripps College, and remanded the matter.

The value of Scripps lies in its insistence that final merits cannot become an opportunity for USCIS to disregard its own favorable findings. Once the agency determined that Dr. Lum met three criteria—one more than the regulatory minimum—those findings formed part of the evidentiary record that USCIS had to evaluate rationally. The agency could consider the quality of the evidence, but it could not treat the satisfied criteria as though they had no cumulative significance.

Scripps thus preserves a meaningful distinction between qualitative review and impermissible re-adjudication. At step one, a criterion may itself contain a qualitative component, and USCIS may assess whether the evidence meets it by a preponderance of the evidence. At final merits, the agency may consider the evidence together to decide whether the statutory standard is met. But it may not quietly heighten the criteria, demand proof that the regulation does not require, or use generalized conclusions to negate findings already made at step one.

This reasoning complements Amin v. Mayorkas. Meeting the minimum number of criteria is not necessarily dispositive, but exceeding the minimum should carry additional evidentiary force. If three satisfied criteria in an EB-1B case are treated no differently from two—or if five satisfied criteria in an EB-1A case are treated no differently from three—the regulatory framework risks becoming largely ornamental. USCIS must explain why the cumulative evidence, including every criterion it found satisfied, still falls short.

Scripps remains especially useful after Agarwal. Even if a court accepts the legality of a two-step framework in principle, a petitioner can still prevail by showing that USCIS applied final merits arbitrarily in the individual case. The litigation should focus on contradictions between the agency’s step-one findings and its final-merits conclusions, failures to address material evidence, extra-regulatory demands, and explanations that run counter to the record.

In this sense, Scripps supplies a narrower but highly practical path. Petitioners need not win the broader argument that final merits is invalid in every case. They can argue that USCIS must conduct the analysis transparently, consistently, and under the preponderance standard—and that a reviewing court should set aside a denial when the agency uses final merits as an unexplained veto over evidence it has already credited.

A Conflict That Invites Circuit Court Resolution

Agarwal is a setback for efforts to dismantle the final merits determination, but it is not the last word. Nor does it erase Mukherji. Instead, the decisions frame the dispute more precisely: where does legitimate evaluation of the statutory standard end, and where does an unlawfully created additional eligibility test begin?

The answer should not turn on whether USCIS calls its analysis “final merits,” “totality of the evidence,” or something else. It should turn on statutory and regulatory text, the procedures through which the agency adopted its framework, and the reasons the agency actually gave for denying the petition. After Loper Bright, courts and not the  USCIS must independently determine the boundaries of INA 203(b)(1)(A) and INA 203(b)(1)(A), and determine whether how much deference to give to the USCIS’s invented final merits determination. Amin v. Mayorkas was decided before Loper Bright. 

Until a circuit court provides authoritative guidance post Loper Bright, EB-1 petitioners should continue to build records capable of surviving USCIS’s two-step review while preserving four complementary arguments. First, Agarwal correctly applied Corner Post and 28 U.S.C. § 2401(a) in rejecting the government’s contention that a challenge to the 2010 Kazarian analysis expired in 2016; the six-year period runs from the final EB-1 denial that first injures the petitioner. Second, under Loper Bright and Kisor, courts must independently determine the governing law, and USCIS cannot obtain deference merely by labeling final merits an interpretation when it does not identify genuinely ambiguous regulatory text that the second step construes. Third, under Amin and Scripps, criteria satisfied beyond the regulatory minimum must carry cumulative evidentiary weight in final merits. Fourth, Scripps College v. Jaddou permits a court to set aside an arbitrary application of final merits when USCIS disregards favorable findings, imposes extra-regulatory demands, or offers an explanation that runs counter to the record—even without invalidating the entire framework. The clash between Mukherji and Agarwal ensures that these issues will remain central to extraordinary ability and outstanding researcher litigation.

Even if a Circuit court rules against the final merits determination, it will not be binding outside the circuit court’s jurisdiction but it will at least provide persuasive authority in those cases until other circuit courts also rule the same way. There is a risk that a  circuit court may also rule the other way and keep the final merits determination intact. Then another plaintiff may have to obtain a decision from another circuit court ruling against the final merits determination, and if there is such a circuit split, the Supreme Court may have the last word. Before we get there, it is important for petitioners to use Loper Bright to get rid of the final merits determination that Mukherji did in as many cases as possible at the district court level.