Think Immigration: One Day Before Taking Effect, Federal Court Postpones the F, J, and I Fixed Admission Period Rule

On September 14, 2026, one day before the rule was set to take effect, Judge F. Dennis Saylor IV of the District of Massachusetts postponed the effective date of the Department of Homeland Security’s (DHS's) Final Rule replacing duration of status with fixed periods of admission for F, J, and I nonimmigrants (international students). Presidents' Alliance on Higher Education and Immigration v. U.S. Department of Homeland Security, No. 1:26-cv-13799-FDS (D. Mass. Sept. 14, 2026). The relief is nationwide, and it reaches the whole rule.

The court found that the plaintiffs had demonstrated a substantial likelihood of success on the merits of their Administrative Procedure Act (APA) claim, that immediate irreparable harm would follow absent relief, and that the balance of equities and the public interest favored the issuing of an injunction. On the first and most important factor, it identified four separate respects in which the rulemaking was arbitrary and capricious.

For practitioners, the bottom line is straightforward: duration of status remains the operative framework for F, J, and I nonimmigrants, for now. The rule is postponed, not vacated, and the case continues on the merits.

What the Rule Would Have Done

The Final Rule, 91 Fed. Reg. 44976 (July 17, 2026), would have ended a framework that has governed F admissions since 1978 and J admissions since 1985. In its place:

• F and J nonimmigrants would be admitted for a fixed period of four years or the program end date, whichever came first, regardless of the actual length of the program.
• I nonimmigrants would be admitted for 240 days or the completion of the assignment, whichever came first.
• Anyone needing more time, which on the record before the court meant most doctoral students and many undergraduates, would file a Form I-539 extension of stay request. The grant or denial would be committed entirely to the discretion of an immigration officer and would be unappealable, even within DHS.
• The departure period would shrink from 60 days to 30 days.
• F nonimmigrants could not pursue a second degree at the same or a lower level, which as a practical matter would foreclose joint and dual degree programs.
• Graduate students could not change educational objectives or transfer institutions, subject only to an extenuating circumstances exception.

DHS quantified compliance costs at roughly $443 million to $449 million overall. It gave the public 32 days to comment, drew nearly 22,000 comments, and finalized the rule largely as proposed.

The Holding: Arbitrary and Capricious in Four Respects

The court was explicit that it was not invalidating the rule because it considered the policy unwise. It stated that it had no power to block the rule on the ground that it is misguided or ill-advised, that the government's candor in promulgating it was questionable, or that it may inflict substantial long-term damage on the United States and its educational institutions. The role of the judiciary, the court said, is not to ensure that the executive branch makes wise decisions, but to ensure that when an agency promulgates a regulation it adheres to the law.

On that question, the court found the rulemaking arbitrary and capricious in at least four respects. Each is worth reading closely, since the analysis tracks familiar State Farm principles and applies well beyond this rule. See Mot. Vehicle Mfrs. Ass’n of U.S. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983)).

1. No reasoned cost-benefit analysis. Commenters submitted an Institute for Progress and NAFSA survey finding that 49 percent of current international students would not have enrolled under a fixed period system, along with an analysis by economist Michael Clemens and immigration policy experts Jeremy Neufeld and Amy Nice projecting annual losses of $72 to $145 billion from a conservative 10 percent enrollment decline. DHS dismissed the surveys as reflecting "hypothetical preferences rather than actual behavior," offered no competing data or methodology, and treated an unquantified cost as effectively zero. Citing Business Roundtable v. SEC and GPA Midstream, the court held that an agency must either quantify costs or explain why it cannot, and that "unquantified factors . . . cannot be dismissed without further inquiry where their impact is both evident and massively significant." See Bus. Roundtable v. S.E.C., 647 F.3d 1144, 1148-49 (D.C. Cir. 2011); see also GPA Midstream Ass’n v. United States Dep’t of Transportation, 67 F.4th 1188, 1200 (D.C. Cir. 2023)
2. No meaningful consideration of alternatives. Commenters proposed upgrading SEVIS to create compliance checkpoints, tying any fixed term to actual program duration, and targeting enhanced screening at sensitive research fields and countries of concern. The court characterized DHS's SEVIS response as circular and its response on program-length alternatives as an ipse dixit. Notably, the court pointed out in a footnote that many of the security incidents DHS cited as evidence of the vulnerabilities of duration of status were themselves discovered through SEVIS.
3. No meaningful response to significant comments. The court catalogued seven areas where DHS responded with statements of belief rather than analysis: enrollment effects, less burdensome alternatives, doctoral completion times, the assumption that students always progress upward through degree levels, reliance interests, USCIS backlogs, and press freedom. On completion times, the court noted that DHS relied on a Department of Education webpage asserting four years for a doctorate, against comment evidence that the median is 5.7 years and 7.3 years for a combined master's and doctoral sequence. Quoting Mayor of Baltimore v. Azar, the court observed that an agency cannot simply say it believes something against the weight of all the evidence before it.
4. No rational connection to the stated objectives. DHS justified the rule on national security and fraud grounds, citing a set of specific incidents. Applying rational basis review under Trump v. Hawaii, the court found the connection to the four-year cap unexplained: the conduct described in the cited examples could be accomplished within a four-year window, or on a six-month B-1, and the Russian operative identified in the rule had been present for two years. On the roughly 77,000 F-1 students who have exceeded ten years in status since 2003, the court noted that DHS never explained why a duration consistent with a doctoral program followed by post-completion training is inherently suspect. And as to the 240-day cap on I nonimmigrants, DHS offered no security or fraud evidence at all.

The court reserved the First Amendment claim and the claim that the academic mobility restrictions are contrary to 8 U.S.C. § 1101(a)(15)(F), reaching neither because the APA holding disposed of the motion.

Irreparable Harm and the Scope of Relief

On irreparable harm, the court found the injury already underway rather than speculative, pointing to student withdrawals at the University of Wisconsin Madison and the University of Maryland Baltimore County, and a 42 percent decline in international applications at the University of Wisconsin Milwaukee. The court also relied on disruption to joint and combined degree programs, canceled international research and conference travel by students already lawfully present, and the administrative upheaval DHS itself projected for institutions, including its estimate that DSOs and ROs would spend roughly 135 hours on rule familiarization in the first year at a cost of up to $267.9 million.

The scope discussion is likely to be the most consequential part of the opinion. Defendants argued that Trump v. CASA limits relief to the parties before the court. Judge Saylor held that APA relief is not coextensive with the equitable authority conferred by the Judiciary Act of 1789, citing footnote 10 of CASA itself, Justice Kavanaugh's concurrence, and the First Circuit's decision in Woonasquatucket River Watershed Council v. USDA, 2026 WL 2279789 (1st Cir. Aug. 7, 2026), which rejected a categorical party limitation on section 705. The distinction is worth noting precisely: footnote 10 of CASA reserved the separate question whether the APA authorizes vacatur under section 706, while the remedy actually granted here is postponement under section 705. What makes this part of the opinion significant is the court's conclusion that the line between APA remedies and traditional equitable authority extends to section 705 as well.

Having found the authority, the court then explained why exercising it was appropriate: plaintiffs represent roughly 600 institutions out of more than 5,000 nationwide, and a party-limited order would require parallel regulatory regimes plus case-by-case determinations of coverage, all complicated by students who transfer or pursue joint degrees.

The court chose postponement under section 705 over vacatur under section 706, denying the vacatur and summary judgment requests without prejudice to renewal.

What This Means for Practice

A few points to carry into client conversations this week.

The comment record did the work. Almost every finding in the opinion traces back to material submitted during the comment period. The enrollment surveys, the Clemens economic analysis, the doctoral completion time data, the observation that students often pursue a second degree at the same level, the reliance interests of institutions and students, the I-539 volume projections, the press freedom concerns, and each of the less restrictive alternatives the court discussed all came from stakeholders who filed within the 32-day window.

That is a function of how APA review works. Merits review ordinarily focuses on the administrative record, and a challenger generally cannot fault an agency for failing to address objections or alternatives that were never meaningfully presented during the rulemaking. Comments serve a purpose before litigation as well, since a well-supported submission can prompt the agency to modify a proposal, as happened here with the extenuating circumstances exceptions for transfers.

The practical point for practitioners is that it is worth encouraging institutional and employer clients to participate in rulemakings that affect them, and to do so substantively: supply data, identify specific alternatives, and document reliance interests. Those submissions become the record on which any later challenge is decided.

The rule is not in effect. F, J, and I nonimmigrants continue to be admitted for duration of status, the 60-day grace period stands, and there is no new I-539 requirement. Institutions do not need to implement the transition provisions.

Nationwide means nationwide. Clients at institutions outside the plaintiff organizations are covered.

This is interim relief. Postponement runs pending final resolution on the merits, and the government may seek review in the First Circuit. The court found the rulemaking deficient in at least four separate respects, each analyzed on its own terms. The scope of section 705 relief after CASA remains an unsettled question on which courts continue to divide, so it is worth tracking how that issue develops.

Advise on planning, not just on status. Students who deferred, declined admission, or restructured joint degree plans in reliance on the rule are the population most affected by the timing here. So are institutions that paid to reprogram batch systems and retrain DSOs and ROs.

A postponed rule is not a withdrawn rule. Importantly, DHS retains the option of further proceedings on remand or a new rulemaking addressing the deficiencies the court identified, so clients planning multiyear programs should not assume the current framework will remain unchanged.