NIH May Not Impose an Across-the-Board Indirect-Cost Cap by “Guidance” Under the Appropriations Rider and HHS Deviation Regulations

Court: U.S. Court of Appeals for the First Circuit (Rikelman, Lipez, Howard, JJ.)
Date: January 5, 2026
Case: Association of American Medical Colleges v. National Institutes of Health (consolidated appeals Nos. 25-1343, 25-1344, 25-1345)
Opinion by: Lipez, Circuit Judge

1. Introduction

These consolidated appeals arose from an abrupt funding-policy change announced by the National Institutes of Health (“NIH”), an agency within the Department of Health and Human Services (“HHS”). On Friday night, February 7, 2025, NIH issued a short “Supplemental Guidance” (Notice Number NOT-OD-25-068) to the 2024 NIH Grants Policy Statement. The Guidance capped reimbursement of research “indirect costs” (also called facilities and administration, “F&A”) at a uniform 15%, effective the next business day.

Plaintiffs—(i) a coalition of states led by Massachusetts and other attorneys general, (ii) major medical and hospital associations, and (iii) higher-education associations and prominent universities—challenged the Guidance in the District of Massachusetts. They contended, among other theories, that the cap violated a recurring congressional appropriations rider that constrained NIH’s indirect-cost practices and that it contravened HHS’s own regulations governing negotiated indirect-cost rates and deviations from them.

The district court issued emergency relief, then a nationwide preliminary injunction, and ultimately (at the parties’ request) converted that injunction into a permanent injunction and vacated the Guidance. The First Circuit affirmed, holding (1) the district court had APA jurisdiction over this challenge to agency-wide guidance, and (2) NIH’s uniform 15% cap was unlawful because it violated both the appropriations rider and governing regulations.

2. Summary of the Opinion

3. Analysis

3.1 Precedents Cited

A. Jurisdictional line between “guidance challenges” and “grant-payment/termination disputes”

The opinion’s jurisdiction analysis is anchored in two recent Supreme Court emergency decisions that the panel treats as controlling guideposts for allocating disputes between district courts (APA) and the Court of Federal Claims (Tucker Act):

• Department of Education v. California, 604 U.S. 650 (2025) (per curiam)
  The Supreme Court stayed a district court order that effectively compelled the Department to restore terminated grants. The First Circuit emphasizes the Court’s distinction between permissible APA vacatur that may incidentally result in disbursement of funds and impermissible orders “to enforce a contractual obligation to pay money.” The opinion quotes the Court’s reliance on Bowen v. Massachusetts, 487 U.S. 879 (1988), and Great-W. Life & Annuity Ins. Co. v. Knudson, 534 U.S. 204 (2002), to draw that boundary.
• NIH v. American Public Health Association ("APHA"), 145 S. Ct. 2658 (2025) (mem.)
  Justice Barrett’s controlling concurrence (treated as the holding under Marks v. United States, 430 U.S. 188 (1977), which in turn quotes Gregg v. Georgia, 428 U.S. 153 (1976)) splits challenges into (i) challenges to “internal agency guidance” (proper in district court) and (ii) challenges to “grant terminations” (proper in the Court of Federal Claims). The First Circuit finds the present case mirrors the “guidance” side of APHA.

The panel also references and integrates lower-court and doctrinal authorities that historically governed APA/Tucker Act boundary disputes:

• Am. Sci. & Eng'g, Inc. v. Califano, 571 F.2d 58 (1st Cir. 1978) (CFC jurisdiction over contract actions is exclusive).
• Albrecht v. Comm. on Emp. Benefits of the Fed. Rsrv. Emp. Benefits Sys., 357 F.3d 62 (D.C. Cir. 2004) (Tucker Act impliedly forbids district-court APA route for essentially contract disputes).
• Megapulse, Inc. v. Lewis, 672 F.2d 959 (D.C. Cir. 1982) (classic framework for identifying APA claims “in disguise”).
• California v. Dep't of Educ., 132 F.4th 92 (1st Cir. 2025) and Am. Pub. Health Ass'n v. NIH, 145 F.4th 39 (1st Cir. 2025) (First Circuit decisions describing the underlying disputes that led to the Supreme Court’s emergency orders).
• Trump v. Boyle, 145 S. Ct. 2653 (2025) (cited for the limited conclusiveness of emergency posture while still informing analysis).

How those precedents influenced the outcome: The First Circuit adopts APHA’s “two-track” allocation rule: district courts can hear challenges to agency-wide guidance as an APA “agency action,” while disputes that effectively compel payment under specific grant agreements (or undo terminations) are Tucker Act matters for the Court of Federal Claims. Because plaintiffs attacked the legality of the Guidance itself—not a particular payment refusal or termination—the case stayed in district court.

B. Interpretive canons and administrative-law framing

• Corley v. United States, 556 U.S. 303 (2009)
  Used to reject NIH’s reading of the rider that would make text superfluous—especially the rider’s “modified approach” clause.
• McKenna v. First Horizon Home Loan Corp., 475 F.3d 418 (1st Cir. 2007)
  Cited for using statutory context as “confirmatory evidence” of congressional intent (here, Congress’s rejection of earlier indirect-cost cap proposals).
• Dep't of Com. v. New York, 588 U.S. 752 (2019)
  Invoked to justify reading statutory constraints in a realistic, non-naïve way given the legislative backdrop.
• U.S. ex rel. Ge v. Takeda Pharm. Co., 737 F.3d 116 (1st Cir. 2013) and United States v. Zannino, 895 F.2d 1 (1st Cir. 1990)
  Used to enforce waiver/forfeiture rules when NIH attempted to re-characterize the Guidance as limited to IHEs.

3.2 Legal Reasoning

A. Jurisdiction: why this was an APA “guidance” case, not a Tucker Act contract case

The court begins from first principles: the United States is immune absent waiver; the APA waives immunity for non-monetary relief, but the waiver is inapplicable when another statute (here, the Tucker Act) “impliedly forbids the relief sought.” NIH argued the suit was “essentially a contract dispute,” because indirect-cost rates are built into grant agreements and relief would affect payments.

Applying Department of Education v. California and NIH v. American Public Health Association ("APHA"), the court draws a practical line:

• If the remedy would operate as an order “to pay money” by enforcing or restoring particular grant obligations (like undoing terminations), the dispute belongs in the Court of Federal Claims.
• If plaintiffs seek vacatur of an agency-wide policy/guidance as unlawful agency action—without directly reinstating or enforcing specific grant payment obligations—district court jurisdiction is proper.

Because plaintiffs challenged NIH’s across-the-board “Supplemental Guidance” for future application (affecting future grants as well as current ones), and did not sue over a concrete withholding decision on a particular award, the court treated the case as the “guidance” side of APHA. The possibility that later disputes about withheld reimbursements could be Tucker Act matters did not convert this guidance challenge into a contract claim.

B. Statutory violation: the appropriations rider foreclosed a uniform indirect-cost cap

The court reads § 224 of the Further Consolidated Appropriations Act, 2024 (reenacted annually since 2018) as containing three independent constraints on NIH:

1. Sentence one (“continue to apply … as … in Q3 FY 2017”): The rider legislatively locks in application of the 45 C.F.R. Part 75 indirect-cost provisions—including deviation approvals—“to the same extent and in the same manner” as in the third quarter of fiscal year 2017. Since the court finds the Guidance violates the deviation regulations, it “necessarily” violates this statutory command.
2. Sentence two, clause one (“no funds … to develop or implement a modified approach”): Even if NIH could nominally comply with deviation mechanics, Congress separately prohibited “a modified approach” to those indirect-cost provisions. NIH’s across-the-board displacement of negotiated rates—something it undisputedly had never done—constituted precisely the forbidden “modified approach.” The court also uses Corley v. United States to reject NIH’s attempt to collapse this clause into mere repetition of sentence one; doing so would render key text superfluous.
3. Sentence two, final clause (“intentionally or substantially expand the fiscal effect”): The rider forbids expanding the “fiscal effect” of deviations beyond the proportional effect seen in Q3 FY 2017. Even assuming “fiscal effect” refers only to the government’s finances (as NIH urged), NIH’s own public statement touted roughly $4 billion in annual savings. NIH offered no record basis to show that magnitude was proportionally consistent with Q3 FY 2017 deviations. The court rejects NIH’s reframing that savings would simply be reallocated to direct costs; the rider focuses on deviation “fiscal effect,” not overall research spending.

Context reinforced the textual reading: Congress had previously rejected an executive proposal to cap NIH indirect-cost reimbursement (e.g., the 2017 proposal) and then enacted—and repeatedly reenacted—this rider to prevent NIH from implementing caps through administrative maneuvering. Citing Dep't of Com. v. New York, the court declined to interpret the rider in a way that would ignore that evident legislative purpose.

C. Regulatory violation: 45 C.F.R. § 75.414(c) does not authorize a universal cap and requires an antecedent, public deviation framework

Independently of the rider, the court holds the Guidance unlawful under HHS’s own regulations governing negotiated indirect-cost rates and deviations.

1) The “class of Federal awards” limit (45 C.F.R. § 75.414(c)(1) and § 75.2)

The regulation provides that negotiated rates “must be accepted,” and an HHS awarding agency “may use a rate different from the negotiated rate” only for “a class of Federal awards” or “a single Federal award,” with required approvals and justification.

The court stresses the defined term “[c]lass of [f]ederal awards” in 45 C.F.R. § 75.2: a “group” awarded under a “specific program or group of programs” or to a “specific type” of entity. This “specific” language restricts NIH from defining the “class” as essentially all awards.

NIH’s Guidance stated a “standard indirect rate of 15% across all NIH grants” and applied to “any new grant issued,” language the court reads as universal. NIH’s late argument that the Guidance applied only to institutions of higher education failed both procedurally (waiver under U.S. ex rel. Ge v. Takeda Pharm. Co.) and on the text, which the court declined to “rewrite” as mere “inartful language.”

2) The deviation procedure and publication requirement (45 C.F.R. § 75.414(c)(3)–(4))

The court reads § 75.414(c)(3) as requiring NIH to have (and to make publicly available) “policies, procedures and general decision making criteria” that its programs “will follow” to seek and justify deviations. It further reads the scheme as sequential and notice-protective:

• NIH must first publish a deviation framework (policies/procedures/criteria).
• NIH must then apply that framework when deciding whether a deviation is justified for a permissible class or single award.
• Under § 75.414(c)(4), NIH must include in the “notice of funding opportunity” the policies relating to indirect-cost reimbursement, enabling applicants to make an “informed decision” under § 75.203(c)(2).

The Guidance’s one-step imposition of a uniform 15% rate—explicitly disavowing individualized redeterminations—was incompatible with a regulatory regime built around negotiated, entity-specific rates and controlled, justified deviations. In the panel’s view, NIH’s interpretation would let a deviation provision swallow the negotiation framework and nullify the applicant-notice requirement.

3.3 Impact

A. Immediate operational impact on NIH and grantees

• NIH cannot implement uniform indirect-cost caps through short-form guidance while a recurring appropriations rider preserves negotiated-rate practices.
• The injunction and vacatur preserve the NICRA-based system for the life of awards (and for future awards) absent lawful, properly channeled change—either from Congress (removing or changing the rider) or from lawful agency regulation consistent with statutory constraints.

B. Structural impact on administrative-law litigation over grants

• The decision operationalizes the post-2025 Supreme Court emergency guidance in NIH v. American Public Health Association ("APHA") and Department of Education v. California: litigants can challenge agency-wide guidance in district court, while disputes that effectively enforce payment obligations under particular awards may be routed to the Court of Federal Claims.
• Practically, plaintiffs may split litigation: (i) district-court APA challenges to policies/guidance; (ii) CFC actions (or other appropriate vehicles) for discrete payment-withholding or termination consequences—subject to remedies and sovereign-immunity limitations.

C. Future constraints on agencies using “deviation” authority as a policy lever

The opinion constrains agencies from using “deviation” clauses to enact broad policy reversals that undermine negotiated-rate systems. Two doctrinal moves are likely to recur in future cases:

• “Class” must be meaningfully limited: the definition in 45 C.F.R. § 75.2 becomes a hard textual checkpoint against universalization.
• Publication/notice provisions have teeth: deviation authority is not self-executing; it must be channeled through publicly available standards and applicant-facing disclosures in funding opportunity notices.

4. Complex Concepts Simplified

• Direct vs. indirect (F&A) costs: Direct costs are tied to a specific project (e.g., a lab technician’s time on that one grant). Indirect costs are shared infrastructure and administration that make research possible but can’t be pinned to one project (e.g., lab buildings, utilities, compliance, IT security).
• NICRA (Negotiated Indirect Cost Rate Agreement): A negotiated percentage rate (often lasting 2–4 years) that federal agencies must generally accept for awards made during its term; it creates predictability for budgeting over a grant’s life.
• “Deviation from negotiated rates” (45 C.F.R. § 75.414(c)): A narrow exception allowing an agency to use a different rate for a particular award or a defined class—only if agency leadership approves it and it is justified under publicly available criteria and properly disclosed in funding announcements.
• Appropriations rider: A condition Congress places on annual spending bills. Even if it’s not permanent legislation, it can be reenacted repeatedly and can tightly control how an agency may use appropriated funds.
• APA jurisdiction vs. Tucker Act jurisdiction: The APA typically allows suits in district court for non-monetary relief (like vacatur of unlawful agency action). The Tucker Act routes contract-based money claims against the United States to the Court of Federal Claims. This case emphasizes that challenging an agency-wide policy is different from suing to enforce payment under a specific grant agreement.
• Vacatur: A remedy that sets aside an agency action (e.g., guidance). It differs from an order compelling the government to pay money under a contract, which triggers Tucker Act concerns.

5. Conclusion

The First Circuit’s decision establishes a clear, enforceable rule in the NIH grants context: NIH cannot use “Supplemental Guidance” and deviation authority to impose a universal indirect-cost cap where Congress has, through a recurring appropriations rider, preserved the negotiated-rate framework and barred “modified approaches” and disproportionate fiscal effects of deviations. In parallel, the opinion adopts and applies the Supreme Court’s recent jurisdictional sorting in NIH v. American Public Health Association ("APHA") and Department of Education v. California, reinforcing that district courts may hear APA challenges to agency-wide guidance even when the subject matter concerns grants, while contract-like disputes over specific award payments may belong in the Court of Federal Claims.

The practical takeaway is twofold: (1) statutory and regulatory guardrails around indirect-cost reimbursement are not merely hortatory—courts will enforce them against rapid, broad policy shifts; and (2) litigants and agencies must attend carefully to forum selection, as “policy” challenges and “payment/termination” disputes may be channeled to different courts with different remedial powers.