Nonretroactive Sentencing Amendments Are Not “Extraordinary and Compelling” Reasons for Compassionate Release

1. Introduction

These consolidated cases asked whether the sentencing disparity created by Congress’s nonretroactive reduction of mandatory penalties—here, the First Step Act of 2018’s change to 18 U. S. C. §924(c) “stacking”—can itself qualify as an “extraordinary and compelling” reason for a sentence reduction under the compassionate-release statute, 18 U. S. C. §3582(c)(1)(A)(i).

Petitioners Daniel Rutherford and Johnnie Carter were sentenced under the pre-2018 §924(c) regime that required a mandatory consecutive 25-year term for a “second or subsequent” §924(c) conviction in the same case (stacking). After Congress eliminated that stacking rule for first-time offenders, they remained ineligible for the new, lower penalties because Congress made the change nonretroactive for defendants already sentenced. They nevertheless sought compassionate release, relying on the resulting disparity—an argument supported by the Sentencing Commission’s 2023 policy amendment adding an “Unusually Long Sentence” category (USSG App. C, Amdt. 814).

The Third Circuit held the nonretroactive change could not be an “extraordinary and compelling” reason and further held the 2023 policy statement conflicted with Congress’s nonretroactivity choice. The Supreme Court granted certiorari to resolve a circuit split.

2. Summary of the Opinion

The Court affirmed. It reasoned that nonretroactive penalty changes are ordinary (thus not “extraordinary”), and Congress’s deliberate decision to withhold retroactive relief cannot be “compelling” as a reason to undo sentences Congress chose to leave intact. The Court emphasized compassionate release’s “gatekeeping” function: eligibility requires an “extraordinary and compelling” reason before considering the §3553(a) factors. The Court further held that the Sentencing Commission’s 2023 policy statement is invalid to the extent it permits consideration of such nonretroactive changes, because policy statements must remain consistent with the governing statute.

3. Analysis

3.1. Precedents Cited

The opinion’s reasoning is built on several strands of prior law: (i) §924(c) stacking doctrine and Congress’s later revision; (ii) the default rule of nonretroactivity; (iii) sentencing-modification “eligibility” gatekeeping; and (iv) the limited force of Sentencing Commission policy statements when they conflict with statutory text.

A. §924(c) stacking and the pre-First Step Act baseline

• Deal v. United States, 508 U. S. 129 (1993): Cited for the pre-2018 interpretation that multiple §924(c) counts in a single prosecution trigger the 25-year mandatory term for the “second” count, enabling stacking. This establishes why Rutherford and Carter’s mandatory minimums were so high and frames the disparity created by §403 of the First Step Act.

B. Nonretroactivity as “ordinary practice” and its line-drawing effects

• Dorsey v. United States, 567 U. S. 260 (2012): The Court quotes Dorsey for the proposition that, even when Congress extends new penalties to certain pending cases, the “ordinary practice” is to “withhol[d] that change from defendants already sentenced.” The Court uses Dorsey to recharacterize disparity as a predictable byproduct of legislative line-drawing rather than a rare injustice.
• Hewitt v. United States, 606 U. S. 419 (2025): Cited repeatedly for (i) the general rule that statutory penalty reductions usually benefit future offenders, and (ii) Congress’s interest in finality and avoiding added litigation when it declines retroactivity. The majority treats Hewitt as confirming that nonretroactivity represents a deliberate congressional tradeoff.
• United States v. Detroit Timber & Lumber Co., 200 U. S. 321 (1906): Appears in the syllabus note (about the nonbinding nature of syllabi), not for substantive sentencing law, but it reflects the Court’s standard practice about what constitutes the authoritative opinion.

C. Statutory interpretation and the limits of negative implication

• NLRB v. SW General, Inc., 580 U. S. 288 (2017) (quoting Marx v. General Revenue Corp., 568 U. S. 371 (2013)): The Court rejects petitioners’ argument that because 28 U. S. C. §994(t) expressly forbids “rehabilitation ... alone” as an extraordinary and compelling reason, all other reasons are implicitly permitted. These cases supply the interpretive principle that negative implication depends on context.
• Robinson v. Shell Oil Co., 519 U. S. 337 (1997): Used to support that the Court need not define the full outer boundary of “extraordinary and compelling” to resolve the “particular dispute” where the statutory meaning is sufficiently clear.

D. Sentencing modification, eligibility “gatekeeping,” and the scope of discretion

• Dillon v. United States, 560 U. S. 817 (2010): Central to the Court’s structure. Dillon supplies the concept that sentence-modification provisions create a “limited class of prisoners” eligible for relief; courts must determine eligibility before exercising discretion about the extent of reduction. The majority imports this gatekeeping logic into compassionate release.
• Concepcion v. United States, 597 U. S. 481 (2022): Distinguished. In Concepcion, eligibility for relief under a different First Step Act provision was conceded, and the question was what information could be considered in recalculating sentence. Here, by contrast, Congress “expressly cabined” discretion by requiring a threshold finding of “extraordinary and compelling reasons.” The Court uses Concepcion to separate “eligibility” constraints from “information considered” once eligibility is established.

E. Separation of powers: punishment choices belong to the legislature

• United States v. Wiltberger, 5 Wheat. 76 (1820): Quoted for the foundational principle that defining crimes and ordaining punishments is for the legislature, not courts. The majority deploys Wiltberger to argue that treating the harshness of mandatory penalties as an “extraordinary and compelling” reason would functionally override Congress’s chosen punishments.

F. The Sentencing Commission’s role after Loper Bright and the “consistency” requirement

• Loper Bright Enterprises v. Raimondo, 603 U. S. 369 (2024): Cited for the proposition that even where Congress authorizes an agency to “give meaning” to statutory terms, courts must independently interpret the statute and ensure the agency stays within statutory bounds.
• Batterton v. Francis, 432 U. S. 416 (1977): Supports the idea that agency definitions cannot exceed statutory authority; invoked to justify rejecting the Commission’s policy statement as inconsistent with §3582(c)(1)(A)(i).
• Koons v. United States, 584 U. S. 700 (2018): Quoted for the principle that policy statements cannot make a defendant eligible where the statute makes him ineligible.

G. Lower-court decisions and the circuit split

• United States v. Andrews, 12 F. 4th 255 (CA3 2021) and United States v. Bryant, 996 F. 3d 1243 (CA11 2021): Andrews reflects the majority circuit trend (pre-2023) rejecting nonretroactive changes as extraordinary/compelling; Bryant reflects the Eleventh Circuit’s different approach to the pre-2023 policy statement’s applicability.
• The Court’s syllabus identifies the split through representative cases rejecting the disparity rationale—United States v. McMaryion; United States v. Jarvis; United States v. Thacker; United States v. Crandall; United States v. Jenkins—and cases allowing it in combination with other circumstances—United States v. Ruvalcaba; United States v. McCoy; United States v. Chen; United States v. McGee.
• Post-2023, several circuits concluded they were not bound by the Commission’s amendment: United States v. Austin; United States v. Bricker; United States v. Black; United States v. Johnson; while United States v. Wilson is noted as reserving the question.

H. Dissent’s doctrinal anchors (and the majority’s implicit rejection)

The dissent’s approach relies heavily on broad-discretion and “totality” concepts, including: Kugler v. Helfant, Welch v. Helvering, Koon v. United States, and Octane Fitness, LLC v. ICON Health & Fitness, Inc. for the flexibility of “extraordinary” concepts; and on sentencing framework decisions like Mistretta v. United States, United States v. Booker, Gall v. United States, and Kimbrough v. United States to emphasize institutional roles and sentencing discretion. The majority’s response is structural: discretion is broad only after the statute’s eligibility condition is met.

3.2. Legal Reasoning

A. Text and ordinary meaning: “extraordinary” + “compelling” + “warrant”

The Court begins with dictionary definitions: “extraordinary” means highly unusual; “compelling” means forcefully convincing. It then adds a crucial contextual hinge: the reasons must “warrant” compassionate release, which signals a threshold justification, not merely relevance.

On that framing, disparity from a nonretroactive penalty change fails twice:

• Not “extraordinary”: nonretroactive changes are common; disparity is an “unexceptional feature” of the default nonretroactivity regime.
• Not “compelling”: Congress’s deliberate decision to withhold retroactivity cannot logically be the “convincing” reason to reduce the very sentences Congress chose to preserve—especially given Congress’s interests in finality and limiting litigation.

B. Statutory structure: eligibility first, then §3553(a) weighing

The Court rejects the petitioners’ effort to fold sentencing-disparity considerations (often associated with §3553(a)(6)) into the “extraordinary and compelling” threshold. Borrowing from Dillon v. United States, the Court treats “extraordinary and compelling reasons” as a gatekeeping requirement that defines a “limited class” eligible for relief; only then does the court consider §3553(a) to decide whether and how much to reduce.

This distinction is the doctrinal pivot that lets the majority cabin Concepcion v. United States: broad resentencing information flows may be permissible where eligibility is conceded or not constrained by limiting language, but compassionate release contains limiting language that must be satisfied first.

C. Avoiding an “end run” around nonretroactivity and mandatory penalties

The majority sees petitioners’ approach as incompatible with Congress’s nonretroactivity line: if sentencing disparity from nonretroactive change can qualify (alone or “in combination”), then compassionate release becomes a vehicle to effectuate retroactivity case-by-case—undermining Congress’s choice.

The opinion presses a separation-of-powers concern via United States v. Wiltberger: if courts can treat “harshness” of mandatory penalties as extraordinary/compelling, they can effectively nullify legislative punishment judgments. The Court treats that as outside the “heartland” of compassionate release and inconsistent with Congress’s institutional allocation.

D. The Sentencing Commission: empowered, but bounded

The Court acknowledges Congress empowered the Commission to “give meaning” to “extraordinary and compelling” (citing Loper Bright Enterprises v. Raimondo), yet insists policy statements must be “consistent with” the statute and cannot expand eligibility beyond statutory limits (citing Koons v. United States and Batterton v. Francis).

Consequently, USSG App. C, Amdt. 814’s “Unusually Long Sentence” category is invalid to the extent it permits nonretroactive legal changes to satisfy the statutory threshold—whether “by itself or in combination with other factors.”

3.3. Impact

• Closes the compassionate-release “retroactivity gap” for §924(c): Prisoners sentenced under pre-First Step Act stacking cannot use the Act’s nonretroactive change as an eligibility trigger for compassionate release.
• Constrains compassionate release more broadly: The holding is not limited to §924(c). It states a general rule: any sentencing amendment Congress makes nonretroactive cannot generate an “extraordinary and compelling” reason based on disparity alone (or as a component of the eligibility mix).
• Invalidates part of the 2023 Commission policy statement in practice: District courts must disregard USSG §1B1.13(b)(6) insofar as it invites consideration of nonretroactive changes in law as extraordinary/compelling.
• Reinforces finality and narrows judicial discretion at the eligibility stage: The decision strengthens a two-step architecture: eligibility is a statutory threshold with “independent and ascertainable limits,” insulating it from §3553(a) balancing impulses.
• Signals a post-Loper Bright posture toward Commission guidance: Even where Congress delegates meaning-making, courts will reject Commission interpretations that the Court views as contradicting statutory structure (especially around eligibility).

4. Complex Concepts Simplified

• §924(c) “stacking”: Under the old regime (per Deal v. United States), multiple firearm counts in the same case could trigger escalating mandatory consecutive sentences (e.g., a 25-year term for the “second” count), producing extremely long mandatory minimums.
• Nonretroactivity: When Congress changes sentencing law but does not apply it to people already sentenced. This creates disparities between earlier- and later-sentenced defendants, but (per the majority) such disparities are routine byproducts of legislative line-drawing.
• Compassionate release (18 U. S. C. §3582(c)(1)(A)(i)): A narrow exception to finality allowing sentence reduction if (i) “extraordinary and compelling reasons” warrant it, (ii) §3553(a) factors support it, and (iii) the reduction is consistent with applicable Commission policy statements—so long as those statements align with the statute.
• Eligibility vs. extent of reduction: The majority separates (1) whether a prisoner can even get in the door (eligibility: extraordinary/compelling), from (2) how much relief is appropriate (discretion: §3553(a)). Petitioners wanted disparity considered at step (1); the Court says it cannot be.
• Sentencing Commission policy statements: Guidance meant to operationalize statutes. The majority holds policy statements cannot override statutory meaning; if they conflict, courts must disregard them.
• “Heartland” of compassionate release: Historically focused on individualized personal circumstances (medical, age, family). The Court uses this history to argue that using nonretroactive sentencing changes would transform compassionate release into an alternate retroactivity mechanism.

5. Conclusion

Rutherford v. United States establishes a categorical interpretive rule at the eligibility stage of compassionate release: sentencing disparities produced by Congress’s nonretroactive sentencing amendments are neither “extraordinary” nor “compelling,” and thus cannot “warrant” relief under §3582(c)(1)(A)(i). The Court frames this as fidelity to statutory text, to Congress’s deliberate nonretroactivity choice, and to separation-of-powers limits on judicial revision of legislatively mandated punishments.

Equally significant is the decision’s institutional holding: even though Congress empowered the Sentencing Commission to “give meaning” to compassionate release standards, courts must independently interpret the statute, and Commission policy statements that expand eligibility beyond what the statute permits are invalid. The practical consequence is a narrower compassionate-release gateway and diminished ability to use compassionate release to mitigate sentences that Congress has chosen not to revisit retroactively.