by Richard Resch

Few features of the federal sentencing system have as much practical impact on defendants’ sentences as the issue of relevant conduct. Codified at §1B1.3 of the United States Sentencing Guidelines, the relevant conduct provision determines which acts and omissions a court may consider when calculating a defendant’s Guidelines range. Its reach extends well beyond the offense of conviction. Under §1B1.3, a defendant can be sentenced based on uncharged conduct, conduct underlying dismissed counts, and even the acts of other people. The U.S. Sentencing Commission’s Primer on Relevant Conduct (2026) offers a detailed survey of the Guidelines provision, commentary, and case law governing this issue. For anyone facing federal sentencing or trying to understand how a Guidelines range was calculated, the primer is essential reading.

The Architecture of
Relevant Conduct

The federal Guidelines occupy a middle ground between two sentencing philosophies. In a pure “charge offense” system, the sentence turns entirely on the elements of the statute the defendant was convicted of violating. In a pure “real offense” system, the sentence reflects everything the defendant actually did, regardless of the formal charges. The relevant conduct framework attempts to blend these approaches. It begins with the offense of conviction and then layers onto it a range of real-world conduct that the Guidelines deem pertinent to sentencing.

Section 1B1.3 identifies five categories of relevant conduct that can affect a defendant’s Guidelines range. The first encompasses the defendant’s own acts and omissions connected to the offense of conviction, including conduct in preparation for the crime and efforts to avoid detection afterward. The second holds a defendant accountable for certain acts of other people, provided those acts occurred as part of a “jointly undertaken criminal activity.” The third extends accountability, for certain offense types, to conduct outside the offense of conviction when it forms part of the “same course of conduct” or a “common scheme or plan.” The fourth attributes to the defendant all harm that resulted from the conduct captured by the first three categories. And the fifth is a catchall, incorporating any other information that the applicable Guidelines specifically direct the court to consider. One limit is built into the harm category. Unless the applicable Guidelines clearly indicate otherwise, harm that is merely risked is not treated as the equivalent of harm that actually occurred.

These categories interact in ways that can dramatically expand a defendant’s sentencing exposure. For example, a person convicted of a single drug sale can be sentenced on the basis of drug quantities from transactions never charged, never proven to a jury, and never admitted in a plea. The relevant conduct framework makes this possible.

Accountability for
the Acts of Others

One of the most consequential features of §1B1.3 is subsection (a)(1)(B), which holds a defendant accountable for acts committed by other people as part of jointly undertaken criminal activity. The Guidelines define this as any criminal plan, scheme, or enterprise undertaken by the defendant together with others, whether or not it was charged as a conspiracy. To attribute another person’s conduct to a defendant under this provision, the government must satisfy a three-part test. The conduct must have been (1) within the scope of the jointly undertaken criminal activity, (2) in furtherance of that activity, and (3) reasonably foreseeable in connection with it. All three requirements must be met.

The scope inquiry is the most fact-intensive of the three, and courts have held that it requires particularized findings about what the individual defendant actually agreed to participate in. This is a critical distinction. A conspiracy charge might encompass the activities of dozens of participants over several years, but the scope of the jointly undertaken criminal activity for sentencing purposes may be far narrower. The Guidelines themselves make this point, i.e., because a count of conviction may be broadly worded, the scope of the jointly undertaken activity “is not necessarily the same as the scope of the entire conspiracy, and hence relevant conduct is not necessarily the same for every participant.”

The Sixth and Seventh Circuits have identified six factors for assessing scope: (1) whether there was a single scheme, (2) similarities in how participants operated, (3) coordination of activities, (4) pooling of resources or profits, (5) the defendant’s knowledge of the scheme’s breadth, and (6) the length and degree of the defendant’s participation. United States v. Bailey, 973 F.3d 548 (6th Cir. 2020); United States v. Salem, 657 F.3d 560 (7th Cir. 2011). Conduct that falls outside what a defendant agreed to is excluded from relevant conduct under subsection (a)(1)(B), even if the defendant knew about it or could have predicted it. The Eleventh Circuit reinforced this point in United States v. Presendieu, 880 F.3d 1228 (11th Cir. 2018), holding that a defendant’s mere awareness of being part of a larger scheme did not make the independently caused losses of an unknown actor part of her relevant conduct.

Attribution of other participants’ conduct under subsection (a)(1)(B) has a firm temporal boundary as well. A defendant’s relevant conduct under that subsection does not include conduct of other conspiracy members before the defendant joined the conspiracy, even if the defendant knew of that conduct. Thus, a person who joins an ongoing drug-distribution conspiracy in its sixth month is not accountable under subsection (a)(1)(B) for quantities distributed by others during the first five months. U.S. Sentencing Guidelines Manual § 1B1.3 cmt. n.3(B) (U.S. Sent’g Comm’n 2025); United States v. Morilla, 177 F.4th 1291 (11th Cir. 2026); United States v. Hunter, 323 F.3d 1314 (11th Cir. 2003); United States v. Carreon, 11 F.3d 1225 (5th Cir. 1994); see also United States v. Longstreet, 603 F.3d 273 (5th Cir. 2010).

Reasonable foreseeability, the third prong, may encompass conduct that the defendant did not specifically agree to commit, but only if that conduct first falls within the scope of the jointly undertaken criminal activity. The Guidelines illustrate the point with a robbery example. A defendant who agrees to participate in an armed bank robbery may be held accountable for a co-defendant’s assault on a teller because the assault is within the scope of and in furtherance of the jointly undertaken robbery and is reasonably foreseeable given the nature of the crime. Courts have applied this principle even where the defendant instructed an accomplice not to harm the victim. But foreseeability cannot substitute for scope. A court must first determine the scope of the criminal activity the defendant agreed to undertake and only then determine whether the conduct of others within that scope and in furtherance of the undertaking was reasonably foreseeable. The reasonable-foreseeability requirement under subsection (a)(1)(B) applies only to the acts and omissions of others; it does not apply to conduct personally undertaken, aided, or abetted by the defendant under subsection (a)(1)(A). U.S. Sentencing Guidelines Manual § 1B1.3 cmt. nn.3(D), 4(B)(i) (U.S. Sent’g Comm’n 2025); United States v. Pica, 106 F.4th 197 (2d Cir. 2024); United States v. Ford, 988 F.3d 970 (7th Cir. 2021); United States v. Patton, 927 F.3d 1087 (10th Cir. 2019); United States v. Donadeo, 910 F.3d 886 (6th Cir. 2018); United States v. Willis, 476 F.3d 1121 (10th Cir. 2007); United States v. Chalarca, 95 F.3d 239 (2d Cir. 1996).

Defense counsel should pay close attention to whether the sentencing court has made individualized foreseeability findings rather than simply attributing conspiracy-wide quantities to the defendant. The Seventh Circuit held in United States v. Granger, 70 F.4th 408 (7th Cir. 2023), that the District Court erred by holding a defendant accountable for the drug quantity of the entire conspiracy without addressing what conduct was reasonably foreseeable to that particular defendant. Granger gives practitioners a concrete, citable basis for challenging presentence reports (“PSRs”) that treat foreseeability as a foregone conclusion once scope has been established. The three-part test requires an independent finding on each prong, and a PSR that skips the foreseeability analysis is vulnerable to objection.

A related procedural point merits attention, because it determines where the relevant conduct fight must be waged. Several circuits have held that a sentencing court satisfies the particularized-findings requirement when it adopts a PSR that itself contains legally sufficient particularized findings on the scope of the jointly undertaken criminal activity. United States v. Ellis, 23 F.4th 1228 (10th Cir. 2022); United States v. Dridi, 952 F.3d 893 (7th Cir. 2020); United States v. McReynolds, 964 F.3d 555 (6th Cir. 2020); United States v. Flores-Alvarado, 779 F.3d 250 (4th Cir. 2015). The practical lesson for defense counsel is twofold. First, if the PSR’s relevant conduct findings are conclusory, attributing conspiracy-wide quantities or scheme-wide losses without analyzing what this particular defendant agreed to, object specifically and on the record, because the adoption of a PSR that lacks particularized findings does not cure the deficiency. Second, do not assume the issue is preserved by a general objection to drug quantity or loss amount. The objection should identify the scope, furtherance, and foreseeability prongs by name and explain why the PSR’s account fails each one. Unobjected-to facts in a PSR are ordinarily deemed admitted, and an unpreserved relevant conduct challenge will be reviewed on appeal, at best, under the unforgiving plain error standard.

Expanded Relevant Conduct: Beyond the Offense of Conviction

For certain categories of offenses, the Guidelines cast an even wider net. Subsection (a)(2) of §1B1.3 authorizes what is commonly called “expanded relevant conduct,” which covers acts and omissions outside the offense of conviction if they formed part of the same course of conduct or a common scheme or plan. This provision applies only to offenses whose Guidelines are listed under §3D1.2(d), a grouping rule that covers offenses measured by aggregate harm: drug trafficking, fraud, money laundering, firearms trafficking and unlawful possession, tax offenses, child pornography distribution, and similar crimes. Offenses involving individual victims of physical harm, such as murder, assault, robbery, and sexual abuse, are excluded from this expanded reach.

One aspect of expanded relevant conduct that defendants and practitioners should not overlook is that subsection (a)(2) does not require multiple counts of conviction. A defendant convicted of a single drug transaction, a single fraud count, or a single firearms offense can still be sentenced on the basis of uncharged conduct that qualifies as the same course of conduct or a common scheme or plan. The Guidelines’ commentary states this explicitly. The practical consequence is that the single-count posture of an indictment provides no insulation from a sentencing calculation that sweeps in an entire course of dealing.

The “same course of conduct” analysis turns on three factors weighed in combination: (1) the similarity of the offenses, (2) their regularity, and (3) their temporal proximity. Courts apply a sliding-scale approach; when one factor is weak, the others must be stronger to compensate. The Seventh Circuit concluded in United States v. Brasher, 105 F.4th 1002 (7th Cir. 2024), that although drug transactions predating the charged offense conduct by almost 15 months lacked temporal proximity to the offense of conviction, there was sufficient similarity between the uncharged transactions and the offense of conviction to sustain the relevant conduct finding under the plain error standard of review. The Eighth Circuit reached a similar conclusion in United States v. Soto, 62 F.4th 430 (8th Cir. 2023), another case reviewed for plain error, holding in a firearms case that “striking” conduct similarity was strong enough to bridge a gap of nearly four years, and the Court went further, concluding that the District Court had committed no error at all, plain or otherwise.

However, the sliding scale cuts both ways, so defense counsel should be familiar with the cases in which courts have rejected the Government’s relevant conduct arguments. The Sixth Circuit provides particularly useful authority on this issue. In United States v. Amerson, 886 F.3d 568 (6th Cir. 2018), the Court held that a prior handgun possession was not relevant conduct for a felon-in-possession offense even though the two incidents were only three and a half months apart, because the Government had failed to show sufficient similarity to compensate for a weak showing on regularity. In United States v. Bowens, 938 F.3d 790 (6th Cir. 2019), the Court rejected a relevant conduct finding where only two instances of illegal gun possession were involved, holding that four months was not a short enough interval to overcome the absence of regularity or similarity. These decisions illustrate that temporal proximity alone does not satisfy the sliding-scale test, and they provide defense-side authority for contesting Government efforts to pull in uncharged conduct based on timing alone.

A “common scheme or plan” requires a different showing. Two or more offenses must be substantially connected by at least one common factor, viz., common victims, common accomplices, common purpose, or similar methods of operation. Courts have cautioned against defining “common purpose” too broadly. The Seventh Circuit rejected the Government’s argument in United States v. Purham, 754 F.3d 411 (7th Cir. 2014), that two separate periods of selling cocaine in the same city constituted a common scheme, where the episodes were not linked by shared accomplices or methods. The Fifth Circuit similarly warned in United States v. Benns, 740 F.3d 370 (5th Cir. 2014), against accepting purposes framed so generally that they would collapse any limits on the principle.

Causation and the Harm Category

The harm category raises a question of its own, and it is one that divides the circuits. That is, how much causation does subsection (a)(3) require? The provision applies to all harm that “resulted from” the conduct captured by the first three categories, and the meaning of that phrase determines how far removed a loss figure or injury finding can be from the defendant’s actual conduct. The Fifth Circuit has held that “resulted from” imposes only a but-for causation requirement. Unless otherwise specified, the relevant conduct must be a but-for cause of a harm for that harm to count in the Guidelines calculation. United States v. Ramos-Delgado, 763 F.3d 398 (5th Cir. 2014). The Sixth and Ninth Circuits have read the phrase to require both but-for causation and proximate cause. United States v. Peppel, 707 F.3d 627 (6th Cir. 2013); United States v. Lonich, 23 F.4th 881 (9th Cir. 2022). The difference is not academic. In fraud and other harm-driven cases, a proximate cause requirement gives the defense a basis to argue that attenuated losses, i.e., harms produced by intervening decisions, market forces, or independent third parties, should be excluded from the calculation, an argument that circuit precedent forecloses in a but-for-only jurisdiction. Counsel litigating loss or injury findings should determine which standard governs in their circuit and, where the question remains open, argue for the proximate cause limitation.

The Burden of Proof
and Its Implications

Every federal circuit except the Fifth has held that relevant conduct need only be established by a preponderance of the evidence. That means the Government must show only that a fact is more likely true than not. The Fifth Circuit has theoretically left open the possibility of a heightened standard but has never actually required one, even when relevant conduct findings increased a defendant’s sentencing range tenfold. United States v. Simpson, 741 F.3d 539 (5th Cir. 2014).

The practical result of this standard deserves careful attention. A defendant can be convicted by a jury under the beyond-a-reasonable-doubt standard for possessing a small quantity of drugs, and then sentenced on the basis of vastly larger quantities that the Government proves only by a preponderance. The same conduct that could not survive the scrutiny of a criminal trial can increase a defendant’s sentence by years or even decades when considered as relevant conduct at sentencing. This asymmetry between the protections afforded at trial and the latitude afforded at sentencing is one of the most contested features of the Guidelines system.

Acquitted Conduct: The 2024 Reform and Its Limits

For decades, federal courts permitted sentencing judges to consider conduct for which a defendant had been acquitted by a jury, so long as the Government proved that conduct by a preponderance of the evidence at sentencing. In 2024, the Sentencing Commission took a significant step toward addressing this practice. Amendment 826, effective November 1, 2024, added subsection (c) to §1B1.3, which now provides that relevant conduct “does not include conduct for which the defendant was criminally charged and acquitted in federal court, unless such conduct also establishes, in whole or in part, the instant offense of conviction.”

The exception is worth discussing. Where conduct underlies both an acquitted charge and the offense of conviction, the sentencing court retains discretion to determine whether the overlapping conduct establishes the current offense. If it does, the court may treat it as relevant conduct despite the acquittal. A 2025 decision by the U.S. District Court for the Northern District of Ohio, United States v. Scott, 779 F. Supp. 3d 937 (N.D. Ohio 2025), illustrated this dynamic in a conspiracy case where the conviction effectively subsumed the substantive counts on which the defendant was acquitted.

However, the amendment did not extend the same protection to dismissed conduct. Charges that are dropped as part of a plea bargain or for other reasons can still be considered at sentencing if the underlying facts are proven by a preponderance of the evidence. Nor did the amendment address conduct that was never charged at all. Uncharged acts remain fully available as relevant conduct, subject only to the preponderance standard. The Seventh Circuit confirmed in United States v. Bridgewater, 950 F.3d 928 (7th Cir. 2020), that dismissed conduct can increase a defendant’s sentence consistent with both the Supreme Court’s decision in United States v. Watts, 519 U.S. 148 (1997), and the Guidelines themselves.

The protection is also confined to federal acquittals. By its terms, subsection (c) excludes only conduct for which the defendant was charged and acquitted “in federal court,” so conduct underlying a state-court acquittal can still be included in the sentencing calculation if proven by a preponderance of the evidence. The Commission tied this limitation to the principles of the dual-sovereignty doctrine and to concerns about administrability, reasoning that federal courts may have greater difficulty parsing acquitted conduct from convicted conduct when the acquittal arose from another jurisdiction’s proceedings at a different time. For defendants with parallel state and federal cases, the practical point is stark. A state acquittal, however complete, does not keep the underlying conduct out of a federal Guidelines range.

One more limitation should be mentioned. Amendment 826 is not retroactive. The Commission took public comment and testimony on whether to designate its 2024 amendments for retroactive application but did not list Amendment 826 in §1B1.10(d), the policy statement identifying the amendments that courts may apply retroactively. Defendants sentenced before November 1, 2024, therefore cannot obtain a sentence reduction under 18 U.S.C. § 3582(c)(2) on the ground that acquitted conduct increased their Guidelines range. And courts have already denied motions premised on the amendment’s supposed retroactive effect. The reform is prospective only.

Enhancements Keyed to the Offense of Conviction

Not every Guidelines provision opens the door to relevant conduct, and practitioners should know how to spot the ones that do not. Some provisions apply only if the “offense of conviction” itself involved the specified conduct, and courts have read that language as excluding reliance on relevant conduct. The Tenth Circuit’s decision in United States v. Ansberry, 976 F.3d 1108 (10th Cir. 2020), is illustrative. The official victim enhancement in subsection (a) of §3A1.2 applies where the victim was affiliated with the government and “the offense of conviction was motivated by such status.” Ansberry held that the enhancement does not apply unless the facts immediately related to the offense of conviction – and not any additional relevant conduct – support its application, and it reversed the enhancement on that basis. The generalized takeaway is that when a PSR applies an enhancement, read the operative Guidelines text carefully. Where the provision is keyed to the “offense of conviction” rather than to the “offense,” a defined term that, under the commentary to §1B1.1, includes the offense of conviction and all relevant conduct, the Government cannot use relevant conduct to fill the gap, and an objection on that ground may defeat the enhancement outright.

Drawing the Line
Between Relevant Conduct
and Criminal History

The Guidelines draw a temporal boundary between conduct that counts as relevant to the current offense and conduct that is treated as criminal history. If a defendant was sentenced for another offense before the events constituting the current federal offense began, the earlier offense is assigned criminal history points rather than treated as expanded relevant conduct, even if the two offenses otherwise share characteristics that would qualify them as the same course of conduct or a common scheme or plan. The Guidelines illustrate this with a straightforward example. A defendant convicted in state court for selling cocaine, who upon release from prison immediately resumes selling cocaine using the same accomplices and methods, cannot have the earlier state offense treated as relevant conduct to the new federal charge. The prior conviction counts as criminal history. U.S. Sentencing Guidelines Manual § 1B1.3 cmt. n.5(C) (U.S. Sent’g Comm’n 2025).

The reverse scenario produces a different result. When a defendant is sentenced for another offense after commencing the current federal offense, and the two offenses share a qualifying connection, the other offense can be treated as relevant conduct. In that situation, the other sentence does not accrue criminal history points. This distinction is important because a defendant serving an undischarged term of imprisonment for relevant conduct is entitled to a sentence adjustment under §5G1.3(b) to account for time already served. Several circuits, including the Second, Third, Seventh, Eighth, and Ninth, have held that this adjustment may bring a sentence below an otherwise applicable statutory minimum, provided no statute specifically requires a consecutive sentence. United States v. Rivers, 329 F.3d 119 (2d Cir. 2003); United States v. Dorsey, 166 F.3d 558 (3d Cir. 1999); United States v. Ross, 219 F.3d 592 (7th Cir. 2000); United States v. Kiefer, 20 F.3d 874 (8th Cir. 1994); United States v. Drake, 49 F.3d 1438 (9th Cir. 1995).

That last point deserves further discussion because of its practical significance. The mechanisms for obtaining a sentence below a statutory mandatory minimum are few: primarily, a motion by the Government under 18 U.S.C. § 3553(e) based on the defendant’s substantial assistance (a §5K1.1 motion standing alone authorizes a departure below the Guidelines range but not below a statutory minimum, Melendez v. United States, 518 U.S. 120 (1996)) and, for qualifying defendants in certain drug cases, the safety valve of 18 U.S.C. § 3553(f). In the five circuits noted above, the §5G1.3(b) adjustment is another. For a defendant who has been arrested, convicted, and sentenced in a state system for conduct that is relevant conduct to a pending federal charge, who is still serving that state sentence at the time of federal sentencing and whose time already served will not be credited to the federal sentence by the Bureau of Prisons, the federal court should adjust the federal sentence downward to account for that time. In those five circuits, the adjustment can produce a federal sentence below the mandatory minimum without requiring the Government’s cooperation. Defense counsel handling cases with parallel state and federal proceedings should evaluate whether the state-court conduct qualifies as relevant conduct under §1B1.3(a)(1) through (a)(3), because that connection is a necessary predicate to a §5G1.3(b) adjustment. Conduct that qualifies only under subsection (a)(4) does not trigger the provision.

Extraterritorial Conduct

At least four federal circuits have held that relevant conduct can include acts committed outside the U.S. The Eleventh Circuit collected and joined these decisions in United States v. Spence, 923 F.3d 929 (11th Cir. 2019). But this position is not unanimous. The Fourth Circuit has cautioned that because relevant conduct must be criminal conduct, a sentencing court may err by considering losses from purely foreign conduct when setting the initial sentencing range. United States v. Elbaz, 52 F.4th 593 (4th Cir. 2022). The Fifth Circuit has held that certain acts a defendant committed in Mexico could not be considered relevant conduct to the federal offenses of illegally importing a firearm and illegal entry. United States v. Levario-Quiroz, 161 F.3d 903 (5th Cir. 1998). And the Second Circuit has held that consideration of a defendant’s foreign activities in calculating his base offense level for drug trafficking was inappropriate. United States v. Azeem, 946 F.2d 13 (2d Cir. 1991). Where a defendant’s criminal activity spanned international borders, the answer to whether foreign conduct counts at sentencing may depend on which circuit hears the case.

Conclusion

The relevant conduct doctrine is one of the most powerful tools in the federal sentencing arsenal. It allows courts to sentence defendants based on conduct that was never charged, never tried before a jury, and never admitted in a guilty plea. It permits punishment based on the acts of other people, so long as those acts were within the scope of a jointly undertaken criminal activity, in furtherance of that activity, and reasonably foreseeable in connection with it. And for a wide range of offenses, it travels backward and forward in time to capture conduct that shares a pattern or purpose with the crime of conviction.

The 2024 exclusion of acquitted conduct from the relevant conduct calculus marked a genuine reform. But the broader architecture remains intact. Defendants continue to face sentencing exposure for uncharged and dismissed conduct proven by only a preponderance of the evidence. The gap between the proof required to convict and the proof required to increase a sentence remains wide. And the relevant conduct rules continue to operate with a complexity that demands careful attention from anyone navigating the federal system. Understanding how §1B1.3 works is not an academic exercise. For defendants facing sentencing, it is often the single most important factor in determining how long they will spend in prison.  

Subscribe today

Already a subscriber? Login