A Call For Clarity: Drug Predicates Under § 4B1.1

Introduction 

In 2023, Ascension Garza Jr. was stopped at an immigration checkpoint in Falfurrias, Texas. A search of his car by Border Patrol agents uncovered sixteen kilograms of cocaine. Garza was arrested and pled guilty to possessing with intent to distribute over five kilograms of cocaine.

At first glance, the facts of USA v. Garza (5th Cir. 2025) are typical of a serious drug offense. But at sentencing, enhancements can dramatically raise the stakes. The length of a defendant’s sentence often depends on prior convictions, guidelines enhancements, and how the court assesses criminal history. A key issue in Garza’s case was whether § 4B1.1 of the Sentencing Guidelines, the “Career Offender” enhancement, applied at sentencing for his prior marijuana convictions. Section 4B1.1 amplifies a defendant’s criminal past, automatically placing them in Criminal History Category VI, the highest possible category, and at or near the statutory maximum term that can be served for their offense. Because hemp had been excluded from the controlled substance schedules for marijuana in 2018, the court determined that the enhancement did not apply for Garza’s prior marijuana convictions from 1992 and 1997, over twenty-five years prior to the instant offense. Without the enhancement, Garza was sentenced to twelve years of incarceration with five years of supervised release. 

If the court had applied the career offender enhancement, Garza’s sentencing range based on the Guidelines would have been 360 months to life, or at a minimum, thirty years. Luckily for Garza, he was sentenced in a time-of-sentencing jurisdiction.1 Garza’s case reflects some circuits’ position on § 4B1.1, which looks to the federal drug schedules at the time of sentencing to determine whether a defendant is designated a “career offender” and subject to a sentencing enhancement. But other circuits have adopted a time-of-conviction approach, consulting past drug schedules from when the defendant was originally convicted of the predicate offense to determine whether the enhancement applies. The result is disparate sentencing practices across jurisdictions,2 placing some defendants in the highest criminal history category but not others with similar records. More troubling is that many of these defendants, like Garza, carry prior marijuana convictions from ten or more years before reoffending.3 Notably, marijuana is a substance that has been reclassified on federal schedules and legalized in several states, calling into question the relevance of these prior convictions at sentencing for the instant offense.

Each year, more than half of criminal defendants subject to the career offender sentencing enhancement are those with prior drug convictions. Because the goal of the Sentencing Guidelines is to “inject transparency, consistency, and fairness” into federal sentencing, clarity on how courts should assess decriminalized drug offenses as § 4B1.1 predicates is needed to restore uniformity to the system and satisfy the Guidelines’ original goals. Courts applying the enhancement also continue to widen significant racial disparities in federal sentencing. In 2024, 59.1% of those subject to the career offender enhancement were Black compared to 22.8% who were white. 

Although the Supreme Court has addressed a similar timing issue for “serious drug crimes” as predicates under the Armed Career Criminal Act (ACCA), some courts misapply that reasoning to the career offender enhancement when it counsels more strongly in favor of a time-of-sentencing approach.4 This Essay argues in favor of a time-of-sentencing approach for § 4B1.1, distinguishing the reasoning used in ACCA cases to arrive at the time-of-conviction holding in that context. Moreover, because amendments to federal drug schedules reflect complex and evolving judgements by policymakers, sentencing should reflect the law at the time a defendant is sentenced to ensure that they are not punished for outdated classifications after already serving time for that offense. The time-of-sentencing approach comports with the Commission’s judgement that sentences should account for policy changes, including Congress’s reassessment of substances or penalties. It also argues that time limits are not foreign to the Guidelines, as the criminal history point system—the standard way to determine a defendant’s criminal history category—ages out prior convictions after ten to fifteen years for even serious offenses. To preserve the goals of the Guidelines, this Essay calls upon the Sentencing Commission to clarify its intent, place time limits on decriminalized drug predicates for § 4B1.1, and restore greater uniformity to the system. 

I.  Similar Defendants in Different Circuits Can Face Vastly Different Sentences 

Section 4B1.1 presents challenges for courts because it imposes harsh recidivist penalties for prior offenses but declines to define the drug offenses that qualify as predicates within its text. The other elements are clearer: A career offender must be at least eighteen at the time they committed the offense of conviction and have two prior felony convictions for a crime of violence or controlled substance offense. The text of § 4B1.1 itself does not incorporate a definition of controlled substances. Instead, it points to § 4B1.2, which states that a controlled substance offense is a federal or state offense, punishable by imprisonment for a year or more. Specifically, controlled substance offenses (1) prohibit the manufacture, distribution, or possession with the intent to distribute a controlled substance or (2) fit the description laid out in 46 U.S.C. § 70503(a) or § 70506(b). Those classified as career offenders are assigned to Criminal History Category VI, the highest category, and to offense levels at or near the statutory maximum for their offenses. Last year, 93% of individuals subject to the career offender enhancement received a longer sentence as a result.

Because the statute is unclear, circuit courts are divided on when to apply the enhancement for prior drug offenses, meaning that mandatory minimum sentences for similar defendants can vary significantly across jurisdictions. These defendants may have the same offense level but very different criminal history categories depending on whether the enhancement is applied. Currently, the Ninth, Fifth, Second, and First circuits support a time-of-sentencing approach when determining whether a defendant’s prior drug convictions count as § 4B1.1 predicates. These courts look to the version of the Controlled Substance Act (CSA) in effect at the time of federal sentencing. They rely on the Guidelines’ instruction to consider the Guidelines “in effect on the date the defendant is sentenced” unless it would violate the Ex Post Facto Clause, a provision that prohibits the government from passing laws that retroactively punish acts predating its passage. Some of these decisions also acknowledge that applying outdated law at sentencing risks undermining Congress’s current policy choices. The Ninth Circuit, for example, reasoned that applying superseded versions of the Code depending on when the defendant acted would create disparities and “illogically attach culpability at sentencing to conduct Congress has concluded is not culpable and dangerous.” This approach, the court explained, would hamper Congress’s ability to revise federal criminal law.

The Sixth, Eleventh, and Third circuits have endorsed a time-of-conviction approach for prior drug offenses.5 In these circuits, drug offenses count as predicates for § 4B1.1 if they were regulated by the CSA or state law when the defendant was convicted of that offense, even when the substance or classes of the substance have since been removed. If Garza was sentenced in any of these circuits, he would be subject to the enhancement for his twenty-five-year-old marijuana priors. Courts using this approach look to the Supreme Court’s ACCA precedent. Part II of this Essay demonstrates that courts’ references to the ACCA are inappropriate because the ACCA’s underlying focus on maximal incarceration is incongruent with § 4B1.1’s underlying focus on proportionality and recidivism. By incorrectly invoking the ACCA, courts have muddled the Sentencing Guidelines goals of fairness and uniformity.

II.  ACCA Precedent Does Not Comport with § 4B1.1’s Statutory Purpose

To answer the timing question, many circuit courts have looked to McNeil v. United States (U.S. 2011) and Brown v. United States (U.S. 2024), Supreme Court cases that dealt with similar timing issues with drug predicates under the ACCA. The ACCA imposes a fifteen-year mandatory minimum on anyone convicted of unlawful firearm possession in violation of 18 U.S.C. § 922(g) and who has three prior convictions for violent felonies or serious drug crimes punishable by a term of ten years or more under federal or state law. This Part focuses on Brown, a more recent case that builds upon similar reasoning in McNeil, examining the innately backward-looking nature of recidivist provisions and the ACCA’s emphasis on dangerousness.

In Brown, the Court held that a state drug conviction counts as an ACCA predicate if it involved a drug on the federal schedules at the time of that conviction. Like § 4B1.1 of the Guidelines, the ACCA focuses on recidivism. But the purpose of the ACCA and the class of offenders it pertains to is distinct because it sets a high bar for past dangerousness. The Court’s reasoning in Brown makes this clear by focusing on the risk of harm posed by armed defendants of this profile, suggesting that the outcome is not directly transferable to drug predicates in the career offender context. The Court pointed to Congress’s approach in crafting the ACCA: to single out “offenses of a certain level of seriousness that involve violence or an inherent risk thereof, and that are likely to be committed by career offenders.” A drug crime punishable by ten or more years of imprisonment, the Court reasoned, “augurs a risk of future dangerousness even if the drug is no longer considered dangerous.” The same is not necessarily true of a drug offense carrying a term of one year or more as covered in § 4B1.1. The Court also explained that defendants who have repeatedly committed the kind of serious crime covered by the ACCA are particularly likely to inflict grave harm when they are in possession of a firearm. Those subject to the ACCA reflect the statute’s concern with violent crime with over 83% of people carrying prior convictions for violent offenses in 2019.

Section 4B1.1 is not premised so heavily on dangerousness. The provision is aimed principally at recidivism and proportionality; it merely seeks to enhance punishments for those who have multiple offenses. For example, 82% of people subject to the career offender enhancement were repeat drug offenders in 2024. Unlike the ACCA, the enhancement does not reflect the same Congressional judgement that certain offenders pose an acute threat to public safety, requiring long-term incapacitation. The design of the ACCA effectuates that judgement by incorporating higher standards. ACCA predicates must be “serious drug offenses” punishable by more than ten years of imprisonment, and the statute requires three successive priors before the mandatory minimum applies. Its scheme in turn reflects a concern with extreme, not routine, recidivism. Section 4B1.1, by comparison, sweeps more broadly: A single drug offense punishable by more than one year can trigger the enhancement. The disparities in threshold and purpose between the ACCA and § 4B1.1 enhancements suggest that there is not a one-to-one match with the Court’s time-of-conviction holding in the ACCA context.

In Brown, the Court also grounded its conclusion in the reference canon, a principle of statutory interpretation used to determine whether a text’s reference to another law is fixed or incorporates subsequent revisions. The canon distinguishes between a statute that incorporates a general subject, which presumptively includes later developments in that area of law, and a statute that incorporates another law by specific reference, which is understood to freeze the law as it existed at the time of enactment. Applying this canon, the Court concluded that the ACCA’s cross-reference to § 102 of the CSA is a specific incorporation, not a general one. Because Congress tethered the ACCA’s definition of a “controlled substance offense” to a section of the CSA, it “cut and pasted” the CSA as it existed in 1986, when the ACCA was enacted, not as later amended. Unlike the ACCA, however, § 4B1.1 contains no express incorporation of the CSA; rather, § 4B1.1’s text references drug offenses more generally. Because the reference canon as applied here incorporates a general subject, the canon should be understood to reflect evolutions in the law, such as the decriminalization or rescheduling of controlled substances. Adhering to the canon respects the statutory intent, and a clarified understanding among circuits would avoid disparate, location-based sentencing practices. 

III.  Arguments for a Time-of-Sentencing Approach Beyond the ACCA Analog

There are other indications in the Guidelines text and compelling policy reasons to prefer time-of-sentencing for § 4B1.1. First, there is a disconnect between the Sixth, Third, and Eleventh circuits’ time-of-conviction approach and the Guidelines mandate to consider the version of the Guidelines “in effect at the time of sentencing.” This language accounts for the fact that there may have been changes to the Guidelines by the time the case arrives at the sentencing phase. In other words, the text affirmatively instructs courts to account for these changes and apply them going forward. Looking to past versions of the drug schedules, which the Guidelines incorporate, would contravene the text as well as the reference canon principle.  

Fairness and double-jeopardy-type concerns also caution against applying a time-of-conviction approach for § 4B1.1. Naturally, if a defendant was formerly convicted of the offense, the schedules in effect at the time of conviction would reflect that. However, those schedules may not reflect the law at the time the defendant committed the instant offense, the conduct for which they are ultimately being sentenced. The defendant has already served their time for the predicate offense. To be sure, patterns of criminal conduct are relevant at sentencing for purposes of deterrence and incapacitation. But offenses where policy has changed should not later place a defendant in the highest possible criminal history category, especially without any discernible benefits from increasing their sentence range by this extreme degree.  

One concern might be that the time-of-sentencing approach arbitrarily affects sentencing. The Court considered this argument in Brown: Looking to the most recent version of the CSA may yield different sentences for defendants merely because of legislative timing relative to the date a defendant is sentenced. Brown provided a hypothetical: Two co-defendants who commit the same state marijuana offense on the same day and engaged in felon-in-possession conduct could receive different sentences. The difference in their sentencing depends on the date of federal sentencing relative to the enactment of the Agricultural Improvement Act, which altered federal marijuana scheduling. Employing a time-of-sentencing approach in the § 4B1.1 context risks the same arbitrary disparity. 

But in reality, this scenario is not uncommon or unique to § 4B1.1. Statutory and regulatory changes occur often, and the federal sentencing system already accommodates evolving law through provisions like § 1B1.11, which directs courts to apply the Guidelines in effect at that time. Sentencing itself is often a highly individualized process, even for identical crimes occurring at the same time. Sentences may diverge due to differences in defendants’ histories, plea agreements, and other Guidelines factors. Adopting a time-of-sentencing approach, then, accords with the Commission’s judgement that sentences should account for the most up-to-date policy landscape, including Congress’s reassessment of substances or penalties, rather than artificially freezing the law at the time of prior convictions. This reasoning is different than “eras[ing]” a defendant’s past conduct. Far from creating arbitrary outcomes, this approach aligns sentencing with the Guidelines’ broader purpose of proportionality, fairness, and the Commission’s forward-looking mandate. 

IV.  When Prior Convictions Age Out at Sentencing: The Criminal History Points System

Aging out prior criminal conduct is not a novel concept for the Guidelines either. In cases in which a sentencing enhancement does not apply, a court calculates the defendant’s criminal history category based on a points system. Points are given to each of the person’s prior convictions. The number of points varies based on the severity of the offense. Notably, older convictions do not count at all; serious convictions expire at fifteen years, and other convictions expire at ten years. The Sentencing Commission created this sunsetting points system because studies demonstrate that the likelihood of recidivism declines over time. In turn, these old convictions lose their predictive value for a defendant’s risk level of offending in the future. Moreover, this time-limit ensures better proportionality in sentencing. Raising the criminal history category to the maximum level for a first-time reoffender for conduct that has since been decriminalized destroys proportionality in sentencing. There is no reason why any of these rationales change in the § 4B1.1 context. 

This argument does not dispute the narrower purpose the enhancement serves beyond criminal history points; there is a clear Congressional judgement that recidivism for drug and violent offenses should be punished more severely. But the enhancement wields too much power when it racks up points for conduct Congress has determined no longer carries the same criminal weight when the defendant is being sentenced. Criminal law is unique in its ability to incorporate evolving ideas about social policy and moral culpability. U.S. sentencing practices should reflect that progress. This age-out analogy adopts the reasoning of the points system to better serve the goals of the Guidelines, rather than undermining the purpose of the career offender enhancement which may still apply to true “career” offenders. 

Conclusion

A persistent circuit split governs how courts determine which prior drug offenses qualify as predicates under § 4B1.1, resulting in disparate sentencing outcomes for similar defendants across jurisdictions. Where the enhancement does apply, it increases sentences for those with prior drug convictions in 93% of cases. By allowing the application of § 4B1.1 to turn on unresolved timing questions, the current practice injects arbitrariness into one of the Guidelines’ more severe sentencing enhancements. The resulting discrepancies deny federal defendants fairness and uniformity at sentencing and fail to advance the Guidelines’ core objectives. Not only does this situation affect individual defendants, but it undermines the legitimacy of the whole sentencing regime: courts rejecting updated Congressional judgements on criminal policy overtly conflicts with the Guidelines instructions to incorporate its text at the time of sentencing. Furthermore, serious doubts about fairness arise if criminal punishment may depend on where a defendant is sentenced. Addressing this issue must be a priority in the Sentencing Commission’s 2025–2026 amendment cycle to restore fairness, uniformity, and transparency to federal sentencing. 

• United States v. Crocco, 15 F.4th 20,25 (1st Cir. 2021) (“Operating under that framework. . . the definition of controlled substance must be keyed to the version of the CSA in effect at the time of the instant federal sentencing–not a prior version.”).
• SeeUnited States v. Minor, 121 F.4th 1085, 1090 (5th Cir. 2024) (surveying the circuit split on when the enhancement applies).
• See, e.g., United States v. Jackson, 2022 WL 303231 at *1 (8th Cir. Feb. 2, 2022); United States v. Minor, 1212 F.4th 1085,1088 (5th Cir. 2024); United States v. Magana, 850 F. App'x 538, 539 (9th Cir. 2021); United States. v. Olivarez, WL 2089419 at *2 (S.D.Tex. 2015).
• See, e.g., United States v. Drake, 126 F.4th 1242, 1245 (6th Cir. 2025); United States v. Nelson, 151 F.4th 577, 584 (4th Cir. 2025); United States v. Harbin, 56 F.4th 843, 849 (10th Cir. 2022). 
• SeeUnited States v. Dubois, 94 F.4th 1284 (11th Cir. 2024) (the “term ‘controlled substance’. . . means a substance regulated by state law when the defendant was convicted of the prior state drug offense”).