In the
United States Court of Appeals For the Seventh Circuit
____________________
No. 21-1385 UNITED STATES OF AMERICA, Plaintiff-Appellee,
v.
CHRISTOPHER L. ASBURY, Defendant-Appellant.
____________________
Appeal from the United States District Court for the Central District of Illinois.
No. 19-cr-20061 — Colin S. Bruce, Judge.
____________________
ARGUED JANUARY 12, 2022 — DECIDED MARCH 3, 2022
____________________
Before FLAUM, EASTERBROOK, and WOOD, Circuit Judges. WOOD, Circuit Judge. Christopher Asbury found himself in federal court facing criminal charges for distribution of methamphetamine, in violation of 21 U.S.C. §§ 841(a)(1), (b)(1)(A)(viii). At his trial, the government introduced evidence showing that he had distributed the equivalent of 82.2 grams of pure (“ice”) methamphetamine; the jury credited that evidence and convicted him. Asbury does not contest his conviction on appeal. He argues only that the 2 No. 21-1385
district court erred in calculating his relevant conduct for purposes of the U.S. Sentencing Guidelines, that this error was prejudicial and not cured by the judge’s statement that any possible error would not have affected the sentence he chose, and that his sentence in any event is substantively unreasonable. The government has conceded the first of these points, but it urges us to uphold the sentence on grounds of harmless error. The judge’s brief statement, however, does not convince us that the guideline error was immaterial. We therefore vacate the sentence and remand this case for resentencing.
I
The underlying details of Asbury’s crime are largely irrelevant to the appeal, and so we omit most of them here. Suffice it to say that federal agents had their eyes on Asbury in June 2019, and they used such common methods as a confidential source and recorded phone calls to nab him. In a controlled buy conducted by the source, Asbury showed up with approximately 83 grams of 99% pure methamphetamine . In September of the same year, a federal grand jury returned an indictment charging Asbury with distributing at least 50 grams of the drug. It also alleged that Asbury had a prior conviction for a serious drug offense. Upon his conviction after a jury trial, he faced a statutory sentencing range of 15 years to life.
The Presentence Investigation Report (PSR) that was prepared for the sentencing proceeding began by noting that the total quantity of pure methamphetamine involved in the offense was 82.2 grams (that is, 99% of 83.1). Using the drug
No. 21-1385 3
conversion tables provided in U.S.S.G. § 2D1.1, cmt. 8(D) (table 2), 1 the PSR calculated a converted drug weight of 1,644 kilograms. But that was nowhere near all. Asserting that it was reflecting reports from others who had furnished information, paragraph 15 of the PSR proposed holding Asbury responsible for a much greater quantity:
According to an interview conducted with a con-
fidential source of information, Asbury distributed approximately 16 ounces of methamphetamine ice in early 2018, approximately 210–250 ounces of methamphetamine ice from early 2018 through July 2018, approximately 30 ounces of methamphetamine ice in late July of 2018, then an additional 72 ounces through late 2018 and 2019, totaling 328 ounces (9,298.8 grams.)
Paragraph 16 added even more: “Additional interviews with Timothy Shannon, Jon Jacobs, and Brandon Hall, show Asbury distributed approximately 230 ounces (6,520.5 grams) from Summer 2017 through February 2019.”
The PSR concluded that Asbury was responsible for 15,819.3 grams of a mixture containing methamphetamine, in addition to the 82.2 grams of the pure drug underlying the
1 From 1991 until 2018, section 2D1.1 of the Guidelines provided “Drug Equivalency Tables,” which converted every drug into its marihuana equivalent. Amendment 808, which was effective November 1, 2018, “replace[d] marihuana as the conversion factor with the new term ‘converted drug weight’ and [changed] the title of the Drug Equivalency Tables to the ‘Drug Conversion Tables.” See U.S.S.G. § 2D1.1, Background at 169 (2021 Manual). Because the district court in this case used the converted weights, we do the same in the interest of consistency.
4 No. 21-1385
offense of conviction. The alleged additional conduct added 31,638.6 kilograms of converted weight to the 1,644 kilograms of converted weight for the ice, for a total of 33,282.6 kilograms of converted weight. Asbury filed a timely objection to those calculations; he insisted that he was not responsible for this alleged relevant conduct. But no additional information or support was forthcoming. The author of the PSR replied only that “[t]hese amounts were obtained from investigative reports prepared by law enforcement that this [probation] office deemed credible.”
Asbury reiterated his objections to paragraphs 15, 16, and 24 of the PSR at the sentencing hearing, but the judge was not moved. That was so even though, when the judge appropriately asked the Assistant U.S. Attorney whether the prosecution intended to introduce any additional evidence, he was told that it did not. Rather than nail down the factual basis for the additional drug-quantity allegations, the court proceeded immediately to the general question whether the distribution of drugs other than those directly involved in the offense could be considered as relevant conduct. Guideline 1B1.3 answers that question in the affirmative, if the relevant conduct is proved by a preponderance of the evidence.
After a brief comment on relevant conduct, the court said that it was “adopt[ing] the probation officer’s position.” This had the effect of raising Asbury’s offense level from 30 to 36. U.S.S.G. §§ 2D1.1(c)(2), (5). The court then added two levels for perjury, raising the offense level to 38. Asbury had a criminal history of VI, which, along with his offense level, gave him an advisory guideline range of 360 months to life. Had the offense level been 32 (i.e., the level without the
No. 21-1385 5
relevant conduct, but with the perjury enhancement), his range would have been 210 to 262 months. Because of Asbury’s prior conviction for a serious felony, his statutory minimum sentence was 15 years (180 months), and the maximum was life.
The court prefaced its discussion of the sentencing factors set forth in 18 U.S.C. § 3553(a) with the following statements:
Let me add one more thing. Let me make the record clear about this: Even if I erred in some way in the calculation of the applicable guidelines as far as the criminal history calculation goes, even if I was wrong, the defendant—and removed a number of points—the defendant would still be a criminal history category VI.
Likewise, as far as the offense level, I’m cognizant of the guidelines. And if I made an error in the guideline calculation in terms of offense level, that would not affect my sentence.
I’m basing my sentence on the Section 3553(a) factors and the exercise of my discretion after placing a lot of thought into this sentencing hearing. After a brief review of the section 3553(a) factors, the court made the following intriguing comment:
[T]here’s no doubt in my mind that any sentence I impose within the statutory range or the guideline range—but, more importantly, within the statutory range—would adequately deter Mr. Asbury from committing any further conduct. He’s 40 years old at this time. Even if I were to impose the statutory mandatory minimum [of 180 months], he would not be 6 No. 21-1385
out until he was in his early 50s. Likewise, I am to protect the public from further crimes of the defendant . The defendant will be out of commission for a lengthy period of time.
With that, the court imposed a sentence of 360 months, double the statutory mandatory minimum. It did not explain why, if a 180-month sentence would have sufficed to deter Asbury from further criminal behavior and to protect the public, the other factors were so compelling that the greater sentence was necessary.
II
Although Asbury has raised a number of arguments on appeal, the only one we need to reach is whether the district court’s erroneous reliance on the unsupported and disputed allegations in the PSR was harmless. The government has “concede[d] that the drug-quantity finding cannot be sustained.” We are not bound to accept such a concession, but in this case we have no trouble doing so. See United States v. Helding, 948 F.3d 864, 869–72 (7th Cir. 2020) (finding error where “no step [was] taken to find some modicum of reliability of the CI information supplied to the probation officer charged with preparing the PSR,” and concluding that this error was not harmless); see also United States v. Gibbs, No. 20-3304, __ F.4th __, 2022 WL 522601, at *3 (7th Cir. Feb. 22, 2022) (“The government bears the burden of proving by a preponderance of the evidence that uncharged drug quantities are attributable to a defendant.”). We therefore focus on the question whether the court’s sentence can be salvaged notwithstanding the error.
No. 21-1385 7
Sentencing proceedings in federal district courts proceed in two steps: first, the judge must calculate the correct advisory sentencing range under the Sentencing Guidelines; and second, the judge must consider the factors set out in section 3553(a). Both of these steps are important—indeed, both are compelled by statute. The requirement of beginning with the guidelines appears in 18 U.S.C. § 3553(a)(4)—a part of the Sentencing Reform Act that was left undisturbed by United States v. Booker, 543 U.S. 220 (2005). It says that “[t]he court, in determining the particular sentence to be imposed, shall consider … the kinds of sentence and the sentencing range established for—(A) the applicable category of offense committed by the applicable category of defendant as set forth in the guidelines—(i) issued by the Sentencing Commission … .” (Emphasis added.) The requirement to take a broader look at the appropriate sentence is found in 18 U.S.C. § 3553(a), which also uses mandatory language in directing the court to take the listed factors into account.
Under this system, the court must start with the guidelines , but it must then weigh the factors set out in section 3553(a). A district court is entitled to disagree with the sentencing philosophy of the guidelines, see Kimbrough v. United States, 552 U.S. 85 (2007), and so it may reject the advisory range. When, and to the extent, it does so, however, it must explain the final sentence and indicate which of the permissible sentencing considerations persuaded it to do so. That explanation must address any “ground of recognized legal merit” raised by the defense that is supported by a factual basis in the record. See United States v. Castaldi, 743 F.3d 589, 595 (7th Cir. 2014).
8 No. 21-1385
The statute does not give the judge the option to bypass the guidelines, and the Supreme Court has underscored this fact. In Rosales-Mireles v. United States, 138 S. Ct. 1897 (2018), it stated that “district courts must begin their analysis with the Guidelines and remain cognizant of them throughout the sentencing proceedings.” Id. at 1904, quoting Peugh v. United States, 569 U.S. 530, 541 (2013), in turn quoting Gall v. United States, 552 U.S. 38, 50 n.6 (2007). Once the judge has done so, however, she has the discretion to explain how much weight (if any) the contested guideline issue had for purposes of the final sentence. If, for example, there is a dispute about drug quantity, the judge must resolve that dispute, but the judge may also reach a final sentence based on the defendant’s criminal history, rather than the difference between the guideline recommendation for different quantities of drugs.
Nor does the statute permit the judge to nullify the guidelines by way of a simple assertion that any latent errors in the guidelines calculation would make no difference to the choice of sentence. In United States v. Abbas, 560 F.3d 660, 667 (7th Cir. 2009), we explained that when the sentencing judge gives “a detailed explanation” of the basis for a “parallel result”—a sentence different from the guideline recommendation—that explanation may render an error in the guidelines calculation harmless. Abbas permits district courts to, in a sense, inoculate their sentences against reversal by giving us the information we need to determine, on appeal, whether an error was harmless without resort to a remand. Such inoculating statements have proliferated since we approved them in Abbas, as this and several other recent cases illustrate. See, e.g., United States v. Bravo, No. 20-1105, __ F.4th __, 2022 WL 420543, at *5–6 (7th Cir. Feb. 11, 2022); United States. v. Skaggs, 25 F.4th 494, 499–500 (7th Cir. 2022).
No. 21-1385 9
Having noticed the frequency with which sentencing judges are relying on inoculating statements, we take this occasion to reiterate what we first said in Abbas and have repeated since: “a conclusory comment tossed in for good measure” is not enough to make a guidelines error harmless. Abbas, 560 F.3d at 667; see also United States v. Loving, 22 F.4th 630, 636 (7th Cir. 2022) (“[W]e cannot infer, based on the district court’s terse comments about the sentencing factors under 18 U.S.C. § 3553(a) that the court believed a 71- month prison sentence would be appropriate regardless of the correct guideline range”). There are no “magic words” in sentencing. If there were, the judge would have no incentive to work through the guideline calculations: she could just recite at the outset that she does not find the guidelines helpful and proceed to sentence based exclusively on her own preferences.
As Abbas suggested, a district court’s statement purporting to inoculate its chosen sentence against errors identified on appeal will be effective only if two conditions are satisfied. First, the inoculating statement must be “detailed.” Abbas, 560 F.3d at 667. By that, we mean that the judge must give specific (though not necessarily lengthy) attention to the contested guideline issue in her explanation. A generic disclaimer of all possible errors will not do. Second, the inoculating statement must explain the “parallel result.” Id. By that, we mean that it must be “tied to the decisions the court made” and account for why the potential error would not “affect the ultimate outcome.” Bravo, 2022 WL 420543, at *5. When an inoculating statement fails to satisfy either of these two criteria, we cannot say with confidence that the district court would have reached the 10 No. 21-1385
same sentence despite the guidelines error. It follows that we cannot say whether the error was harmless.
To illustrate: a judge might explain that she would have reached the same sentence whether the defendant had an offense level of 30 or an offense level of 34, coupled with a criminal history of V. But the judge would need to give a reason for such a conclusion, explaining why the difference between an advisory range of 151 to 188 months (level 30, history V) and a range of 235 to 293 months (level 34, history V) did not provide useful guidance for sentencing that particular defendant. Our recent opinion in Bravo, in which we reversed and remanded for resentencing despite an inoculating statement, provides an example along these lines. See id. at *5–6.
Abbas itself provides an example of an effective inoculating statement. The error in that case was the district court’s reliance on guideline 2C1.1, which applies to crimes constituting “extortion under color of official right,” in sentencing a defendant who had impersonated an FBI agent. Abbas, 560 F.3d at 661. We said that “[t]he issue was clear—whether Abbas was subject to extortion under color of official right liability; the judge ruled that he was, but then indicated that, even if he was not, she would have exercised her discretion to apply the same enhancement based on the [section] 3553(a) factors.” Id. at 667. Because the judge had already addressed and accounted for the specific possible error, her inoculating statement rendered that error harmless.
Expanding the Abbas principle beyond its original contours would be inconsistent with a long line of Supreme Court decisions. The Court has said that the “[f]ailure to calculate the correct Guidelines range constitutes procedural
No. 21-1385 11
error[.]” Peugh, 569 U.S. at 537. In general, this type of error is prejudicial. As the Court put it, “[w]hen a defendant is sentenced under an incorrect Guidelines range—whether or not the defendant’s ultimate sentence falls within the correct range—the error itself can, and most often will, be sufficient to show a reasonable probability of a different outcome absent the error.” Rosales-Mirales, 138 S. Ct. at 1907 (emphasis added). Before finding such an error harmless, therefore, we must be sure that it “did not affect the district court’s selection of the sentence imposed.” United States v. Jett, 982 F.3d 1072, 1078 (7th Cir. 2020).
III
This record does not give us the necessary confidence that the conceded guideline error that occurred here was harmless. At no point did the district court squarely address the possibility that it might have erred in its drug-quantity finding—indeed, if it had done so, it undoubtedly would have noticed that there was no evidentiary support for the allegations in paragraphs 15, 16, and 24 of the PSR. Nor did the court explain why it would have given exactly the same sentence even if the relevant conduct were excluded. Indeed, in the passage we quoted earlier, the court seemed to admit that a 360-month sentence was greater than necessary to comply with the purposes set forth in section 3553(a)(2)— actually twice as long, since the 15-year statutory minimum is only 180 months. That is inconsistent with section 3553(a), which states that “[t]he court shall impose a sentence sufficient, but not greater than necessary, to comply with the purposes set forth in paragraph (2) of this subsection.” (Emphasis added.)
12 No. 21-1385
We realize that the judge did say that “if I made an error in the guideline calculation in terms of offense level, that would not affect my sentence.” But that is not the sort of specific explanation of a parallel result that Abbas calls for. Possible problems on appeal might have included disputes over relevant conduct, disputes over drug purity, disputes over the perjury enhancement, and maybe others. The district court’s generic disclaimer sheds no light on which, if any, of those potential errors it had in mind. It is thus not specific enough to permit a finding of harmless error.
Nor did the court provide us with an adequate explanation of the parallel result. We recognize that it emphasized Asbury’s extensive criminal history when it announced its sentence. (Asbury’s score was 34, more than 2.5 times the 13 points needed to fall within the highest category, VI.) But the court offered almost no explanation related to the severity of the offense of conviction. The judge just said: “This is a drug distribution offense. That means it’s a serious offense. It is not a petty offense, by any means; it’s serious.” This explained nothing. We can assume that crimes with a statutory minimum of 15 years are serious, rather than petty. But the precise severity depends on the quantity of drugs that was attributable to the scheme—the very point that the court brushed by. We do not know if it would have made the same assessment of severity if it thought it was dealing with only 5% as much methamphetamine as the PSR had found.
Finally, the court had nothing to say about the need to avoid unwarranted sentencing disparities. If it meant to indicate that some upward deviation from the statutory minimum of 15 years (180 months) was warranted to account for Asbury’s recidivism (and this is just a guess), the court could
No. 21-1385 13
have achieved that goal within the proper guideline range of 210 to 262 months. It is worth noting, in this connection, that the sentence the court chose was one at the bottom of the range that it adopted (360 months to life). This suggests that the guideline calculation did affect its sentencing decision.
IV
As we noted earlier, the Supreme Court has said that an error in calculating a defendant’s guideline range “can, and most often will, be sufficient to show a reasonable probability of a different outcome absent the error.” Rosales- Mireles, 138 S. Ct. at 1907. Asbury’s case is not one of the rare ones that falls outside the general rule. The district court did not assure itself that there was an adequate factual basis for the drug quantity reported in the PSR, and its brief comment to the effect that the guidelines had not driven its sentence was not enough to fix this problem. In fact, the court’s later remarks suggesting that even a 180-month sentence would have sufficed confused things further.
We therefore VACATE the sentence and REMAND this case for re-sentencing using an offense level that does not rely on the contested relevant conduct.