United States v. Emmanuel Trencell Merritt

RECOMMENDED FOR PUBLICATION Pursuant to Sixth Circuit I.O.P. 32.1(b) File Name: 24a0117p.06

UNITED STATES COURT OF APPEALS FOR THE SIXTH CIRCUIT

┐ UNITED STATES OF AMERICA, │ Plaintiff-Appellee, │ > No. 23-1216 │ v. │ │ EMMANUEL TRENCELL MERRITT, │ Defendant-Appellant. │ ┘

Appeal from the United States District Court for the Western District of Michigan at Grand Rapids. No. 1:21-cr-00051-1—Jane M. Beckering, District Judge.

Decided and Filed: May 21, 2024

Before: BUSH, NALBANDIAN, and MURPHY, Circuit Judges.

_________________

COUNSEL

ON BRIEF: Paul L. Nelson, OFFICE OF THE FEDERAL PUBLIC DEFENDER, Grand Rapids, Michigan, for Appellant. Erin K. Lane, UNITED STATES ATTORNEY’S OFFICE, Grand Rapids, Michigan, for Appellee. _________________

OPINION _________________

NALBANDIAN, Circuit Judge. Emmanuel Merritt pleaded guilty to being a felon in possession of firearms and was sentenced to 120 months in prison. But two twists underlie this seemingly straightforward case. First, before Merritt pleaded guilty, he spent several months evading authorities. So the district court faulted him for not accepting responsibility and denied him a two-point reduction in his total offense level, which Merritt challenges as an error.

No. 23-1216 United States v. Merritt Page 2

Second, the district court gave Merritt a three-point increase in his criminal history score because one of his past state offenses included a total prison sentence of 870 days. But Merritt believes that his sentence was for only 330 days, because of credits he received, and the district court was thus wrong in calculating his score. Both claims fail, however, so we AFFIRM.

I.

A.

On January 12, 2021, Grand Rapids police were searching for Emmanuel Merritt, who had two outstanding arrest warrants, when they received information that he had checked into a hotel. Video surveillance showed Merritt arriving with a female in a white Kia Sportage the day before. When officers approached Merritt outside the hotel, he ran but was soon apprehended. A search revealed that Merritt was carrying $1,724 and a cell phone.

Officers spoke with the woman accompanying Merritt, and she told them she drove Merritt to the hotel the night before with Merritt riding in the front passenger seat. Because the Kia Sportage was her car, she permitted officers to search it, which revealed two handguns under the front passenger seat. The woman said she knew nothing about the guns and did not know they were there. Officers discovered evidence connecting Merritt to the guns on his cell phone, including photos and videos of him holding the guns recovered from the Kia Sportage and another firearm that police had recovered earlier from an abandoned vehicle. That abandoned vehicle contained two guns alongside AT&T store receipts bearing Merritt’s name.

On March 10, 2021, a federal grand jury indicted Merritt on one count of being a felon in possession of firearms—the two pistols found in the Kia Sportage—in violation of 18 U.S.C. § 922(g)(1). Merritt was arrested on October 18, 2021, and released on bond a few days later.1 But less than one month into his pretrial release, Merritt allegedly “threatened to shoot up the residence of [his girlfriend] and her family members.” R.49, PSR, p.5 ¶9, PageID 193. Merritt

1 “Merritt was arrested on state charges mirroring the instant offense on January 12, 2021. He was released on bond in that case on February 10, 2021. The state charges related to firearms were later dismissed” because of the federal indictment. R.49, pp.4 ¶7, 15 ¶55, PageID 192, 203.

No. 23-1216 United States v. Merritt Page 3

later showed up to his girlfriend’s workplace, which she reported to the police. He admitted to violating the conditions of his release but denied making threats.

As a result, Pretrial Services modified his bond conditions to allow Merritt to move out from living with his girlfriend. He gave authorities a new address, but when Pretrial Services conducted a home visit on November 17, 2021, Merritt was not there. The owner of the residence told them that she knew Merritt but he had never lived there. So “Pretrial Services was unaware of Mr. Merritt’s current whereabouts.” R.49, p.6 ¶9, PageID 194. Perhaps unsurprisingly, when it came time for Merritt’s final pretrial conference two days later, Merritt was nowhere to be found. So the district court issued a warrant for Merritt’s arrest.

Two days later, officers pulled Merritt over in a rental car in Sugarland, Texas. Once they discovered the federal warrant for Merritt’s arrest, they asked him to exit the vehicle, but he sped off and evaded arrest. Eight months later, in July 2022, Ohio State Troopers arrested Merritt in northern Ohio.

B.

In August 2022, Merritt pleaded guilty to one count of felony possession of firearms. A probation officer prepared an initial presentence report (PSR) for the sentencing. This initial PSR did not include an adjustment for obstruction of justice, but it did include a three-level reduction for acceptance of responsibility. The government objected to both the inclusion of an acceptance-of-responsibility reduction and the exclusion of an obstruction-of-justice adjustment. For his part, Merritt objected to the scoring of a prior state offense in his criminal history calculation and argued that he should receive credit for accepting responsibility.

In response, the probation officer prepared a final PSR, this time adding a two-level enhancement for obstruction of justice and removing the previous acceptance-of-responsibility reduction. This put Merritt’s total offense level at 28, and with a criminal history category IV, Merritt’s imprisonment range was 110–137 months. Because the then-existing statutory maximum for his crime was 120 months, see 18 U.S.C. § 924(a)(2) (2018), his range was reduced to 110–120 months. Merritt again objected to the denial of the acceptance-of- responsibility reduction and to the calculation of his criminal history score.

No. 23-1216 United States v. Merritt Page 4

At sentencing, the district court disagreed with both objections.2 As for Merritt’s argument that he accepted responsibility after he had been recaptured, the court found that “given the whole history of [Merritt’s] conduct in this case,” Merritt had “not clearly demonstrated acceptance of responsibility,” so the district court did not grant the two-level reduction. R.65, Sent. Tr., p.30, PageID 347. And on Merritt’s challenge to his criminal history score, the district court found that the PSR scored it “correctly.” Id. at pp.40–41, PageID 357– 58. The district court sentenced Merritt to a within-Guidelines sentence of 120 months in prison. Id. at p.58, PageID 375. Merritt timely appealed. He argues that the district court (1) erred in denying the acceptance-of-responsibility reduction and (2) did not properly calculate the criminal history score. We examine each in turn.

II.

A.

The Sentencing Guidelines tell district courts to “decrease the offense level by 2 levels” if “the defendant clearly demonstrates acceptance of responsibility for his offense.” U.S.S.G. § 3E1.1(a). When a defendant has also obstructed justice, both adjustments may be warranted in “extraordinary cases,”3 but “courts have employed an exacting standard to determine whether a defendant has accepted responsibility after having obstructed justice.” United States v. Williams, 940 F.2d 176, 183 (6th Cir. 1991).

Regarding our standard of review when considering these types of issues, our court pivoted from de novo to clear-error review nearly two decades ago. Before Buford v. United

2 Merritt presented four claims at sentencing, but we discuss only the two relevant on appeal. Merritt also sought a downward variance based on the criminal history scoring. But he does not challenge the denial of the variance on appeal. 3 This “extraordinary cases” language comes from the commentary to the Guidelines. U.S.S.G. § 3E1.1 cmt. n.4. But the district court below found the text of § 3E1.1(a) unambiguous and therefore did not resort to the commentary. See R.65, p.25, PageID 342 (explaining that the judge would not rely on the commentary because the “extraordinary cases” language “basically add[s] an additional substantive rule” that isn’t in the text of the Guidelines); see also id. at p.29, PageID 346. To the extent the commentary adds to the Guideline, the court was correct to not consider it. Commentary cannot “replace or modify” the text itself. United States v. Havis, 927 F.3d 382, 386 (6th Cir. 2019) (en banc) (per curiam); see also United States v. Riccardi, 989 F.3d 476, 479 (6th Cir. 2021). And we agree with the district court that the text of the Guidelines is clear enough that we need not rely on the commentary, and Merritt does not challenge this on appeal.

No. 23-1216 United States v. Merritt Page 5

States, 532 U.S. 59 (2001), we consistently reviewed district courts’ application of an acceptance-of-responsibility enhancement to undisputed facts de novo. See United States v. Holland, 1 F. App’x 266, 267 (6th Cir. 2001); United States v. Whitman, 209 F.3d 619, 622 (6th Cir. 2000); United States v. Tilford, 224 F.3d 865, 867 (6th Cir. 2000); United States v. Jeter, 191 F.3d 637, 638 (6th Cir. 1999); United States v. Childers, 86 F.3d 562, 563 (6th Cir. 1996). But we changed in 2002, after Buford said that district courts were entitled to deference in their application of the career-offender enhancement to undisputed facts. See United States v. Miller, 45 F. App’x 359, 364 (6th Cir. 2002) (referencing Buford, 532 U.S. at 63–66). Cases acknowledging that pivot have since applied clear-error review to district courts’ application of the acceptance-of-responsibility enhancement to undisputed facts. See United States v. Searer, 636 F. App’x 258, 260 n.1 (6th Cir. 2016); United States v. Arzola, 528 F. App’x 487, 495 (6th Cir. 2013); United States v. Fortner, 491 F. App’x 692, 695 (6th Cir. 2012); United States v. Redmond, 475 F. App’x 603, 612 (6th Cir. 2012); United States v. Genschow, 645 F.3d 803, 813 (6th Cir. 2011); United States v. Balboa-Gallardo, 417 F. App’x 459, 460 (6th Cir. 2011) (per curiam); United States v. Clements, 142 F. App’x 223, 226 (6th Cir. 2005); United States v. Brown, 367 F.3d 549, 556 (6th Cir. 2004); United States v. Webb, 335 F.3d 534 (6th Cir. 2003). We’ve reflected this shift away from de novo review in recent cases emphasizing the “great deference” given to district courts on this question. United States v. Histed, 93 F.4th 948, 961 (6th Cir. 2024) (quoting United States v. McCloud, 730 F.3d 600, 605 (6th Cir. 2013)).

For example, we explained in Webb that we afford “great deference” or “due deference” to the district court’s decision. 335 F.3d at 537–38 (first quoting U.S. Sentencing Guidelines Manual § 3E1.1 cmt. n.5 (2000); and then quoting 18 U.S.C. § 3742(e)). This tracks our general tendency to defer to a district court’s application of Guidelines to facts, which arises because most of these application questions are fact-bound. See United States v. McCullough, No. 22- 3984, 2023 WL 8717228, at *2 (6th Cir. Dec. 18, 2023) (citing United States v. Wallace, 51 F.4th 177, 183 (6th Cir. 2022)). Although that tendency does not apply across the board. See United States v. Thomas, 933 F.3d 605, 608–09 (6th Cir. 2019) (collecting cases). Moreover, as we note above, we have gone so far as to say that we review the acceptance-of-responsibility question for “clear error”—even though that standard is generally reserved for straight fact questions. Genschow, 645 F.3d at 813; Webb, 335 F.3d at 537.

No. 23-1216 United States v. Merritt Page 6

All of this seems straightforward and would point indisputably to clear-error review of acceptance-of-responsibility questions. But we have not always been consistent on this point. See Thomas, 933 F.3d at 611 (“[W]e note that our cases have also diverged on the standard of review for § 3E1.1.”). Sometimes, even post-Buford, we have “said that ‘if the only issue presented is the propriety of applying the reduction to the uncontested facts, the decision is reviewed de novo.’” Id. (quoting United States v. Denson, 728 F.3d 603, 614 (6th Cir. 2013)).4

When we confronted this issue in Thomas, we explained that Buford—which our circuit still applies—used “guideline-specific logic.” 933 F.3d at 609. So rather than make categorial pronouncements about standards of review, we decide standard-of-review questions “guideline- by-guideline.” Id. This is consistent with our practice in past cases. See United States v. Shanklin, 924 F.3d 905, 919 (6th Cir. 2019) (according due deference to whether firearm was possessed in connection with another felony under U.S.S.G. § 2K2.1(b)(6)(B)); United States v. Whited, 473 F.3d 296, 297 (6th Cir. 2007) (reviewing de novo whether a crime involved “a substantial risk of harm to human life” under U.S.S.G. § 2D1.1(b)(5)(B)). And it comports with the Supreme Court’s recent adoption of context-specific logic like Buford’s in U.S. Bank National Association ex rel. CWCapital Asset Management LLC v. Village at Lakeridge, LLC, 583 U.S. 387, 395–96 (2018). Thomas, 933 F.3d at 609.

With each Guideline requiring its own context-specific inquiry, we can glean from Buford where we look to answer this question. We ask whether the sentencing or appellate court is “better position[ed]” to answer the question posed by the Guideline at issue. Id. at 609–11 (quoting Buford, 532 U.S. at 64). And for the answer, we consider the “relative institutional advantages enjoyed by the district court in making the type of determination at issue,” such as whether this Guideline implicates the values of sentencing courts’ deeper experience applying the Guidelines to various facts and their more intimate familiarity with the facts of each case. Buford, 532 U.S. at 64–65.

4 Perhaps just as many of our cases still apply de novo review as those that apply clear-error review—even after the court’s pivot in 2002, thus adding to the confusion. See United States v. Smith, 782 F. App’x 383, 386 (6th Cir. 2019); United States v. Hollis, 823 F.3d 1045, 1047 (6th Cir. 2016) (per curiam); United States v. Kennedy, 595 F. App’x 584, 590 (6th Cir. 2015); Denson, 728 F.3d at 614; United States v. Hakley, 101 F. App’x 122, 125–26 (6th Cir. 2004); United States v. Reaume, 338 F.3d 577, 582 (6th Cir. 2003).

No. 23-1216 United States v. Merritt Page 7

Applying Buford’s approach here, multiple considerations cut in favor of deferring to the lower court on questions of acceptance of responsibility. A “district judge sees many more” of these challenges “than does an appellate judge,” id. at 64, and is thus better equipped to tell when a defendant has sincerely accepted responsibility. And when assessing acceptance of responsibility, “factual nuance may closely guide the legal decision, with legal results depending heavily upon an understanding of the significance of case-specific details,” id. at 65, like the understanding of whether the defendant was on the run, whether he willfully turned himself in, and if or when he decided to finally comply with authorities. Lower courts are best positioned to apply the Guidelines to such historical facts. See U.S. Bank, 583 U.S. at 394; Miller v. Fenton, 474 U.S. 104, 114 (1985) (explaining that if an “issue falls somewhere between a pristine legal standard and a simple historical fact,” the appropriate standard of review will often reflect which “judicial actor is better positioned” to make the decision). Plus, judging whether an individual has accepted responsibility for his crimes seems akin to a factual finding “based on credibility,” which “demand[s] great deference to the district court because ‘only the trial judge can be aware of the variations in demeanor and tone of voice that bear so heavily on the listener’s understanding of and belief in what is said.’” United States v. Ruiz-Lopez, 53 F.4th 400, 402–03 (6th Cir. 2022) (quoting Anderson v. City of Bessemer City, 470 U.S. 564, 575 (1985)). So the sentencing court sits in the best institutional posture to answer this fact-bound question of responsibility.

Because the Supreme Court’s reasoning in Buford explains the several advantages district courts have in evaluating fact-intensive questions like this one, and because our early published decisions used deference language—even explicitly rejecting de novo review, see Genschow, 645 F.3d at 813 n.7; see also Webb, 335 F.3d at 537–38—the correct course appears to be deferring to the trial court. So we review the question of whether a defendant has demonstrated that he accepted responsibility for clear error, a standard that reflects the great deference we grant a district court’s decision on this factual matter. And Merritt bears the burden of demonstrating that he accepted responsibility. United States v. Guerrero, 76 F.4th 519, 531 (6th Cir. 2023).

No. 23-1216 United States v. Merritt Page 8

B.

The district court found that “the whole history of [Merritt’s] conduct in this case” shows that he had “not clearly demonstrated acceptance of responsibility.” R.65, p.30, PageID 347. Merritt believes his conduct since he was returned to custody demonstrates otherwise. The government defends the district court’s conclusion, arguing that it did not err in denying the two- point reduction. We agree with the government and affirm the district court on this question.

The district court probed the record to determine whether Merritt had accepted responsibility for his crime. Walking through the series of events at sentencing, the district court explained that “Merritt[] was on . . . unsecured bond, a federal detention bond, and he absconds.” Merritt then has two problems while his bond was being modified: “He’s going back and following his girlfriend to her place of work, which he admits,” and he “gives our detention office a fake address for someone who knows him, but he’s never living there, and he’s gone.” He’s then “found a few days later” in “Texas at a traffic stop” where he’s “asked to pull over,” but he “doesn’t obey” and instead “flees, and he gets away.” Then “[e]ight months later we find him in Ohio.” The district court concluded “that under no paradigm” do these actions “manifest[] clear acceptance of responsibility for the offense. It’s the opposite.” Id.

Precedent supports the district court’s conclusion. In United States v. Watkins, we found the denial of an acceptance-of-responsibility reduction was proper because “the district court was permitted to consider [the defendant’s] absconding from supervised release prior to sentencing, remaining at-large for three months on an outstanding arrest warrant, and failing to appear for sentencing.” 86 F. App’x 934, 938 (6th Cir. 2004); see also United States v. Anderson, No. 23- 1127, 2024 WL 209225, at *2 (6th Cir. Jan. 19, 2024) (upholding denial of reduction where defendant fled “from pretrial supervision” and “remained a fugitive from justice for over fourteen months”).

And we have upheld a district court’s denial of the acceptance-of-responsibility reduction when the defendant “willfully fail[ed] to appear” at the “final pretrial conference.” United States v. Heighton, 272 F. App’x 469, 472 (6th Cir. 2008). That is exactly what happened here— Merritt was presumably heading for Texas at the time of his final pretrial conference. Far from

No. 23-1216 United States v. Merritt Page 9

showing a desire to accept responsibility, this behavior reveals “a desire to avoid punishment for the charged offense.” Watkins, 86 F. App’x at 938.

What’s more, we have held that simply pleading guilty does not demonstrate acceptance of responsibility. In Anderson, after running from authorities for several months, the defendant was eventually apprehended and pleaded guilty. 2024 WL 209225, at *2. But we held that the only evidence the defendant mustered was “his guilty plea, his participation in the presentence investigation process, and his expressions of remorse,” id., which is just like Merritt’s conduct here. “These acts, though praiseworthy, constitute an entirely ordinary course of conduct for those faced with a criminal indictment.” Id.; see also United States v. Starks, 64 F. App’x 501, 502 (6th Cir. 2003) (holding defendant did not deserve reduction where he was “a fugitive from justice for nearly nine months” then “merely pleaded guilty following his apprehension”).

Merritt counters that United States v. Gregory, 315 F.3d 637 (6th Cir. 2003), cuts in his favor. But his reliance on Gregory is unavailing. There, the defendant’s “obstructive conduct predated the indictment and guilty plea,” and he had “fully cooperated subsequently.” Id. at 640. So by the time he was charged, “Gregory cooperated with prison officials,” and then pleaded guilty, and was “entitled to a downward adjustment.” Id. at 641.5

Contrast that with Merritt’s case. Merritt was indicted on March 10, 2021, and did not plead guilty until August 11, 2022, roughly seventeen months later. In that intervening time, Merritt failed to comply with his bond conditions, gave a false address once he moved, fled from the police once found in Texas, and was finally arrested in Ohio. “On any one” of the “days he spent at large,” Merritt “could have turned himself in and willingly accepted responsibility for his actions.” Anderson, 2024 WL 209225, at *2. But he did not, which shows he did not accept responsibility. See United States v. Harper, 246 F.3d 520, 528 (6th Cir. 2001) (explaining that “if a defendant awaiting trial escapes from custody but then immediately turns himself into authorities, this could constitute” accepting “responsibility subsequent to an attempt to obstruct

5 Interestingly, in Gregory, the “United States agreed to recommend that Gregory be granted a reduction in offense level for acceptance of responsibility as part of his plea agreement, provided that he continue to cooperate. He did so, and the United States made its recommendation.” 315 F.3d at 640. That did not happen here. In fact, the government objected to the initial inclusion of the acceptance-of-responsibility reduction. But the Gregory court did not rely on that fact in its analysis, so we hesitate to rely on it too much here.

No. 23-1216 United States v. Merritt Page 10

justice”), overruled on other grounds by United States v. Leachman, 309 F.3d 377 (6th Cir. 2002).

One, perhaps the key, feature that distinguishes this case from Gregory is that Merritt made all his evasive and violative moves after he had been indicted. See Anderson, 2024 WL 209225, at *3 (holding that defendant was “unlike Gregory” because “most of [the] obstructive conduct postdated his indictment”); see also United States v. Zerpa-Ruiz, 784 F. App’x 353, 357 (6th Cir. 2019) (Defendant “engaged in obstructive conduct after his indictment, thereby failing to fully cooperate and genuinely accept responsibility.”). Indeed, Merritt gives up the game when he says that once he “was returned to custody, he took steps to accept responsibility.” Appellant Br. at 12 (emphasis added). It was too late by then. See Harper, 246 F.3d at 528 (“[I]t is the ordinary case in which a defendant pleads guilty after his attempts to obstruct justice are thwarted.”).6

The district court therefore did not err in denying the two-point reduction for acceptance of responsibility.7

6 Merritt cites Gregory for the following standard: “Appropriate considerations for determining whether a reduction is warranted include the defendant’s truthful admission of the offense conduct, the defendant’s voluntary assistance to authorities in resolving the offense, and the timeliness of the defendant’s conduct in affirmatively accepting responsibility for his actions.” 315 F.3d at 640. But Gregory cites no authority for these three considerations. And later cases split on following this standard. Compare United States v. Angel, 355 F.3d 462, 477–78 (6th Cir. 2004) (quoting Gregory for these three factors), with Histed, 93 F.4th at 961 (quoting Gregory only for the rule that a “defendant found to have obstructed justice can still receive an acceptance-of-responsibility reduction if his ‘obstructive conduct predated the . . . guilty plea’”). Rather than rely on an infrequently invoked list of factors of unknown origin, we elect to stick to the text of § 3E1.1(a). 7 At sentencing and in his appellate brief, Merritt focused much of his argument on the incentives it would create to deny him the acceptance-of-responsibility reduction. “If the reduction for acceptance of responsibility is regularly denied to defendants who have engaged in pre-plea obstructive conduct it is a disincentive to plead guilty. A defendant who gets no benefit for pleading guilty might think it is worth ‘rolling the dice’ at trial since there is little additional risk of doing so.” Appellant Br. at 13–14. The government responds that awarding Merritt the two- point reduction here “would set a perverse incentive encouraging defendants to flee justice without the risk of losing two levels for acceptance of responsibility upon their eventual arrest.” Appellee Br. at 19. At best, the incentive arguments are a wash. At worst, they cut against Merritt because the benefit of pleading guilty is not undercut when the acceptance-of-responsibility reduction is denied. Criminal defendants still benefit from pleading guilty rather than going to trial by receiving the certainty of a plea agreement and assurances that the government will drop or reduce certain charges. In any event, we do not interpret laws based on the incentives our rulings would or would not create. Judges are not economists or sociologists. That is why we should stick to the job we are good at—interpreting the

No. 23-1216 United States v. Merritt Page 11

III.

Merritt also argues that the district court miscalculated his criminal history category, constituting reversible procedural error. We review sentencing decisions “under a deferential abuse-of-discretion standard” when the defendant properly objects below, as Merritt did when challenging his criminal-history calculation. United States v. Bolds, 511 F.3d 568, 578 (6th Cir. 2007) (quoting Gall v. United States, 552 U.S. 38, 41 (2007)). And we look for both procedural and substantive reasonableness. United States v. Wagner, No. 22-1880, 2024 WL 328841, at *3 (6th Cir. Jan. 29, 2024). Merritt challenges only the procedural reasonableness of his sentence, and a sentencing court commits procedural error by, for example, improperly calculating the Guidelines range. Gall, 552 U.S. at 51; United States v. Johnson, 79 F.4th 684, 705 (6th Cir. 2023). “In reviewing the district court’s calculation of the Guidelines, we still review the district court’s factual findings for clear error and its legal conclusions de novo.” Bolds, 511 F.3d at 579.

Before we consider this claim, the complexities in the record surrounding this issue warrant some explanation.

A.

In calculating Merritt’s criminal history points, the PSR added three points for a previous state crime—carrying a concealed weapon and police officer assault, resistance, and obstruction—that Merritt committed when he was 17. The probation officer will “[a]dd 3 points for each prior sentence of imprisonment exceeding one year and one month” in the PSR. U.S.S.G. § 4A1.1(a). “The term ‘sentence of imprisonment’ means a sentence of incarceration and refers to the maximum sentence imposed.” U.S.S.G. § 4A1.2(b)(1). The Guidelines require the addition of three points even though a defendant committed this crime before he turned 18 because a 15-year lookback period applies if the sentence “exceed[ed] one year and one month.” U.S.S.G. § 4A1.2(e)(1). But if the sentence ranged from 60 days to 13 months, only a five-year

text of statutes and guidelines and reviewing determinations by the district courts. And here, the plain text decides this question in the government’s favor.

No. 23-1216 United States v. Merritt Page 12

lookback period applies, so Merritt’s crime would not score in this case. See U.S.S.G. § 4A1.2(d)(2)(A).

Merritt was never sentenced to more than one year and one month in a single sentence for his state offense. But repeated probation violations led Merritt to be sentenced to a 270-day sentence with 8 days credited for time served, another 270-day sentence with 140 days credited, then a 330-day sentence with 330 days credited, totaling 870 days—all related to the same original offense.8

To “compute the criminal history points for § 4A1.1(a),” the probation officer must, “[i]n the case of a prior revocation of probation, parole, supervised release, special parole, or mandatory release, add the original term of imprisonment to any term of imprisonment imposed upon revocation.” U.S.S.G. § 4A1.2(k)(1). Because this provision incorporates § 4A1.1(a), the phrases “term of imprisonment” and “sentence of imprisonment” are interchangeable. See United States v. Jones, 107 F.3d 1147, 1163 (6th Cir. 1997) (discussing the two interchangeably in the context of § 4A1.2(k), which talks of “term of imprisonment,” and how it relates to state definitions of “sentence of imprisonment”).

The district court found that the three-point scoring was correct based on the plain text of the Guidelines, because the sum of the probation violation-sentence and the original sentence far exceeded 13 months. So this crime scored three points against Merritt in calculating his criminal history score, which significantly contributed to Merritt’s overall score of seven and placed him in criminal history category IV. Without these three points, Merritt would have been in criminal history category III and his Guidelines range would have been reduced from 110–137 months to 97–121 months.

Merritt argues the “330 days were the cumulative sentence for the conviction and both [probation] violations” and therefore did not exceed one year and one month. Appellant Br. at

8 The PSR explains that Merritt initially “violated bond on January 16, 2014,” because he was “out after curfew.” R.49, p.11 ¶47, PageID 199. He again “violated his probation in March 2015,” and though the PSR reported that the “nature of that violation was not specified in available records,” id., the parties provided the district court with information that Merritt violated probation by “affiliating with a gang member,” R.65, pp.38–39, PageID 355–56. Merritt then violated probation again “in July 2015 by engaging in new criminal activity.” R.49, p.11 ¶47, PageID 199; see also R.65, pp.39–40, PageID 356–57.

No. 23-1216 United States v. Merritt Page 13

19. “If these sentences reflected additional time instead of cumulative time,” Merritt attests that he “would have been incarcerated for 862 days on the same case between March 11, 2014, and October 6, 2015,” but “there are only 574 days between those dates.” Id. at 20–21. So the “only explanation is that each judgment was intended to replace the prior judgments,” and the “330- day sentence had to have replaced the prior sentences.” Id. at 21.

B.

The court below did not err in calculating Merritt’s Guidelines range. The court was correct to include the three-point past state offense within Merritt’s criminal history score because, under the plain text of the Guidelines, each new “sentence” the state court imposed on Merritt for his probation violations adds up, no matter how much of that sentence Merritt actually served because of time-served credit. See United States v. Ramirez-Perez, 643 F.3d 173, 176 (6th Cir. 2011); United States v. Galvan, 453 F.3d 738, 741 (6th Cir. 2006).

To explain why, we must recognize the distinction between the sentence imposed and the sentence actually served. The Guidelines tell us that a “‘sentence of imprisonment’ means a sentence of incarceration and refers to the maximum sentence imposed.” U.S.S.G. § 4A1.2(b)(1).9 This is distinct from the sentence actually served, which may be as much as the maximum sentence imposed or as little as no time at all. Indeed, sentences could end early for various reasons—parole, compassionate release, good behavior credits, etc.—that do not necessarily have any bearing on the original sentence. See United States v. Katsma, 147 F. App’x 561, 564 (6th Cir. 2005) (explaining that the “fortuitous happenstance that allowed the defendant to begin service of his sentence at [a halfway house] almost immediately did not change the fact that the actual punishment imposed upon [defendant] was a jail sentence of nine months’ duration”). So an early release from prison does not change the sentence that was formally “pronounced” by the court (unless, of course, the sentence is vacated, in which case it is

9 This definition is consistent with the common understanding of “sentence.” See Black’s Law Dictionary 1636 (11th ed. 2019) (defining “sentence” as the “judgment that a court formally pronounces after finding a criminal defendant guilty; the punishment imposed on a criminal wrongdoer,” for example, “a sentence of 20 years in prison”). And this closely tracks the definition of “sentence” in place in 1987 when the pertinent Guidelines provisions were first adopted. See Black’s Law Dictionary 1222 (5th ed. 1979) (“The judgment formally pronounced by the court or judge upon the defendant after his conviction in a criminal prosecution, imposing the punishment to be inflicted.”).

No. 23-1216 United States v. Merritt Page 14

treated as if it never existed). See Ramirez-Perez, 643 F.3d at 176 (“When a court gives credit for time served, it affects the length of time the defendant will be required to serve. It does not, however, affect the sentence that is pronounced.”).

So in this context, we do not look to the sentence served but to the “maximum sentence imposed.” U.S.S.G. § 4A1.2(b)(1). And “to compute the criminal history points” when dealing with “a prior revocation of probation,” we “add” the sentence of imprisonment to any sentence of imprisonment imposed on revocation. U.S.S.G. §§ 4A1.2(k)(1) & 4A1.1(a). So each time Merritt violated probation, the state court imposed a “sentence,” which we “add” to the previous sentence.

This matches how the state court described what it was doing each time Merritt violated probation. The first “sentence” was for 270 days. R.65, p.38, PageID 355. Then when Merritt violated parole, the court explained it would be “the sentence of this court” that Merritt “do 270 days in the county jail,” and this was “a final judgment and sentence of the Court.” R.52-1, 3/20/15 Prob. Viol. Tr., pp.10–11, PageID 255–56. Of course, Merritt did not actually serve 270 days in jail for this violation because he was credited for time served (140 days of that time). And when Merritt violated probation again in July, the judge revoked his probation and ordered him “incarcerated in the Kent County Correctional Facility for a period of 330 days with credit for 330 days[’] time served.” R.52-2, 10/6/15 Sent. Tr., p.6, PageID 265. This was the “final judgment[] and sentence[] of the Court.” Id. at p.8, PageID 267. So adding up the “maximum sentence imposed” for each, Merritt was sentenced to 270 days plus 270 days plus another 330 days, equaling 870 days, which exceeds the one-year-and-one-month threshold. The state offense was thus properly scored in the PSR.

Though Merritt doesn’t direct us to any cases that hold otherwise, he does raise a state constitutional wrinkle. He argues that, under the double jeopardy clause of the Michigan constitution, “when a defendant in Michigan is sentenced for a parole violation, he must receive credit for the time he previously served on the conviction or on prior violations; ‘[t]o hold otherwise could lead to the anomalous result of a defendant suffering longer incarceration as a result of having been placed on probation than if initially sentenced to the maximum prison term possible for the offense.’” Appellant Br. at 19–20 (quoting Michigan v. Sturdivant, 312 N.W.2d

No. 23-1216 United States v. Merritt Page 15

622, 625 (Mich. 1981)); see also Michigan v. Wagner, 485 N.W.2d 133, 135 (Mich. App. 1992) (“[D]ouble jeopardy principles require an award of credit for time spent in jail as a condition of probation when the defendant is later sentenced to prison upon revocation of probation.”).

But, as Michigan law makes clear, Michigan courts comply with this double-jeopardy limitation by crediting the defendant for time served, not refraining from imposing a new sentence. That’s why the Michigan state court explained it was imposing a “sentence” but credited Merritt for time already served—in the last instance, crediting Merritt the full amount of the sentence to comply with constitutional restraints. And we have held that even under Michigan law, “for the purposes of Guidelines criminal history calculation, it matters not whether a defendant’s sentence included credit for time served presentence.” Galvan, 453 F.3d at 741; see also Ramirez-Perez, 643 F.3d at 176.

And nothing in the Guidelines or our caselaw directs us to decrease the sentence based on credit for time served. When the Guidelines wish to exclude parts of a sentence, they do so explicitly. Right below the definition for “sentence of imprisonment,” U.S.S.G. § 4A1.2(b)(1), the following provision explains that “[i]f part of a sentence of imprisonment was suspended, ‘sentence of imprisonment’ refers only to the portion that was not suspended,” U.S.S.G. § 4A1.2(b)(2); see also Ramirez-Perez, 643 F.3d at 176 (noting that the “Guidelines expressly exclude parts of a sentence in other circumstances”).10 Merritt has never argued that the grant of credit for time served can amount to the “suspension” of a sentence under this provision. We thus need not consider that issue. So the PSR was correct to note that the “[S]entencing [G]uidelines do not direct that credit be subtracted when determining the cumulative sentence

10 Ramirez-Perez also held that the “sentence of imprisonment for Guidelines purposes is ‘the sentence pronounced, not the length of time actually served.’” 643 F.3d at 176 (quoting U.S.S.G. § 4A1.2, cmt. n.2). And two other circuits have relied on the commentary to hold that a “sentence” is the formal “sentence pronounced, not the length of time actually served.” United States v. Gonzalez-Arias, 946 F.3d 17, 40 (1st Cir. 2019); see also United States v. Jesse, 950 F.3d 552, 555 (8th Cir. 2020). Of course since we decided Ramirez-Perez, we decided Riccardi, which held that we don’t look at the commentary unless the text of the Guidelines is “genuinely ambiguous.” 989 F.3d at 485–86 (quoting Kisor v. Wilkie, 139 S. Ct. 2400, 2415 (2019)). Here, we don’t find the Guidelines ambiguous, because the text and structure tell us that we look at the maximum sentence formally imposed and do not subtract for time served. So we don’t need to defer to the commentary.

No. 23-1216 United States v. Merritt Page 16

under USSG § 4A1.2(k)(1).” R.49, PageID 223. And the district court was correct to conclude that Merritt’s offense “has to be scored.” R.65, pp.38–41, PageID 355–58.11

We therefore affirm the district court on the question of whether Merritt’s criminal history score was properly calculated.

IV.

For the reasons set forth above, we AFFIRM.

11 Merritt concedes that the text of the Guidelines dictates this result but argues that “nothing prevents the court from examining the entire record when determining the cumulative sentence for this conviction and subsequent violations.” Appellant Br. at 21–22. This is not an argument about the proper interpretation of § 4A1.2; this is an argument for a variance, as the district court recognized. See R.65, pp.38–41, PageID 355–58 (“I know you’ve really argued for me to consider that in the variance, but I saw no law in support of any other form of interpretation of (k)(1).”). But the district court rejected Merritt’s request for a downward variance, and Merritt did not appeal that issue, so it is not before us now. We therefore affirm the district court’s holding on the criminal history calculation question. Because we see no error here, we do not reach the government’s argument that any error was harmless.