UNITED STATES OF AMERICA, Appellee, v. DAVID A. FRATES, Defendant, Appellant.
No. 16-1933
United States Court of Appeals For the First Circuit
July 18, 2018
APPEAL FROM THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF MASSACHUSETTS [Hon. Richard G. Stearns, U.S. District Judge]
Before Torruella, Lipez, and Kayatta, Circuit Judges.
Ian Gold, on brief for appellant.
Mark T. Quinlivan, Assistant United States Attorney, and William D. Weinreb, Acting United States Attorney, on brief for appellee.
LIPEZ,
Frates appeals this sentence, challenging his classification as a career offender, and alternatively asking us to vacate his sentence in light of a recently enacted amendment to the Guidelines. We find no error with the district court‘s application of the Guidelines. Nonetheless, we exercise our discretion under United States v. Godin (Godin II), 522 F.3d 133 (1st Cir. 2008), and United States v. Ahrendt, 560 F.3d 69 (1st Cir. 2009), to vacate Frates‘s sentence and remand to allow the district court to consider the United States Sentencing Commission‘s current policy рosition on who qualifies as a career offender.
I.
This case arises at a peculiar moment in the history of the Sentencing Guidelines’ career offender enhancement. That enhancement increases the sentencing ranges of certain defendants whose offense of conviction was “either a crime of violence or a controlled substance offense,” and who have at least two such prior convictions.
The term “crime of violence” means any offense under federal or state law, punishable by imprisonment for a term exceeding one year, that—
(1) has as an element the use, attempted use, or threatened use of physical force against the person of another, or
(2) is burglary of a dwelling, arson, or extortion, involves use of explosives, or otherwise involves conduct that presents a serious potential risk of physical injury to another.
The Guidelines’ “crime of violence” definition mirrored the Armed Career Criminal Act‘s (“ACCA“) definition of “violent felony.”
In June 2015, the Supreme Court held that the residual clause of the ACCA‘s violent felony definition was unconstitutional. Johnson v. United States, 135 S. Ct. 2551, 2563 (2015). It reasoned that “the indetеrminacy of the wide-ranging inquiry required by the residual clause both denies fair notice to defendants and invites arbitrary enforcement by judges.” Id. at 2557. Sentencing judges interpreting the residual clause faced “grave uncertainty” about how to estimate the risk of injury involved in a crime, and also what level of risk sufficed to qualify a crime as a violent felony. Id. at 2257-58. These vagaries were more than the strictures of due process could tolerate: “Invoking so shapeless a provision to condemn someone to prison for 15 years to life does not comport with the Constitution‘s guarantee of due process.” Id. at 2560.
Not surprisingly, in the wake of Johnson, there were challenges to the constitutionality of the Guidelines’ crime of violence definition. Most of the circuit courts to address the issue held that section 4B1.2(a)‘s identically-worded residual clause was unconstitutionally vague. See United States v. Hurlburt, 835 F.3d 715 (7th Cir. 2016); United States v. Calabretta, 831 F.3d 128 (3d Cir. 2016); United States v. Pawlak, 822 F.3d 902 (6th Cir. 2016); United States v. Madrid, 805 F.3d 1204 (10th Cir. 2015). But see United States v. Matchett, 802 F.3d 1185 (11th Cir. 2015). In the First Circuit, the government routinely took the position that Johnson‘s reasoning extended to the crime of violence definition, and conceded that section 4B1.2(a)‘s residual clause was void. See, e.g., Ball, 870 F.3d at 3; United States v. Thоmpson, 851 F.3d 129, 131 (1st Cir. 2017).
This “ongoing litigation and uncertainty resulting from the Johnson decision” prompted the United States Sentencing Commission to adopt an amendment eliminating the residual clause from the crime of violence definition.
Four months after Amendment 798 took effect, the Supreme Court rejected a void-for-vagueness challenge to the crime of violence definition‘s residual clause. Beckles v. United States, 137 S. Ct. 886, 890 (2017). Distinguishing Johnson, the Court explained that the ACCA “fix[ed] the permissible sentences for criminal offenses,” while the Guidelines “merely guide the exercise of a court‘s discretion in choosing an appropriate sentence within the statutory range.” Id. at 892. Since the Guidelines are discretionary, they are “not amenable to a vagueness challenge,” and thus “§ 4B1.2(a)‘s residual clause is not void for vagueness.” Id. at 894-95.
The result in Beckles creates a quirk for defendants (1) sentenced pursuant to section 4B1.2(a)‘s residual clause prior to Amendment 798, and (2) whose appeals were pending when the amendment became effective. Although stricken by the Sentencing Commission, the residual clause remains valid as applied to them. Hence, they will be the last group subjected to the disfavored -- yet constitutional -- residual clause. This is the context in which Frates appeals his sentence.
II.
Frates asserts that neither his offense of conviction nor his prior convictions qualify as crimes of violence. As to his present conviction for federal armed bank robbery, he suggests that the crime does not fit within the force clause, and that we should remand to give the district court the opportunity to determine in the first instance whether the residual clause covers the crime. Regarding his prior convictions, he argues that his four Massachusetts unarmed robbery convictions do not count as crimes of violencе under any of section 4B1.2(a)‘s clauses, leaving him without the two requisite crimes of violence necessary to trigger the career offender enhancement. We address these contentions in turn.
A. Frates‘s Offense of Conviction
It is axiomatic that in determining whether a crime fits within the force clause, we look to the elements that comprise the offense, rather than the defendant‘s conduct in committing the crime. See, e.g., United States v. Ramos-González, 775 F.3d 483, 504 (1st Cir. 2015). This analysis involves taking a “categorical approach” and determining whether the elements оf the defendant‘s crime of conviction necessarily require the use, attempted use, or threatened use of physical force against another person. See, e.g., United States v. Martinez, 762 F.3d 127, 133 (1st Cir. 2014).1
An offense qualifies as a crime
The federal armed bank robbery statute, in relevant part, penalizes “[w]hoever, by force and violence, or by intimidation, takes, or attempts to take, . . . any property or money . . . belonging to . . . any bank . . . ,” and who, in committing such an offense, “assaults any person, or puts in jeopardy the life of any person by the use of a dangerous weapon or device.”
Shortly after Frates filed his opening brief staking out these positions, we rebuffed identical arguments in Ellison, 866 F.3d at 36-38. The defendant in Ellison argued that his conviction for federal unarmed robbery did not qualify as a crime of violence under the force clause. Id. at 34. We squarely rejected his argument, concluding that
Ellison thus undermines Frates‘s assertion that federal armed bank robbery does not require the use, attempted use, or threatened use of force. Indeed, a conviction for federal unarmed bank robbery -- at issue in Ellison -- is a lesser included offense to federal armed bank robbery. See United States v. Spinney, 65 F.3d 231, 235 n.3 (1st Cir. 1995). As we are bound by this prior panel decision, e.g., Wurie, 867 F.3d at 34, we need not probe the matter further: federal armеd bank robbery is a crime of violence under section 4B1.2(a)‘s force clause.
B. Frates‘s Prior Convictions
The Massachusetts statute criminalizing unarmed robbery provides:
Whoever, not being armed with a dangerous weapon, by force and violence, or by assault and putting in fear, robs, steals or takes from the person of another, or from his immediate control, money or other property which may be the subject of larceny, shall be punished by imprisonment in the state prison for life or for any term of years.
Our precedent compels us to conclude that Massachusetts unarmed robbery is a crime of violence under the residual clause. In United States v. De Jesus, we held that the Massachusetts offense of larceny from the person qualified as a crime of violence under the residual clause. 984 F.2d 21, 22 (1st Cir. 1993).
Massachusetts classifies larceny from the person as a lesser included offense of unarmed robbery. See Commonwealth v. Glowacki, 499 N.E.2d 290, 294 (Mass. 1986); Commonwealth v. Sheppard, 537 N.E.2d 583, 585 (Mass. 1989). Hence, Massachusetts unarmed robbery is necessarily a crime of violence under the residual clause.
Frates urges us to abandon this otherwise straightforward analysis and overrule De Jesus. He argues that De Jesus was based on the “ordinary case” method for determining whether an offense fits within the residual clause, and that Johnson rendered that methodology invalid. We disagree.
Under the ordinary case method, we assess whether the elements of the crime, in the ordinary case, “(1) present a risk of physical injury similar to the risk presented by the clause‘s enumerated offenses and (2) [are] similar ‘in kind’ to those offenses.” United States v. Holloway, 630 F.3d 252, 260 (1st Cir. 2011) (quoting United States v. Giggey, 551 F.3d 27, 41-42 (1st Cir. 2008) (en banc)); see also Ramírez, 708 F.3d at 305 (applying the ordinary case method). Although Johnson was critical of this approach, see 135 S. Ct. at 2557-58, we recently rejected the contention that the Court‘s criticism in the ACCA context allows us to overrule prior decisions applying the ordinary case method to section 4B1.2(a)‘s residual clause. In Wurie, the defendant asked us to reconsider our holding in United States v. Glover, 558 F.3d 71, 80 (1st Cir. 2009), that Massachusetts assault and battery with a dangerous weapon was a crime of violence under the residual clause. 867 F.3d at 32. We explained that Johnson did not “necessarily reject[] the ‘ordinary case’ analysis in all of its applications.” Id. at 35. Rather, the Court‘s criticism of that methodology was “only one part of its conclusion as to why the residual clause of the ACCA was unconstitutionally vague.” Id. at 34. Since the Court later declined to extend Johnson‘s holding to the crime of violence residual clause, Beckles, 137 S. Ct. at 890, we were “not persuaded that Johnson ‘offers a sound reason for believing’ that the panel in Glover ‘would change its collective mind’ in light of Johnson.” Wurie, 867 F.3d at 35 (quoting United States v. Rodríguez, 527 F.3d 221, 225 (1st Cir. 2008)).
Wurie thus undermines Frates‘s attempt to rely on Johnson to circumvent De Jesus. Instead, De Jesus remains controlling, and requires us to conclude that Massachusetts unarmed robbery is a crime оf violence under the residual clause. As both Frates‘s offense of conviction and his prior unarmed robbery convictions were crimes of violence under the version of the Guidelines applicable at the time of his sentencing, the district court did not err in applying the career offender enhancement.2
III.
Once we have concluded that a district court did not err in sentencing a defendant, it is ordinarily the end of the matter. In a narrow category of cases, however, we have discretion to vacatе a correctly imposed
A. The Godin/Ahrendt Doctrine
Our discretion to remand in such situations derives from a pair of cases involving Amendment 709 to the Guidelines. Godin II, 522 F.3d at 133; Ahrendt, 560 F.3d at 69. Amendment 709 “restat[ed] the rules for determining when multiple crimes are counted as one for criminal history purposes.” Godin II, 522 F.3d at 135; see also
Nonetheless, we vacated and remanded both of their sentences to give the district courts the opportunity to weigh the Sentencing Commission‘s revised policy as a discretionary factor in imposing sentence. Under the Cоmmission‘s changed thinking, Godin and Ahrendt would have been subject to significantly lower guideline ranges. Godin had two prior burglary convictions for which she was sentenced on the same date. Id. at 134. Counting these offenses as a single sentence would have removed her from the career offender category and decreased her guideline sentencing range from 262-327 months’ imprisonment to 121-130 months. Id. Ahrendt had committed three offenses in the same week, and he was sentenced for those crimes on the same date. Counting his offenses as a single sentence wоuld have decreased his guideline range from 210-262 months’ imprisonment to 168-210 months. See Ahrendt, 560 F.3d at 73;
Our decisions to vacate and remand in those cases were animated by two principal factors. The first involved the posture of the cases and the manner in which the Sentencing Commission chose to amend the Guidelines. In Godin II, we explained that “the posture of this case is peculiar: the amendment is not applicable retroactively, but neither has the pending appeal yet resulted in a final disposition.” 522 F.3d at 135. The non-finality of Godin‘s
The second factor that led us to vacate and remand in Godin and Ahrendt was the discretionary nature of the federal sentencing regime after United States v. Booker, 543 U.S. 220 (2005). Under federal sentencing procedures post-Booker, district courts begin by calculating a defendant‘s guideline sentencing range. See, e.g., Molina-Martinez v. United States, 136 S. Ct. 1338, 1345 (2016). This range is merely advisory. See Booker, 543 U.S. at 245. Courts then exercise their discretion to select a sentence -- either inside or outside of the advisory guideline range -- that is “sufficient, but not greаter than necessary” to fulfill certain sentencing objectives.
Three more recent cases provide guidance on when we will choose to exercise our discretion under the Godin/Ahrendt doctrine. The defendant in United States v. Matos, 611 F.3d 31 (1st Cir. 2010), asked us to vacate his sentence and remand to allow the district court to consider Amendment 709. Distinguishing Godin II and Ahrendt, we explained that “it was evident” in those cases “that if the Guidelines as amended . . . had been in effect at the time of the defendant‘s sentencing,” the defendant would have been subjected to a lower sentencing range. Id. at 39. However, it was “far from clear” that Amendment 709 would have had any effect on Matos‘s sentence. Id. The district court would have had to “engage in fact-finding to determine whether the Amendmеnt applie[d].” Id. Given the complexity of the district court‘s task on remand, we “conclude[d] that Godin and Ahrendt [did] not advocate in favor of remanding for resentencing.” Id. at 39-40.
We recently applied this reasoning from Matos in a case involving Amendment 798. In Wurie, we explained that it was unclear whether the defendant would have benefitted from the intervening amendment. 867 F.3d at 36. On remand, the district court would have had to consider whether “at least two of Wurie‘s prior offenses . . . qualify as crimes of violence under the force clause.” Id. This analysis would have been “much more complicated” than the “simple mechanistic change” involved in Godin II and Ahrendt. Id. at 36-37. We accordingly declined to vacate thе defendant‘s sentence and remand for resentencing.
Lastly, we ordinarily will not use our discretion under the Godin/Ahrendt doctrine
Godin II and Ahrendt thus establish a narrow doctrine that gives us discretion to vacate a defendant‘s sentence and remand when: (i) the Sentеncing Commission adopts a substantive, non-retroactive amendment to the Guidelines; (ii) the amendment is adopted before the defendant‘s sentence becomes final on appeal; and (iii) the amendment would have lowered the defendant‘s guideline range if it had been in effect at the initial sentencing. However, we will ordinarily not exercise our discretion under Godin/Ahrendt when the district court‘s analysis on remand would be complex, or when the district court was aware of the amendment during the initial sentencing.
If we do remаnd a case pursuant to the Godin/Ahrendt doctrine, the district court is prohibited from recalculating the defendant‘s guideline range in light of the intervening amendment, lest it circumvent the Sentencing Commission‘s non-retroactivity determination. It may, however, consider the Commission‘s revised policy position in exercising its discretion to select an appropriate sentence for the defendant.
B. Vacating Frates‘s Sentence and Remanding for Resentencing
The facts of this case squarely implicate our Godin/Ahrendt doctrine, and counsel in favor of exercising our discretion to vacate Frates‘s sentence and remand to the district court. Amendment 798 is a substantive, non-retroactive amendment that the Sentencing Commission enacted while Frates‘s appeal was pending. If the amendment had been in effect at the time of Frates‘s sentencing, his guideline range would have decreased from 188-235 months’ imprisonment to 92-115 months. Indeed, the government concedes that Frates‘s prior convictions for Massachusetts unarmed robbery would not qualify as сrimes of violence under Amendment 798.4
Given the government‘s concession, the sentencing process on remand will be “mechanistic,” not complex. Wurie, 867 F.3d at 37. The district court will need to consider only whether the Sentencing Commission‘s current policy about who qualifies as a career offender affects its discretionary choice of sentence. Finally, there is no indication that the district court was aware of Amendment 798 at the time of sentencing.
The government advances three primary reasons why we should nonethelеss decline to vacate Frates‘s sentence and remand for resentencing. All three are unpersuasive. First, the government argues that a line from the Supreme Court‘s decision in Dillon v. United States, 560 U.S. 817 (2010), undermines the Godin/Ahrendt doctrine. Dillon involved the question of whether a provision of the Guidelines that limited a court‘s discretion in sentence modification proceedings remained mandatory -- rather than advisory -- after Booker. 560 U.S. at 819-822. In holding that the provision was mandatory, the Court sought to establish that other provisions of the
The Godin/Ahrendt doctrine is entirely consistent with the Supreme Court‘s observation that the Sentencing Commission‘s retroactivity determinations are binding. Both Godin II and Ahrendt acknowledge that the Commission‘s pronouncement that Amendment 709 was non-retroactive bound the court. Godin II, 522 F.3d at 136 (“The original guideline range . . . remains applicable, because the amendment was substantive and non-retroactive.“); Ahrendt, 560 F.3d at 79 (“Because Amendment 709 is non-retroactive, . . . Ahrendt is not entitled to the benefit of [the] amendment . . . .“). Indeed, the Godin/Ahrendt doctrine necessarily presupposes that the Commission‘s retroactivity determinations are binding. It only allows district courts to consider the Sentencing Commission‘s revised policies as a discretionary factor in resentencing defendants.
Second, the government contends that vacating and remanding in this case will effectively open the floodgates for defendants challenging their sentences “in light of Johnson.” This position is vastly overstated. As we detailed above, the Godin/Ahrendt doctrine applies only to a narrow category of cases involving certain non-retroactive amendments that are adopted before a defendant‘s sentence becomes final on appeal. Our holding here is thus potentially relevant to defendants sentenced pursuant to section 4B1.2(a)‘s residual clause only if their sentences were not yet final when Amendment 798 went into effect in November 2017. We fail to see how -- as the government seems to suggest -- this case would apply more broadly to defendants challenging their sentences based on Johnson‘s invalidation of the ACCA‘s residual clause.
Third, the government contends that the district court‘s decision to vary downward from Frates‘s guideline sentencing range makes remanding his case for resentencing unnecessary. Though “perhaps not irrelevant,” a district court‘s decision to depart from a defendant‘s guideline range will not ordinarily be a significant factor in determining whether to remand under Godin/Ahrendt. Ahrendt, 560 F.3d at 80 (reasoning that the amount an amendment reduces a defendant‘s guideline range is not dispositive in deciding whether to remand). A downward variance does not necessarily alter our “judgment that a diffеrent result might well be reached on remand,” Adams, 640 F.3d at 43, as the Sentencing Commission‘s revised policy may lead a district court to vary further from a defendant‘s guideline range. Indeed, the guideline range “anchor[s] . . . the district court‘s discretion,” such that even when “the sentencing judge sees a reason to vary from the Guidelines . . . the Guidelines are in a real sense [still] the basis for the sentence.” Molina-Martinez, 136 S. Ct. at 1346 (emphasis omitted) (quoting Peugh v. United States, 569 U.S. 530, 542, 549 (2013)). The knowledge that the Sentencing Commission would choose a different “anchor” thus remains a relevant discretionary factor for district courts to consider even where they initially varied downward.
This case provides a prime example of this principle. The district court varied downward based on Frates‘s history of substance abuse and mental health issues, and imposed a sentence of 132 months’ imprisonment. Under the Sentencing Commission‘s current thinking, that sentence would constitute an upward variance of 17 months. The district court may well view
IV.
For these reasons, we “think it prudent to allow the [district] court thе opportunity to consider the Sentencing Commission‘s updated views.” Ahrendt, 560 F.3d at 80. We therefore vacate Frates‘s sentence and remand for resentencing consistent with this opinion. On remand, the court‘s initial calculation of Frates‘s guideline range remains in effect. The court is under no obligation to modify Frates‘s sentence if, in its discretion, it does not feel that modification is warranted.
So ordered.