UNITED STATES of America v. Corey GRANT, Appellant
Corey Grant was sixteen years old when he committed various crimes that led to his ultimate incarceration. He was convicted in 1992 of conspiracy and racketeering under the Racketeer Influenced and Corrupt Organizations Act (“RICO”), as well as of various drug trafficking charges and a gun charge. At sentencing, the District Court determined that Grant would never be fit to reenter society and sentenced him to life in prison without the possibility of parole (“LWOP”) for the RICO conspiracy and racketeering convictions. He received a concurrent forty-year term for the drug convictions and a mandatory consecutive five-year term for the gun conviction.
In 2012, the Supreme Court decided Miller v. Alabama, which held, inter alia, that only incorrigible juvenile homicide offenders who have no capacity to reform may be sentenced to LWOP. 567 U.S. 460, 479-80, 132 S.Ct. 2455, 183 L.Ed.2d 407 (2012). It also extended the Court's earlier holding in Graham v. Florida, 560 U.S. 48, 75, 130 S.Ct. 2011, 176 L.Ed.2d 825 (2010)—that juvenile non-homicide offenders are entitled to a “meaningful opportunity to obtain release based on demonstrated maturity and rehabilitation”—to all non-incorrigible juvenile homicide offenders. 567 U.S. at 479, 132 S.Ct. 2455 (quoting Graham, 560 U.S. at 75, 130 S.Ct. 2011); see also id. at 473, 132 S.Ct. 2455; Graham, 560 U.S. at 82, 130 S.Ct. 2011. In light of Miller, the District Court granted Grant's 28 U.S.C. § 2255 motion. At resentencing, the District Court determined that Grant's upbringing, debilitating characteristics of youth, and post-conviction record demonstrated that he had the capacity to reform and that a LWOP sentence was therefore inappropriate under Miller. Instead, the District Court sentenced Grant to a term of sixty-five years without parole.
On appeal, Grant challenges the constitutionality of his new sentence. He contends that he will be released at age seventy-two at the earliest, which he purports to be the same age as his life expectancy. In Grant's estimation, his sentence violates the Eighth Amendment to the Constitution of the United States because it constitutes de facto LWOP and therefore fails to account for his capacity for reform and to afford him a meaningful opportunity for release.
This case presents several difficult challenges for this Court. It calls upon us to decide a novel issue of constitutional law: whether the Eighth Amendment prohibits a term-of-years sentence for the duration of a juvenile homicide offender's life expectancy (i.e., “de facto LWOP”) when the defendant's “crimes reflect transient immaturity [and not] ․ irreparable corruption.” Montgomery v. Louisiana, ––– U.S. ––––, 136 S.Ct. 718, 734, 193 L.Ed.2d 599 (2016). Next, if we find that it does, then we must decide what framework will properly effectuate the Supreme Court's determination that the Eighth Amendment affords non-incorrigible juvenile offenders a right to a meaningful opportunity for release. Furthermore, we must take great pains throughout our discussion to account for the substantive distinction that the Supreme Court has made between incorrigible and non-incorrigible juvenile offenders in order to ensure that the latter is not subjected to “a punishment that the law cannot impose upon [them].” Id. (quoting Schriro v. Summerlin, 542 U.S. 348, 352, 124 S.Ct. 2519, 159 L.Ed.2d 442 (2004) ).
Our decision today therefore represents an incremental step in the constitutional discourse over the unique protections that the Eighth Amendment affords to juvenile homicide offenders.
I. FACTS AND PROCEDURAL HISTORY
In March 1987, local law enforcement authorities in Elizabeth, New Jersey became aware of an organized gang of teenagers called the E-Port Posse, led by Bilal Pretlow. The Posse operated a narcotics network that would regularly buy multi-kilogram amounts of cocaine in New York City, cut and package the cocaine in stash houses, and sell it on the streets of Elizabeth. Its members had access to firearms and they regularly used threats, physical violence and murder to carry out their objectives. Appellant Corey Grant—who was thirteen when he joined the Posse in 1986—was employed as one of the Posse's main enforcers.
On January 25, 1991, a superseding indictment charged Grant with RICO conspiracy (Count 1), in violation of 18 U.S.C. § 1962(d); racketeering (Count 2), in violation of 18 U.S.C. § 1962(c); conspiracy to possess with the intent to distribute cocaine (Count 4), in violation of 21 U.S.C. § 846; two counts of possession with the intent to distribute cocaine (Counts 5 and 6), in violation of 21 U.S.C. § 841(a)(1); and two counts of possession of a weapon in relation to a crime of violence or drug trafficking, in violation of 18 U.S.C. § 924(c) (Counts 10 and 11), one of which was dismissed prior to the return of a verdict.2
Grant, who was below the age of eighteen during his tenure with the Posse, proceeded to trial as an adult in February 1992. The jury returned a partial verdict finding him guilty of the RICO conspiracy, racketeering, and drug and gun possession counts (Counts 1, 2, 4, 5, 6, and 11), and—as predicates for the racketeering charge—found that he murdered Mario Lee and attempted to murder Dion Lee.
Dion Lee was a former member of the E-Port Posse who continued to individually sell drugs after leaving the gang. In August 1989, Grant, who was sixteen years old at the time, encountered a group of rival drug dealers while delivering drugs for Pretlow, including Lee. Grant warned Lee at gunpoint not to be in Pretlow's territory unless he was working for Pretlow. Lee refused, and Grant struck him in the head with a gun while another Posse member assaulted him. When Lee retreated, Grant and an associate shot him in the leg. Lee ultimately survived.
Later that month, Grant encountered Dion's brother, Mario Lee, another independent drug dealer who was warned by the Posse not to operate within its territory. Grant confronted Lee in an apartment courtyard where drugs were commonly sold and tried to force Lee into the building. Lee broke free and began to retreat, but Grant ordered his associate to shoot Lee to prevent any escape. The associate killed Lee.
At sentencing, the District Court denied Grant's departure motion and imposed a sentence within the then-mandatory Sentencing Guidelines of LWOP on the two RICO counts, a concurrent forty-year term of imprisonment on the drug-trafficking counts, and a five-year consecutive term of imprisonment on the gun possession count. The convictions and sentence were affirmed on direct appeal. United States v. Grant, 6 F.3d 780 (3d Cir. 1993) (unpublished table decision), cert. denied, 510 U.S. 1061, 114 S.Ct. 732, 126 L.Ed.2d 696 (1994).
Twelve years later, Grant sought a writ of habeas corpus pursuant to 28 U.S.C. § 2241, which was dismissed for lack of jurisdiction. We affirmed. Grant v. Williamson, 198 Fed.Appx. 263, 264 (3d Cir. 2006). Grant then filed a § 2255 motion, which was dismissed as untimely. Grant v. United States, No. CIV. A. 06-5952 HAA, 2008 WL 360982 (D.N.J. Feb. 8, 2008).
In 2012, the Supreme Court decided Miller, which held that mandatory LWOP sentences for juvenile homicide offenders violated the Eighth Amendment. 567 U.S. at 479, 132 S.Ct. 2455. Grant subsequently sought and received leave from this Court to file a second § 2255 motion. In re Pendleton, 732 F.3d 280, 282 (3d Cir. 2013) (per curiam). He argued that his LWOP sentence was imposed without consideration of mitigating circumstances related to his age at the time of his crimes. The District Court agreed and ordered that Grant be resentenced. Grant v. United States, No. CIV. A. 12-6844 JLL, slip. op. at 7, 2014 WL 5843847 (D.N.J. Nov. 12, 2014).
At resentencing, the District Court limited the scope of its review to the RICO conspiracy and racketeering counts, the charges for which Grant received a mandatory life sentence, thereby leaving in place the forty-year sentence for drug crimes and the mandatory consecutive five-year sentence for illegal gun possession. It determined that Grant's upbringing, debilitating characteristics of youth, and post-conviction record sufficiently evidenced that he was not incorrigible and that an LWOP sentence was therefore inappropriate under Miller. However, the District Court also emphasized that it would issue a sentence that “promote[s] respect for the law,” “provide[s] just punishment,” and “protect[s] the public.” App. 154-55. It then imposed a term of sixty-years' imprisonment on Counts 1 and 2 to run concurrently with the drug charges, resulting in a new effective sentence of sixty-five years without parole.3
Under this sentence, assuming good time credit, see 18 U.S.C. § 3624(b)(1), Grant will be eligible for release at age seventy-two, which he contends is the same age as his life expectancy. Grant now appeals his new sentence to this Court.
II. JURISDICTION AND STANDARD OF REVIEW
The District Court had jurisdiction under 18 U.S.C. § 3231, and we have jurisdiction under 28 U.S.C. § 1291 and 18 U.S.C. § 3742(a). “We employ a plenary standard of review to a defendant's Eighth Amendment challenge to his sentence.” United States v. Miknevich, 638 F.3d 178, 185 (3d Cir. 2011).
III. SUPREME COURT PRECEDENT
The Supreme Court has long grappled with the societal bounds of imposing the most severe punishments. It has maintained that the scope of what is considered cruel and unusual punishment under the Eighth Amendment is not fixed, but instead depends on “the evolving standards of decency that mark the progress of a maturing society.” Miller, 567 U.S. at 469, 132 S.Ct. 2455 (quoting Estelle v. Gamble, 429 U.S. 97, 102, 97 S.Ct. 285, 50 L.Ed.2d 251 (1976) ). Of equal importance, sentencing must also be individualized to account for the defendant's mitigating circumstances to ensure that the most serious punishments are “reserved only for the most culpable defendants committing the most serious offenses.” Id. at 476, 132 S.Ct. 2455. The Court therefore has categorically prohibited the imposition of the most severe punishments on classes of defendants that have diminished culpability due to immutable characteristics and where state practice and legislative enactments demonstrate a national consensus against imposing those punishments on members of that class. See, e.g., Atkins v. Virginia, 536 U.S. 304, 321, 122 S.Ct. 2242, 153 L.Ed.2d 335 (2002) (holding that Eighth Amendment proscribes death penalty for intellectually disabled offenders).
This case requires us to further consider the societal boundaries of punishing juvenile homicide offenders. We therefore feel it necessary to inform our forthcoming analysis by detailing the line of Supreme Court cases that, under the Eighth Amendment, has proscribed the most severe punishments from being imposed on juvenile offenders: (1) Roper v. Simmons, 543 U.S. 551, 568, 125 S.Ct. 1183, 161 L.Ed.2d 1 (2005) (prohibiting death penalty for juvenile offenders); (2) Graham, 560 U.S. at 82, 130 S.Ct. 2011 (prohibiting LWOP for juvenile non-homicide offenders); (3) Miller, 567 U.S. at 479, 132 S.Ct. 2455 (prohibiting mandatory LWOP for juvenile homicide offenders); and (4) Montgomery, 136 S.Ct. at 729, 736 (holding that Miller applies retroactively on collateral review). At bottom, we must consider whether the logic of these cases—that the debilitating characteristics of youth make juveniles less deserving of the most severe punishments—forecloses de facto LWOP for juvenile offenders whose crimes do not reflect “irreparable corruption.” Montgomery, 136 S.Ct. at 726 (quoting Miller, 567 U.S. at 480-81, 132 S.Ct. 2455). As such, this case requires us to consider the next incremental step in the constitutional dialogue over the contours of the Eighth Amendment's protections, as applied to juvenile homicide offenders.
A. Roper v. Simmons
In Roper, the Supreme Court held that the Eighth Amendment prohibits the death penalty for defendants who committed their crimes before the age of eighteen. 543 U.S. at 578, 125 S.Ct. 1183. After determining that there existed a national consensus against the death penalty for juvenile offenders, the Court, relying on science and social science, reasoned that, relative to adults, juveniles have a “lack of maturity and an underdeveloped sense of responsibility”; that they “are more vulnerable or susceptible to negative influences and outside pressures, including peer pressure”; and that their characters and personality traits are “more transitory, less fixed.” Id. at 569-70, 125 S.Ct. 1183 (quoting Johnson v. Texas, 589 U.S. 350, 367 (1993) ); accord Graham, 560 U.S. at 68, 130 S.Ct. 2011. The failings of a minor therefore are not the moral equivalent of those of an adult because there is a greater possibility that a minor's character deficiencies “will be reformed.” 543 U.S. at 570, 125 S.Ct. 1183. Accordingly, Roper established the principle that juvenile offenders are not deserving of the most severe punishments because they are innately less culpable than adults. Id. at 569-70, 125 S.Ct. 1183 (“[J]uvenile offenders cannot with reliability be classified among the worst offenders.”).
In light of these innate characteristics, the Court also determined that the penological justifications for the death penalty—i.e., retribution and deterrence—have diminished applicability to juvenile offenders. Retribution, the Court noted, “is not proportional if the law's most severe penalty is imposed on one whose culpability or blameworthiness is diminished, to a substantial degree, by reason of youth and immaturity.” Id. at 571, 125 S.Ct. 1183. As for deterrence, the Court determined that “the same characteristics that render juveniles less culpable than adults suggest as well that juveniles will be less susceptible to deterrence.” Id. Thus, because juveniles are less culpable than adults and there is diminished justification for imposing severe punishments on juvenile offenders, the Court concluded that the Eighth Amendment does not permit juvenile offenders to be sentenced to death. Id. at 578, 125 S.Ct. 1183.
B. Graham v. Florida
Building on the logic of Roper, Graham held that the Eighth Amendment bars juvenile offenders from being sentenced to LWOP for a non-homicide crime. 560 U.S. at 74-75, 82, 130 S.Ct. 2011. As with Roper, the Graham Court first determined that a national consensus had developed against LWOP for juvenile non-homicide offenders.4 Id. at 62-67, 130 S.Ct. 2011. It then reaffirmed Roper's fundamental principle that juveniles are less culpable than adults, id. at 68, 130 S.Ct. 2011, and concluded that “compared to an adult murderer, a juvenile offender who did not kill or intend to kill has a twice diminished moral culpability. The age of the offender and the nature of the crime each bear on the analysis,” id. at 69, 130 S.Ct. 2011.
The Court reasoned that LWOP was an overly severe punishment because it uniquely shares particular characteristics with capital punishment. Like the death penalty, LWOP “alters the offender's life by a forfeiture that is irrevocable” and “deprives the convict of the most basic liberties without giving hope of restoration.” Id. at 69-70, 130 S.Ct. 2011; see also id. at 69, 130 S.Ct. 2011 (“[LWOP] is the second most severe penalty permitted by law.” (quoting Harmelin v. Michigan, 501 U.S. 957, 1001, 111 S.Ct. 2680, 115 L.Ed.2d 836 (1991) ) ). Put differently:
Life in prison without the possibility of parole gives no chance for fulfillment outside prison walls, no chance for reconciliation with society, no hope. Maturity can lead to that considered reflection which is the foundation for remorse, renewal, and rehabilitation. A young person who knows that he or she has no chance to leave prison before life's end has little incentive to become a responsible individual.
Id. at 79, 130 S.Ct. 2011. Also, according to the Court, LWOP is an even harsher punishment for juveniles than it is for adults because “a juvenile offender will on average serve more years and a greater percentage of his life in prison than an adult offender.” Id. at 70, 130 S.Ct. 2011. Thus, “[a] 16-year-old and a 75-year-old each sentenced to life without parole receive the same punishment in name only.” Id. at 70, 130 S.Ct. 2011. “This reality,” the Court stated, “cannot be ignored.”
Next, the Court extended the penological reasoning of Roper to LWOP, noting that “none of the goals of penal sanctions that have been recognized as legitimate—retribution, deterrence, incapacitation, and rehabilitation—provides an adequate justification” and that “[a] sentence lacking any legitimate penological justification is by its nature disproportionate to the offense.” Id. at 71, 130 S.Ct. 2011. Regarding retribution and deterrence, the Court reiterated its reasoning from Roper. Id. at 71-72, 130 S.Ct. 2011. But unlike in Roper, the Graham Court also addressed incapacitation and rehabilitation, concluding that neither justified the imposition of LWOP on juvenile non-homicide offenders.
As to the former, the Court explained:
To justify life without parole on the assumption that the juvenile offender forever will be a danger to society requires the sentencer to make a judgment that the juvenile is incorrigible. The characteristics of juveniles make that judgment questionable. It is difficult even for expert psychologists to differentiate between the juvenile offender whose crime reflects unfortunate yet transient immaturity, and the rare juvenile offender whose crime reflects irreparable corruption․ [I]ncorrigibility is inconsistent with youth.
Id. at 72-73, 130 S.Ct. 2011 (citations and internal quotation marks omitted). Regarding rehabilitation, the Court reasoned that LWOP “forswears altogether the rehabilitative ideal” because “[b]y denying the defendant the right to reenter the community, the State [impermissibly] makes an irrevocable judgment about that person's value and place in society” that fails to account for his or her “capacity for change and limited moral culpability.” Id. at 74, 130 S.Ct. 2011.
Critical to this case, in order to effectuate its holding that the Eighth Amendment forbids LWOP sentences for juvenile non-homicide offenders, the Court mandated that such offenders be afforded a “meaningful opportunity to obtain release” during their lifetime:
A State is not required to guarantee eventual freedom to a juvenile offender convicted of a nonhomicide crime. What the State must do, however, is give defendants like Graham some meaningful opportunity to obtain release based on demonstrated maturity and rehabilitation․ The Eighth Amendment does not foreclose the possibility that persons convicted of nonhomicide crimes committed before adulthood will remain behind bars for life. It does prohibit States from making the judgment at the outset that those offenders never will be fit to reenter society.
Id. at 75, 130 S.Ct. 2011; see also id. at 82, 130 S.Ct. 2011 (“A State need not guarantee the offender eventual release, but if it imposes a sentence of life it must provide him or her with some realistic opportunity to obtain release before the end of that term.”). It is the scope of this mandate, which the Court reiterated in Miller, that we consider today. See Miller, 567 U.S. at 479, 132 S.Ct. 2455 (quoting Graham, 560 U.S. at 75, 130 S.Ct. 2011).
C. Miller v. Alabama
Relying on Graham as its “foundation stone,” Miller held that mandatory LWOP sentences for juvenile homicide offenders violate the Eighth Amendment. Id. at 470 n.4, 479, 132 S.Ct. 2455; see also id. at 473, 132 S.Ct. 2455 (“Graham's reasoning implicates any life-without-parole sentence imposed on a juvenile, even as its categorical bar relates only to nonhomicide offenses.”). Mandatory LWOP, the Court reasoned, contradicts Graham's and Roper's core principle—“that imposition of a State's most severe penalties on juvenile offenders cannot proceed as though they were not children”—because it prevents sentencing judges from taking into account the juvenile's youth and attendant circumstances. Id. at 473-74, 132 S.Ct. 2455.
Furthermore, having found in Graham that juvenile life sentences were analogous to capital punishment, the Court in Miller concluded that a line of cases that requires individualized sentencing when imposing the death penalty also applies to mandatory LWOP for juvenile offenders. Id. at 475-76, 132 S.Ct. 2455. Accordingly, a sentencing court must have the ability to consider the “mitigating qualities of youth” because “youth is more than a chronological fact.” Id. at 476, 132 S.Ct. 2455 (citations omitted).
Notably, however, the Court did not categorically ban LWOP for juvenile homicide offenders. Rather, it required courts to conduct individualized sentencing hearings that “take into account how children are different, and how these differences counsel against irrevocably sentencing them to a lifetime in prison” before imposing LWOP. Id. at 480, 132 S.Ct. 2455. In discussing the deficiencies of a scheme that treats every child as an adult, the Court enumerated various considerations that doubtlessly can be used to determine whether a juvenile offender is incorrigible:
• “[C]hronological age and its hallmark features—among them, immaturity, impetuosity, and failure to appreciate risks and consequences.” Id. at 477, 132 S.Ct. 2455.
• “[T]he family and home environment that surrounds [the juvenile offender]—and from which he cannot usually extricate himself—no matter how brutal or dysfunctional.” Id.
• “[T]he circumstances of the homicide offense, including the extent of his participation in the conduct and the way familial and peer pressures may have affected him.” Id.
• “[T]hat he might have been charged and convicted of a lesser offense if not for incompetencies associated with youth—for example, his inability to deal with police officers or prosecutors (including on a plea agreement) or his incapacity to assist his own attorneys.” Id. at 477-78, 132 S.Ct. 2455.
• “[T]he possibility of rehabilitation․” Id. at 478, 132 S.Ct. 2455.
The Court, however, cautioned that the bar for imposing LWOP is high. It predicted that LWOP would “be uncommon” and reserved only for “the rare juvenile offender whose crime reflects irreparable corruption.” Id. at 479-80, 132 S.Ct. 2455 (quoting Roper, 543 U.S. at 573, 125 S.Ct. 1183). Thus, while not a categorical bar, Miller effectively prohibits LWOP for nearly all juvenile offenders. Only those who are permanently incorrigible may receive such a sentence.
D. Montgomery v. Louisiana
Montgomery held that Miller applied retroactively on collateral review because it announced a new substantive rule of constitutional law:
Even if a court considers a child's age before sentencing him or her to a lifetime in prison, that sentence still violates the Eighth Amendment for a child whose crime reflects unfortunate yet transient immaturity. Because Miller determined that sentencing a child to life without parole is excessive for all but the rare juvenile offender whose crime reflects irreparable corruption, it rendered life without parole an unconstitutional penalty for a class of defendants because of their status—that is, juvenile offenders whose crimes reflect the transient immaturity of youth. As a result, Miller announced a substantive rule of constitutional law. Like other substantive rules, Miller is retroactive․
Montgomery, 136 S.Ct. at 734 (internal quotation marks and citations omitted).
Montgomery also reiterated that “Miller requires a sentencer to consider a juvenile offender's youth and attendant characteristics before determining that life without parole is a proportionate sentence” and that “[a] hearing where ‘youth and its attendant characteristics’ are considered as sentencing factors is necessary to separate those juveniles who may be sentenced to life without parole from those who may not.” Id. at 734-35. Thus, after echoing Miller's admonition that imposition of LWOP on a juvenile homicide offender will be “rare,” the Court made clear that “Miller drew a line between children whose crimes reflect transient immaturity and those rare children whose crimes reflect irreparable corruption.” Id. at 734.
Several state legislatures have reacted to Miller and Montgomery by either affording juvenile homicide offenders an early opportunity to seek parole, capping the length that a juvenile offender may be sentenced for homicide, or both. See, e.g., N.J. Stat. Ann. § 2C:11-3(b)(5) (requiring courts to either sentence juvenile homicide offenders to 30 years without parole or to make them eligible for parole after 30 years). Conversely, Congress—which abolished parole in the federal system 5 —has not enacted any legislation to date to effectuate the Supreme Court's holdings. Our task, therefore, is to determine what minimum protections the Supreme Court's Eighth Amendment jurisprudence requires in the absence of such congressional action.6
IV. DE-FACTO LWOP
Grant challenges the constitutionality of his sixty-five year sentence, arguing that it violates the Eighth Amendment under Miller because he will be released no earlier than at age seventy-two. Citing various life expectancy estimates, Grant argues that his life expectancy is also seventy-two, and that he is therefore likely to die in prison since decades of imprisonment diminish life expectancy. This case raises an issue of first impression for this Court: does the Eighth Amendment prohibit term-of-years sentences for the entire duration of a juvenile homicide offender's life expectancy when the defendant's “crimes reflect transient immaturity [and not] ․ irreparable corruption,” Montgomery, 136 S.Ct. at 734?
We hold that it does.7 A term-of-years sentence without parole that meets or exceeds the life expectancy of a juvenile offender who is still capable of reform is inherently disproportionate and therefore violates the Eighth Amendment under both Miller and Graham. We reach this conclusion for three reasons. First, Miller reserves the sentence of LWOP only for juvenile homicide offenders “whose crimes reflect permanent incorrigibility.” Id. Second, the Supreme Court's concerns about the diminished penological justification for LWOP sentences for juvenile offenders apply with equal strength to de facto LWOP sentences. Third, de facto LWOP is irreconcilable with Graham and Miller's mandate that sentencing judges must provide non-incorrigible juvenile offenders with a “meaningful opportunity to obtain release based on demonstrated maturity and rehabilitation.” Graham, 560 U.S. at 75, 130 S.Ct. 2011; see also Miller, 567 U.S. at 479, 132 S.Ct. 2455.8
A. Miller reserved LWOP only for “incorrigible” juvenile homicide offenders.
To fully appreciate why LWOP sentences for juvenile homicide offenders who are capable of reform violate the Eighth Amendment, we must first consider the genesis of the Supreme Court's distinction between incorrigible and non-incorrigible juveniles. Miller, like Graham and Roper, is based on the principle that the debilitating characteristics of youth—namely that children have heightened immaturity, increased vulnerability to peer pressure, and more transient identities—make “children ․ constitutionally different from adults for purposes of sentencing.” Miller, 567 U.S. at 471, 132 S.Ct. 2455. Miller therefore requires sentencing courts to “take into account how children are different, and how those differences counsel against irrevocably sentencing them to a lifetime in prison.” Id. at 480, 132 S.Ct. 2455.
To effectuate this constitutional principle that youth mitigates against the imposition of the most severe punishments on children, the Court “drew a line” between two classes of juvenile homicide offenders. Montgomery, 136 S.Ct. at 734. The first is non-incorrigible juvenile offenders who are capable of reform and “whose crimes reflect transient immaturity.” Id. (quoting Miller, 567 U.S. at 479, 132 S.Ct. 2455). The second is “rare” incorrigible juvenile offenders who have no capacity for change and “whose crimes reflect irreparable corruption.” Id. (quoting Miller, 567 U.S. at 479-80, 132 S.Ct. 2455). Only the second class of homicide offenders may be sentenced to LWOP. That this distinction is of constitutional magnitude was made incontrovertibly clear in Montgomery, where the Court held that Miller announced a substantive rule of constitutional law because “juvenile offenders whose crimes reflect the transient immaturity of youth” constitute a class of defendants upon which LWOP cannot be imposed. Id. (“Before Miller, every juvenile convicted of a homicide offense could be sentenced to life without parole. After Miller, it will be the rare juvenile offender who can receive that same sentence.”). We must therefore give effect to the Supreme Court's pronouncement that children who are found to have the capacity for change are to be treated differently than those who are not.
A sentence for a juvenile offender who is not incorrigible but that still results in him spending the rest of his life in prison does not appreciate the categorical differences between children and adults and between children who are incorrigible and those that have “diminished culpability and greater prospects for reform.” Miller, 567 U.S. at 471, 132 S.Ct. 2455. Moreover, there is no indication that Miller's holdings depended on a sentence formally being designated as LWOP. For example, the Court's categorization of LWOP as a particularly harsh sentence for juveniles, and even as one akin to the death penalty, see id. at 474, 132 S.Ct. 2455, did not turn on the sentence's formal designation. Both punishments “[i]mprison[ ] an offender until he dies” and “alters the remainder of his life by a forfeiture that is irrevocable.” Id. at 474-75, 132 S.Ct. 2455. Indeed, it would make little sense if sentencing courts could circumvent Miller and eradicate this constitutionally required distinction simply by imposing extraordinarily high term-of-years sentences. Thus, a sentence that treats a non-incorrigible juvenile offender as if he or she were an incorrigible one is irreconcilable with Miller.
B. The Court's penological concerns regarding juvenile LWOP sentences apply with equal strength to de facto LWOP sentences.
A de facto LWOP sentence for a non-incorrigible juvenile offender also violates the Eighth Amendment because it lacks an adequate constitutional justification to make it a proportionate sentence. See Graham, 560 U.S. at 71, 130 S.Ct. 2011 (“A sentence lacking any legitimate penological justification is by its nature disproportionate to the offense.”). “Miller ․ did more than require a sentencer to consider a juvenile offender's youth before imposing life without parole; it established that the penological justifications for life without parole collapse in light of ‘the distinctive attributes of youth.’ ” Montgomery, 136 S.Ct. at 734 (quoting Miller, 567 U.S. at 472, 132 S.Ct. 2455). Indeed, all of the reasons provided in Graham for why traditional penological justifications cannot validate LWOP against nonhomicide offenders, see 560 U.S. at 71-75, 130 S.Ct. 2011, apply with equal strength to non-incorrigible juvenile homicide offenders: the impotence of deterring juveniles, the shortcomings of retribution as a result of diminished culpability, the increased opportunity for reform that vitiates incapacitation, and the irreconcilable tension between LWOP sentences and rehabilitation. These distinctive attributes are equally relevant regardless of the crime or of the formal distinction between de facto and de jure LWOP sentences:
[N]one of what it said about children—about their distinctive (and transitory) mental traits and environmental vulnerabilities—is crime-specific. Those features are evident in the same way, and to the same degree, when ․ a botched robbery turns into a killing. So Graham's reasoning implicates any life-without-parole sentence imposed on a juvenile, even as its categorical bar relates only to nonhomicide offenses.
Miller, 567 U.S. at 473, 132 S.Ct. 2455; see also id. at 471, 132 S.Ct. 2455 (“Our decisions rest[ ] ․ on common sense—on what ‘any parent knows’․” (quoting Roper, 543 U.S. at 569, 125 S.Ct. 1183) ).9 Thus, without an adequate constitutional justification, de facto LWOP remains a disproportionate sentence for a non-incorrigible juvenile offender, rendering it unconstitutional under the Eighth Amendment. Id. at 473, 132 S.Ct. 2455 (“The characteristics of youth, and the way they weaken rationales for punishment, can render a life-without-parole sentence disproportionate.”); see also id. at 469, 132 S.Ct. 2455 (“[T]he concept of proportionality is central to the Eighth Amendment.” (quoting Graham, 560 U.S. at 59, 130 S.Ct. 2011) ).
C. De facto LWOP violates Graham's and Miller's “meaningful opportunity to obtain release” mandate.
For the purposes of considering whether a de facto LWOP sentence for a non-incorrigible juvenile offender affords him or her a “meaningful opportunity to obtain release based on demonstrated maturity and rehabilitation,” Graham, 560 U.S. at 75, 130 S.Ct. 2011, we feel it only necessary to state the obvious: a de facto LWOP sentence cannot possibly provide a meaningful opportunity for release because it relegates the juvenile offender to spending the rest of his or her life behind prison bars and prohibits him or her from ever reentering society. As the Graham Court stated:
[LWOP] forswears altogether the rehabilitative ideal. By denying the defendant the right to reenter the community, the State makes an irrevocable judgment about that person's value and place in society. This judgment is not appropriate in light of a juvenile ․ offender's capacity for change and limited moral culpability.
560 U.S. at 74, 130 S.Ct. 2011. The Court's reasoning in Graham applies to de facto LWOP sentences with the same force as it does to de jure ones. Like de jure LWOP, de facto LWOP is entirely incompatible with Graham's mandate that those juvenile offenders capable of reform be afforded a meaningful opportunity for release. See Moore v. Biter, 725 F.3d 1184, 1194 (9th Cir. 2013) (“[De facto LWOP] is irreconcilable with Graham's mandate that a juvenile nonhomicide offender must be provided ‘some meaningful opportunity’ to reenter society.” (quoting Graham, 560 U.S. at 75, 130 S.Ct. 2011) ); see also Miller, 567 U.S. at 473, 132 S.Ct. 2455 (“Life without parole ․ [is] at odds with a child's capacity for change.”).10
D. The Decisions of Other Circuits
Our holding that Miller and its antecedents prohibit sentencing non-incorrigible juvenile offenders to term-of-years sentences that meet or exceed their life expectancy has also been adopted by a plurality of our sister circuits. Notably, in McKinley v. Butler, 809 F.3d 908 (7th Cir. 2016), Judge Posner applied the logic of Miller to vacate a 100-year sentence imposed on a non-incorrigible juvenile offender, reasoning that the District Court “did not consider the Supreme Court's ‘children are different’ statement in Miller,” and that:
[I]t is such a long term of years (especially given the unavailability of early release) as to be—unless there is a radical increase, at present unforeseeable, in longevity within the next 100 years—a de facto life sentence, and so the logic of Miller applies․
[T]he “children are different” passage ․ from Miller v. Alabama cannot logically be limited to de jure life sentences, as distinct from sentences denominated in numbers of years yet highly likely to result in imprisonment for life.
Id. at 911.
Similarly, in Moore, the Ninth Circuit reviewed under the Antiterrorism and Effective Death Penalty Act of 1996 (“AEDPA”), Pub. L. No. 104-132, 110 Stat. 1214, the California Court of Appeal's decision that Graham did not apply to a 254-year sentence for multiple crimes on the basis that it was a term-of-years sentence.11 725 F.3d at 1187. The Ninth Circuit held that the state court's failure to apply Graham was “contrary to ․ clearly established Federal law,” id. at 1186 (quoting 28 U.S.C. § 2254(d)(1) ), because (1) “Graham's focus was not on the label of a ‘life sentence,’ ” id. at 1192; (2) both LWOP and de facto LWOP “deny the juvenile the chance to return to society,” id.; and (3) the sentence “is irreconcilable with Graham's mandate that a juvenile nonhomicide offender must be provided ‘some meaningful opportunity’ to reenter society,” id. at 1194 (quoting Graham, 560 U.S. at 75, 130 S.Ct. 2011).
Moreover, in Budder v. Addison, 851 F.3d 1047 (10th Cir. 2017), the Tenth Circuit held, also on AEDPA review, that a sentence of 155 years violated the Eighth Amendment under Graham. Id. at 1053 n.4 (“Graham addressed any sentence that would deny a juvenile nonhomicide offender a realistic opportunity to obtain release, regardless of the label a state places on that sentence.”). It reasoned that Graham created a categorical rule, which a state cannot escape “merely because [it] does not label this punishment as ‘life without parole.’ ” Id. at 1056.
The Eighth Circuit is the only federal court of appeals to date to hold otherwise. In United States v. Jefferson, 816 F.3d 1016 (8th Cir. 2016), the Eighth Circuit determined that Miller did not apply to de facto life sentences because “[t]he Court in Miller did not hold that the Eighth Amendment categorically prohibits imposing a sentence of life without parole on a juvenile offender.” Id. at 1018-19. But Jefferson misses the point of Graham and Miller: a juvenile homicide offender may be sentenced to LWOP only if he or she is determined to be incorrigible at sentencing, otherwise the State “must provide him or her with some realistic opportunity to obtain release before the end of that term.” Graham, 560 U.S. at 82, 130 S.Ct. 2011 (emphasis added); see also Miller, 567 U.S. at 479, 132 S.Ct. 2455. That Miller did not categorically prohibit LWOP altogether does not mean that it permits de facto LWOP sentences for juvenile offenders who are not incorrigible. We therefore decline to follow Jefferson.
The weight of authority supports our conclusion that the Eighth Amendment prohibits de facto LWOP sentences for juvenile offenders that are not incorrigible. Here, the District Court found that Grant is capable of reform, and that determination is not before us on appeal. Under Miller and our holding today, the District Court's finding therefore categorically forecloses a sentence of LWOP, whether de jure or de facto, and requires the District Court to sentence Grant in a manner that allows for “some meaningful opportunity to obtain release based on demonstrated maturity and rehabilitation.” Graham, 560 U.S. at 75, 130 S.Ct. 2011; see also Miller, 567 U.S. at 479, 132 S.Ct. 2455. We will therefore vacate Grant's sentences as to Counts 1 and 2, and remand to the District Court for further proceedings consistent with this opinion.
V. JUVENILE SENTENCING AND MEANINGFUL OPPORTUNITY FOR RELEASE
Having held that a term-of-years sentence that meets or exceeds the life-expectancy of a non-incorrigible juvenile offender violates the Eighth Amendment, we must now consider the contours of the offender's right to a meaningful opportunity for release. See Graham, 560 U.S. at 75, 130 S.Ct. 2011; see also Miller, 567 U.S. at 479, 132 S.Ct. 2455. Grant contends that his sentence does not afford him such an opportunity. Relying on various mortality estimates and social scientific studies, he argues that the District Court determined him to be capable of reform, but that he was nonetheless still sentenced to a term-of-years that allows release only when he is seventy-two, the same age as his estimated life expectancy.12 Even if his life expectancy exceeds his sentence by some years, he contends that a meaningful opportunity for release must afford him an opportunity for “[p]ersonal [f]ulfillment,” which release at age seventy-two does not provide. Appellant Reply Br. at 1.
The Government disagrees. It contends that seventy-two is Grant's life expectancy measured from birth, but that his life expectancy measured from his current age of forty-four is actually 76.7 years. Grant's sentence is constitutional, the Government argues, because he should live 4.7 years past his release at age seventy-two. Thus, in the Government's estimation, Grant's sentence sufficiently provides for “hope for some years of life outside prison walls,” which is all that Graham and Miller require. Government Br. at 29 (emphasis omitted) (quoting Montgomery, 136 S.Ct. at 737).
To determine what constitutes a meaningful opportunity for release, we look to the Supreme Court's original diagnosis of the constitutional infirmity that plagues juvenile LWOP. See Casey v. Planned Parenthood of Se. Pa., 14 F.3d 848, 857 (3d Cir. 1994) (“We must look to the language of the Supreme Court's opinion to see what it intend[s]․”). In holding that juvenile LWOP is not an appropriate sentence in light of an offender's capacity for change and limited culpability, the Court viewed the problem with the punishment as more profound than just denial of release:
[A] categorical rule [barring LWOP] gives all juvenile nonhomicide offenders a chance to demonstrate maturity and reform. The juvenile should not be deprived of the opportunity to achieve maturity of judgment and self-recognition of human worth and potential․ Life in prison without the possibility of parole gives no chance for fulfillment outside prison walls, no chance for reconciliation with society, no hope.
Graham, 560 U.S. at 79, 130 S.Ct. 2011; see also id. at 69-70, 130 S.Ct. 2011 (“[LWOP] deprives the convict of ․ hope of restoration”); id. at 73, 130 S.Ct. 2011 (“A life without parole sentence improperly denies the juvenile offender a chance to demonstrate growth and maturity.”). This passage conveys the essence of what a “meaningful opportunity for release” is: a non-incorrigible juvenile offender must be afforded an opportunity for release at a point in his or her life that still affords “fulfillment outside prison walls,” “reconciliation with society,” “hope,” and “the opportunity to achieve maturity of judgment and self-recognition of human worth and potential.” Id. at 79, 130 S.Ct. 2011. That is, the mandate encompasses more than mere physical release at a point just before a juvenile offender's life is expected to end.
The contours of the requirement are also informed by the Court's concern that “defendants serving life without parole sentences are often denied access to vocational training and other rehabilitative services that are available to other inmates.” Id. at 74, 130 S.Ct. 2011; see also id. at 79, 130 S.Ct. 2011 (“[I]t is the policy in some prisons to withhold counseling, education, and rehabilitation programs for those who are ineligible for parole consideration.”). This view illustrates the Court's belief that—in order to afford “hope” and a chance for “fulfillment outside prison walls,” “reconciliation with society,” and “self-recognition of human worth and potential,” consistent with the Eighth Amendment, id. at 79, 130 S.Ct. 2011—the State must give non-incorrigible juvenile offenders the opportunity to meaningfully reenter society upon their release.13 See id. at 75, 130 S.Ct. 2011 (“[The Eighth Amendment] prohibit[s] States from making the judgment at the outset that those offenders never will be fit to reenter society.”); see also id. at 74, 130 S.Ct. 2011 (stating that it is “not appropriate” for sentencing courts to “deny[ ] the [non-incorrigible juvenile] defendant the right to reenter the community” in light of his or her “capacity for change and limited moral culpability” (emphasis added) ).
We must therefore effectuate the Court's mandate and adopt a broader conception of what constitutes a “meaningful opportunity for release” than what the Government puts forth. The Government's “hope for some years outside prison walls” standard is too narrow in light of the Court's statements that the Eighth Amendment requires mitigating the pernicious long term effects that LWOP has on juvenile offenders who still have the capacity to reform. However, we agree with the Government that the Supreme Court has not gone as far as to say that juvenile offenders must be afforded a right to a “meaningful life” after prison—in fact, neither Miller nor Graham even guarantees that a juvenile offender will ever be released from prison during his or her lifetime. Graham, 560 U.S. at 75, 130 S.Ct. 2011; see also Miller, 567 U.S. at 479, 132 S.Ct. 2455.
With this in mind, we elect to fashion a principled legal framework that carries out the Supreme Court's holdings but goes no further. We do so for three reasons. First, as always, we are “bound to follow the mandate of the Supreme Court as embodied in its opinion.” Casey, 14 F.3d at 859. Second, “juvenile offenders whose crimes reflect the transient immaturity of youth” are now a constitutionally recognized class of defendants that are afforded a right to a meaningful opportunity for release. Montgomery, 136 S.Ct. at 734. The distinction between incorrigible and non-incorrigible juvenile homicide offenders is undoubtedly substantive, and we must therefore take great precautions to ensure that courts properly account for this feature once they have determined that a juvenile offender is capable of reform. See id. (“Miller announced a substantive rule of constitutional law.”). And third, by providing principled guidance to lower courts on how to effectuate the Court's meaningful opportunity for release mandate, we “ensure[ ] ‘careful observation of [the] allocation of authority’ established by the three-tier system of federal courts which ‘is necessary for a properly functioning judiciary.’ ” Casey, 14 F.3d at 857 (first alteration added) (quoting Litman v. Mass. Mut. Life Ins. Co., 825 F.2d 1506, 1508 (11th Cir. 1987) ). Accordingly, for all of these reasons, we resolve to provide such guidance in order to fulfill our judicial responsibility to give life to the minimum constitutional protections that the Supreme Court has found the Eighth Amendment requires.
To begin, a sentencing process that effectuates both our holding that de facto LWOP for non-incorrigible juvenile offenders violates the Eighth Amendment and a meaningful opportunity for release must start with a factual determination of the juvenile offender's life expectancy. We shall require sentencing courts to make this determination so that a juvenile offender who is capable of reform is not sentenced to a term-of-years beyond his or her expected mortality. Additionally, the juvenile offender's life expectancy provides an informed estimate that allows sentencing courts to calculate the amount of time that he or she will have to reenter society after an opportunity for release.
How, then, does one measure life expectancy? One source could be actuarial tables or life expectancy data. In addition to accounting for gender, life expectancy tables can “focus more granularly on numerous other distinctions that impact the life expectancy of a particular individual, such as race, income, geography, education, family history, medical history, and other factors.” United States v. Mathurin, 868 F.3d 921, 932 (11th Cir. 2017). But reliance solely on life expectancy tables is problematic. In Mathurin, the Eleventh Circuit identified the serious constitutional issues that relying on such data can raise:
[This] approach does raise some questions, including whether it would be constitutional to rely on a person's race in determining how long a sentence to impose on that individual. By Defendant's reasoning, and based on the mortality table he cited in the district court, Hispanics should receive longer sentences than either whites or blacks solely because they generally live longer, and Hispanic females should receive the longest sentences of all due to their longer average life expectancy․
Further, mortality tables shed no light on whether it is one's membership in a certain racial or ethnic population that, as a biological matter, determines life expectancy or whether instead it is the social, economic, medical, and cultural factors associated with a particular ethnic identity that primarily determine how long an individual can be expected to live.
Id. (citations omitted). These concerns are not confined to the context of race, either. By virtue of having a longer life expectancy based on an actuarial table, women would be sentenced to longer prison terms than men, the richer longer than the poorer, and the well-educated longer than the lesser educated, to name a few. We therefore decline to advise that life expectancy be measured based solely on actuarial tables alone. See O'Toole v. United States, 242 F.2d 308, 309 (3d Cir. 1957) (“[L]ife tables are a guide, not a formula which a judge ․ is compelled to apply.”).
Rather, to avoid the aforementioned constitutional problems, we hold that a sentencing judge must conduct an individualized evidentiary hearing to determine the non-incorrigible juvenile homicide offender's life expectancy before sentencing him or her to a term-of-years sentence that runs the risk of meeting or exceeding his or her mortality. Such hearings are already a familiar exercise for lower courts, which routinely measure life expectancy in various tort, contract, and employment disputes. See, e.g., Anastasio v. Schering Corp., 838 F.2d 701, 709 (3d Cir. 1988) (“A claimant's work and life expectancy are pertinent factors in calculating front pay, just as they are in assessing damages for future loss of earnings in breach of employment contract and personal injury cases.”). Critically, in addition to actuarial tables, lower courts should consider any evidence made available by the parties that bears on the offender's mortality, such as medical examinations, medical records, family medical history, and pertinent expert testimony. Our foregoing constitutional concerns are dispelled by consideration of such evidence at an individualized hearing, which affords lower courts substantial discretion to “make an individualized assessment based on the facts presented,” Gall v. United States, 552 U.S. 38, 50, 128 S.Ct. 586, 169 L.Ed.2d 445 (2007), so that “the punishment ․ fit[s] the offender and not merely the crime,” Pepper v. United States, 562 U.S. 476, 487-88, 131 S.Ct. 1229, 179 L.Ed.2d 196 (2011) (quoting Williams v. People of State of N.Y., 337 U.S. 241, 247, 69 S.Ct. 1079, 93 L.Ed. 1337 (1949) ).
Once a non-incorrigible juvenile offender's life expectancy has been determined, the next step is for a sentencing court to shape a sentence that properly accounts for a meaningful opportunity for release. As discussed, a “meaningful opportunity for release” must provide for “hope” and a chance for “fulfillment outside prison walls,” “reconciliation with society,” and “the opportunity to achieve maturity of judgment and self-recognition of human worth and potential.” Graham, 560 U.S. at 79, 130 S.Ct. 2011. This mandate, therefore, raises a challenging question for this Court: at what age is one still able to meaningfully reenter society after release from prison? Is there a principled reason for why, say, a juvenile offender can properly reenter society at age fifty but not at age sixty? At age sixty but not at age seventy? We believe not. Unlike in Roper, where the Supreme Court relied on scientific and social scientific scholarship to proscribe the death penalty for anyone who commits a crime before the age of eighteen, see 543 U.S. at 569-70, 125 S.Ct. 1183, we are not aware of any widely accepted studies to support such precise line drawing on a principled basis in the prison release context.
However, what is clear is that society accepts the age of retirement as a transitional life stage where an individual permanently leaves the work force after having contributed to society over the course of his or her working life. See, e.g., Retirement, Black's Law Dictionary (10th ed. 2014) (“Termination of one's own employment or career, esp. upon reaching a certain age․”). It is indisputable that retirement is widely acknowledged as an earned inflection point in one's life, marking the simultaneous end of a career that contributed to society in some capacity and the birth of an opportunity for the retiree to attend to other endeavors in life.
As we stated above, a non-incorrigible juvenile offender is not guaranteed an opportunity to live a meaningful life, and certainly not to a meaningful retirement. Nevertheless, in order to effectuate the Eighth Amendment's requirement of meaningful opportunity for release, a juvenile offender that is found to be capable of reform should presumptively be afforded an opportunity for release at some point before the age of retirement. Cf. Graham, 560 U.S. at 58, 130 S.Ct. 2011 (“To determine whether a punishment is cruel and unusual, courts must look beyond historical conceptions to the evolving standards of decency that mark the progress of a maturing society.” (internal quotation marks omitted) (quoting Estelle, 429 U.S. at 102, 97 S.Ct. 285) ). A sentence that preserves the juvenile offender's opportunity to contribute productively to society inherently provides him or her with “hope” to “reconcil[e] with society” and achieve “fulfillment outside prison walls.” Id. at 79, 130 S.Ct. 2011. It also accounts for the Court's trepidation that LWOP sentences deprive non-incorrigible juvenile offenders of vocational training opportunities, which presumably otherwise prepare them to become productive members of society's working class. See id. at 74, 130 S.Ct. 2011.
Accordingly, lower courts must consider the age of retirement as a sentencing factor, in addition to life expectancy and the § 3553(a) factors, when sentencing juvenile offenders that are found to be capable of reform. Critically, under all circumstances, lower courts must only consider the uniform national age of retirement. Otherwise, estimates of retirement ages that account for locality, state, gender, race, wealth or other differentiating characteristics raise similar constitutional concerns to those plagued by reliance on life-expectancy tables alone. Without fixing the age of retirement to a uniform standard, classes of juvenile defendants that retire on average later in life would unreasonably be subjected to longer sentences. Cf. Mathurin, 868 F.3d at 932 (sentencing juveniles based solely on mortality tables “would unquestionably lead to challenges from defendants from longer-living ethnic groups who would be subject to longer sentences based on that ethnicity”).
We cannot say with certainty what the precise national age of retirement is, as it is a figure that incrementally fluctuates over time. The Social Security Administration (“SSA”) provides for early retirement at age sixty-two and—for people born after 1960—for full retirement at age sixty-seven. See 20 C.F.R. § 404.409. Over half of Americans that retired in 2016 did so early and before their full retirement age.14 A series of polls conducted by Gallup News since April 2014 have indicated sixty-two, sixty-five, and sixty-six as either the mean or median ages of retirement or expected ages of retirement.15 Yet another study concluded that the average age of retirement is sixty-four for men and sixty-two for women. See Alicia H. Munnell, Ctr. for Ret. Research At Bos. Coll., The Average Age of Retirement: An Update (2015). Thus, by all accounts, the national age of retirement to date is between sixty-two and sixty-seven inclusive.
However, the age of sixty-five appears to be the commonly accepted age of retirement in the national conscience. It was set as the original normal retirement age when the Social Security Act was enacted in 1935, and remains one of the most—if not the most—frequent ages of retirement. See Social Security Act of 1935, Pub. L. No. 74-271, § 210(c), 49 Stat. 620 (defining “qualified individual” for Social Security purposes in part as “any individual with respect to whom ․ is at least sixty-five years of age”); id. § 202(a) (authorizing “qualified individual[s]” to receive Social Security payments “on the date [they] attain[ ] the age of sixty-five”).16 Today, pension plans must begin to distribute benefits by age sixty-five to qualify for various significant tax benefits, see 26 U.S.C. § 401(a)(14)(A), and the Employment Retirement Security Act defines the term “normal retirement age” in part as “the time a plan participant attains age 65,” 29 U.S.C. § 1002(24); see also 26 U.S.C. § 411(a)(8)(B)(i) (same); 26 C.F.R. § 1.411(a)-7(b)(ii)(A) (same). Without definitively determining the issue, we consider sixty-five as an adequate approximation of the national age of retirement to date. However, district courts retain the discretion to determine the national age of retirement at sentencing, and remain free to consider evidence of the evolving nature of this estimate.
We do not, however, categorically foreclose the possibility that a district judge may sentence a non-incorrigible juvenile offender beyond the national age of retirement, subject to the § 3553(a) factors. A sentencing judge has “greater familiarity with ․ the individual case and the individual defendant before him than the ․ appeals court.” Kimbrough v. United States, 552 U.S. 85, 109, 128 S.Ct. 558, 169 L.Ed.2d 481 (2007) (quoting Rita v. United States, 551 U.S. 338, 357-58, 127 S.Ct. 2456, 168 L.Ed.2d 203 (2007) ). “He is therefore ‘in a superior position to find facts and judge their import under § 3553(a)’ in each particular case.” Id. (quoting Gall, 552 U.S. at 51, 128 S.Ct. 586); see also United States v. Booker, 543 U.S. 220, 233, 125 S.Ct. 738, 160 L.Ed.2d 621 (2005) (“We have never doubted the authority of a judge to exercise broad discretion in imposing a sentence within a statutory range.”). In light of the fact that Miller did not categorically bar a sentencing court from imposing LWOP on juvenile homicide offenders to begin with, we believe that it would be inappropriate to restrict a district court's discretion to fashion an appropriate term-of-years sentence in the alternative.
We therefore adopt only a rebuttable presumption that a non-incorrigible juvenile offender should be afforded an opportunity for release before the national age of retirement, not a hard and fast rule. While we believe that this presumption is necessary to give life to the Supreme Court's holdings in Graham and Miller, it also affords lower courts the discretion to depart from it in the exceptional circumstances where a juvenile offender is found to be capable of reform but the § 3553(a) factors still favor a sentence beyond the national age of retirement. We believe that such instances will be rare and unusual, and that, even then, a term-of-years sentence cannot meet or exceed the juvenile offender's life expectancy. However, given the “discrete institutional strengths” of district courts to provide for individualized sentencing, Kimbrough, 552 U.S. at 109, 128 S.Ct. 558, we see no reason why they should not retain such discretion in the juvenile context, so long as their departure is consistent with Miller's Eighth Amendment guarantee for a meaningful opportunity for release.17
To summarize, we hold that: (1) a sentence that either meets or exceeds a non-incorrigible juvenile offender's life expectancy violates the Eighth Amendment; (2) courts must hold evidentiary hearings to determine the non-incorrigible juvenile homicide offender's life expectancy before sentencing him or her to a term-of-years that may meet or exceed his or her expected mortality; and (3) when sentencing the juvenile homicide offender, a court must consider as sentencing factors his or her life expectancy and the national age of retirement, in addition to the § 3553(a) factors, to properly structure a meaningful opportunity for release. A non-incorrigible juvenile offender should presumptively be sentenced below the national age of retirement, unless the remaining sentencing factors strongly mitigate against doing so. Sentencing judges therefore retain the discretion to sentence incorrigible juvenile offenders to LWOP and non-incorrigible ones to a term-of-years beyond the national age of retirement but below life expectancy, although we believe that either of these circumstances will be rare and exceptional.18
Our decision today effectuates Miller and its antecedents—as we are required to do—which provide that non-incorrigible juvenile offenders must be afforded a meaningful opportunity for “fulfillment outside prison walls,” “reconciliation with society,” “hope,” and the “opportunity to achieve maturity of judgment and self-recognition of human worth and potential.” Graham, 560 U.S. at 79, 130 S.Ct. 2011. We will vacate Grant's sixty-year sentences for the RICO conspiracy and racketeering convictions, and remand to the District Court to resentence Grant consistent with this opinion.19
VI. SENTENCING PACKAGE DOCTRINE
Separately, Grant argues that the sentencing package doctrine requires vacatur of his forty-year concurrent sentence for the drug convictions and the five-year consecutive sentence for the gun offense. That doctrine states:
[W]hen a defendant is found guilty on a multicount indictment, there is a strong likelihood that the district court will craft a disposition in which the sentences on the various counts form part of an overall plan. When a conviction on one or more of the component counts is vacated, common sense dictates that the judge should be free to review the efficacy of what remains in light of the original plan, and to reconstruct the sentencing architecture upon remand ․ if that appears necessary in order to ensure that the punishment still fits both crime and criminal.
United States v. Miller, 594 F.3d 172, 180 (3d Cir. 2010) (alteration in original) (quoting United States v. Davis, 112 F.3d 118, 122 (3d Cir. 1997) ); see also id. at 181-82 (“When a conviction on one or more interdependent counts is vacated on appeal, the resentencing proceeding conducted on remand is de novo unless we specifically limit the district court's authority.”). This argument is before us on plain error review because defense counsel did not explicitly raise the sentencing package doctrine below, a point which Grant concedes.20 United States v. Price, 458 F.3d 202, 206 (3d Cir. 2006) (“We apply plain error review when an issue was not brought to the attention of the district court.”); see Appellant Reply Br. at 22 (“[C]ounsel did not use the magic words ‘sentencing package doctrine’․”).
Grant's sentencing package contention fails on plain error review. The sentencing package doctrine provides a basis for a de novo resentencing when “a conviction on one or more interdependent counts is vacated.” Miller, 594 F.3d at 181-82 (emphasis added); see also Dean v. United States, ––– U.S. ––––, 137 S.Ct. 1170, 1176, 197 L.Ed.2d 490 (2017) (explaining that sentencing package cases “typically involve ․ a successful attack by a defendant on some but not all of the counts of conviction.” (quoting Greenlaw v. United States, 554 U.S. 237, 253, 128 S.Ct. 2559, 171 L.Ed.2d 399 (2008) ) ). Here, all of Grant's convictions were affirmed on direct review, and the sentencing package doctrine, as we have previously defined it, is therefore inapplicable. Whether the doctrine should also apply to vacated sentences raises a substantial question that merits further consideration.21 See United States v. Catrell, 774 F.3d 666, 670 (10th Cir. 2014) (applying sentencing package doctrine to vacated sentence). However, we decline to reach that issue today because Grant did not adequately preserve it, see supra note 20, and it is therefore not properly before us. See United States v. Dupree, 617 F.3d 724, 727 (3d Cir. 2010) (“[A]rguments not raised in the district courts are waived on appeal.”). The alleged error that Grant identifies is not plain, and his sentences on the drug and gun possession convictions remain unchanged.22
For the aforementioned reasons, we will vacate Grant's sentences as to his RICO conspiracy and racketeering convictions (Counts 1 and 2). We will also vacate Grant's sentence for drug conspiracy (Count 4) so that the District Court may correct its inadvertent sentencing error and reinstate the original forty-year concurrent sentence for this conviction. This case is remanded to the District Court to resentence Grant consistent with this opinion.
United States v. Grant, No. 16-3820, concurring in part and dissenting in part.
I join in full Parts III through V of the majority. I completely agree with the majority's approach to the challenging yet critical issues that this appeal raises with respect to the Eighth Amendment and juvenile sentencing and that we must vacate Grant's sixty-year sentences as to Counts 1 and 2. Furthermore, the sixty-year sentence imposed on Count 4 constituted an inadvertent error and must be vacated. However, I cannot join Part VI of the majority opinion. In short, Grant raised with sufficient specificity the argument that the sentencing package doctrine applied in this case, and the District Court thereby committed reversible error by failing to apply this doctrine. Accordingly, I would vacate all of the sentences imposed by the District Court and remand for resentencing de novo on Counts 4, 5, 6, and 11 as well as on Counts 1 and 2.
As the majority notes, the sentencing package doctrine generally states:
“[W]hen a defendant is found guilty on a multicount indictment, there is a strong likelihood that the district court will craft a disposition in which the sentences on the various counts form part of an overall plan. When a conviction on one or more of the component counts is vacated, common sense dictates that the judge should be free to review the efficacy of what remains in light of the original plan, and to reconstruct the sentencing architecture upon remand ․ if that appears necessary in order to ensure that the punishment still fits both crime and criminal.
(Maj. Op. at 153–54 (quoting United States v. Miller, 594 F.3d 172, 180 (3d Cir. 2010) ).) “When a conviction on one or more interdependent counts is vacated on appeal, the resentencing proceeding conducted on remand is de novo unless we specifically limit the district court's authority.” Miller, 594 F.3d at 181-82. It appears uncontested that this doctrine applies in the § 2255 context. See, e.g., United States v. Davis, 112 F.3d 118, 120-24 (3d Cir. 1997).
While the majority relies on the plain error standard of review, I believe that Grant adequately preserved his argument regarding the applicability of the sentencing package doctrine. Although (at least in retrospect) Grant should have explicitly referred to the doctrine, there is no “magic words” requirement for deciding whether a party has sufficiently raised a particular theory or argument. Instead, “the ‘crucial question regarding waiver is whether [Grant] presented the argument with sufficient specificity to alert the district court.’ ” (Id. at 44 n.20, 128 S.Ct. 586 (quoting Brennan v. Norton, 350 F.3d 399, 418 (3d Cir. 2003) ).) Grant clearly argued that the District Court was required to resentence de novo on all counts. At the resentencing hearing ordered by the District Court, his attorney's statements implicated the basic notion of a single sentencing “package” by characterizing the individual sentences as “part and parcel” of one overall sentence:
[T]his was all part and parcel of one sentence. I don't think anybody looked upon this as somehow a breakdown of you got 40 on this, you got 40 on that and five on that. This was a life sentence” (A40); “[I]t is really part and parcel of the entire sentence that was imposed here, Judge. [To now say] you really got this 40, and you got this five, I mean really is not the spirit of Miller” (A43); “If you parcel out the 40 at this time, Judge, is not really consistent with what Judge Ackerman was doing. Ackerman knew, Judge Ackerman that he was giving him life without parole. So, I mean, to say now that, well, this part should stand, I mean, it is not really consistent with what the sentence was. The sentence was life without parole. I submit to your Honor that really what we are here for today is a new sentencing hearing” (A44); “[I]t should be clear that really it is a whole new sentencing. Everything was part and parcel of imposing a sentence that the Court thought was the correct sentence” (A85).
I do not believe that anything more was required to raise the sentencing package doctrine (after all, it is Grant's position that the doctrine clearly applies where a sentence is vacated and not merely where the underlying conviction is vacated, and, as I explain below, I agree with Grant). In fact, the District Court told defense counsel that “I understand your point.” (A42.) “You are saying that I should look at this as one cohesive sentence of life and treat it that way in determining what is an appropriate total sentence.” (Id.) “So the 40-year sentence—anyway, I understand your point. You say it is part and parcel of all one sentence, and that the sentence as a whole was offensive to the Miller concept, right?” (A44.) The District Court then expressly disposed of Grant's argument that “this is an entirely new sentence” (A151) and essentially applied law of the case principles instead of the sentencing package doctrine.
According to the majority, “[t]he sentencing package doctrine ․ has been applied in our precedential opinions only to vacated convictions—not vacated sentences—because ‘[w]hen a conviction on one or more of the component counts is vacated, common sense dictates that the judge should be free to review the efficacy of what remains in light of the original plan ․ in order to ensure that the punishment still fits both crime and criminal.’ ” (Maj. Op. at 154 n.20 (quoting Miller, 594 F.3d at 180).) However, I see no reason why the doctrine does not apply to vacated sentences. In other words, what real difference is there between a vacated sentence and a vacated conviction for purposes of the sentencing package doctrine? A vacated sentence on one or more counts may mean that “what remains” no longer fits the “criminal.” “Because a district court's ‘original sentencing intent may be undermined by altering one portion of the calculus,’ United States v. White, 406 F.3d 827, 832 ( [7th Cir. 2005] ), an appellate court when reversing one part of a defendant's sentence ‘may vacate the entire sentence ․ so that, on remand, the trial court can reconfigure the sentencing plan ․ to satisfy the sentencing factors in 18 U.S.C. § 3553(a).’ Greenlaw v. United States, [554 U.S. 237, 253, 128 S.Ct. 2559, 171 L.Ed.2d 399 (2008) ].” Pepper v. United States, 562 U.S. 476, 507, 131 S.Ct. 1229, 179 L.Ed.2d 196 (2011) (emphasis added); see also, e.g., 28 U.S.C. § 2255(a) (“A prisoner in custody under sentence of a court established by Act of Congress claiming the right to be released upon the ground that the sentence was imposed in violation of the Constitution or laws of the United States, or that the court was without jurisdiction to impose such sentence, or that the sentence was in excess of the maximum authorized by law, or is otherwise subject to collateral attack, may move the court which imposed the sentence to vacate, set aside or correct the sentence.” (emphasis added) ). In United States v. Guevremont, 829 F.2d 423 (3d Cir. 1987), we allowed the district court to correct an illegal sentence:
In addition to allowing an increase in sentence when the sentence is less than the statutory minimum, courts have also held that, where the sentencing judge's intention is clear, an increase of the sentence to make it conform with that intention is constitutional. In United States v. Busic, 639 F.2d 940 (3d Cir.), cert denied, [452 U.S. 918] [101 S.Ct. 3055, 69 L.Ed.2d 422] (1981), we rejected the argument that the Constitution does not allow an increase of sentence in a case where the sentencing judge's intention is clear. In Busic, we ruled that, where one count of an interdependent sentencing plan has been vacated on appeal, the entire plan should be vacated and the defendant should be resentenced according to the initial intent of the court. We found that, under the circumstances, concerns of judicial vindictiveness were removed and to hold otherwise would allow “the court's sentencing plan ․ [to be] thwarted.” 639 F.2d at 947.
Id. at 428 (citations omitted).1 Accordingly, it is not surprising that the Tenth Circuit “appl[ied] [the] sentencing package doctrine to [a] vacated sentence.” (Id. at 46-47 (citing United States v. Catrell, 774 F.3d 666, 670 (10th Cir. 2014) ).) While the Catrell court acknowledged that it (like both the Third Circuit and the Supreme Court) has stated that the doctrine applies when one of the counts is set aside, “[t]his language is best viewed as descriptive rather than prescriptive.” Catrell, 774 F.3d at 670.
Furthermore, the specific circumstances of this case clearly favor the application of the sentencing package doctrine. This case involves more than, to give two examples, mere errors in calculating the defendant's offense level or range under the Sentencing Guidelines. (See, e.g., Appellee's Brief at 20-21 (“Since, then, however, at least one panel, citing United States v. Ciavarella, 716 F.3d 705, 735 (3d Cir. 2013), has opined that de novo resentencing still would not be required, even when a conviction is vacated unless the vacated count affects the defendant's total offense level or guideline range.” (citing United States v. Walpole, 599 Fed.Appx. 56, 58 (3d Cir. 2015) ) ).) Instead, the District Court determined that Grant's original LWOP sentences were unconstitutional under the Supreme Court's recent rulings applying the Eighth Amendment to juvenile sentencing and ultimately resentenced him to serve concurrent terms of sixty years on these counts. We then have considered in this appeal novel issues of constitutional law regarding the unique protections that the Eighth Amendment affords to juvenile offenders. Based on our holdings, “[w]e will vacate Grant's sixty-year sentences for the RICO conspiracy and racketeering convictions, and remand to the District Court to resentence Grant consistent with this opinion.” (Maj. Op. at 153 (footnote omitted).) In turn, the government asserts that Grant's emphasis on the interconnected nature of the sentences does not alter the analysis because it is undisputed that a defendant's sentences, “when ‘collect[ed] ․ in the aggregate, [Appellant's Brief at 51-52], constitute an entire package.’ ” (Appellee's Brief at 22 (citing United States v. Pimienta-Redondo, 874 F.2d 9, 14 (1st Cir. 1989) (en banc ) ).) Specifically, the five-year consecutive sentence on the gun conviction was clearly implicated by this constitutional analysis. For instance, the majority acknowledged in its introduction as well as its summary of the factual and procedural history that “the District Court sentenced Grant to a term of sixty-five years without parole” (Maj. Op. at 135) and, “[u]nder this sentence, assuming good time credit, see 18 U.S.C. § 3624(b)(1), Grant will be eligible for release at age seventy-two, which he contends is the same age as his life expectancy” (id. at 137). Both Grant and the government relied on this sixty-five-year term in their respective life expectancy arguments. In addition, the original LWOP sentences affected the sentencing decisions on the remaining counts by rendering such decisions merely symbolic. Whatever sentence the sentencing court imposed on the drug (and gun) counts, it was clear that Grant would die in prison. Accordingly, the District Court indicated at its original sentencing that it “wants to send a message” about the “plague” of drugs. (A450-A451.) “On resentencing, of course, with the LWOP sentences on Counts 1 and 2 vacated, the 40 years on Counts 4-6 were highly consequential, establishing a substantial floor for the overall sentence.” (Appellant's Brief at 52.)
Eighth Amendment case law likewise indicates that the sentencing package doctrine governs this proceeding. “Notably, in McKinley v. Butler, 809 F.3d 908 (7th Cir. 2016), Judge Posner applied the logic of Miller to vacate a 100-year sentence imposed on a non-incorrigible juvenile offender.” (Maj. Op. at 145.) However, this offender was actually sentenced to “consecutive 50-year prison terms.” Id. at 909. The majority also relies on AEDPA rulings from the Ninth and the Tenth Circuit concluding that the respective aggregate sentences violated the Eighth Amendment under Miller and Graham, respectively. See Budder v. Addison, 851 F.3d 1047, 1049 (10th Cir.) (three life sentences and additional sentence of twenty years all to be served consecutively and petitioner will not be eligible for parole until he served 131.75 years), cert. denied sub nom. Byrd v. Budder, ––– U.S. ––––, 138 S.Ct. 475, 199 L.Ed.2d 374 (2017); Moore v. Biter, 725 F.3d 1184, 1187 (9th Cir. 2013) (consecutive sentences totaling 254 years and four months with parole eligibility after petitioner served 127 years and two months). The Massachusetts Supreme Judicial Court determined that “a trial court judge, in resentencing a juvenile offender originally sentenced to multiple consecutive terms of life without the possibility of parole, may conduct a sentencing hearing to consider resentencing the juvenile offender to concurrent terms.” Commonwealth v. Costa, 472 Mass. 139, 33 N.E.3d 412, 415 (2015). Like this Court, Massachusetts recognizes the sentencing package doctrine. See, e.g., id. at 417. While the practical consequences of deciding between consecutive and concurrent sentences were originally limited to the defendant's treatment in prison, the state's intervening case law applying Miller “transformed a choice that could be regarded as ‘somewhat symbolic’ into one of some consequence since a consecutive sentence doubles the amount of time the defendant must serve before he becomes eligible for parole.” Id.
Significantly, the Wyoming Supreme Court expressly invoked the sentencing package doctrine in reversing a “de facto” LWOP sentence and “remand[ing] to the district court with instructions to resentence on all counts.” Bear Cloud v. State, 334 P.3d 132, 135 (Wyo. 2014). The state supreme court originally affirmed sentences of “20-25 years in prison for Aggravated Burglary, life in prison ‘according to law’ for first-degree murder [i.e., LWOP], to be served consecutively to the aggravated burglary sentence; and 20-25 years in prison for conspiracy to commit aggravated burglary, to be served concurrently with the first-degree murder sentence.” Id. (citation omitted). The United States Supreme Court granted certiorari, ordering “ ‘Judgment vacated, and case remanded to the Supreme Court of Wyoming for further consideration in light of [Miller].’ ” Id. (quoting Bear Cloud v. Wyoming, 568 U.S. 802, 133 S.Ct. 183, 183-84, 184 (L.Ed.2d 5 2012) ). The Wyoming Supreme Court originally determined on remand from the United States Supreme Court that only the life sentence was at issue, and the state trial court accordingly resentenced Bear Cloud to life imprisonment with the possibility of parole after serving twenty-five years, to run consecutive to the previously imposed sentence of twenty to twenty-five years for aggravated burglary and concurrently to the additional twenty to twenty-five sentence for conspiracy. Id. at 136. Bear Cloud appealed, and the Wyoming Supreme Court acknowledged that its earlier remand for resentencing solely on the LWOP sentence was inconsistent with Pepper as well as the state supreme court's own holding in an appeal filed by Bear Cloud's co-defendant, see Sen v. State, 301 P.3d 106 (Wyo. 2013). Bear Cloud, 334 P.3d at 141. “When the United States Supreme Court vacated the judgment in [the first Bear Cloud disposition], it wiped the slate clean. We remand for the district court to consider the entire sentencing package—that is, the sentences for all three counts—when it resentences Mr. Bear Cloud.” Id. at 142; see also, e.g., id. at 141 (“That process must be applied to the entire sentencing package, when the sentence is life without parole, or when aggregate sentences result in the functional equivalent of life without parole.”); Sen, 301 P.3d at 127 (“Further, because Sen's sentence of life without the possibility of parole may have impacted the sentencing decisions with respect to his conspiracy and aggravated burglary convictions, which resulted in an additional 40 to 50 years imprisonment beyond his life term, we think the appropriate course is to vacate those sentences and remand for resentencing on all counts in order to give full effect to our decision.”).
The District Court, however, did not “consider the entire sentencing package” when it resentenced Grant. As the government acknowledges, the doctrine “leaves a judge ‘free to review,’ ‘entitled to reconsider’ and with jurisdiction to recalculate' [the] § 2255 petitioner's entire sentence.” (Appellee's Brief at 21 (quoting Davis, 112 F.3d at 122-23).) However, the District Court instead considered and applied the law of the case doctrine, explaining that “it would be almost unfair to the system and unfair to Judge Ackerman all of these years later for me to sort of sit in his shoes to figure out, with him having the feel of the case, having listened to the evidence of the distribution of the drugs, the extent of the drugs, the nature of the drug trafficking and enterprise that was involved here” and that there was nothing in the record indicating “that this was some kind of clear manifest injustice by Judge Ackerman” with respect to the drug and gun counts. (A152.) Accordingly, I would vacate the sentences on the drug and gun counts (as well as the RICO conspiracy and racketeering counts) and remand for resentencing de novo.
2. The indictment charged multiple individuals involved in the E-Port Posse. The charges discussed here are limited to those made against Grant.
3. The District Court also unwittingly imposed a sixty-year sentence on Count 4, a drug offense for which Grant was originally sentenced to forty years. The District Court decided—just prior to issuing this sentence—that the scope of its review was limited to Counts 1 and 2, and that it would leave intact the original sentence for Grant's drug convictions. App. 152 (“[T]here is nothing in the record before me that would indicate that [there] was some kind of clear manifest injustice by [the original District Court] with the sentence that [it] issued with regard to the drug conviction․”). The District Court's sixty-year resentence to Count 4 was therefore undoubtedly inadvertent error. We will vacate that sentence and instruct the District Court to reinstate the original forty-year concurrent sentence for Count 4.
4. The Court reasoned that “only 11 jurisdictions nationwide in fact impose life without parole sentences on juvenile nonhomicide offenders—and most of those do so quite rarely—while 26 States, the District of Columbia, and the Federal Government do not impose them despite apparent statutory authorization.” Graham, 560 U.S. at 64, 130 S.Ct. 2011.
5. See Comprehensive Crime Control Act of 1984, Pub. L. No. 98-473, § 218(a)(5), 98 Stat. 1837, 2027 (repealing 18 U.S.C. §§ 4201-4218).
6. Our decision today is therefore not the only constitutionally permissible remedy to this case, as Congress retains the prerogative to afford additional protections to juvenile homicide offenders beyond the minimal safeguards that the Eighth Amendment provides. See Miknevich, 638 F.3d at 186 (“[Congress] possesses broad authority to determine the types and limits of punishments for crimes.”).
7. This holding extends to all sentencings of juvenile non-homicide offenders because, by definition, such offenders are not incorrigible. See Graham, 560 U.S. at 69, 130 S.Ct. 2011 (“[W]hen compared to an adult murderer, a juvenile offender who did not kill or intend to kill has a twice diminished moral culpability.”). Since we hold that non-incorrigible juvenile homicide offenders cannot be sentenced to de facto LWOP, the same must be true for juvenile non-homicide offenders as well. See id. (“The Court has recognized that defendants who do not kill, intend to kill, or foresee that life will be taken are categorically less deserving of the most serious forms of punishment than are murderers.”).
8. Notably, at oral argument, the Government conceded that a sentence that exceeds the life expectancy of a non-incorrigible juvenile homicide offender violates the Eighth Amendment. The Government contends, however, that in this case, Grant's life expectancy is 76.7 and that he is therefore likely to be released before his death, consistent with Miller.
9. Incapacitation is the only conceivable penological justification that could apply with more force to a non-incorrigible juvenile homicide offender than to a juvenile nonhomicide offender. The logic there would be that recidivism by the homicide offender poses an enhanced risk to public safety than does that of a non-homicide offender. However, Graham squarely forecloses this argument: “To justify life without parole on the assumption that the juvenile offender forever will be a danger to society requires the sentencer to make a judgment that the juvenile is incorrigible.” 560 U.S. at 72, 130 S.Ct. 2011. Under Miller, a judge can still make a determination that the juvenile offender is incorrigible and sentence him to LWOP, thereby incapacitating the offender and removing him or her from society. Thus, a juvenile homicide offender who is not incorrigible by definition does not pose a permanent danger to society, making perpetual incapacitation an inappropriate penological justification for the sentence.
10. As a secondary argument, the Government contends that geriatric release under 18 U.S.C. § 3582(c)(1)(A)(ii)—which Grant could be eligible for when he reaches 70—satisfies Graham's requirement of meaningful opportunity for release.It does not. A decision under this provision is entirely discretionary with the Bureau of Prisons and does not assure, subject to judicial review, consideration of youth and attendant circumstances. See 18 U.S.C. § 3582(c)(1)(A) (“[T]he court, upon motion of the Director of the Bureau of Prisons, may reduce the term of imprisonment․” (emphasis added) ).
11. The Ninth Circuit's review of the California Court of Appeal's decision—i.e., that of an intermediate state court—arose from litigation over Moore's federal habeas petition:Moore filed pro se state habeas petitions in the Los Angeles County Superior Court, the California Court of Appeal, and the California Supreme Court, arguing that his sentence was unconstitutional under Graham. The Los Angeles County Superior Court summarily denied Moore's petition. The California Court of Appeal held that Graham does not apply to Moore's sentence. The California Supreme Court summarily denied review.On May 10, 2011, Moore filed a timely federal habeas petition. The district court summarily dismissed Moore's federal petition on the ground that Moore had not exhausted his available state remedies․Moore timely filed a notice of appeal and applied for a certificate of appealability.Moore, 725 F.3d at 1187.
12. Grant's expected age of release accounts for good time credit. See 18 U.S.C. § 3624(b)(1).
13. This same concern—lack of vocational training—also animated the Court to adopt a categorical rule in Graham, rather than a case-by-case approach, in order to “avoid[ ] the perverse consequence in which the lack of maturity that led to an offender's crime is reinforced by the prison term.” Graham, 560 U.S. at 79, 130 S.Ct. 2011.
14. The SSA reported that 1,647,370 of the 2,910,752 Americans who claimed Social Security retirement benefits had their benefits reduced for early retirement. See Soc. SEC. Admin., Annual Statistical Supplement 2017, Table 6.B3, Number and Percentage Distribution With and Without Reduction for Early Retirement, by Sex and Monthly Benefit, 2016 (2017).
15. Lydia Saad, Three in 10 U.S. Workers Foresee Working Past Retirement Age, Gallup News (May 13, 2016), http://news.gallup.com/poll/191477/three-workers-foresee-working-past-retirement-age.aspx; Rebecca Rifkin, Americans Settling on Older Retirement Age, Gallup News (Apr. 29, 2015), http://news.gallup.com/poll/182939/americans-settling-older-retirement-age.aspx; Rebecca Rifkin, Average U.S. Retirement Age Rises to 62, Gallup News (Apr. 28, 2014), http://news.gallup.com/poll/168707/average-retirement-agerises.aspx?g_source=position5&g_medium=related%20&g%20 campaign=tiles.
16. See also Norma B. Coe et. al., Sticky Ages: Why Is Age 65 Still a Retirement Peak? (Ctr. for Ret. Research at Bos. Coll., Working Paper No. 2013-2, 2013) (explaining why retirees commonly elect to retire at age sixty-five even after SSA increased full retirement age to sixty-six); Wojciech Kopczuk & Jae Song, Stylized Facts and Incentive Effects Related to Claiming of Retirement Benefits Based on Social Security Administration Data 14 (Univ. of Mich. Ret. Research Ctr., Working Paper No. 2008-200, 2008) (concluding from SSA administrative data that “retiring around [one's] 65th birthday has intrinsic value to individuals”); id. at 13 (“It is clear that following the increase in the full retirement age, the number of retirements at the 65th birthday remains elevated․”).
17. However, as with all sentences, district courts are required to take into account the “overarching provision” of § 3553(a), which compels them to “ ‘impose a sentence sufficient, but not greater than necessary,’ to accomplish the goals of sentencing.” Kimbrough, 552 U.S. at 101, 128 S.Ct. 558 (quoting 18 U.S.C. § 3553(a) ); see also United States v. Olhovsky, 562 F.3d 530, 552 (3d Cir. 2009) (“The ‘overarching principle’ of parsimony that Congress included in § 3553 directs the courts to impose a sentence ‘sufficient, but not greater than necessary, to comply with the purposes set forth in [§ 3553(a) ].’ ” (quoting 18 U.S.C. § 3553(a) ) ).
18. Nothing in our opinion today disturbs a district court's ability to determine in the first instance that a juvenile is incorrigible and therefore eligible for LWOP under Miller.
19. Because we vacate for the aforementioned reasons, we need not consider Grant's alternative contention that the District Court did not adequately consider his youth and attendant circumstances.
20. Grant argues that de novo review should nonetheless apply because defense counsel repeatedly argued below that all of the sentences across the multiple counts were “all part and parcel of one sentence [of life without parole].” Appellant Reply Br. at 22 (quoting A40). Grant relies on Brennan v. Norton for the proposition that the “crucial question regarding waiver is whether [Grant] presented the argument with sufficient specificity to alert the district court.” 350 F.3d 399, 418 (3d Cir. 2003) (quoting Keenan v. City of Phila., 983 F.2d 459, 471 (3d Cir. 1992) ).The sentencing package doctrine, however, has been applied in our precedential opinions only to vacated convictions—not vacated sentences—because “[w]hen a conviction on one or more of the component counts is vacated, common sense dictates that the judge should be free to review the efficacy of what remains in light of the original plan ․ in order to ensure that the punishment still fits both crime and criminal.” Miller, 594 F.3d at 180 (quoting Davis, 112 F.3d at 122). Grant's request that the District Court resentence him de novo in light of his vacated sentences did not invoke the doctrine because the “crime and criminal” remained unchanged given that his underlying convictions were still intact. He also did not contend to the District Court that the doctrine should be extended to vacated sentences.Grant therefore failed to adequately raise his sentencing package argument before the District Court, and this issue is not preserved. See Doe v. Mercy Catholic Med. Ctr., 850 F.3d 545, 558 (3d Cir. 2017) (“Theories not raised squarely [before the district court] cannot be surfaced for the first time on appeal.”); United States v. Joseph, 730 F.3d 336, 337 (3d Cir. 2013) (“We hold that for parties to preserve an argument for appeal, they must have raised the same argument in the District Court—merely raising an issue that encompasses the appellate argument is not enough.”); United States v. Dupree, 617 F.3d 724, 728 (3d Cir. 2010) (“[V]ague allusion[s] to an issue will not suffice to preserve it for appeal[.]” (first and second alterations added) (quoting In re Ins. Brokerage Antitrust Litig., 579 F.3d 241, 262 (3d Cir. 2009) ) ).
21. Grant incorrectly contends that our precedent already dictates that the sentencing package doctrine applies to instances where a sentence alone is vacated. Two of the three cases that he relies on—United States v. Fumo, 513 Fed.Appx. 215 (3d Cir. 2013), and United States v. Brown, 385 Fed.Appx. 147 (3d Cir. 2010)—are not precedential, and as such, are of no effect. See Internal Operating Procedures of the United States Court of Appeals for the Third Circuit, 5.7 (January 2017) (“The court by tradition does not cite to its not precedential opinions as authority.”). The third case, United States v. Guevremont, 829 F.2d 423 (3d Cir. 1987), does not explicitly invoke the doctrine. Furthermore, we adopted the sentencing package doctrine in 1997, ten years after Guevremont was decided. See Davis, 112 F.3d at 122; see also Miller, 594 F.3d at 180 (“In Davis, we endorsed the ‘sentencing package doctrine’․”).
22. We do, however, exempt Grant's sentence to Count 4 from this holding. As discussed above, supra note 3, we will vacate this sentence for the sole purpose of allowing the District Court to correct its inadvertent sentencing error.
1. The majority asserts that Guevremont did not explicitly invoke the sentencing package doctrine and was decided ten years before this Court adopted the doctrine. However, Guevremont did, at the very least, rely on our language in Busic, which, in turn, “gave rise to what has since been termed the sentencing package doctrine,” United States v. Murray, 144 F.3d 270, 273 (3d Cir. 1998). The majority also recognizes that the Court has applied this doctrine in the context of vacated sentences in two non-precedential decisions. See United States v. Fumo, 513 Fed.Appx. 215, 218-19 (3d Cir. 2013); United States v. Brown, 385 Fed.Appx. 147, 148 (3d Cir. 2010).
GREENAWAY, JR., Circuit Judge.