SHWARTZ, Circuit Judge.

OPINION

Drunk driving is a dangerous and often deadly crime. "Approximately a quarter million people are injured annually in alcohol-related crashes," Begay v. United States, 553 U.S. 137, 156-57, 128 S.Ct. 1581, 170 L.Ed.2d 490 (2008) (Alito, J., dissenting), and the number "who are killed ... by drunk drivers is far greater than the number of murders committed" during many other violent crimes, id. at 157 & n.4, 128 S.Ct. 1581. "[F]rom 1982 to 2016, alcohol-related accidents took roughly 10,000 to 20,000 lives in this Nation every single year. In the best years, that would add up to more than one fatality per hour." Mitchell v. Wisconsin, ––– U.S. ––––, 139 S. Ct. 2525, 2536, 204 L.Ed.2d 1040 (2019) (emphasis omitted) (citations omitted).

Today, we consider whether Pennsylvania’s driving under the influence ("DUI") law, which makes a DUI at the highest blood alcohol content ("BAC") a first-degree misdemeanor that carries a maximum penalty of five years’ imprisonment, see 18 Pa. Cons. Stat. Ann. § 1104 ; 75 Pa. Cons. Stat. Ann. §§ 3802(c), 3803(b)(4), constitutes a serious crime that requires disarmament. Plaintiff Raymond Holloway, Jr., was convicted under this statute, and by the terms of 18 U.S.C. § 922(g)(1), he is prohibited from possessing a firearm. Holloway claims this prohibition violates his Second Amendment rights. The District Court agreed and enjoined applying § 922(g)(1) to him. Because Holloway was convicted of a serious crime as contemplated by Binderup v. Attorney General United States of America, 836 F.3d 336 (3d Cir. 2016) (en banc), applying § 922(g)(1) to him does not offend the Second Amendment. Therefore, we will reverse the District Court’s order and remand for the entry of judgment in favor of the Government.

I

In 2002, Holloway was convicted of a DUI at the highest BAC, but the charge was dismissed upon his completion of an accelerated rehabilitation program. In 2005, Holloway was again arrested for driving under the influence and registered a BAC of 0.192%. Holloway pled guilty to violating 75 Pa. Cons. Stat. Ann. § 3802(c) for driving under the influence at the highest BAC (greater than 0.16%). He received a sentence of 60 months’ "Intermediate Punishment," including 90-days’ imprisonment that allowed him work release, a $1,500 fine, and mandatory drug and alcohol evaluation.

In 2016, Holloway sought to purchase a firearm but was unable to do so because of his disqualifying DUI conviction. Holloway sued the Attorney General of the United States and other federal officials (the "Government") in the United States District Court for the Middle District of Pennsylvania, claiming that § 922(g)(1) is unconstitutional as applied to him and seeking declaratory and permanent injunctive relief. The parties filed cross-motions for summary judgment.

The District Court granted Holloway’s motion for summary judgment, awarded him a declaratory judgment, and entered a permanent injunction barring the Government from enforcing § 922(g)(1) against him. Holloway v. Sessions, 349 F. Supp. 3d 451, 463 (M.D. Pa. 2018). Applying Binderup, the Court held that § 922(g)(1) is unconstitutional as applied to Holloway because (1) Holloway’s DUI offense was a non-serious crime that has not historically been a basis for the denial of Second Amendment rights, id. at 459-60, and (2) the Government failed to demonstrate that disarmament of individuals like Holloway would promote the public safety, particularly given his decade of crime-free behavior, id. at 460-62. The Government appeals.

II

A

The sole issue on appeal is whether applying 18 U.S.C. § 922(g)(1) to Holloway, which makes it unlawful for him to possess a firearm due to his prior conviction, violates his Second Amendment rights.

In District of Columbia v. Heller, the Supreme Court held that the Second Amendment protects the right of "law-abiding, responsible citizens to use arms in defense of hearth and home." 554 U.S. 570, 635, 128 S.Ct. 2783, 171 L.Ed.2d 637 (2008). This right, however, "is not unlimited." Id. at 626, 128 S.Ct. 2783. Indeed, the Court cautioned that "nothing in [its] opinion should be taken to cast doubt on longstanding prohibitions on the possession of firearms by felons." Id. The Court described the felon ban as just one "example[ ]" of "presumptively lawful regulatory measures." Id. at 627 n.26, 128 S.Ct. 2783.

Since Heller, we have been called upon to determine whether various laws unlawfully infringe the Second Amendment. Some of these laws regulate who can possess firearms, see, e.g., Beers v. Att’y Gen. U.S., 927 F.3d 150, 155-56 (3d Cir. 2019) (ban on possession by those adjudicated mentally defective or committed to mental institution); Binderup, 836 F.3d 336 (ban on possession by certain convicts). Other laws regulate the type of firearms that may be possessed. See, e.g., Ass’n of N.J. Rifle & Pistol Clubs, Inc. v. Att’y Gen. N.J. (N.J. Rifle ), 910 F.3d 106 (3d Cir. 2018) (large capacity magazines). In each instance, we examined the challenged law by applying the two-part test first articulated in United States v. Marzzarella, 614 F.3d 85 (3d Cir. 2010). Under that test, we first "ask whether the challenged law imposes a burden on conduct falling within the scope of the Second Amendment’s guarantee." Id. at 89. "If it does not, our inquiry is complete." Id. If it does, we move to the second step: we evaluate the law under some form of heightened scrutiny. See id. at 96-97.

After Marzzarella, we addressed a constitutional challenge to § 922(g)(1) in United States v. Barton, 633 F.3d 168 (3d Cir. 2011). Barton recognized that § 922(g)(1) was one of the "presumptively lawful" measures referenced in Heller, id. at 172, 128 S.Ct. 2783, but that individuals could challenge § 922(g)(1) on an as-applied basis, id. at 173. Barton, however, did not expressly apply Marzzarella’s two-step framework. Id. Rather, Barton held that a challenger could rebut the presumption that § 922(g)(1) constitutionally applied to him by "present[ing] facts about himself and his background that distinguish his circumstances from those of persons historically barred from Second Amendment protections." Id. at 174. The "historically barred" class, Barton concluded, was individuals "likely to commit violent offenses." Id. at 173-74. Thus, Barton held that if an individual could show that he posed no threat of future violence, then § 922(g)(1) could not constitutionally apply to him. Id. at 174. We revisited Barton and as-applied challenges to § 922(g)(1) as an en banc Court in Binderup. Binderup resulted in several opinions from fifteen judges: (1) an opinion by Judge Ambro, joined in full by two judges and joined additionally in part by four other judges; (2) an opinion by Judge Hardiman, joined in full by four judges, and which concurred in part with Judge Ambro and concurred in the judgment; and (3) an opinion by Judge Fuentes, joined by six judges (some of whom joined parts of Judge Ambro’s opinion), which concurred in part, dissented in part, and dissented from the judgment.

There are no specific rules for how to identify the holdings and legal standards from split circuit opinions. We can, however, look to the rules we use to identify such standards in fractured Supreme Court opinions, as set forth in Marks v. United States, 430 U.S. 188, 97 S.Ct. 990, 51 L.Ed.2d 260 (1977), and its progeny. We need not conduct an explicit Marks analysis of the Binderup opinions here because we already recited its holdings, as expressed by Judge Ambro’s controlling opinion, in Beers, 927 F.3d at 155-56 ; see also N.J. Rifle, 910 F.3d at 130 (Bibas, J., dissenting) (describing Judge Ambro’s Binderup opinion as the "controlling opinion"), and it binds us. Mateo v. Att’y Gen. U.S., 870 F.3d 228, 231 n.6 (3d Cir. 2017) (prior panel’s precedential opinion "binding on subsequent panels"); see also Jackson v. Danberg, 656 F.3d 157, 165 n.10 (3d Cir. 2011) (applying a legal standard derived from a previous panel opinion’s Marks analysis as the law of our Circuit).

Nevertheless, both Beers and Marks reveal the following relevant Binderup holdings agreed to by a majority of judges:

(1) Marzzarella’s two-step test—and not the test articulated in Barton—governs Second Amendment challenges, 836 F.3d at 346-47 (Ambro, J.); id. at 387 (Fuentes, J.);

(2) At Marzzarella step one for challenges to § 922(g)(1), we determine whether an individual has committed a "serious" offense, and thus was an "unvirtuous citizen[ ]" who was historically barred from possessing firearms and fell out of the Second Amendment’s scope, id. at 348-49 (Ambro, J.); id. at 387 (Fuentes, J.);

(3) Barton’s focus on whether the challenger’s crime was violent or whether the challenger poses a threat of violence is overruled, id. at 348-49 (Ambro, J.); id. at 387 n.72 (Fuentes, J.);

(4) a challenger, otherwise barred from possession by § 922(g)(1), can make a factual showing that he falls outside of the historically barred class, id. at 347 & n.3, 349 (Ambro, J.); id. at 365-67 (Hardiman, J.);

(5) intermediate scrutiny applies at Marzzarella step two, id. at 353 (Ambro, J.); id. at 396-97 (Fuentes, J.). Thus, as we said in Beers, 927 F.3d at 155, Binderup held that "the two-step Marzzarella framework controls all Second Amendment challenges, including as-applied challenges to § 922(g)(1)," 836 F.3d at 356 (Ambro, J.). At step one, the challenger must "identify the traditional justifications for excluding from Second Amendment protections the class of which he appears to be a member[.]" Id. at 347. When the class at issue is historically excluded convicts, as here and in Binderup, the challenger must show that he was not previously convicted of a serious crime. Id. at 350. A crime is "serious" based on circumstances related to the offense, id. at 350-53, and so evidence of a challenger’s rehabilitation or his likelihood of recidivism is not relevant, id. at 349-50. There are no fixed rules for determining whether an offense is serious but various factors may be informative including, but not limited to, whether the crime poses a danger or risk of harm to self or others, whether the crime involves violence or threatened violence, the classification of the offense, the maximum penalty, the penalty imposed, and how other jurisdictions view the crimes. See id. at 351-52. If a challenger makes a "strong" showing that the regulation burdens his Second Amendment rights and that he has not committed a "serious" crime, and thus is different from those historically barred from possessing firearms, then "the burden shifts to the Government to demonstrate that the regulation satisfies" intermediate scrutiny. Id. at 347.

We apply this framework to determine whether § 922(g)(1) as applied to Holloway violates his Second Amendment rights.

B

At the first step of the analysis, we must determine whether the application of § 922(g)(1) burdens Holloway’s Second Amendment rights by considering the traditional justifications for denying certain criminals Second Amendment rights and then examining whether Holloway’s offense is disqualifying. We "presume the judgment of the legislature is correct and treat any crime subject to § 922(g)(1) as disqualifying unless there is a strong reason to do otherwise." Id. at 351.

1

As previously stated, Heller embraced the "longstanding prohibitions on the possession of firearms by felons." 554 U.S. at 626, 128 S.Ct. 2783. Because Holloway’s DUI misdemeanor conviction carries a maximum penalty of five years’ imprisonment, it is deemed a disqualifying felony under § 922(g)(1). Thus, the application of § 922(g)(1) is presumptively lawful. See Binderup, 836 F.3d at 348 (Ambro, J.).

2

We next examine whether Holloway’s crime was nonetheless "not serious enough to strip [him] of [his] Second Amendment rights." Id. at 351. Under Binderup, "a person who did not commit a serious crime retains his Second Amendment rights," because "a non-serious crime does not demonstrate a lack of ‘virtue’ that disqualifies an offender from exercising those rights." Id. at 349.

A crime that presents a potential for danger and risk of harm to self and others is "serious." See "Serious," Black’s Law Dictionary (11th ed. 2019) (defining "serious" as, among other things, "dangerous; potentially resulting in death or other severe consequences"). "There is no question that drunk driving is a serious and potentially deadly crime .... The imminence of the danger posed by drunk drivers exceeds that at issue in other types of cases." Virginia v. Harris, 558 U.S. 978, 979-80, 130 S.Ct. 10, 175 L.Ed.2d 322 (2009) (Mem.) (Roberts, C.J., dissenting from denial of writ of certiorari); see Mitchell, 139 S. Ct. at 2541 (Sotomayor, J., dissenting) ("[D]runk driving poses significant dangers that [states] must be able to curb."); Begay, 553 U.S. at 141, 128 S.Ct. 1581 ("Drunk driving is an extremely dangerous crime.").

All three branches of the federal government have recognized as much. The Supreme Court has described individuals "who drive with a BAC significantly above the ... limit of 0.08% and recidivists" as "the most dangerous offenders." Birchfield v. North Dakota, ––– U.S. ––––, 136 S. Ct. 2160, 2179, 195 L.Ed.2d 560 (2016). Congress and the Executive Branch have also recognized the dangers posed by drunk driving. Congress requires states to implement highway safety programs "to reduce injuries and deaths resulting from persons driving motor vehicles while impaired by alcohol." 23 U.S.C. § 402(a)(2)(A)(iii). The Secretary of Transportation conditions the receipt of certain highway-related funds on states’ implementation of programs with impaired driving countermeasures that will "effective[ly]" "reduce driving under the influence of alcohol." § 405(a)(3), (d). Thus, all branches of the federal government agree that DUIs are dangerous, and those who present a danger may be disarmed.

While use or the threatened use of violence is not an element of a DUI offense, see 75 Pa. Cons. Stat. Ann. § 3802(c) (providing "[a]n individual may not drive, operate or be in actual physical control of the movement of a vehicle after imbibing a sufficient amount of alcohol such that the alcohol concentration in the individual’s blood or breath is 0.16% or higher"), a showing of violence is not necessary for a crime to be deemed serious, see, e.g., Binderup, 836 F.3d at 348 (Ambro, J.); id. at 390-91 (Fuentes, J.); Medina v. Whitaker, 913 F.3d 152, 160 (D.C. Cir. 2019) (holding that fraud, by lying on mortgage documents, is "a serious crime"). Thus, the fact that an offense does not include the use or threatened use of violence does not mean it is not serious.

Moreover, though labeled as a first-degree misdemeanor, Holloway’s DUI crime carries a three-month mandatory minimum prison term and a five-year maximum prison term. See 18 Pa. Cons. Stat. Ann. § 1104 ; 75 Pa. Cons. Stat. Ann. § 3803(b)(4) ; 75 Pa. Cons. Stat. Ann. § 3804(c)(2). While "generally the misdemeanor label ... in the Second Amendment context, is ... important" and is a "powerful expression" of the state legislature’s view, it is not dispositive. Binderup, 836 F.3d at 352. First, not only is the distinction "minor and often arbitrary," Tennessee v. Garner, 471 U.S. 1, 14, 105 S.Ct. 1694, 85 L.Ed.2d 1 (1985) ; see also Burgess v. United States, 553 U.S. 124, 132, 128 S.Ct. 1572, 170 L.Ed.2d 478 (2008), some states do not use the distinction at all, see, e.g., N.J. Stat. Ann. § 2C:1-4 (dividing offenses into "crimes," "disorderly persons offenses," and "petty disorderly persons offenses"); § 2C:43-1(a) (dividing "crimes" further into four degrees); State v. Doyle, 42 N.J. 334, 200 A.2d 606, 613 (1964) ("Criminal codes in New Jersey have not utilized the felony-misdemeanor nomenclature or classification of the English common law."). Second, "numerous misdemeanors involve conduct more dangerous than many felonies." Garner, 471 U.S. at 14, 105 S.Ct. 1694. Indeed, giving dispositive weight to the felony/misdemeanor nomenclature for determining whether an offense is serious would mean that the following offenses, labeled under Pennsylvania law as misdemeanors and carrying a five-year maximum penalty (the maximum Holloway faced), 18 Pa. Cons. Stat. Ann. § 1104(1), would not qualify as serious crimes: involuntary manslaughter, § 2504(b), terrorism, § 2717(b)(1), assaulting a child, § 2701(b)(2), abusing a care-dependent person, § 2713.1(b)(1), making terroristic threats, § 2706(d), threatening to use weapons of mass destruction, § 2715(b)(1), shooting a fire bomb into public transportation, § 2707(a), indecent assault by forcible compulsion, § 3126(a)(2), concealing the murder of a child, § 4303(a), luring a child into a motor vehicle or structure, § 2910(a), restraining a person "in circumstances exposing him to risk of serious bodily injury," § 2902(a)(1), and stalking, § 2709.1(c)(1). At bottom, Heller emphasized that the Second Amendment right belongs to "law-abiding, responsible citizens," 554 U.S. at 635, 128 S.Ct. 2783, and whether labeled a felon or misdemeanant, those who commit serious crimes are not "the kinds of ‘law-abiding’ citizens whose rights Heller vindicated," Binderup, 836 F.3d at 392 (Fuentes, J.).

Furthermore, the maximum penalty that may be imposed often reveals how the legislature views an offense. Put succinctly, "the maximum possible punishment is certainly probative of a misdemeanor’s seriousness." Id. at 352 (Ambro, J.). "[T]he category of serious crimes changes over time as legislative judgments regarding virtue evolve," id. at 351, and here, the Pennsylvania legislature has demonstrated an evolution in judgment. Pennsylvania’s DUI laws were amended in 2003 when state legislators observed that "[t]oo many people have been injured and killed on our highways," H.R. Legis. Journal, 187th Gen. Assemb., Reg. Sess. 1443 (Pa. 2003) (statement of Rep. Turzai), and unlike in other states, which saw an eleven percent decrease in deaths caused by drunk drivers, such deaths "continue to rise" in Pennsylvania with a five percent increase, H.R. Legis. Journal, 187th Gen. Assemb., Reg. Sess. 1444 (Pa. 2003) (statement of Rep. Harper); S. Legis. Journal, 187th Gen. Assemb., Reg. Sess. 981 (Pa. 2003) (statement of Sen. Williams). At the time of the amendment, thirteen individuals were killed every two weeks in Pennsylvania from alcohol-related accidents. H.R. Legis. Journal, 187th Gen. Assemb., Reg. Sess. 1445 (Pa. 2003) (statement of Rep. Harper). "[M]ore than half of all fatal alcohol-related accidents [were] caused by hardcore drunken drivers, those people whose BACs are .16 or above," H.R. Legis. Journal, 187th Gen. Assemb., Reg. Sess. 1444 (Pa. 2003) (statement of Rep. Harper), and "one-third of drunk driving arrests involve[d] repeat offenders," S. Legis. Journal, 187th Gen. Assemb., Reg. Sess. 981 (Pa. 2003) (statement of Sen. Williams). To address this "very serious matter," H.R. Legis. Journal, 187th Gen. Assemb., Reg. Sess.1445 (Pa. 2003) (statement of Rep. Harper), the legislature "provid[ed] for tough civil and criminal penalties together with mandatory treatment," H.R. Legis. Journal, 187th Gen. Assemb., Reg. Sess. 1443 (Pa. 2003) (statement of Rep. Turzai), to "mak[e] it clear that if you are under the influence of alcohol or drugs and behind the wheel in Pennsylvania, you will be punished," H.R. Legis. Journal, 187th Gen. Assemb., Reg. Sess. 1445 (Pa. 2003) (statement of Rep. Harper). Therefore, despite the misdemeanor label, Pennsylvania’s decision to impose a mandatory minimum jail term and a maximum penalty of up to five years’ imprisonment for a second DUI at the highest BAC reflects the seriousness of the offense.

Holloway received the statutory minimum sentence of 90 days’ imprisonment, 75 Pa. Cons. Stat. Ann. § 3804(c)(2), and although he was permitted to work, he received a custodial sentence unlike either of the challengers in Binderup. 836 F.3d at 352 ("With not a single day of jail time, the punishments here reflect the sentencing judges’ assessment of how minor the violations were."). The legislature’s mandate that repeat DUI offenders receive at least three months in jail reflects its judgment that such offenses are serious.

Pennsylvania is not alone in its decision to severely punish repeat DUI offenders. Mitchell, 139 S. Ct. at 2536 ("[M]any States ... have passed laws imposing increased penalties for recidivists or for drivers with a BAC level that exceeds a higher threshold." (citations omitted)). Although most states do not impose penalties for second DUI offenses that subject an offender to disarmament under § 922(g)(1), three states impose penalties that subject misdemeanants who commit a second DUI at a higher BAC to § 922(g)(1) disarmament. Moreover, several states grade a second DUI offense as a felony, thus triggering disarmament. The absence of a cross-jurisdictional consensus regarding the punishment for such conduct does not mean the conduct is not serious. Indeed, states unanimously agree that DUIs are crimes subject to punishment.

Holloway suggests that his crimes cannot be so serious to justify federal disarmament and that to apply § 922(g)(1) to him would be overinclusive because Pennsylvania law only disarms DUI offenders at their third offense and permits them to apply for relief after ten years. This argument ignores the gradations in Pennsylvania’s DUI laws. In fact, Pennsylvania’s prohibition may be broader than § 922(g)(1) because it applies to all DUIs under 75 Pa. Cons. Stat. Ann. § 3802, regardless of punishment. For example, an individual who commits a third DUI, none at the high or highest BAC, within a five-year period, is convicted of a second-degree misdemeanor under 75 Pa. Cons. Stat. Ann. § 3803(a)(2) and subject to up to two years’ imprisonment under 18 Pa. Cons. Stat. Ann. § 1104(2). This individual’s third DUI triggers Pennsylvania’s disarmament statute under 18 Pa. Cons. Stat. Ann. § 6105(c), but does not trigger § 922(g)(1) because it falls within § 921(a)(20)(B) ’s exception for state misdemeanors subject to a term of imprisonment of two years or less. Holloway’s second DUI, however, subjects him to the federal provision but not the state provision because his offense was at the highest BAC, which enhanced the grading of his offense to a first-degree misdemeanor and exposed him to five years’ imprisonment. Thus, Pennsylvania’s disarmament statute captures offenders who may not face § 922(g)(1) ’s bar and shows that Pennsylvania meant to disarm a broader swath of offenders than § 922(g)(1).

Together, these considerations demonstrate that Holloway’s DUI conviction constitutes a serious crime, placing him within the class of "persons historically excluded from Second Amendment protections." Binderup, 836 F.3d at 347. Because Holloway has not met his burden at the first step of the analysis to overcome the presumptive application of § 922(g)(1), § 922(g)(1) is constitutional as applied to him, and he is not entitled to relief. III

For the foregoing reasons, we will reverse the order granting Holloway summary judgment, a declaratory judgment, and an injunction and remand for the entry of judgment in favor of the Government.

FISHER, Circuit Judge, dissenting.

Driving under the influence of alcohol is undoubtedly a significant offense deserving of punishment. Yet the principal question in this case is not whether that offense is "serious" in the abstract or even as a matter of ordinary understanding. "Seriousness" here has a discrete legal meaning—that a conviction of the crime deprives in perpetuity an individual of an enumerated constitutional right. Under our precedent, these two categories are distinct, and they must be treated as such. Just because this question arises under the Second Amendment does not make our decision any less weighty. If the circumstances were different, we would assuredly consider very carefully the legal standard for depriving an individual of his right to free speech. The majority incorrectly, in my view, holds that Holloway has not carried his burden at Step One of the two-step framework established in United States v. Marzzarella , 614 F.3d 85 (3d Cir. 2010). Further, because I conclude that at Step Two, 18 U.S.C. § 922(g)(1) as applied here does not survive intermediate scrutiny, I must respectfully dissent.

I

Under the Marzzarella framework, we first determine "whether the challenged law imposes a burden on conduct falling within the scope of the Second Amendment’s guarantee." 614 F.3d at 89. In particular, our precedent requires the challenger to satisfy the two elements articulated in United States v. Barton , 633 F.3d 168 (3d Cir. 2011). He must "identify the traditional justifications for excluding from Second Amendment protections the class of which he appears to be a member," and then "present facts about himself and his background that distinguish his circumstances from those of persons in the historically barred class." Binderup v. Attorney Gen. United States of America , 836 F.3d 336, 347 (3d Cir. 2016) (en banc) (plurality opinion) (citing Barton , 633 F.3d at 173-74 ); see id. at 366 (Hardiman, J., concurring in part and concurring in the judgments); see also Beers v. Attorney Gen. United States of America , 927 F.3d 150, 157 (3d Cir. 2019) (adopting this test for an as-applied challenge to 18 U.S.C. § 922(g)(4) ).

In Binderup , ten judges on the fifteen-member en banc court agreed that, in the context of as-applied challenges to 18 U.S.C. § 922(g)(1), the "historically barred class" is those who are "unvirtuous" because they have "committed a serious criminal offense, violent or nonviolent." 836 F.3d at 348 (plurality opinion); see id. at 387 (Fuentes, J., concurring in part, dissenting in part, and dissenting from the judgments); see also United States v. Donovan , 661 F.3d 174, 182 (3d Cir. 2011) (noting that when this Circuit confronts a fractured decision, we "look[ ] to the votes of dissenting [judges] if they, combined with votes from plurality or concurring opinions, establish a majority view on the relevant issue"). "Seriousness"—and by extension "unvirtuousness"—therefore has no independent legal significance. It is a gloss on the first part of the Barton test—a way of describing the offenses committed by those historically barred from possessing a firearm. The principal question before us today concerns the application of Barton ’s second prong in the § 922(g)(1) context—that is, how to evaluate whether a challenger’s crime is sufficiently similar to crimes of the historically barred class such that he is not entitled to Second Amendment protection. The Binderup Court divided on this issue, and, for the reasons detailed below, it remains an open question whether the multifactor test used in Binderup is binding precedent in our Circuit—despite the lower courts’ application of it as such. See, e.g. , Williams v. Barr , 379 F. Supp. 3d 360, 370-74 (E.D. Pa. 2019) ; Holloway v. Sessions , 349 F. Supp. 3d 451, 457-60 (M.D. Pa. 2018). Nevertheless, for reasons I also state, the test is an appropriate means under our precedent of determining whether a challenger’s crime is "serious" for purposes of Marzzarella Step One.

It is on this latter point—the application of the multifactor test—that I break with my colleagues in the majority. They interpret the test’s list of factors to be non-exhaustive, Majority Op. at II.A, and so they supplement their analysis of the factors with additional considerations. The majority appears to concede that at least three of the four Binderup factors are in Holloway’s favor, but still concludes that Holloway is not entitled to Second Amendment protection. Although I agree that we are not bound to consider the four factors exclusively, I disagree with my colleagues in how they have applied and supplemented those factors. Simply because our precedent does not require us to apply the four factors alone does not mean the determination of "seriousness" is open to any legal content. Our precedent does require us to follow the doctrinal structure established in Barton and adopted in Binderup . The "seriousness" inquiry is a comparison of the challenger’s circumstances with those of the historically barred class. The majority’s analysis, in my view, diverges too far from this requirement.

A

As it was applied in Binderup , the multifactor test contains four factors for determining whether an individual’s crime is sufficiently "serious" to deprive him of his Second Amendment right. First, the court considers whether the state classifies the challenger’s disqualifying crime under § 922(g)(1) as a felony or a misdemeanor. 836 F.3d at 351 (opinion of Ambro, J.). Second, it determines whether the challenger’s crime "had the use or attempted use of force as an element." Id. at 352. Third, also relevant is the sentence the challenger in fact received. Although the maximum possible sentence determines whether the crime triggers the § 922(g)(1) bar, the crime’s "seriousness" for purposes of Second Amendment analysis turns, in part, on the challenger’s actual punishment. Finally, the court considers whether there exists a "cross-jurisdictional consensus regarding the seriousness of the [challenger’s] crimes." Id. Although this multifactor test garnered the support of only three judges, it was declared "the law of our Circuit" under the Supreme Court’s Marks rule. Id. at 356. My review of our case law leads me to question this conclusion. Courts and legal scholars disagree as to the nature of the Marks rule and how it is to be applied. In particular, there are multiple possible versions of the rule, and the Supreme Court’s most recent statement on the matter acknowledged but declined to resolve this debate. See Hughes v. United States , ––– U.S. ––––, 138 S. Ct. 1765, 1771-72, 201 L.Ed.2d 72 (2018). On my assessment, the multifactor test would be Circuit precedent under only one of these versions, and our Court has not adopted this interpretation of the Marks rule above the others. As a result, despite the declaration in Binderup to the contrary, I do not think Marks requires us to treat the multifactor test as controlling authority.

B

Nevertheless, like the District Court, I believe that the multifactor test should guide the Step One analysis in this case. On my reading, the four factors reflect an underlying logic that is consistent with our precedent in Barton and Binderup . Those cases require us to assess the relation between the challenger’s "circumstances [and] those of persons historically barred from Second Amendment protections." Barton , 633 F.3d at 174 ; see also Binderup , 836 F.3d at 346-47 (plurality opinion); id. at 366 (Hardiman, J., concurring in part and concurring in the judgments). This comparative exercise demands certain measures of "seriousness," and those measures should naturally be the features—the classification, elements, and punishments—common to the crimes that traditionally have qualified the individuals convicted of them for firearm dispossession. These crimes include felonies, crimes of violence, and (as Binderup held) some nonviolent misdemeanors. Further, because neither courts nor scholars have agreed on the precise contours of this category—and in particular how "longstanding" a regulation must be for its violators to be considered part of the historically barred class, see, e.g. , United States v. Skoien , 614 F.3d 638, 640-41 (7th Cir. 2010) (en banc); C. Kevin Marshall, Why Can’t Martha Stewart Have a Gun? , 32 HARV . J.L. & PUB. POL’Y 695 (2009) —the multifactor test has the virtue of permitting a number of different measures of "seriousness" without making any one factor dispositive.

A methodical evaluation of each factor, consistent with this logic, compels the conclusion reached by the District Court: that Holloway’s conduct has not removed him from the scope of Second Amendment protection. In conducting this analysis, I shall also address the majority’s additional considerations—the "potential for danger and risk of harm" posed by the challenger’s crime, Majority Op. at II.B.2, and the maximum level of punishment Pennsylvania imposes for Holloway’s second DUI offense, id. While, as noted, I do not dispute that the majority may supplement the four factors, any such additions must be—as the four factors are—consistent with the comparative exercise required by Barton and Binderup .

1

The first factor asks whether the challenger’s crime is a felony or a misdemeanor. The majority acknowledges that Pennsylvania classifies Holloway’s second DUI offense as a misdemeanor, but it points out that the offense "carries ... a five-year maximum prison term." Majority Op. at II.B.2. Yet, under our precedent, the potential prison term cannot nullify the relevance of the felony/misdemeanor distinction for determining whether a crime is "serious" enough to deprive an individual of his Second Amendment right. A common feature of the crimes that traditionally have barred an individual from owning a firearm is that they are classified as felonies.

For example, in Heller , the Supreme Court warned specifically that its opinion should not be read to question "longstanding prohibitions on the possession of firearms by felons ." District of Columbia v. Heller , 554 U.S. 570, 626, 128 S.Ct. 2783, 171 L.Ed.2d 637 (2008) (emphasis added); see also McDonald v. City of Chicago , 561 U.S. 742, 786, 130 S.Ct. 3020, 177 L.Ed.2d 894 (2010) (plurality opinion). Congress itself recognized the relevance of the distinction when it excluded from § 922(g)(1) ’s reach misdemeanors punishable by imprisonment of two years or less. See 18 U.S.C. § 921(a)(20)(B). If, as the majority suggests, the maximum length of the sentence rather than the classification of the crime is what really matters, then Congress would never have made an exception for misdemeanors alone. It would either have amended § 922(g)(1) to cover all crimes punishable by more than two years’ incarceration or never added § 921(a)(20)(B) in the first place.

Further, the classification of a crime as a felony has profound implications for whether a person may possess a firearm under state law. On my assessment, thirty-two out of fifty-one jurisdictions (the fifty states and the District of Columbia) disarm individuals because of a felony conviction. That is, they bar for at least some time the possession of a firearm precisely because the person was convicted of a crime labeled a felony. The distinction therefore matters for defining the historically barred class, regardless of jurisdictional diversity in the sentence ranges for various crimes.

As noted, in evaluating the relevance of the felony/misdemeanor distinction, the majority lends great weight to the maximum punishment that Pennsylvania imposes for Holloway’s offense. See Majority Op. at II.B.2. However, a majority of the en banc Court in Binderup rejected the significance of that consideration. As Judge Ambro noted there, prohibitions on the possession of firearms by criminals are only "presumptively lawful." Binderup , 836 F.3d at 350 (opinion of Ambro, J.) (emphasis added) (citing Heller , 554 U.S. at 626-27 & n.26, 128 S.Ct. 2783 ), and in the absence of an explicit declaration to the contrary, all presumptions are rebuttable. To hold otherwise would constitute "an end-run around the Second Amendment," in effect subjecting such prohibitions to rational-basis review rather than the heightened scrutiny demanded when a constitutional right is at stake. Id. at 351-52. As a result, the maximum possible sentence for Holloway’s crime, although a valid consideration, cannot detract from the relevance of a factor that is consistent with our precedent in Barton and Binderup .

In saying this, I do not question the Pennsylvania legislature’s judgment that an offense such as Holloway’s should be punishable by a lengthy prison term. But for the purposes of answering the question before us today—whether that offense is "serious" enough to deprive Holloway of his Second Amendment right—we must look to how his offense compares with those of the historically barred class. That involves giving weight to the felony/misdemeanor distinction. In addition to the sentence it permitted, the Pennsylvania legislature also chose to punish Holloway’s crime as a misdemeanor. Indeed, the sentence and the classification are inseparable—all such misdemeanors in Pennsylvania carry Holloway’s maximum possible prison term. See 18 PA. CONS. STAT. § 1104(1) (2019). Even as a simple matter of statutory interpretation, then, the classification of the crime matters. This factor therefore weighs in Holloway’s favor.

2

The second factor asks whether the "[c]hallenger’s offense had the use or attempted use of force as an element." Binderup , 836 F.3d at 352 (opinion of Ambro, J.). The majority concedes that Holloway’s DUI offense does not fulfill this criterion, see Majority Op. at II.B.2, but it supplements its analysis by considering the crime’s "potential for danger and risk of harm to self and others," id. Although the Marks rule does not foreclose additions to the multifactor test by a panel majority, our precedent demands that the "seriousness" inquiry be a comparative exercise involving the challenger’s offense and the characteristic features of those crimes that traditionally have disqualified persons from owning firearms. The virtue of the second Binderup factor is that it crystallizes in a clear legal standard the evident historical concern with force and violence. By contrast, the relevant historical and contemporary authorities do not support a standard focusing on all conduct that poses a "potential for danger and risk of harm to self and others." Id.

The most prominent late eighteenth-century sources supporting legislative power to bar certain individuals from owning firearms are the proposals made in the ratifying conventions of Pennsylvania, New Hampshire, and Massachusetts. The first of these provides that "no law shall be passed for disarming the people or any of them, unless for crimes committed, or real danger of public injury from individuals." THE ADDRESS AND REASONS OF DISSENT OF THE MINORITY OF THE CONVENTION, OF THE STATE OF PENNSYLVANIA, TO THEIR CONSTITUENTS 1 (Phila., E. Oswald 1787), https://www.loc.gov/item/90898134. It is important to note that the two categories are interlocking—the provision captures both convicted criminals and those non-criminals who pose a "real danger of public injury." Id. The inclusion of the latter phrase in turn suggests that the drafters did not necessarily have in mind all crimes, but rather those that manifest a real danger to the public. To this extent, I agree with the majority’s reading of the text. See Majority Op. at II.B.2 n.11.

Yet the provision alone does not tell us what "real danger of public injury" means. Perhaps the best way of interpreting this historical term is to look to the dispossessory provisions proposed at the other two conventions. In voting to ratify the Federal Constitution, New Hampshire’s delegates also recommended certain amendments to it. Among these was a provision that "Congress shall never disarm any citizen, unless such as are or have been in actual rebellion." 1 JONATHAN ELLIOT, THE DEBATES IN THE SEVERAL STATE CONVENTIONS ON THE ADOPTION OF THE FEDERAL CONSTITUTION 326 (2d ed. 1836). Although the Pennsylvania minority’s "real danger of public injury" was likely meant to sweep more broadly than New Hampshire’s "in actual rebellion," insofar as we are attempting to discover the limitations the ratifying public would have implicitly placed on the Second Amendment, the New Hampshire provision suggests a concern with armed conflict or violence against the government, rather than with all dangerous acts. In this context, it is noteworthy that the Pennsylvania minority speaks of the danger of public, rather than private, injury—a distinction it explicitly makes elsewhere in the document. See, e.g. , ADDRESS AND REASONS OF DISSENT , at 3 ("The absolute unqualified command that congress have over the militia may be made instrumental to the destruction of all liberty, both public and private ...."). From this perspective, it appears the Pennsylvania antifederalists had in mind something narrower than the majority’s standard of "risk of harm to self and others." This understanding is also found in Samuel Adams’s proposal to the Massachusetts ratifying convention. The Constitution, he suggested, should never be "construed to authorize Congress ... to prevent the people of the United States who are peaceable citizens from keeping their own arms." 3 WILLIAM V. WELLS, THE LIFE AND PUBLIC SERVICES OF SAMUEL ADAMS 267 (Bos., Little, Brown & Co. 1865). What Adams meant by "peaceable" can be determined from the rest of his proposal. He also thought the Constitution should not be construed "to prevent the people from petitioning, in a peaceable and orderly manner, the Federal Legislature for a redress of grievances." Id. The right to keep arms was linked to the assembly and petitioning right not only in Adams’s proposal but also in the Bill of Rights itself. To many late-eighteenth-century Americans, the arms right in the Second Amendment helped to ensure that the liberties guaranteed in the First Amendment would not be eroded by a tyrannical central government. See AKHIL REED AMAR, THE BILL OF RIGHTS: CREATION AND RECONSTRUCTION 47-48 (1998). Thus, in both Adams’s proposal and the Bill of Rights, it is "the people" who are given the right to petition their government and to possess arms. That entity, of course, is the one that (as the Preamble declares) alone has the power to form the government, and concomitantly to alter or abolish it. In this context, "peaceable" refers to those individuals who remain a part of "the people," and do not independently disturb or take up arms against its legitimate government. Only "the people" itself has that ability.

In sum, the principal historical evidence from the Founding period suggests that the majority’s "risk of harm" standard is too broad to serve as a basis for comparison under our precedent. The correct standard appears to be something closer to the one used in Binderup , focusing on the presence of force or violence in the challenger’s conduct. Notably, in a part of Barton that remains good law, our Court summarized the ratifying convention proposals as "confirm[ing] that the common law right to keep and bear arms did not extend to those who were likely to commit violent offenses." 633 F.3d at 173 (emphasis added); see also Kanter v. Barr , 919 F.3d 437, 456 (7th Cir. 2019) (Barrett, J., dissenting) (concluding that "[t]he concern common to all three" proposals is "threatened violence and the risk of public injury").

Further, although the majority cites contemporary authorities to support its standard, these seem to me inapt for conducting the comparison required by Barton and Binderup . On my reading, the majority principally relies on an inference from a colloquial understanding of drunk driving’s "seriousness" to that offense’s "seriousness" for purposes of depriving a person of Second Amendment protection. See Majority Op. at II.B.2. This is a category mistake. If we conducted a poll of a representative sample of Americans, asking them whether drunk driving is a serious crime, it is likely that most would answer affirmatively. Such an appeal to ordinary meaning has legal purchase in the context of statutory interpretation because a court there confronts words as adopted by a procedurally established majority of the people’s elected representatives. But "serious" for present purposes is not a statutory, let alone a constitutional, term. It is how a majority of this Circuit’s judges in Binderup summarized the crimes that historically have deprived persons convicted of them of the right to own a firearm. "Serious," therefore, has a discrete legal meaning, and the "seriousness" inquiry must be given content consistent with that meaning. It is a determination of whether a challenger’s offense is sufficiently similar to those committed by the historically barred class. Evaluation of the second factor should be grounded in this legal framework.

Given the indeterminate nature of the historically barred class, I do not dispute that current authorities may assist us in measuring the "seriousness" of a challenger’s offense. But any such measurement must be consistent with our precedent. To me, the most relevant contemporary authorities for measuring "seriousness" are in fact included in the third and fourth factors: the actors within the criminal-justice system who confronted the challenger’s offense and imposed a punishment, and the jurisdictions that penalize the challenger’s conduct as a crime. As a result, I must conclude that the second factor weighs in Holloway’s favor.

3

Although the preceding factors support Holloway, they are insufficient in themselves to establish whether he is entitled to Second Amendment protection. Because a majority of the judges in Binderup held that a nonviolent misdemeanor may be "serious," the preceding factors, while probative measures of "seriousness," are not dispositive. Yet in the absence of common features of "serious" nonviolent misdemeanors—and Binderup did not specify any—we must compare the punishment for the challenger’s crime with the punishments for the crimes of the historically barred class. See 836 F.3d at 352 (opinion of Ambro, J.). The third and fourth Binderup factors both accomplish this end.

The third factor looks to the sentence the challenger received. It directs our attention to the unique circumstances of the challenger’s offense and conviction. Holloway was arrested in January 2005 after a police officer witnessed him driving the wrong way down a one-way street. Holloway , 349 F. Supp. 3d at 454. He registered a blood alcohol concentration (BAC) at the "highest rate" under Pennsylvania law, and because this was his second DUI offense, he was convicted of a first-degree misdemeanor, punishable by up to five years in prison. Id. However, he received the mandatory minimum sentence, which included three months of confinement on a work-release program. Id. 454-55.

The majority finds this factor against Holloway, emphasizing that, unlike the challengers in Binderup , he received a punishment that deprived him of his liberty. See Majority Op. at II.B.2. While this fact is certainly evidence that Pennsylvania considers Holloway’s offense more significant than that of Binderup (which was also committed in Pennsylvania), it does not measure Holloway’s offense against those of the historically barred class. A factor that considers the punishment received suggests some deference to the decisions of those within the criminal-justice system. See Binderup , 836 F.3d at 352 (opinion of Ambro, J.) ("[P]unishments are selected by judges who have firsthand knowledge of the facts and circumstances of the cases and who likely have the benefit of pre-sentence reports prepared by trained professionals."). Here, the actors on the ground did not deem Holloway’s offense "serious" enough to warrant the maximum penalty that Pennsylvania law permitted. Rather, the sentencing judge imposed the lightest punishment that the law allowed—a term of imprisonment, with work release, considerably shorter than the qualifying sentences under either § 922(g)(1) or § 921(a)(20)(B). As the District Court noted, Holloway’s assignment to a work-release program "undergirds the relatively minor nature of his sentence and suggests that the sentencing judge did not find Holloway to pose a significant risk to public safety." Holloway , 349 F. Supp. 3d at 457.

For the purposes of the Barton and Binderup comparison, then, I conclude that those who administered Pennsylvania’s law did not deem Holloway’s offense "serious" enough to merit imposition of a sentence on a par with those of the historically barred class. The argument that Holloway’s punishment was greater than anything received by the Binderup challengers bears more on the final factor than on the present one. The latter supports Holloway’s claim to Second Amendment protection.

4

The fourth factor asks whether there is a "cross-jurisdictional consensus regarding the seriousness of the [challenger’s] crime[ ]." Id. Like the sentence actually received, the challenger’s maximum possible punishment similarly provides a point of comparison with the historically barred class, but it cannot be assessed by looking to the challenger’s jurisdiction alone. The fact that the challenger’s crime is punishable by more than one or two years is the very reason he is in court; it demonstrates only that one jurisdiction has chosen to punish his conduct on terms comparable to the punishments of the historically barred class. More significant is how jurisdictions generally punish the challenger’s conduct because such a measure permits a comparison of current appraisal of the significance of the challenger’s crime with the punishments imposed on the historical class.

My review of the DUI laws in all fifty states and the District of Columbia reveals a notable consensus in how these jurisdictions punish Holloway’s conduct. Most importantly, only twelve of these jurisdictions punish such conduct with a maximum term of imprisonment exceeding one year. Of these twelve jurisdictions, seven provide for a maximum punishment exceeding two years, and only four of these seven classify such a crime as a misdemeanor. The other three jurisdictions, as well as the remaining five that punish the crime by more than one year of imprisonment, classify it as a felony. Given these statistics, there is no cross-jurisdictional consensus that a second DUI offense with a BAC at 0.192% is "serious" for purposes of Second Amendment analysis. In fact, the consensus lies in the other direction: a significant majority of jurisdictions—thirty-nine out of fifty-one—do not consider Holloway’s second DUI offense to be a crime worthy of punishment in accord with that of a traditional felony.

The majority finds it sufficient that "states unanimously agree that DUIs are crimes subject to punishment." Majority Op. at II.B.2. Yet as I have emphasized, our precedent dictates that the relevant measures of "seriousness" are those indicating how the challenger’s circumstances compare with the circumstances of the historically barred class. The fact of punishment alone should not render a crime "serious" enough to deprive an individual of a constitutional right. In the light of the evidence presented above, I must conclude that under the fourth factor, Holloway is not removed from the scope of Second Amendment protection.

***

Drunk driving is a dangerous crime. Declaring it not "serious" for purposes of the Second Amendment in no way detracts from its "seriousness" in the ordinary understanding of that word. But that is my point—the two categories are distinct, and our analysis should reflect that fact. Although Binderup did not create controlling precedent on the nature of the "seriousness" inquiry, the legal content of that inquiry must fulfill the requirements established in Barton and Binderup . Properly understood and applied, the multifactor test meets these demands. And in the context of the present case, it leads me to agree with the District Court that § 922(g)(1) burdens Holloway’s constitutional right to own a firearm. In this way, I part with the majority in this case.

II

If a court determines, as I do here, that the challenged law burdens protected conduct, then Marzzarella ’s second step requires the court to "evaluate the law under some form of means-ends scrutiny." 614 F.3d at 89. In Binderup , the same ten judges who agreed to adopt Marzzarella ’s two-step framework and the "seriousness" standard also accepted the application of intermediate scrutiny in as-applied challenges to § 922(g)(1). See Binderup , 836 F.3d at 353 (opinion of Ambro, J.); id. at 398 (Fuentes, J., concurring in part, dissenting in part, and dissenting from the judgments). Therefore, our precedent requires the application of intermediate scrutiny in the present case. See Donovan , 661 F.3d at 182.

Following a long line of Supreme Court case law, Marzzarella enumerated two elements of intermediate-scrutiny review. First, the government interest in the enforcement of the challenged regulation must be "significant, substantial, or important." 614 F.3d at 98 (internal quotation marks omitted). Second, there must be a "reasonable" fit between the asserted government interest and the regulation as written or applied. Id. ; see also Ass’n of N.J. Rifle & Pistol Clubs, Inc. v. Attorney Gen. New Jersey , 910 F.3d 106, 119 (3d Cir. 2018) (adopting this two-part test); Drake v. Filko , 724 F.3d 426, 436 (3d Cir. 2013) (same). I will consider each in turn.

A

The parties do not contest that the government has a substantial interest in "protecting the public from people who cannot be trusted to use firearms responsibly." Appellants’ Br. at 29. Neither Holloway’s brief nor the District Court’s opinion even mention this element. Thus, there is no reason to question whether the government has a substantial interest in enforcing § 922(g)(1).

B

Our primary difficulty lies in determining how to apply the second element of intermediate-scrutiny review to § 922(g)(1). Binderup established no precedent for how to decide whether there is a "[reasonable] fit between [ § 922(g)(1) ] and the asserted governmental end." Marzzarella , 614 F.3d at 98. Moreover, the standards applied by the judges in that case are not the same as the standard applied by the Court in Marzzarella . Yet as I detail in Section II.B.1, these standards are in fact doctrinally consistent with each other. If the government presents sufficient evidence to support its enforcement of the regulation at issue, we are then to evaluate how closely the regulation has been drawn to advance that interest. This is the standard I apply in Section II.B.2, concluding that § 922(g)(1) as applied in the present case fails intermediate scrutiny.

1

There is no binding precedent in our Circuit for the proper application of intermediate scrutiny to § 922(g)(1). In Binderup , the opinion announcing the Court’s judgment said the government "must ‘present some meaningful evidence, not mere assertions, to justify its predictive ... judgments’ " regarding the danger presented by the challengers and others like them. 836 F.3d at 354 (opinion of Ambro, J.) (quoting Heller v. District of Columbia , 670 F.3d 1244, 1259 (D.C. Cir. 2011) ). By contrast, in Marzzarella , the Court held that 18 U.S.C. § 922(k) "fits reasonably with [the government’s asserted] interest in that it reaches only conduct creating a substantial risk of rendering a firearm untraceable." 614 F.3d at 98. Whereas in Binderup , then, the judges were concerned with the evidence the government put forward, in Marzzarella the Court focused on the relation between the government’s asserted interest and the statute’s actual operation.

Despite this ostensible difference, these standards are in fact consistent with each other as a doctrinal matter. Marzzarella followed Heller in looking to the Supreme Court’s First Amendment case law for guidance, calling that doctrine "the natural choice" for "evaluating Second Amendment challenges." 614 F.3d at 89 n.4. In particular, for the second prong of intermediate-scrutiny review—that "the fit between the challenged regulation and the asserted objective be reasonable, not perfect"— Marzzarella referred to two of the Supreme Court’s commercial-speech cases. See id. at 98 (citing Lorillard Tobacco Co. v. Reilly , 533 U.S. 525, 556, 121 S.Ct. 2404, 150 L.Ed.2d 532 (2001) ; and Bd. of Trs. of State Univ. of N.Y. v. Fox , 492 U.S. 469, 480, 109 S.Ct. 3028, 106 L.Ed.2d 388 (1989) ). A brief consideration of commercial-speech doctrine allows us to see how our Circuit’s Second Amendment precedent in fact dictates a single standard for subjecting § 922(g)(1) to intermediate scrutiny.

The Supreme Court applies a four-step test for determining whether a regulation of commercial speech violates the First Amendment. A court must first "determine whether the expression is protected by the First Amendment," and then "ask whether the asserted governmental interest is substantial." Cent. Hudson Gas & Elec. Corp. v. Pub. Serv. Comm’n of N.Y. , 447 U.S. 557, 566, 100 S.Ct. 2343, 65 L.Ed.2d 341 (1980) ; see also Fox , 492 U.S. at 475, 109 S.Ct. 3028. If the answer to both inquires is affirmative, the government must then show "that the statute directly advances a substantial governmental interest and that the measure is drawn to achieve that interest." Sorrell v. IMS Health Inc. , 564 U.S. 552, 572, 131 S.Ct. 2653, 180 L.Ed.2d 544 (2011) (citing Fox , 492 U.S. at 480-81, 109 S.Ct. 3028 ; and Cent. Hudson , 447 U.S. at 566, 100 S.Ct. 2343 ).

This test bears notable resemblance to our Circuit’s developing Second Amendment doctrine. For our purposes here, the third and fourth steps are especially remarkable: they resemble the standards applied in Binderup and Marzzarella , respectively. Both are essential means of measuring the fit between the interest and the regulation. Indeed, the Supreme Court has said that these steps are not necessarily distinct inquiries. In as-applied challenges, the question posed at step three "cannot be answered by limiting the inquiry to whether the governmental interest is directly advanced as applied to a single person or entity." United States v. Edge Broad. Co. , 509 U.S. 418, 427, 113 S.Ct. 2696, 125 L.Ed.2d 345 (1993). The court must also consider "the regulation’s general application to others" with the same relevant characteristics as the challenger. Id. As a result, the validity of the regulation’s application to the challenger "properly should be dealt with under the fourth factor of the Central Hudson test." Id. This means that, regardless of the nature of the challenge, the third and fourth steps "basically involve a consideration of the fit between the legislature’s ends and the means chosen to accomplish those ends." Id. at 427-28, 113 S.Ct. 2696 (internal quotation marks omitted).

This background clarifies the standard to apply in the present case. In effect, Binderup concerned the correct application of the third step—whether the regulation "directly advances a substantial governmental interest." Sorrell , 564 U.S. at 572, 131 S.Ct. 2653. The three-judge opinion announcing the judgment of the Court did not need to advance its inquiry any further, because it concluded that § 922(g)(1) already failed as applied. In Marzzarella , however, there was no question whether the government had presented sufficient evidence to justify its enforcement action, and so the Court looked to how closely § 922(k) was drawn to achieve the government’s stated interest, holding that the statute is not impermissibly overinclusive because "it reaches only conduct creating a substantial risk of rendering a firearm untraceable." 614 F.3d at 98. As a result, to my mind Binderup and Marzzarella are doctrinally consistent, or at least reconcilable, in the light of how the Supreme Court has elaborated the final two steps of the commercial-speech test. At Marzzarella Step Two, if we are satisfied with the evidence supporting the statute’s application, we must then consider how closely the statute has been drawn to advance the government’s substantial interest.

2

Applying that standard in the present case, I conclude that § 922(g)(1) does not survive intermediate scrutiny. I disagree with the District Court, however, that the government has failed to produce evidence demonstrating that its enforcement of the statute directly advances its stated substantial interest. Rather, the flaw with the government’s case is that the statute as applied here is "wildly underinclusive." Nat’l Inst. Fam. & Life Advocs. v. Becerra , ––– U.S. ––––, 138 S. Ct. 2361, 2375, 201 L.Ed.2d 835 (2018) (quoting Brown v. Entm’t Merchs. Ass’n , 564 U.S. 786, 802, 131 S.Ct. 2729, 180 L.Ed.2d 708 (2011) ).

a

In finding that § 922(g)(1) ’s enforcement in this case does not directly advance the government’s substantial interest, the District Court demanded an excessively particularized connection between the evidence proffered and Holloway’s circumstances. Yet, as explained above, we should not limit our "inquiry to whether the governmental interest is directly advanced as applied to a single person or entity," but also consider "the regulation’s general application to others." Edge Broad. Co. , 509 U.S. at 427, 113 S.Ct. 2696. The government’s studies in Binderup were "obviously distinguishable." 836 F.3d at 354 (opinion of Ambro, J.). They concerned the likelihood of incarcerated felons to reoffend, though the Binderup challengers were neither incarcerated nor felons under state law. And the studies cited recidivism rates not applicable to individuals in the challengers’ situation. More compelling studies would have presented evidence relating to individuals "with the Challengers’ backgrounds." Id. at 355.

The government’s expert report in the present case does exactly that. It offers evidence relating to the features of Holloway’s biography that are at issue in this case. It refers to the likelihood of drug and alcohol abuse among repeat DUI offenders. D. Ct. Docket No. 61-4, at 4. It refers to firearm purchasers with prior alcohol-related convictions. Id. at 9. These are the features of Holloway’s biography at issue here. For the purposes of government policy, barring individuals with those characteristics from possessing a firearm is reasonable.

b

As explained above, our inquiry into "reasonable fit" does not end here. The question is not merely whether it is reasonable to disarm the challenger because of his conviction, but whether "the fit between the challenged regulation and the asserted objective [is] reasonable." Marzzarella , 614 F.3d at 98 (emphasis added). As a result, we must consider, in the context of this as-applied challenge, how closely § 922(g)(1) has been drawn to achieve the government’s substantial interest.

Under this standard, the law appears to be significantly underinclusive. Holloway’s crimes—a first DUI offense at a BAC of 0.131%, and a second DUI offense less than three years later with a BAC of 0.192%—implicate § 922(g)(1) in only eight of fifty-one jurisdictions (the fifty states and the District of Columbia). These eight jurisdictions account for approximately 21% of the United States population. On average, then, only about one in five individuals behaving exactly as Holloway did would be barred from possessing a firearm under § 922(g)(1). The statute’s dependence on state criminal classifications and punishments results in an underinclusive application that raises constitutional concerns, regardless of the reasonableness of disarming recidivist DUI offenders.

c

The next question is whether this underinclusivity renders § 922(g)(1) as applied here unconstitutional under intermediate scrutiny. To my mind, there are two principal counterarguments to answering this question affirmatively. Both of them fail.

First, it might be argued that our precedent remains unsettled regarding whether underinclusivity is a valid consideration in the Second Amendment context. Although Marzzarella allowed that a regulation’s "underinclusiveness can be evidence that the interest is not significant enough to justify the regulation," 614 F.3d at 99, the Court was there referring to underinclusivity in the context of strict, rather than intermediate, scrutiny. As a result, a future panel majority may reject a consideration of underinclusivity in intermediate-scrutiny review. See, e.g. , Ass’n of N.J. Rifle & Pistol Clubs , 910 F.3d at 122 n.28 ("While our Court has consulted First Amendment jurisprudence concerning the appropriate level of scrutiny to apply to a gun regulation, we have not wholesale incorporated it into the Second Amendment." (citations omitted)).

Yet, in constitutional law, underinclusivity follows necessarily from the evaluation of a fit between means and ends. And in Marzzarella we explicitly adopted a test that considers "the fit between the challenged regulation and the asserted objective." 614 F.3d at 98 ; see also Reilly , 533 U.S. at 556, 121 S.Ct. 2404 ; Fla. Bar v. Went For It, Inc. , 515 U.S. 618, 632, 115 S.Ct. 2371, 132 L.Ed.2d 541 (1995) ; Fox , 492 U.S. at 480, 109 S.Ct. 3028. The assessment of fit looks to the relation between the class of persons who come within the scope of the regulation’s stated objective, and the class of persons actually affected by the regulation. See, e.g. , Joseph Tussman & Jacobus tenBroek, The Equal Protection of the Laws , 37 CALIF. L. REV. 341, 344-53 (1949). Under this standard, what matters is not whether a regulation is specifically overinclusive, but rather by how much it is either over- or underinclusive. See, e.g. , City of Cincinnati v. Discovery Network, Inc. , 507 U.S. 410, 428, 113 S.Ct. 1505, 123 L.Ed.2d 99 (1993) (holding a city ordinance intended to advance safety and aesthetic interests unconstitutional because it unjustifiably affected only a small fraction of operating newsracks, thus constituting an unreasonable fit between ends and means).

This generalized inquiry encompasses both intermediate and strict scrutiny. The difference between those standards is the degree , rather than the type , of fit—whether the fit is either "reasonable" or "perfect." Marzzarella , 614 F.3d at 98 ; see McCutcheon v. Fed. Election Comm’n , 572 U.S. 185, 218, 134 S.Ct. 1434, 188 L.Ed.2d 468 (2014) (plurality opinion) ("Even when the Court is not applying strict scrutiny, we still require ‘a fit that is not necessarily perfect, but reasonable; ... that employs not necessarily the least restrictive means but ... a means narrowly tailored to achieve the desired objective.’ " (alterations in original) (quoting Fox , 492 U.S. at 480, 109 S.Ct. 3028 )). Intermediate scrutiny simply requires less of a fit between the governmental interest and the challenged regulation than strict scrutiny does.

It would be contrary to the logic of this analysis to hold that under intermediate scrutiny alone a court may not consider a regulation’s underinclusivity. To be sure, there may be a compelling reason why the Second Amendment context precludes such a consideration, but, to my mind, even that determination must now be left either to this Court sitting en banc or to the Supreme Court. Because our Court in Marzzarella adopted a means-ends fit analysis, we have already decided that underinclusivity is at least a valid consideration.

Second, it might be argued that § 922(g)(1) as applied here falls into one of the contexts in which the Supreme Court has upheld a regulation despite claims of underinclusivity. In particular, the Court has acknowledged two principal defenses—that a distinction drawn by a lawmaking body is in itself legitimate, and that a legislature is permitted to address a problem incrementally. See Williams-Yulee v. Fla. Bar , 575 U.S. 433, 135 S. Ct. 1656, 1668-69, 191 L.Ed.2d 570 (2015) (highlighting these two defenses); see also McConnell v. Fed. Election Comm’n , 540 U.S. 93, 207-08, 124 S.Ct. 619, 157 L.Ed.2d 491 (2003) ("[R]eform may take one step at a time, addressing itself to the phase of the problem which seems most acute to the legislative mind." (quoting Buckley v. Valeo , 424 U.S. 1, 105, 96 S.Ct. 612, 46 L.Ed.2d 659 (1976) )); Burson v. Freeman , 504 U.S. 191, 207, 112 S.Ct. 1846, 119 L.Ed.2d 5 (1992) ("States adopt laws to address the problems that confront them. The First Amendment does not require States to regulate for problems that do not exist."). These defenses do not support § 922(g)(1) as applied in the present case. Congress has drawn no distinction between different types of conduct—the same behavior may activate § 922(g)(1) or not based merely on where that behavior occurred. See City of Cincinnati , 507 U.S. at 428, 113 S.Ct. 1505 (declaring a city ordinance unconstitutionally underinclusive under intermediate scrutiny "[b]ecause the distinction [the city] has drawn has absolutely no bearing on the interests it has asserted"). For this same reason, it is hard to see how the statute represents Congress addressing problems as they arise. Section 922(g)(1) sweeps so broadly, covering any person convicted under state law of a felony or a misdemeanor carrying a sentence that exceeds two years, that in particular applications it is underinclusive, curtailing the constitutional rights of some and not others for the exact same conduct. Far from regulating for problems that do not exist, Congress is here not even regulating the vast majority of conduct it apparently deems problematic.

***

Ultimately at stake in this case is whether the government may arbitrarily burden the constitutional right of some citizens and not others. This equality concern goes to the heart of constitutional adjudication, regardless of the nature of the right at issue. As Justice Jackson put it in a different context:

The framers of the Constitution knew, and we should not forget today, that there is no more effective practical guaranty against arbitrary and unreasonable government than to require that the principles of law which officials would impose upon a minority must be imposed generally. Conversely, nothing opens the door to arbitrary action so effectively as to allow those officials to pick and choose only a few to whom they will apply legislation and thus to escape the political retribution that might be visited upon them if larger numbers were affected. Courts can take no better measure to assure that laws will be just than to require that laws be equal in operation.

Ry. Express Agency v. New York , 336 U.S. 106, 112-13, 69 S.Ct. 463, 93 L.Ed. 533 (1949) (Jackson, J., concurring). When a law, for reasons unrelated to enforcement discretion, on average punishes the same conduct only one in five times, such that those chosen individuals are deprived in perpetuity of a constitutional right, there is not a reasonable fit between the legislature’s asserted interest and the challenged regulation. If Congress wants to bar all individuals convicted of a second DUI offense with a BAC above 0.16% of owning a firearm, then it must do so through the ordinary channels of democratic lawmaking. At least then all persons’ constitutional right will be treated equally.

For these reasons, I respectfully dissent.