69 Am. U. L. Rev. F. 233 (2020).

*Executive Editor, American University Law Review, Volume 70; J.D. Candidate, May 2021, American University Washington College of Law. Thank you to the following people for helping me make this Comment a reality: Elizabeth Davis, Note and Comment Editor, American University Law Review, Volume 69, for her endless optimism and support; Professor Stephen Wermiel, for fostering my interest in constitutional law and serving as my faculty advisor for this piece; the Law Review staff, for dedicating their time and hard work to improve the piece; and my family, for encouraging me in my academic and professional endeavors.

December 21, 2018 marked a significant achievement in modern criminal justice reform: the First Step Act of 2018, designed to help reverse negative consequences of mass incarceration in the United States, became law. A primary feature of the First Step Act is an earned time credits system that allows eligible prisoners to reduce the length of their incarceration by participating in rehabilitative programs. Concerned that early release of some prisoners might endanger the public, lawmakers debated prisoner eligibility for this system through the last days of the law’s development, and ultimately excluded sixty-eight groups of prisoners from eligibility based on their convictions.

These decisions—like any government action that results in different treatment of similarly situated individuals—required compliance with the government’s constitutional obligation to provide equal protection under the laws. However, Congress did not perfectly comply with this mandate in designing the First Step Act. The Act excludes all prisoners convicted for offenses under 18 U.S.C. § 32 from eligibility for earned time credits, while it allows prisoners convicted under 18 U.S.C. § 2291—a nearly identical statute—the opportunity to receive the credits. As this Comment explains, this decision cannot survive constitutional scrutiny, and Congress must reexamine the earned time credits eligibility provisions to ensure its efforts to improve the criminal justice system do not jeopardize prisoners’ constitutional rights.

“Prison walls do not form a barrier separating prison inmates from the protections of the Constitution.”

—Justice Sandra Day O’Connor1Turner v. Safley, 482 U.S. 78, 84 (1987).

“[The First Step Act] will not allow dangerous, violent criminals to be released early. That is pure fiction.”

—United States Senator John Cornyn2164 Cong. Rec. S7642 (daily ed. Dec. 17, 2018) (statement of Sen. Cornyn).

Introduction

On December 7, 2007, a jury found David Cain Jr. guilty of sixteen counts of racketeering activity and related offenses.3United States v. Cain, 671 F.3d 271, 278 (2d Cir. 2012), remanded to United States v. Soha, No. 05-CR-360-A, 2017 WL 2222244 (W.D.N.Y. May 22, 2017); Jury Verdict, United States v. Cain, No. 1:05-cr-00360-RJA-HKS (W.D.N.Y. Dec. 7, 2007), ECF No. 251. During the trial, the government presented evidence that Cain, once the owner of a tree-trimming and logging business in New York, recruited and directed people to intentionally destroy or steal his competitors’ property to force the competitors out of business.4David Cain, Jr. Sentenced to 55 Years, Fed. Bureau of Investigation (Jan. 13, 2009), https://archives.fbi.gov/archives/buffalo/press-releases/2009/bffo011309.htm [https://perma.cc/JQX8-X55Y]. The resulting crimes included three arsons for which the jury convicted Cain, one of which targeted an airplane and its hangar at a local airport5Id. in violation of 18 U.S.C. § 32.6See 18 U.S.C. § 32(a)(1), (4) (2018) (establishing criminal penalties for setting fire to aircraft or their storage facilities). Following the jury’s verdict on the charges against Cain, the court sentenced Cain to a fifty-five year sentence.7David Cain, Jr. Sentenced to 55 Years, supra note 4.

About three years later, on November 24, 2010, a jury convicted Abdi Wali Dire on charges8United States v. Dire, 680 F.3d 446, 450 (4th Cir. 2012); Jury Verdict, United States v. Dire, No. 2:10-cr-00056-MSD-FBS (E.D. Va. Nov. 24, 2010), ECF No. 215. for his involvement in an attack on the USS Nicholas while the ship was on a counter-piracy mission on the Indian Ocean.9Dire, 680 F.3d at 449–50. In an attempt to seize the vessel to later negotiate a ransom, Dire and two others attacked the ship by firing AK-47 assault rifles at its crew, who successfully defended against the attack and captured Dire.10Id. at 449. The government charged Dire with committing an act of violence against persons on board a vessel11Id. at 450. in violation of 18 U.S.C. § 2291,12See 18 U.S.C. § 2291(a)(6) (2018) (establishing criminal penalties for acts of violence against individuals on board vessels). for which the judge sentenced Dire to twenty years of incarceration.13Dire, 680 F.3d at 450–51.

In 2018, the statutes that Cain and Dire violated became part of Congress’s discussion about a long-awaited criminal justice reform bill. The resulting law—the First Step Act of 201814Pub. L. No. 115-391, 132 Stat. 5194 (2018) (codified at scattered sections of 18 and 34 U.S.C.A. (West 2020)). (the First Step Act or “the Act”)—aims to reduce federal incarceration levels while maintaining public safety,15See 164 Cong. Rec. S7746–47 (daily ed. Dec. 18, 2018) (statement of Sen. Cornyn) (explaining Congress’s intent to provide opportunities for early release to prisoners “who would take advantage of” them, “but . . . not . . . at the cost of public safety”). in part through earned time credits that allow eligible prisoners to obtain early release from prison.16See 18 U.S.C.A. § 3632(d)(4)(C) (West 2020) (providing that prisoners may apply time credits toward time in prerelease custody or supervised release). In finalizing the bill, Congress decided that prisoners serving sentences for § 32 and § 2291 crimes should have different eligibility criteria for earned time credits.17See id. § 3632(d)(4)(D)(i), (xlvi) (foreclosing earned time credits to all prisoners convicted under 18 U.S.C. § 32 but only foreclosing the credits to prisoners convicted under § 2291 if their conduct involved “a substantial risk of death or serious bodily injury”). Under the law that resulted from these decisions, Cain is ineligible for the credits, while Dire is eligible as long as the conduct leading to his conviction did not put anyone at substantial risk of death or serious bodily injury.18See supra notes 6, 12, and 17 and accompanying text (demonstrating that Cain, as a prisoner serving a sentence for a § 32 conviction, is ineligible for earned time credits, while Dire, as a prisoner serving a sentence for a § 2291 conviction, is only ineligible if his conduct meets the required standard for ineligibility).

Dire’s conduct undoubtedly involved such a risk,19Supra note 10 and accompanying text. making him ineligible for earned time credits. For other current or future inmates who receive sentences for § 2291 violations, however, the nature of their conduct might be the difference between remaining behind bars or being released. Meanwhile, inmates serving time for § 32 violations—despite the nearly identical nature of § 32 and § 2291—lack such opportunities under all circumstances.20See supra note 17 (explaining that the Act prohibits all prisoners convicted under § 32 from accruing earned time credits).

This Comment argues that the First Step Act’s eligibility provisions for earned time credits violate the Fifth Amendment’s guarantee of equal protection of federal laws by arbitrarily limiting eligibility for prisoners with separate but substantively identical underlying convictions as other eligible prisoners.21See Bolling v. Sharpe, 347 U.S. 497, 499 (1954); infra notes 79–88 and accompanying text. The equal protection principle of the Fifth Amendment requires the federal government to have a rational basis for treating similarly situated groups of prisoners differently22See infra notes 79–88 and accompanying text.—which Congress lacked in differentiating between two groups of prisoners under the First Step Act. With implementation of the earned time credits system underway,23The first required step for implementation of the earned time credits system was the Attorney General’s creation of a risk and needs assessment tool to assess the recidivism risk for each federal prisoner. See § 18 U.S.C.A. 3632(a) (West 2020) (giving the Attorney General 210 days to complete the tool). The statute gave the Director of the Bureau of Prisons 180 days following the Attorney General’s completion of the risk and needs assessment tool to assess each prisoner’s recidivism risk and begin to assign prisoners to appropriate recidivism reduction programs through which eligible prisoners can earn time credits. Id. § 3621(h)(1). The Department of Justice announced the completion of the risk and needs assessment tool on July 19, 2019. Department of Justice Announces the Release of 3,100 Inmates Under First Step Act, Publishes Risk and Needs Assessment System, Dep’t of Just. (July 19, 2019), https://www. justice.gov/opa/pr/department-justice-announces-release-3100-inmates-under-first-step-act-publishes-risk-and [https://perma.cc/7JFZ-DKRW]. By January 15, 2020, the Bureau of Prisons had “completed an initial assessment for each inmate and assigned applicable programs of identified needs.” FSA Update, Fed. Bureau of Prisons (Jan. 15, 2020, 5:30 PM), https://www.bop.gov/resources/news/20200115_fsa_update.jsp [https://perma.cc/3FN7-6QJY]. The Bureau also announced plans to “review current risk scores and levels to determine where adjustments may need to be made” following the Department of Justice’s release of an updated risk and needs assessment tool. Id. The Department of Justice released the updated assessment tool on January 15, 2020. Id. and as demands for expanded criminal justice reform continue,24See, e.g., 164 Cong. Rec. H10398–99 (daily ed. Dec. 20, 2018) (letter from the ACLU and The Leadership Conference on Civil and Human Rights) (“[F]or all [the First Step Act’s] benefits, much more needs to be done. . . . The bill continues to exclude many federal prisoners from earning time credits and . . . ‘cash[ing] in’ the credits they earn.”); 164 Cong. Rec. S7764 (daily ed. Dec. 18, 2018) (statement of Sen. Booker) (“This legislation is just one step in the right direction. . . . [A]nd I hope [it] will be the momentum for greater, urgently needed reforms that will be supported by conservatives in this country and progressives.”); see also Ames Grawert & Tim Lau, How the First Step Act Became Law—and What Happens Next, Brennan Ctr. for Just. (Jan. 4, 2019), https://www.brennancenter.org/blog/how-first-step-act-became-law-and-what-happens-next [https://perma.cc/5CYN-H3GH] (outlining the next steps lawmakers should take to combat mass incarceration). confronting unconstitutional provisions within the Act is essential to ensuring that the Act achieves its intended goals.

Part I of this Comment provides relevant background on (1) the First Step Act and the earned time credits system that it established; (2) two criminal statutes—18 U.S.C. § 32 and § 2291—on which eligibility for earned time credits is based; (3) the U.S. Constitution’s guarantee of equal protection of the laws; and (4) the standard of review that courts apply to prisoners’ assertions of violations of their constitutional rights, along with cases in which parties have challenged sentencing laws on equal protection grounds. Part II argues that the First Step Act’s eligibility provisions for earned time credits violate the equal protection principle of the Fifth Amendment because they are not rationally related to any legitimate government interest.

I. Background

The First Step Act’s earned time credits allow eligible prisoners the significant benefit of reducing their time behind bars.25See 18 U.S.C.A § 3632(d)(4) (listing time credits as an incentive for prisoners to participate in recidivism risk reduction programs and thereby shorten their time in prison). As a result, eligibility for the credits was a key issue throughout the First Step Act’s development and resulted in compromises between lawmakers about how to classify federal prisoners for eligibility.26See infra notes 43–47 and accompanying text (discussing Congress’s debates about which prisoners should be eligible for earned time credits). This Part provides a brief history of the First Step Act, including the development of the earned time credits system within it, and discusses two criminal statutes that the Act uses to classify prisoners for earned time credits eligibility. This Part also describes the constitutional principle of equal protection and discusses cases in which courts have analyzed equal protection claims in the context of prisoners’ rights.

A. First Step Act of 2018

For about three decades beginning around 1980, Congress passed several laws that “consistently made federal criminal justice significantly more punitive.”27Shon Hopwood, The Effort to Reform the Federal Criminal Justice System, 128 Yale L.J. Online 791, 791 (2019). This trend was largely a response to growing concerns about illicit drug use in the United States,28See Sheldon Whitehouse, Foreword, 11 Harv. L. & Pol’y Rev. 359, 359 (2017) (“For the last three decades, no single factor has affected our criminal justice system more profoundly than our policy with respect to illegal drugs.”). along with a “shift in sentencing philosophy and practice . . . [toward] a more determinate and more punitive direction.”29Marc Mauer, Why Are Tough on Crime Policies so Popular?, 11 Stan. L. & Pol’y Rev. 9, 11 (1999). Based at least in part on post-1980 punitive criminal justice laws, the federal prison population increased from 24,260 in 1980 to 185,617 in 2017,30Hopwood, supra note 27, at 793; see also Whitehouse, supra note 28, at 359 (“From increased reliance on mandatory minimum sentences . . . to a prioritization of law enforcement over [drug use] prevention and treatment, a series of policy choices since the 1970s have led to a fivefold explosion in the nation’s prison population.”). contributing to the United States’ status as the country with twenty-five percent of the world’s prisoners, despite representing only five percent of the world’s population.31164 Cong. Rec. S7644 (daily ed. Dec. 17, 2018) (statement of Sen. Durbin).

The federal government’s “tough on crime” approach had some undesirable consequences. First, the rise in incarceration rates was costly: while the annual adjusted-for-inflation cost of incarceration per federal prisoner remained nearly constant between 1980 and 2010, the number of people incarcerated drove a significant increase in government spending for corrections, which “outpaced budget increases for nearly all other key government services . . . including education, transportation, and public assistance.”32Nat’l Research Council, Nat’l Acad. of Scis., The Growth of Incarceration in the United States: Exploring Causes and Consequences 314–16 (Jeremy Travis et al. eds., 2014). Second, a focus on punishment through lengthy prison sentences diminished other potential responses to crime, such as rehabilitation33See id. at 129 (“Unlike many other Western countries, the United States responded to escalating crime rates by enacting highly punitive policies and laws and turning away from rehabilitation and reintegration.”).—a tradeoff that research suggests might further exacerbate the costs of the criminal justice system.34See Jacob Reich, The Economic Impact of Prison Rehabilitation Programs, Wharton Pub. Pol’y Initiative Blog (Aug. 17, 2017), https://publicpolicy.wharton.upenn.edu /live/news/2059-the-economic-impact-of-prison-rehabilitation/for-students/blog/news.php#_edn2 [https://perma.cc/N72X-B395] (summarizing research demonstrating that educational opportunities, job training, and alcohol and drug addiction programs in prisons reduce recidivism and suggesting that the upfront costs of rehabilitation programs are worth the ultimate cost savings of reducing recidivism). Third, policies emphasizing incarceration disproportionately affected minorities;35See Nat’l Research Council, supra note 32, at 5 (“[T]he effects of harsh penal policies in the past 40 years have fallen most heavily on [B]lacks and Hispanics, especially the poorest.”). for instance, between 1980 and 2010, the imprisonment rate for Blacks was at least 4.6 times the imprisonment rate for whites.36Id. at 58–59 (noting that while the [B]lack-white ratio in incarceration rates was the lowest in 2010 that it had been “over the entire period for which race-specific incarceration rates are available,” “it was still very large” and “exceed[ed] racial differences for many other common social indicators”). Finally, while passage of tougher criminal laws coincided with a declining crime rate, the laws’ impact on that decline was modest.37See id. at 155 (“On balance, . . . data . . . support[s] the conclusion that the growth in incarceration rates reduced crime, but the magnitude of the crime reduction remains highly uncertain and the evidence suggests it was unlikely to have been large.”); Criminal Justice Facts, The Sent’g Project (citing Nat’l Research Council, supra note 32), https://www.sentencingproject.org/criminal-justice-facts [https://perma.cc/64RM-VL6V] (explaining possible factors for the limited impact that increasing incarceration rates have had on the reduction of crime rates, including the ineffectiveness of incarceration at reducing certain kinds of crimes and the tendency for people to commit fewer crimes as they age).

By 2010, Congress had become more open to a new type of reform aimed at combatting the negative effects of mass incarceration.38See Hopwood, supra note 27, at 793 (summarizing the evolution toward “bipartisan agreement that [the criminal justice] system desperately needs reform”). Eight years later, on December 21, 2018—following a 2010 law with a narrow scope,39See Fair Sentencing Act of 2010, Pub. L. No. 111-220, 124 Stat. 2372 (no currently effective sections). While the Fair Sentencing Act was an early outcome of Congress’s shifted mindset toward favoring less punitive criminal justice laws, many criticized it for not going far enough. See, e.g., Tyler B. Parks, The Unfairness of the Fair Sentencing Act of 2010, 42 U. Mem. L. Rev. 1105, 1107–08 (2012) (“[I]t may have been more appropriate to call the [Fair Sentencing Act] the ‘Slightly Fairer Sentencing Act of 2010,’ because . . . [w]ithout applying the [legislation] retroactively, [its] whole purpose—to restore fairness—[could] not be achieved.” (emphasis omitted) (footnote omitted)). failed attempts at subsequent legislation,40See Hopwood, supra note 27, at 794 (“Between 2015 and 2018, Congress could not pass the Sentencing Reform and Corrections Act, which stalled in the Senate. And many other reforms died on the vine after being introduced in Senate and House committees.” (citations omitted)). and extensive compromise by lawmakers in finalizing the law that eventually passed41The first iteration of the bill that became the First Step Act was focused on improving conditions in federal prisons. Grawert & Lau, supra note 24. A coalition of civil rights groups resisted the bill in this initial stage, arguing that it would do little to reduce the size of the federal prison population. Id. In November 2018, lawmakers compromised to enhance the Act with sentencing reform—but this version of the bill swung too far in the opposite direction as it started, losing the support of some conservative members. See id. (explaining that the Act initially stalled in the Senate due to right-wing opposition after it was amended to incorporate four provisions from the Sentencing Reform and Corrections Act). Further compromise—and the defeat of a series of conservative amendments proposed days before the Senate and House votes—yielded the version of the bill that President Donald Trump signed into law. Id.—President Donald Trump signed the First Step Act of 2018 into law.42Id.

Through the First Step Act, Congress intended to resolve some of the problems within the criminal justice system,43See supra notes 32–37 and accompanying text (describing negative consequences of mass incarceration in the United States). but not at the expense of public safety.44See supra note 15. These dual goals shaped the debate around the Act’s provisions on eligibility for earned time credits,45See, e.g., 164 Cong. Rec. S7758 (daily ed. Dec. 18, 2018) (statement of Sen. Toomey) (supporting the First Step Act’s inclusion of an amendment that would “ensure that all violent felons . . . will not be eligible for the earned time credit”); Andrea Drusch & Lesley Clark, Ted Cruz Rallies Conservatives with Changes to Criminal Justice Reform Plan, Fort Worth Star-Telegram (Dec. 6, 2018, 12:51 PM), https://www.star-telegram.com/latest-news/article222714545.html (“Though [U.S. Senator Ted] Cruz has long been a proponent of changes to the criminal justice system, he explained of his proposal [to expand the list of prisoners who are ineligible for early release programs]: ‘My principal concern is that we should not be releasing violent offenders.’”). which allow prisoners who participate in recidivism reduction programs to apply the time spent in such programs to transition from prison to prerelease custody or supervised release.4618 U.S.C.A. § 3632(d)(4)(C) (West 2020). Prisoners who are ineligible for earned time credits may receive other rewards for participation in recidivism reduction programs, such as phone and visitation privileges and transfer to prisons that are closer to the prisoner’s release residence. See id. § 3632(d)(1)–(2) (imposing no eligibility restrictions for incentives other than the earned time credits). The earned time credits system was a contentious element of the Act among legislators due to the payoff that the system can provide to prisoners; debates about the system ultimately contributed to Congress’s decision to exclude several groups of prisoners from earned time credits eligibility.47See supra notes 2 and 45 and accompanying text.

In its final form, the Act excluded sixty-eight groups of prisoners from the earned time credits system based on the criminal statute that they violated.48See § 3632(d)(4)(D) (excluding, among others, prisoners convicted of crimes related to “female genital mutilation”; “domestic assault by a habitual offender”; “criminal street gangs”; “gathering or delivering defense information to aid a foreign government”; “unlawful possession or use of a firearm during and in relation to any crime of violence or drug trafficking crime”; and various other offenses under the Controlled Substances Act, 21 U.S.C. § 841 (2018)). Generally, prisoners are ineligible for earned time credits if they are serving sentences for crimes of “violen[ce], terrorism, espionage, human trafficking, sex and sexual exploitation, [being a] repeat felon in possession of firearm, . . . fraud, [and] high-level drug offenses.” Nathan James, Cong. Research Serv., R5558, The First Step Act: An Overview 5 (2019). Among the prisoners who are ineligible for earned time credits under the Act are those convicted of crimes under 18 U.S.C. § 32, “relating to [the] destruction of aircraft or aircraft facilities,”49§ 3632(d)(4)(D)(i). and 18 U.S.C. § 2291, “relating to the destruction of a vessel or maritime facility, but only if the conduct that led to the conviction involved a substantial risk of death or serious bodily injury.”50§ 3632(d)(4)(D)(xlvi). The next section discusses the provisions and history of these sections of the U.S. criminal code.

B. Convictions Under 18 U.S.C. §§ 32, 2291

Section 32 of 18 U.S.C. provides criminal penalties for interfering with the safe operation of aircraft or aircraft facilities.51See 18 U.S.C. § 32(a) (2018) (criminalizing various acts related to aircraft and aircraft facility safety). For simplicity, this Comment only analyzes § 32(a) and § 2291(a) and possible convictions under those sub-sections. Those sub-sections include the majority of the offenses that § 32 and § 2291 prohibit, and the variety of offenses that the sub-sections describe are sufficient for this Comment’s analysis of the similarities between prisoners convicted under § 32 and § 2291. The statute imposes a fine, imprisonment for up to twenty years, or both on anyone who willfully performs acts such as “set[ting] fire to, damag[ing], destroy[ing], disabl[ing], or wreck[ing] any aircraft . . . or any appliance or structure” used to operate, maintain, or store the aircraft; interfering with the operation of aircraft in flight; and other related acts, including conspiracy to commit the prohibited acts.52Specifically, 18 U.S.C. § 32(a) imposes fines or imprisonment on “[w]hoever willfully”: (1) sets fire to, damages, destroys, disables, or wrecks any aircraft . . .; (2) places or causes to be placed a destructive device or substance in, upon, or in proximity to, or otherwise makes or causes to be made unworkable or unusable or hazardous to work or use, any such aircraft, or any part or other materials used or intended to be used in connection with the operation of such aircraft, if such . . . [conduct] is likely to endanger the safety of any such aircraft; (3) sets fire to, damages, destroys, or disables any air navigation facility, or interferes by force or violence with the operation of such facility, if such . . . [conduct] is likely to endanger the safety of any such aircraft in flight; (4) . . . sets fire to, damages, destroys, or disables or places a destructive device or substance in, upon, or in proximity to, any appliance or structure . . . used, or intended to be used, in connection with the operation, maintenance, loading, unloading or storage of any such aircraft or any cargo carried . . . on any such aircraft; (5) interferes with or disables, with intent to endanger the safety of any person or with a reckless disregard for the safety of human life, anyone engaged in the authorized operation of such aircraft . . .; (6) performs an act of violence against or incapacitates any individual on any such aircraft, if such . . . [conduct] is likely to endanger the safety of such aircraft; (7) communicates information, knowing [it] to be false and under circumstances in which [it] may reasonably be believed, thereby endangering the safety of any such aircraft in flight . . . .

Congress enacted § 32 in 1956 in response to a plane crash near Longmont, Colorado resulting from the explosion of a time bomb in luggage on the airplane.53See S. Rep. No. 84-1472, at 1 (1956) (“The recent disaster at Longmont, Colo. . . . has focused attention on the need for such legislation.”). Federal lawmakers were concerned that then-existing legislation was insufficient to address the magnitude of the criminal conduct that led to the Longmont incident.54See id. (“[N]one of [the federal statutes then in effect were] drafted with this type of problem specifically in mind, with the result that the statutes now in effect either do not adequately cover the subject matter or do not provide penalties commensurate with the magnitude of the offense.”); see also His Mother Among 44 Killed in Liner Crash, Wash. Post, Nov. 15, 1955, at 1 (“Officials in Washington explained that in many respects, [f]ederal criminal law has not yet caught up with the air age . . . .”). Congress modeled § 32 after an existing statute related to the destruction of trains in an effort to cure the deficiencies they saw in the criminal code related to violent conduct on airplanes.55S. Rep. No. 84-1472, at 1.

While the statute was a response to a tragedy that occurred when an aircraft was in flight, resulting in the deaths of forty-four people—and while several provisions of the statute include likelihood of harm to individuals as an element56See supra note 52 (noting that § 32 criminalizes, among other things, conduct that endangers or risks endangering passengers on aircraft in flight).—the statute also criminalizes destruction or disabling of stationary aircraft or aircraft facilities from which no substantial risk of or actual injury or death results.57See § 32(a); see, e.g., United States v. Bishop, 66 F.3d 569, 589 n.32 (3d Cir. 1995) (noting that § 32 is an example of Congress’s ability to “regulate vehicles used in interstate commerce, i.e. that have traveled, do travel, or will travel in interstate commerce whether or not they are actually traveling in interstate commerce when regulated”); United States v. Heightland, 865 F.2d 94, 96 (6th Cir. 1989) (quoting United States v. Hume, 453 F.2d 339, 340 (5th Cir. 1971)) (“Congress intended not only to protect civil aircraft while actually operating in interstate commerce, but also to protect such aircraft as [are] used or employed in interstate commerce, and also the parts, materials and facilities used by such aircraft.”). This prevented the petitioner in United States v. Cain,58671 F.3d 271 (2d Cir. 2012). David Cain Jr., from successfully arguing that § 32 did not apply to his actions because the aircraft that he destroyed was not in flight at the time of its destruction.59See id. at 298 (interpreting the statute’s phrase “any civil aircraft used, operated, or employed in interstate . . . air commerce” as encompassing aircraft that are not in flight at the time of the destructive conduct). Cain, a business owner in upstate New York, led a series of arsons and other acts intended to intimidate his competitors and force them out of business.60David Cain, Jr. Sentenced to 55 Years, supra note 4; see also Dan Herbeck, Niagara Man Convicted of Racketeering Charges[,] Jury Also Convicts Brother, Cousin as Gang Members, Buff. News (Dec. 7, 2007), https://buffalonews.com/2007/12/07/niagara-man-convicted-of-racketeering-charges-jury-also-convicts-brother-cousin-as-gang-members (“Cain was determined to run the largest tree-trimming and logging business in Niagara and Orleans counties, and he decided to accomplish his goal by driving others out of business . . . .”). One arson targeted a 1949 vintage airplane that belonged to Cain’s competitor; the aircraft “erupted in flames inside its hangar” after Cain and the aircraft’s owner had a disagreement.61Herbeck, supra note 60. There is no record of physical injuries resulting from the fire or indication that it involved substantial risk of death of or serious bodily injury to any person.62See Indictment at 17, United States v. Cain, No. 1:05-cr-00360-RJA-HKS (W.D.N.Y. Nov. 19, 2007), ECF No. 250 (describing the July 9, 2004 arson of Cain’s competitor’s aircraft without mention of any risk of, or actual, injury to or death of anyone as part of the 18 U.S.C. § 32 grand jury charge); see also David Cain, Jr. Sentenced to 55 Years, supra note 4 (reporting on the arson without mention of any risk of injury or death from the incident); Herbeck, supra note 60; cf. Dan Herbeck, Tree Trimmer Characterized as Gangster[,] Business Rival Tells of Threats, Fear He Felt, Buff. News (Nov. 14, 2007), https://buffalonews.com/2007/11/14/tree-trimmer-characterized-as-gangster-business-rival-tells-of-threats-fear-he-felt (describing the fear that the aircraft owner felt toward Cain due to incidents in which Cain “ran [him] off the road” but not reporting on any risk of injury or death as part of the actual arson of the aircraft). Nonetheless, the court upheld the application of § 32 to Cain’s conduct.63Cain, 671 F.3d at 298. The court’s interpretation of the statute underscores the statute’s broad prohibition of destruction of or damage to aircraft or aircraft facilities in any form, regardless of whether risk of death or serious bodily injury is substantial.

Similar to § 32, § 2291 criminalizes the destruction of vessels or maritime facilities or violent acts on board vessels.64See 18 U.S.C. § 2291(a) (2018) (prohibiting various acts related to vessel and maritime facility safety). For simplicity, this Comment only analyzes § 32(a) and § 2291(a) and possible convictions under those sub-sections. See supra note 51 (explaining the focus § 32(a) and § 2291(a) rather than all sub-sections within § 32 and § 2291). Congress enacted § 2291 when it passed the Uniting and Strengthening America by Providing Appropriate Tools Required to Intercept and Obstruct Terrorism (USA PATRIOT) Improvement and Reauthorization Act of 200565Pub. L. No. 109-177, 120 Stat. 192 (codified at scattered sections of 8, 15, 18, 21, 28, and 34 U.S.C.A. (West 2020)). See Brian T. Yeh & Charles Doyle, Cong. Research Serv., RL33332, USA PATRIOT Improvement and Reauthorization Act of 2005 (H.R. 3199): A Legal Analysis 33–34 (2006) (identifying 18 U.S.C. § 2291 as a statute that the Act established). The USA PATRIOT Improvement and Reauthorization Act modified and made permanent several provisions of the USA PATRIOT Act of 2001. See id. at 1–4. to support the federal government in combatting terrorism.66See Pub. L. No. 109-177, 177 (identifying the USA PATRIOT Improvement and Reauthorization Act’s purpose as “extend[ing] and modify[ing] authorities needed to combat terrorism”); see also Yeh & Doyle, supra note 65, at 1 (describing the new legislation’s functions, including making permanent expiring provisions of the USA PATRIOT Act, allowing judicial review of national security provisions, and adding to available tools and checks on abuse of the law). Section 2291 imposes fines, prison sentences for up to twenty years, or both on individuals who knowingly destroy vessels or equipment used to operate, maintain, or store the vessel; interfere with the operation of maritime facilities; or perform other related acts, including conspiracy to commit the prohibited acts.67 Specifically, § 2291(a) imposes fines or imprisonment on “[w]hoever knowingly”: (1) sets fire to, damages, destroys, disables, or wrecks any vessel; (2) places or causes to be placed a destructive device . . . [or] substance . . . in, upon, or near, or otherwise makes or causes to be made unworkable or unusable or hazardous to work or use, any vessel, or any part or other materials used or intended to be used in connection with the operation of a vessel; (3) sets fire to, damages, destroys, or disables or places a destructive device or substance in, upon, or near, any maritime facility, including any aid to navigation, lock, canal, or vessel traffic service facility or equipment; (4) interferes by force or violence with the operation of any maritime facility, including any aid to navigation, lock, canal, or vessel traffic service facility or equipment, if such action is likely to endanger the safety of any vessel in navigation; (5) sets fire to, damages, destroys, or disables or places a destructive device or substance in, upon, or near, any appliance [or] structure . . . used, or intended to be used, in connection with the operation, maintenance, loading, unloading, or storage of any vessel or any passenger or cargo carried . . . on any vessel; (6) performs an act of violence against or incapacitates any individual on any vessel, if such . . . [conduct] is likely to endanger the safety of the vessel or those on board; . . . (8) communicates information, knowing [it] to be false and under circumstances in which [it] may reasonably be believed, thereby endangering the safety of any vessel in navigation . . . .

Section 2291 “harmonize[d] the somewhat outdated maritime provisions [of the U.S. criminal code] with the existing criminal sanctions for destruction [of] or interference with an aircraft or aircraft facilit[y]” in § 32,68Yeh & Doyle, supra note 65, at 35 (quoting H.R. Rep. No. 109-333, at 104 (2005)). thus largely mirroring § 32’s substantive provisions.69See id. (noting that § 2291 “mirrors the substantive provisions for the destruction of aircraft and their facilities” in § 32). Despite the similar nature of § 32 and § 2291, they differ in some ways, including the general requirement in § 32 that the prohibited damage adversely impacts safe operation, which § 2291 is “less likely” to require. Id. at 36. But see infra notes 141–48 and accompanying text (describing scenarios in which § 32 offenses, like § 2291 offenses, might not pose harm to individuals). In this vein, the conduct that § 2291 prohibits is broad in scope and includes any type of interference with vessels or maritime facilities.70Compare § 2291(a)(6) (criminalizing acts of violence against individuals on board a vessel that are likely to jeopardize the vessel’s safety), with § 2291(a)(1) (prohibiting the disabling of vessels even without any risk to human safety). In reality, most § 2291 charges have arisen from acts of piracy that involved substantial risks to those on board the targeted vessels.71See, e.g., United States v. Dire, 680 F.3d 446, 449–50 (4th Cir. 2012) (reviewing petitioners’ challenges of their convictions that resulted from an act of piracy in which they attacked a Navy vessel with AK-47s); Cabaase v. United States, 395 F. Supp. 3d 724, 728 (E.D. Va. 2019) (reviewing petitioner’s motion to vacate, set aside, or correct his sentence for acts of piracy, one of which involved firing of AK-47s at and return fire from the vessel). This was the case in United States v. Dire,72680 F.3d 446 (4th Cir. 2012). a case that arose when Abdi Wali Dire and his co-defendants attacked a U.S. Navy vessel on the Indian Ocean with AK-47 assault rifles, attempting to seize the vessel and secure ransom.73Id. at 449–50. But § 2291 also prohibits acts that are less dangerous to human life or safety than Dire’s were, and it would prohibit individuals from setting fire to stationary vessels as Cain did with his competitor’s aircraft.74Compare § 2291(a) (prohibiting “set[ting] fire to . . . any vessel”), with § 32(a) (prohibiting “set[ting] fire to . . . any civil aircraft”), and United States v. Cain, 671 F.3d 271, 298 (2d Cir. 2012) (upholding the defendant’s conviction for setting fire to a civil aircraft that was not in flight).

Between FY1994 and FY2016, approximately sixty individuals were convicted of offenses under § 32, and approximately eight individuals were convicted under § 2291.75These figures are based on the Bureau of Justice Statistics (BJS) Federal Justice Statistics Resource Center’s “Federal Criminal Case Processing Statistics” tool, available at https://www.bjs.gov/fjsrc [https://perma.cc/M889-PRLH] [hereinafter BJS data]. Because source agencies provide data to BJS with varying levels of specificity of sub-sections within criminal codes, see https://www.bjs.gov/fjsrc/index.cfm? p=help&topic=t_sec_caution [https://perma.cc/5C7Q-QXN2], these figures are approximations of the total convictions under § 32(a) and § 2291(a), which are the sub-sections of the statutes on which this Comment is focused. See supra note 51. Of those convicted under the two statutes, most received sentences to terms of imprisonment without any probation.76BJS data, supra note 75. Most sentences that courts imposed upon prisoners convicted under § 2291 between FY1994 and FY2016 were shorter than sentences imposed upon prisoners convicted under § 32 during the same period. Id. However, because the possible range of sentences under the two statutes is identical, individuals who violate § 2291 may face identical or even longer sentences than those who violate § 32. See supra notes 52, 67 and accompanying text. The total federal prison population as of March 2019 of 221,000 individuals77Wendy Sawyer & Peter Wagner, Mass Incarceration: The Whole Pie 2019, Prison Pol’y Initiative (Mar. 19, 2019), https://www.prisonpolicy.org/reports/pie2019.html [https://perma.cc/GR45-XMEZ]. indicates that sixty-eight convictions for offenses under § 32 and § 2291 between FY1994 and FY2016 is a small percentage of the criminal convictions at large that would have occurred during that timeframe. Even so, the individuals serving sentences for violations of these statutes have the same rights as all prisoners within the criminal justice system, including the equal protection of the laws.78See infra Part I.C (defining the U.S. Constitution’s equal protection guarantee).

C. Equal Protection Principle of the Fifth Amendment

The Fifth Amendment provides that no person shall be “deprived of life, liberty, or property, without due process of law.”79U.S. Const. amend. V. While not explicitly stated, the Fifth Amendment forbids the federal government from denying “any person within its jurisdiction the equal protection of the laws,”80Id. amend. XIV. just as the Fourteenth Amendment forbids such actions by the states.81See Weinberger v. Wiesenfeld, 420 U.S. 636, 638 n.2 (1975) (“This Court’s approach to Fifth Amendment equal protection claims has always been precisely the same as to equal protection claims under the Fourteenth Amendment.”). The command that the federal government guarantee equal protection of the laws to those in its jurisdiction “is essentially a direction that all persons similarly situated should be treated alike.”82City of Cleburne v. Cleburne Living Ctr., Inc., 473 U.S. 432, 439 (1985) (citing Plyer v. Doe, 457 U.S. 202, 216 (1982)). Courts define “similarly situated” as alike “in all relevant respects.”83Nordlinger v. Hahn, 505 U.S. 1, 10 (1992) (citing F.S. Royster Guano Co. v. Virginia, 253 U.S. 412, 415 (1920)). See Robbins v. Becker, 794 F.3d 988, 996 (8th Cir. 2015) (quoting Reget v. City of La Crosse, 595 F.3d 691, 695 (7th Cir. 2010)) (“To be similarly situated . . . the persons alleged to have been treated more favorably must be identical or directly comparable to the plaintiff in all material respects.”); Marrero-Gutierrez v. Molina, 491 F.3d 1, 9 (1st Cir. 2007) (“[T]he test is whether an objective person would see two people similarly situated based upon the incident and context in question.”).

The legislature is responsible for determining whether different people are similarly situated for the purposes of a law and has wide latitude in making such decisions.84See Plyer, 457 U.S. at 216 (“A legislature must have substantial latitude to establish classifications that roughly approximate the nature of the problem perceived, that accommodate competing concerns both public and private, and that account for limitations on the practical ability of the [government] to remedy every ill.”). Unless the basis for different treatment under a law is one that the Supreme Court has deemed “suspect” or “quasi-suspect,” such as race or gender, courts will uphold classifications within legislation that are rationally related to legitimate government interests.85City of Cleburne, 473 U.S. at 440. This analysis is referred to as the “rational basis test.”86See, e.g., Smith v. Shalala, 5 F.3d 235, 239 (7th Cir. 1993) (“Because this classification . . . does not involve a suspect class . . . we must examine it under the rational basis test. Therefore, we must determine whether the regulatory scheme bears a rational relationship to legitimate legislative goals.”).

D. Equal Protection Challenges to Sentencing Laws

In Turner v. Safley,87482 U.S. 78 (1987). the Court established that prisoners’ claims of violations of their constitutional rights are subject to the rational basis test.88See id. at 89 (“[W]hen a prison regulation impinges on inmates’ constitutional rights, the regulation is valid if it is reasonably related to legitimate penological interests.”). Turner involved two regulations at a Missouri prison, one of which did not allow prison inmates to marry unless there was a “compelling” reason for doing so, and the other which limited the exchange of mail correspondence between inmates.89Id. at 81–82. The lower courts applied strict scrutiny analysis—a higher standard than the rational basis test that requires classifications to be narrowly tailored to achieve a compelling government interest—to the prisoners’ claims; the courts held that the provisions were unconstitutional, finding that neither regulation was “the least restrictive means of achieving the security [and rehabilitation] goals of the regulation[s].”90Id. at 83 (describing the standard of strict scrutiny that the lower courts applied as requiring that the law furthers “an important or substantial governmental interest” for which “the limitation is no greater than necessary or essential” (citation omitted)).

However, the Supreme Court held that the lower courts applied the wrong test to determine whether the regulations violated the prisoners’ constitutional rights.91Id. at 81. Rational basis scrutiny of prisoners’ rights is the appropriate analysis, the Court explained, because the judiciary is “ill equipped to deal with the increasingly urgent problems of prison administration and reform,” which are a better match for the legislative and executive branches of government.92Id. at 84–85 (quoting Procunier v. Martinez, 416 U.S. 396, 405 (1974)).

In applying rational basis scrutiny to laws or regulations that prisoners argue are unconstitutional, courts should consider four factors: (1) whether the connection between the law and the government’s interest is “valid [and] rational” or rather so remote that the law can only be seen as arbitrary or irrational; (2) whether prisoners can exercise their rights through alternative means if the law remains intact; (3) the impact that accommodating the prisoners’ asserted rights would have on prison resources, guards, and inmates; and (4) whether the government can achieve its goal with means beyond the law.93Id. at 89–91. Applying these factors, the Court upheld the correspondence regulation at issue in Turner but held that the marriage regulation was unconstitutional.94See id. at 91–93, 97–99 (holding that the correspondence regulation was logically connected to prison security concerns, lacked easy alternatives, and prevented negative impacts on prison administration, while the marriage regulation was not reasonably related to any penological interests).

Under the framework the Supreme Court articulated in Turner, prisoners’ equal protection challenges to sentencing laws almost invariably fail. In resolving prisoners’ equal protection claims, courts have recognized various types of classifications of prisoners through sentencing laws that are rationally related to legitimate government interests and constitutional under rational basis review.95See, e.g., Robinson v. Deloach, No. 1:06cv1061-MHT, 2009 WL 1116838, at *3 (M.D. Ala. Apr. 24, 2009) (holding a sentencing law valid when it classified prisoners on the basis of whether the prisoner was serving a life sentence); Jones v. Bruce, 921 F. Supp. 708, 711 (D. Kan. 1996) (on the basis of the seriousness of a prisoner’s underlying offense); Kansas v. Perez, 11 P.3d 52, 54–55 (Kan. 2000) (on the basis of the type of unlawful act a prisoner committed while on parole).

Criminal statutes routinely classify convicted defendants and treat classes differently based on the nature of the conduct that led to their convictions, as is evident by different crimes carrying different penalties. Hence, not surprisingly, the seriousness of an underlying offense has served as sufficient basis for several laws that classify individuals in the sentencing context.96See, e.g., Robinson, 2009 WL 1116838, at *4 (upholding a sentencing law in an equal protection case when the law classified prisoners on the basis of the seriousness of a prisoner’s underlying offense). Courts have also upheld laws that classify individuals for non-sentencing purposes based on the rational connection between the government’s desire to protect society from offenders of more serious crimes and the different treatment. See, e.g., Matter of A.C., No. 18 CA 1, 2018 WL 5309946, at *3 (Ohio Ct. App. Oct. 22, 2018) (upholding a statute that “restrict[ed] the availability of [record] expungement in cases where the [l]egislature has found the crimes sufficiently serious to warrant not extending the privilege of expungement to those who have committed those offenses”). In Robinson v. Deloach,97No. 1:06cv1061-MHT, 2009 WL 1116838, at *1 (M.D. Ala. Apr. 24, 2009). an Alabama state inmate challenged the Alabama Correctional Incentive Time Act98Ala. Code §§ 14-9-40 to -44 (2020). (ACITA), alleging that its eligibility scheme for sentence reductions violated the equal protection clause of the Fourteenth Amendment.99Robinson, 2009 WL 1116838, at *1. Under the ACITA, eligible prisoners in Alabama institutions may receive reductions from the terms of their sentences for good behavior in prison.100See § 14-9-41(a) (specifying calculations of reductions from prisoners’ sentences). However, prisoners who are sentenced to terms of more than fifteen years are ineligible for such reductions.101Id. § 14-9-41(e). As an inmate sentenced to a term of imprisonment for more than fifteen years, the petitioner in Robinson was ineligible for the reductions.102See id.; Robinson, 2009 WL 1116838, at *4. The court rejected the equal protection claim based on “well settled” law that the ACITA’s eligibility provisions are “rationally related to the legitimate state interest of preventing the early release of serious offenders.”103Robinson, 2009 WL 1116838, at *4 (citing Thornton v. Hunt, 852 F.2d 526, 527 (11th Cir. 1988) (per curiam)). Robinson demonstrates that different treatment of prisoners who have varying sentences corresponding with crimes of varying severity in the eyes of the legislature is acceptable to courts under the rational basis test.

In reviewing sentencing laws that classify prisoners based on their conviction type, courts have emphasized that the legislature is not required to draft laws that perfectly achieve the legislature’s goal; lawmakers may exclude prisoners from sentence reductions even if such prisoners argue they should be eligible.104See, e.g., Jones v. Bruce, 921 F. Supp. 708, 711 (D. Kan. 1996) (upholding a sentencing law despite the petitioner’s argument that it was under-inclusive). In Jones v. Bruce,105921 F. Supp. 708 (D. Kan. 1996). the petitioner challenged the denial of his petition to convert his sentence under the Kansas Sentencing Guidelines Act106Kan. Stat. Ann. § 21-4701 to -4729 (West 2020) (repealed 2011). (KSGA), which allowed some prisoners to modify the sentences they received prior to the KSGA’s enactment.107Jones, 921 F. Supp. at 709. The petitioner, who alleged he was not violent or dangerous to the public but was ineligible for the sentence modification, argued that conversion of his sentence would promote the legislature’s stated goal of easing prison overcrowding while not jeopardizing public safety.108See id. at 710–11 (“Petitioner argues [that] the goals of the KSGA are thwarted by denying him retroactive application of the Act and, moreover, the conversion of his sentence would promote the goals of the KSGA because the prison population would be reduced by releasing an offender who is not violent or a danger to the public.”). According to the court, however, rational basis review does not turn on “whether the legislation perfectly accomplishes its goals but whether the legislation was prompted by a legitimate goal and whether there is a rational basis for the classification.”109Id. at 711. Thus, over-inclusiveness or under-inclusiveness is not fatal to a law that classifies prisoners for sentencing in an effort to balance prison conditions with public safety.110See Vance v. Bradley, 440 U.S. 93, 108 (1979) (“Even if the classification involved . . . is to some extent both underinclusive and overinclusive, and hence the line drawn by Congress [is] imperfect, it is nevertheless the rule that in a case like this ‘perfection is by no means required.’” (quoting Phillips Chem. Co. v. Dumas Sch. Dist., 361 U.S. 376, 385 (1960))).

Beyond the nature of prisoners’ underlying offenses, courts have upheld classifications based on other criteria, including whether prisoners are serving life sentences, in reviewing sentencing laws.111See, e.g., Robinson v. Deloach, No. 1:06cv1061-MHT, 2009 WL 1116838, at *3 (M.D. Ala. Apr. 24, 2009) (holding that limiting eligibility of sentence reductions to prisoners serving life sentences was rationally related to ensuring equity in sentencing). In addition to his claim related to the ACITA, the petitioner in Robinson also challenged a different statute that did not allow him to petition for a sentence reduction because he was not serving a life sentence, despite him being a nonviolent offender as required by the statute for sentence reductions.112Id. at *2. The court rejected the petitioner’s argument that he must be afforded the same opportunity for a sentence reduction as prisoners serving life sentences, reasoning that the eligibility requirements were rationally related to the state’s interest in ensuring equitable sentencing for prisoners who had been sentenced to life under previously mandatory guidelines.113Id. at *3. Hence, when there is a fundamental difference in the nature of a sentence—in other words, a life sentence versus a sentence for a precise number of years not intended to last a lifetime—the government has a rational basis for treating prisoners serving such sentences differently.

Another legitimate government interest that courts have recognized in reviewing sentencing laws is reducing sentencing disparities for repeat criminals who commit crimes while on parole. Kansas v. Perez11411 P.3d 52 (Kan. 2000). involved a provision of the KSGA that allowed some parolees who had been sentenced under old sentencing guidelines to reduce their sentences based on the new guidelines.115See id. at 53–54 (providing criteria for whether a parolee may obtain a sentence reduction under the KSGA). The provision applied to parolees who committed crimes while on parole for which they were convicted and sentenced; such parolees could ask a court to reduce the sentence for the prior crime and have it run concurrently with the sentence for the subsequent crime.116See id. In contrast, parolees who did not commit crimes while on parole, but rather violated the terms of their parole, could not receive such a sentence reduction.117See id.; see also U.S. Dep’t of Justice, The United States Parole Commissions Expedited Revocation Procedure (2004), https://www.justice.gov/sites/default/ files/uspc/legacy/2004/05/11/expedited_apai1.pdf (citing “administrative violations (such as alcohol abuse, drug use, or failure to report) or misdemeanor offenses (such as simple possession of a controlled substance, petit larceny, disorderly conduct, or driving while intoxicated)” as common parole violations). The petitioner moved for a sentence conversion under the KSGA after violating the terms of his parole.118Perez, 11 P.3d at 53. The lower court denied his motion, and the Supreme Court of Kansas affirmed.119Id. at 53, 55.

The court explained that parolees who committed crimes while on parole after the passage of the new sentencing guidelines were differently situated from parolees who violated the terms of their parole.120Id. at 54. The first group would be subject to two sentencing schemes—one for the prior crime under the old sentencing guidelines, and one for the new crime under the new guidelines.121See id. at 54 (“[A] defendant convicted of a crime committed while on parole . . . will face an additional sentence . . . .”). The second group, even if they returned to prison because of their parole violations, would not receive new sentences and thus would not encounter two separate sentencing schemes.122See id. (“[A] technical parole violator . . . will not face a new sentence, [and] after being found in violation of the conditions of parole will continue on parole, [have the parole] revoked, or be subjected to any order the Kansas parole board sees fit to enter.” (citation omitted)). Therefore, the Kansas legislature’s different treatment of these groups was “rationally related to the KSGA’s goals of consistency and proportionality, as well as public safety, by diminishing the disparity in sentences where the disparity would most prominently appear.”123Id. Perez suggests that distinguishing between parolees based on whether they simply violate the terms of their parole or instead commit new substantive crimes is rationally related to legitimate penological goals.

Collectively, these cases establish several legitimate government interests—including protecting the public from offenders of serious crimes, treating prisoners serving life sentences fairly, and minimizing sentencing disparities for repeat criminals—that legislators might seek to achieve through sentencing schemes that treat prisoners differently. All statutory provisions that the plaintiffs challenged in these cases prevailed under the courts’ application of rational basis scrutiny.

II. Analysis

The First Step Act’s eligibility provisions for earned time credits are the product of government decisions regarding which prisoners are eligible to receive earned time credits and have an opportunity to obtain early release and which prisoners are not.124See supra notes 43–48 and accompanying text (describing the debates within Congress that shaped provisions that limited the groups of prisoners who were eligible to receive earned time credits). The Constitution requires that the government have a sufficient basis for treating these groups of federal prisoners differently.125See supra note 96 and accompanying text (articulating courts’ use of the rational basis test when determining the validity of different classifications of prisoners). This Part explains why the equal protection principle of the Fifth Amendment applies to earned time credits eligibility and the basis for an equal protection claim by a § 32 prisoner who is ineligible for earned time credits. This Part then argues that the First Step Act’s provisions on earned time credits eligibility for § 32 and § 2291 prisoners fail the rational basis test because they lack any rational relationship to legitimate government interests. Finally, this Part highlights a policy consideration for whether lawmakers should base federal prisoners’ eligibility for incentives like earned time credits on the nature of their convictions.

A. The Eligibility Requirements for Earned Time Credits Under the First Step Act Invoke the Equal Protection Principle by Classifying Similarly Situated Groups Differently

The First Step Act treats prisoners convicted under 18 U.S.C. § 32 and § 2291 differently by specifying different circumstances under which the prisoners are eligible for earned time credits.126Compare 18 U.S.C.A. § 3632(d)(4)(D)(i) (West 2020) (excluding all prisoners serving sentences for convictions under § 32 from earned time credits eligibility), with 18 U.S.C.A. § 3632(d)(4)(D)(xlvi) (excluding prisoners serving sentences for convictions under § 2291 “only if the conduct that led to the conviction involved a substantial risk of death or serious bodily injury”). Ineligibility of § 2291 prisoners requires a triggering condition—that the conduct leading to the conviction involved substantial risk of death of or serious bodily injury to another person127§ 3632(d)(4)(D)(xlvi).—while ineligibility of § 32 prisoners lacks that condition.128See id. § 3632(d)(4)(D)(i) (noting that prisoners serving sentences for § 32 convictions are ineligible for earned time credits with no exceptions). As a result, prisoners who are serving sentences for violating § 32 and whose conduct did not involve substantial risk of death or serious bodily injury are ineligible for earned time credits,129See id. while prisoners serving sentences for § 2291 violations whose conduct was of the same character are eligible.130See id. § 3632(d)(4)(D)(xlvi). Thus, if the prisoners are similarly situated, the Act’s eligibility provisions that apply to them only comport with the equal protection principle of the Fifth Amendment if a rational basis for treating the groups differently exists.131See City of Cleburne v. Cleburne Living Ctr., Inc., 473 U.S. 432, 439–40 (1985) (explaining the federal government’s constitutional obligation to treat all similarly situated persons alike unless the classification “is rationally related to a legitimate state interest”).

The statutory scheme, plain text, and legislative history of § 32 and § 2291 demonstrate that prisoners convicted under those statutes whose conduct leading to their convictions was of a similar nature are similarly situated. As a preliminary matter, both statutes appear within the same title of the U.S. Code that proscribes conduct that Congress has deemed a criminal offense. Unlike the individuals that the court compared in Perez—the petitioner, who had violated the terms of his parole, and other parolees who committed substantive crimes while on parole132See Kansas v. Perez, 11 P.3d 52, 54 (Kan. 2000) (“The situation of a defendant convicted of a crime committed while on parole is different than that of a defendant returned to prison as a technical parole violator.”).—prisoners convicted under § 32 or § 2291 have all committed a substantive crime. Further, individuals found guilty of § 32 or § 2291 crimes are subject to the same range of punishments: a fine, imprisonment for not more than twenty years, or both.13318 U.S.C. §§ 32(a), 2291(a) (2018).

Sections 32 and 2291 both consist of a series of provisions specifying proscribed actions related to damage to or destruction of property employed in travel and commerce.134See id. § 32(a)(1) (defining the subject of the statute as “any aircraft in the special aircraft jurisdiction of the United States or any civil aircraft used, operated, or employed in interstate, overseas, or foreign air commerce”); § 2291(a)(1) (defining the subject of the statute as “vessel[s]”). Each statute prohibits several similar categories of conduct related to that property. Specifically, both statutes include some provisions covering conduct that involves endangerment of human life or safety and some that do not.135See supra notes 56–63, 68–73 and accompanying text (comparing a case in which a defendant was convicted under § 32 for conduct that did not endanger human life with a case in which a court sentenced a defendant under § 2291 for conduct that did endanger human life). The First Step Act acknowledges this variable with respect to § 2291 by subdividing prisoners sentenced under that statute and basing earned time credits eligibility on whether the conduct leading to the conviction “involved a substantial risk of death or serious bodily injury.”13618 U.S.C.A. § 3632(d)(4)(D)(xlvi) (West 2020).

The plain text of § 2291 confirms the existence of a basis for this distinction. For example, the government might prosecute an individual for violating § 2291 by “mak[ing] or caus[ing] to be made unworkable or unusable . . . any vessel, or any part or other materials used or intended to be used in connection with the operation of a vessel.”13718 U.S.C. § 2291(a)(2). Someone might violate this provision by removing or tampering with a component of the vessel that is essential for its operation, thereby preventing the vessel from departing.138Similarly, an individual might violate § 2291 by disabling “any appliance, structure, property, machine, or apparatus . . . used, or intended to be used, in connection with the operation . . . of any vessel,” which would also prevent the vessel from traveling. Id. § 2291(a)(5). Separately, an individual is subject to prosecution under § 2291 for “interfer[ing] by force or violence with the operation of any maritime facility . . . if such action is likely to endanger the safety of any vessel in navigation.”139Id. § 2291(a)(4). The text of this clause indicates that the provision would not apply unless the perpetrator used violence and the vessel or vessels affected by the individual’s conduct were in motion, thus with people on board.140Other provisions within § 2291 also specifically outlaw conduct that poses substantial risk of death or injury to individuals. See id. § 2291(a)(6) (involving acts of violence “likely to endanger the safety of the vessel or those on board”); § 2291(a)(7) (involving acts of violence “likely to cause serious bodily injury”); § 2291(a)(8) (involving communication of false information that “endanger[s] the safety of any vessel in navigation”). The situation contemplated by the former provision is less likely to involve substantial risk of harm to human life or safety than the situation that would invoke the latter provision. Thus, the First Step Act’s eligibility scheme for prisoners convicted under § 2291 reflects realistic distinctions in conduct leading to such convictions.

The text of § 32, as in § 2291, reflects provisions that differ with respect to whether the proscribed conduct endangers human life or safety. Section 32 covers conduct related to “any aircraft in the special aircraft jurisdiction of the United States or any civil aircraft used, operated, or employed in interstate, overseas, or foreign air commerce.”141§ 32(a)(1) (emphasis added). Aircraft within the “special aircraft jurisdiction of the United States” are defined as aircraft that are in flight.14249 U.S.C. § 46501 (2012) (using § 46051’s definition in § 31 and defining “special aircraft jurisdiction of the United States” as including five categories of aircraft, all of which are “in flight”). Hence, any conduct that § 32 describes, if performed in relation to an aircraft within the special aircraft jurisdiction of the United States, necessarily involves substantial risk of death or serious bodily injury to those on board the aircraft. However, courts have interpreted the second category of aircraft to which § 32 conduct applies as including aircraft that are not necessarily in flight at the time of the conduct.143See supra note 59 (highlighting a case interpreting § 32 to include conduct committed while an aircraft is on the ground). Hence, if the conduct leading to conviction under § 32 involved a “civil aircraft used, operated, or employed in interstate, overseas, or foreign air commerce,” the conduct would not necessarily have posed risks to human life or safety.144§ 32(a)(1).

For example, the government might charge an individual with violating § 32 if the individual “disables . . . any civil aircraft used, operated, or employed in interstate, overseas, or foreign air commerce” that is not in flight at the time the individual acts.145Id. Just as someone might tamper with a stationary vessel and make it incapable of operation, a person could tamper with equipment in a stationary aircraft at its storage location to prevent it from operating. Either scenario poses a limited, if any, risk of death or serious bodily injury to anyone.146Section 32 also has a provision that proscribes “set[ting] fire to, damag[ing], destroy[ing], or disabl[ing] or plac[ing] a destructive device or substance in, upon, or in proximity to, any appliance or structure, ramp, landing area, property, machine, or apparatus, or any facility or other material used, or intended to be used, in connection with the operation, maintenance, loading, unloading, or storage of any . . . aircraft.” § 32(a)(4). This is nearly identical to § 2291(a)(5). See supra note 67. Neither § 32(a)(4) nor § 2291(a)(5) require that the defendant’s conduct involved substantial risk of death or serious bodily injury for the government to prosecute the defendant. David Cain Jr.’s role in the destruction of his competitor’s vintage airplane by arson,147David Cain, Jr. Sentenced to 55 Years, supra note 4. while a more extreme example, also qualifies as a § 32 violation in which the underlying conduct did not involve substantial risk of death or serious bodily injury, as nobody was on board or near the stationary airplane in its hangar at the time it burst into flames.148Supra note 62.

The legislative history of § 32 and § 2291 also suggests that Congress viewed individuals who commit crimes under those statutes as similarly situated. While Congress passed the two laws nearly fifty years apart,149Congress originally enacted § 32 in 1956. 18 U.S.C. § 32. It later established § 2291 as part of the USA PATRIOT Improvement and Reauthorization Act of 2005. Yeh & Doyle, supra note 63, at 37. the laws had the shared purpose of combatting violent acts in transportation settings, whether an incident like the Longmont, Colorado explosion on a commercial airline150Supra note 53. or the 9/11 terrorist attacks.151Supra note 66 and accompanying text. This shared intent for the laws shows that the textual similarities between § 32 and § 2291 are no accident: recognizing that the federal government lacked sufficiently punitive laws related to conduct on vessels, Congress intentionally modeled § 2291 after § 32.152See supra notes 68–69 and accompanying text. Because the “general[] require[ment]” of § 32 convictions that the conduct leading to the conviction impact the safe operation of the aircraft is not required in all cases, Yeh & Doyle, supra note 65, at 38–39, the lesser likelihood that § 2291 convictions require impact to safe operation of a vessel is not dispositive of whether § 32 and § 2291 convictions could each involve conduct that does not impact operational safety.

The inclusion of § 32 and § 2291 in the federal criminal code, textual parallels between the two statutes, and legislative history of each law suggest that prisoners serving sentences for violations of either statute whose conduct did not involve a substantial risk of death or serious bodily injury are similarly situated. Therefore, the equal protection principle applies to the earned time credits eligibility provisions for prisoners serving sentences under § 32 and § 2291, and the provisions must comport with rational basis requirements.

B. The First Step Act’s Earned Time Credits Eligibility Provisions Fail the Rational Basis Test and Violate the Fifth Amendment’s Guarantee of Equal Protection

No existing cases that have considered equal protection challenges to sentencing statutes can be used to justify the First Step Act’s dissimilar treatment of prisoners serving sentences under § 32 and § 2291. The Act’s provisions on these prisoners’ eligibility for earned time credits are not rationally related to the government interests that have been recognized in prior cases.

Convictions under § 32 and § 2291 can be equally as serious—or “harmless”—with respect to human safety. As the numerous provisions within each statute demonstrate, conduct leading to conviction under either statute might be tremendously dangerous to other people and necessarily involve substantial risk of death or serious bodily injury—such as Abdi Wali Dire’s attack of a Navy vessel with gunfire in United States v. Dire153United States v. Dire, 680 F.3d 446, 449 (4th Cir. 2012).—or it might involve the disabling of an aircraft or vessel under less risky circumstances—such as David Cain Jr.’s setting fire to a stationary aircraft while it sat unoccupied in its hangar in United States v. Cain.154United States v. Cain, 671 F.3d 271, 297–98 (2d Cir. 2012). This Comment discusses other examples of § 32 conduct that would not involve substantial risk of death or serious bodily injury in Part I.B.

Congress equated the seriousness of crimes under § 32 and § 2291 by incorporating identical punishments for the crimes.155See 18 U.S.C. §§ 32(a), 2291(a) (2018) (imposing the same punishment of a fine, imprisonment for not more than twenty years, or both). Unlike Robinson, in which the prisoners that the statute treated differently were serving sentences of different maximum lengths for crimes of varying severity,156Robinson v. Deloach, No. 1:06cv1061-MHT, 2009 WL 1116838, at *4 (M.D. Ala. Apr. 24, 2009). individuals imprisoned for violations of § 32 and § 2291 may be serving identical sentences for conduct involving equal risk of death or serious bodily injury or lack thereof. The court in Robinson easily identified a rational relationship between the decision to foreclose sentence reductions to prisoners serving sentences for more serious offenses and the goal of maintaining public safety when releasing prisoners.157See id. (“[T]he ACITA’s categorization of prisoners based on the length of their sentences is rationally related to the legitimate state interest of preventing the early release of serious offenders.”). If a prisoner sentenced under § 32 for conduct that did not involve a substantial risk of death or serious bodily injury challenged the First Step Act’s earned time credits eligibility provisions, the reviewing court would not be able to draw the same conclusion to justify the First Step Act that the court in Robinson drew to uphold the ACITA.

Beyond serving as a proxy for the potentially equal seriousness of the crimes, the identical punishment provisions for convictions under § 32 and § 2291 also preclude justifying different treatment by the nature of the sentence, as the Robinson court did in resolving the petitioner’s second claim.158See id. at *3 (upholding an Alabama law that made prisoners eligible for sentence reductions based on whether they were serving a life sentence or not). While the Alabama legislature’s decision to make sentence reductions available to prisoners who had been sentenced to life imprisonment under previously mandatory guidelines was rationally related to the desire to achieve more equitable sentencing for the state’s prisoners,159Id. no such argument can be made for offering different opportunities to prisoners serving identical sentences.

The holding in Perez, which upheld different sentencing treatment of parolees based on actions they commit while on parole, does not apply to the First Step Act’s provisions that classify prisoners for earned time credits availability based on their original crime.160Kansas v. Perez, 11 P.3d 52, 54–55 (Kan. 2000). In Perez, the court explained that parolees serving sentences for prior crimes can be given different opportunities to reduce their sentences based on whether they commit a crime while on parole or merely violate the terms of their parole.161See id. at 54 (holding that the distinction between parolees who violate their parole and who commit additional crimes is “reasonably related to legitimate state interests”). In contrast, two prisoners that face sentences for distinct substantive crimes—one under § 32 and one under § 2291—should have the same opportunities to reduce their time in prison assuming they are also similarly situated in all other material respects.

To defend its different treatment of § 32 prisoners and § 2291 prisoners, the government might argue that the exception to earned time credits ineligibility for § 2291 that it carved out would never apply to § 32 prisoners. To make this argument, the government would need to demonstrate that some convictions under § 2291 might not involve a substantial risk of death or serious bodily injury, while all convictions under § 32 will.162See supra notes 49–50 and accompanying text. Only some of the provisions of § 2291—those that do not include as an element substantial likelihood of risk to human life or safety163See 18 U.S.C. § 2291(a)(1)–(3), (5), (9) (2018) (listing provisions lacking the requirement that the conduct jeopardizes the safety of the vessel or those on board).—would cover such circumstances.

Here is where the problem with this argument would arise: the provisions of § 2291 that contemplate conduct that does not involve substantial risk to human life or safety also appear in § 32.164See id. § 32(a)(1), (4), (8) (delineating provisions lacking the requirement that the conduct jeopardizes the safety of the aircraft or those on board). Compare 18 U.S.C. § 2291(a)(5) (prohibiting damaging, destroying, or disabling any appliance or structure used for vessel operation, maintenance, or storage), with 18 U.S.C. § 32(a)(4) (outlawing the same activities with respect to aircraft). By arguing that § 2291 applies in situations with no such risks, the government must simultaneously acknowledge that § 32 applies in similar situations. Hence, the government cannot defend an exception for § 2291 prisoners’ ineligibility by asserting that the exception would be applicable for them but inapplicable for § 32 prisoners.

To counter this analysis, the government might further contend that while some § 32 provisions do not facially require substantial risk of death or serious bodily injury, the conduct these provisions proscribe would implicitly involve such harm. Using David Cain Jr.’s conviction as an example,165See United States v. Cain, 671 F.3d 271, 298 (2d Cir. 2012) (upholding Cain’s § 32 conviction). the government could argue that arson that targets an aircraft is necessarily more dangerous to human life or safety than arson that targets a vessel, and that Cain’s arson of his competitor’s aircraft endangered the lives of the people who, for example, attempted to put the fire out. But an arson of a vessel would be equally, if not potentially more, dangerous due to the size of vessels—like those belonging to the U.S. Navy, the subject of Dire.166United States v. Dire, 680 F.3d 446, 449 (4th Cir. 2012). Thus, any implicit substantial risk of death or serious bodily injury that the provisions of § 32 include would also necessarily be encompassed by the parallel provisions of § 2291.

Application of the factors that the Supreme Court articulated in Turner v. Safley167See supra note 93 and accompanying text (listing the factors that the Court identified as central to rational basis review of laws impacting prisoners’ rights). underscores the failure of the First Step Act’s earned time credits eligibility provisions to overcome rational basis scrutiny. As discussed, it is difficult to discern a rational connection between the eligibility provisions for § 32 and § 2291 prisoners and the government’s interest in balancing rehabilitation opportunities for prisoners with public safety.168See supra notes 163–64 and accompanying text. Prisoners to whom the First Step Act denies earned time credits are unable to exercise their rights through other means, as the Act is the first of its kind to establish such a system. Furthermore, accommodating § 32 prisoners’ asserted rights to earned time credits if their conduct is similar to the conduct that makes § 2291 prisoners eligible would not significantly impact criminal justice system resources for two reasons: (1) § 32 prisoners are eligible to participate in recidivism reduction programs regardless of whether they can earn time credits from participation; and (2) the small number of § 32 prisoners who might be eligible for transition to prerelease custody or supervised release would have a negligible impact on the prison system’s monitoring of individuals in these environments. Finally, the government can arguably best ensure a balance between rehabilitation opportunities for prisoners and public safety by reallocating resources from incarcerating all § 32 prisoners to monitoring those prisoners as they transition out of incarceration and into a different type of sentence.

The rational basis test requires the most deference to the legislature of any other equal protection review.169See City of Cleburne v. Cleburne Living Ctr., Inc., 473 U.S. 432, 440 (1985) (explaining the general rule of presuming legislation to be valid when it is rationally related to a legitimate state interest). But the test does not require courts to turn a blind eye to arbitrary legislative decisions or oversights in lawmaking that threaten individuals’ constitutional rights,170See id. at 446 (“The [government] may not rely on a classification whose relationship to an asserted goal is so attenuated as to render the distinction arbitrary or irrational.”). like the earned time credits eligibility provisions under the First Step Act for § 32 and § 2291 prisoners.

C. Denying Some Prisoners the Same Opportunities as Others Based on Underlying Convictions Might Have Only a Minimal Impact on Recidivism

Even if the government could overcome the equal protection issues discussed in this Comment by arguing that § 32 prisoners are categorically more dangerous than prisoners who are eligible for earned time credits, its decision to limit eligibility to only the least dangerous prisoners who it views as capable of rehabilitation might be flawed.171See, e.g., Letter from the Leadership Conference on Civil and Human Rights, to the House Judiciary Comm. (May 8, 2018) [hereinafter Vote “No” Letter], https://civilrights docs.info/pdf/policy/letters/2018/not-retroactive-Sign-On-Letter-Oppose-First%20Step %20Act-5.8.18-FINAL.pdf [https://perma.cc/GV85-HSJT] (urging members of Congress to incorporate broader rehabilitative opportunities for prisoners into the First Step Act).

A prisoner’s eligibility for earned time credits requires that the risk and needs assessment system has determined that the prisoner has a low risk of recidivating, or that a prison warden has determined that the prisoner is unlikely to recidivate.172See 18 U.S.C.A. § 3624(g)(1)(D) (West 2020); supra note 23 (explaining the creation and implementation of the risk and needs assessment system). Congress’s exclusion of some prisoners from eligibility represents its belief that those prisoners are incapable of ever obtaining a status as unlikely to recidivate.173See 164 Cong. Rec. S7641 (daily ed. Dec. 17, 2018) (statement of Sen. Cornyn) (“There are some people, sadly, who will never take advantage of the opportunity to transform their lives through [rehabilitative] programs . . . in other words, there are some people, unfortunately, we can’t save, but there are others who . . . want to turn their lives around. Those are the type of people this criminal justice reform bill speaks to.”). However, data suggests that “effectively reducing recidivism requires focusing programs, jobs, and real and meaningful incentives [to rehabilitate] on those most likely, not least likely, to reoffend.”174Vote “No” Letter, supra note 171; cf. Shon Hopwood, Second Looks & Second Chances, 41 Cardozo L. Rev. 83, 111–12 (2019) (“Most people convicted of violent or sex offenses, which represents a sizable number of people in federal prison, will one day be released. It is thus incumbent on Congress to incentivize their participation in behavioral therapy and educational programs designed to reduce recidivism.” (footnotes omitted)).

As long as criminal justice reform limits rehabilitation incentives for a significant number of prisoners—based on their status as a violent offender or other criteria like immigration status175See 164 Cong. Rec. H10399 (daily ed. Dec. 20, 2018) (letter from the ACLU and The Leadership Conference on Civil and Human Rights) (“The continued exclusion of immigrants from the many benefits of the bill [including earned time credits] simply based on immigration status is deeply troubling.”).—it may be unlikely to reduce recidivism. Thus, setting aside any legal issues with the First Step Act’s earned time credits eligibility provisions, it might be worthwhile for Congress to further consider the policy debate and available research on the benefits of focusing rehabilitative programs on offenders of more serious crimes.176But cf. Reginald Dwayne Betts, Could an Ex-Convict Become an Attorney? I Intended to Find out, N.Y. Times Mag. (Oct. 16, 2018), https://www.nytimes.com/2018/ 10/16/magazine/felon-attorney-crime-yale-law.html [https://perma.cc/Y694-L994] (suggesting that, regardless of an individual’s prior conviction or efforts to successfully reenter society—as rehabilitative programs are designed to facilitate—a status as a convicted felon significantly inhibits the individual’s opportunities to work, vote, rent property, and otherwise live the way non-felons do).

Conclusion

The First Step Act represents a bipartisan agreement on needed changes to the criminal justice system and a compromise that many thought was impossible.177See Grawert & Lau, supra note 24 (describing the outlook at the beginning of President Trump’s presidency that federal criminal justice reform would face too many challenges to succeed). Concerns about public safety reasonably arose as lawmakers considered offering some prisoners the significant benefit of reducing their time in prison.178See supra notes 43–47 and accompanying text (quoting members of Congress who have expressed concern about allowing violent criminals to be released from prison early). Basing earned time credits eligibility for § 2291 prisoners on whether their conduct posed substantial risks of death or serious bodily injury appears to be a logical consequence of public safety concerns. However, by not making the same distinction for prisoners serving sentences for violating § 32, the government has unconstitutionally treated one group of prisoners differently than another without a rational basis.

Individuals who have been convicted of crimes under § 32 and § 2291 are similarly situated in all material respects. They have all committed crimes that Congress has written into the U.S. criminal code,17918 U.S.C. §§ 32, 2291 (2018). rather than breaking other rules such as parole requirements. The statutes’ text and cases interpreting the statutes show that conduct that does not pose a substantial risk of death or serious bodily injury is possible within the context of each, and thus that the underlying conduct could involve equally minimal harm to others.180See supra Part II.B. The range of punishment for acts under each statute is identical.181§§ 32(a), 2291(a). Therefore, the various arguments that courts upheld as justifications for other sentencing schemes are insufficient to support the constitutionality of the First Step Act provisions that this Comment discusses.

Furthermore, the specific arguments that the government would likely put forth to frame the equal protection issue in this context would also be inadequate. The government cannot reasonably argue, and a court could not find, that § 32 offenses are inherently more dangerous than § 2291 offenses without ignoring the textual parallels between the two statutes. The extremely small percentage of federal prisoners who are serving sentences for § 32 convictions precludes reliance on the argument that the First Step Act’s earned time credits eligibility provisions are permissibly imperfect by including too few people in its earned time credits eligibility scheme.

The First Step Act’s unequal treatment of prisoners convicted under § 32 and § 2291 is arbitrary, contravening the equal protection principle of the Fifth Amendment. As momentum for criminal justice reform continues, Congress must protect the constitutional rights of the prisoners to which potential reforms apply and continue to consider the effectiveness of its policy of foreclosing rehabilitative opportunities to prisoners deemed too hopeless to save.

Share this post