In 2003, the Supreme Court rejected a constitutional challenge to the Alaska Sex Offender Registration Act,1×1. 1994 Alaska Sess. Laws ch. 41, § 4 (codified at Alaska Stat. §§ 12.63.010–.100 (2018)). finding that its requirement that people who have committed sex crimes register with law enforcement does not constitute punishment.2×2. See Smith v. Doe, 538 U.S. 84, 89, 105–06 (2003). Since then, laws regulating sex offenders have multiplied and expanded. In many states, these laws affect virtually every aspect of the lives of those subject to them, from where they can live and work,3×3. E.g., Ga. Code Ann. § 42-1-15(a)–(c) (2014) (prohibiting some registered sex offenders from living or working within 1000 feet of any “child care facility, church, school, or area where minors congregate,” id. at § 42-1-15(b)); Va. Code Ann. § 46.2-116(C) (2017) (prohibiting people required to register as sex offenders from obtaining registrations to drive tow trucks). to their freedom on the internet,4×4. E.g., N.Y. Exec. Law § 259-c(15) (McKinney 2018) (prohibiting some sex offenders on parole from “using the internet to . . . access a commercial social networking website”). to how they can spend their holidays.5×5. E.g., 720 Ill. Comp. Stat. 5/11-9.3(c-2) (2017) (“It is unlawful for a child sex offender to participate in a holiday event involving children under 18 years of age . . . .”). Recently, in Vasquez v. Foxx,6×6. 895 F.3d 515 (7th Cir. 2018), cert. denied, 139 S. Ct. 797 (2019). the Seventh Circuit upheld an amended Illinois law that restricts where those convicted of sex offenses involving minors can reside, dismissing an as-applied argument that it violates the Ex Post Facto Clause.7×7. Id. at 518, 522. The Constitution contains two clauses barring ex post facto laws, one at the federal level, see U.S. Const. art. I, § 9, cl. 3, and one at the state level, see id. § 10, cl. 1. In this case, the court cited the former. Vasquez, 895 F.3d at 520 n.3 (citing U.S. Const. art. I, § 9, cl. 3). The court merged two of the five independent factors laid out in the relevant test from Smith v. Doe8×8. 538 U.S. 84 (2003). and thereby functionally eliminated any independent inquiry into the consequences and scope of the statute. In so doing, it set a lower bar for states to meet in defending even the most onerous restrictions on those convicted of sex offenses.
The Illinois legislature enacted a law in 2000 restricting where “child sex offenders” could live.9×9. Vasquez, 895 F.3d at 518; 720 Ill. Comp. Stat. 5/11-9.3(b-10). Defining this term through a list of enumerated offenses, most of which involve sex-based crimes with minor victims,10×10. 720 Ill. Comp. Stat. 5/11-9.3(d)(2.5). the statute made it unlawful for anyone who qualified as a child sex offender to “knowingly reside within 500 feet of a playground or a facility providing programs or services exclusively directed toward persons under 18 years of age.”11×11. Vasquez, 895 F.3d at 518 (quoting Act of July 7, 2000, Pub. Act No. 91-911, 2000 Ill. Laws 2051 (codified as amended at 720 Ill. Comp. Stat. 5/11-9.3(b-10))). Eight years later, the legislature amended the statute, adding day-care homes to the list of places included.12×12. Id. at 518; see 720 Ill. Comp. Stat. 5/11-9.3(b-10). Violating the restriction is punishable by up to three years in prison.13×13. 720 Ill. Comp. Stat. 5/11-9.3(f); 730 Ill. Comp. Stat. 5/5-4.5-45(a) (2017).
The Vasquez plaintiffs, Joshua Vasquez and Miguel Cardona, both qualified as “child sex offenders” in this statutory scheme.14×14. Vasquez, 895 F.3d at 518–19. The two men were convicted for possession of child pornography and indecent solicitation of a child, respectively. Id. at 518. In August 2016, both men completed their annual sex offender registration with the Chicago Police Department and were told that a child day-care home had opened within 500 feet of their established residences.15×15. Id. at 518–19. Because the 2008 amendment was enacted before they acquired their residences, both men were subject to the statute’s restrictions.16×16. Id. Police officers told them that they were required to move within thirty days or risk arrest and prosecution.17×17. Brief of Plaintiffs-Appellants Joshua Vasquez and Miguel Cardona at 4–6, Vasquez, 895 F.3d 515 (No. 17-1061).
In response, Vasquez and Cardona brought suit in the Northern District of Illinois against Kimberly M. Foxx, in her capacity as the State’s Attorney of Cook County, and the City of Chicago.18×18. Vasquez v. Foxx, No. 16-CV-8854, 2016 WL 7178465, at *1–2 (N.D. Ill. Dec. 9, 2016). Among their claims, they argued that the application to them of the day-care home rule violated the Ex Post Facto Clause, since they were convicted for their triggering offenses before this restriction was in place.19×19. Id. at *2. The plaintiffs also brought procedural due process, Takings Clause, and substantive due process challenges. Id. at *2, *3–9. The defendants moved to dismiss the plaintiffs’ complaint.20×20. Id. at *1. Granting the defendants’ motion, the district court held, in regard to the ex post facto claim, that the law did not operate retroactively and therefore did not run afoul of the Ex Post Facto Clause.21×21. Id. at *4–5.
The Seventh Circuit affirmed on all counts.22×22. The focus here will be on the plaintiffs’ ex post facto claim. Vasquez, 895 F.3d at 520–22. On the other claims, the court ruled that (1) the plaintiffs had failed to exhaust state court remedies for their Takings Clause argument, id. at 523, and the restrictions did not constitute a taking, id. at 524; (2) the lack of an individualized hearing did not violate the plaintiffs’ procedural due process rights, id.; and (3) the plaintiffs’ substantive due process rights were not violated because the statute passed rational basis review, id. at 525. Writing for the panel, Judge Sykes23×23. Judge Sykes was joined by Judges Bauer and Rovner. noted that “a statute is not an impermissible ex post facto law unless it is both retroactive and penal.”24×24. Vasquez, 895 F.3d at 520. First, the court upheld the district court’s conclusion that the penalties for violating the residency restriction did not apply retroactively.25×25. Id. Judge Sykes explained that although the amendment applied to people whose relevant convictions occurred before 2008, only conduct that occurred after its enactment was punishable.26×26. Id. Accordingly, she concluded that the statute “merely create[d] new, prospective legal obligations based on the person’s prior history.”27×27. Id. (quoting United States v. Leach, 639 F.3d 769, 773 (7th Cir. 2011)).
The court then considered whether the statute was penal in nature. To do this, it analyzed whether the restrictions effectively constituted punishment for the underlying offense, such that they functioned as a criminal penalty imposed after conviction.28×28. Id. at 520–21. It initially rejected this claim out of hand,29×29. Id. at 521. as it found the residency statute to be substantially similar to other statutes regulating sex offenders that were deemed nonpenal by the Seventh Circuit30×30. Id. (citing Leach, 639 F.3d at 773). and the Supreme Court.31×31. Id. (citing Smith v. Doe, 538 U.S. 84, 105–06 (2003)).
However, it then went on to independently analyze the statute, applying the Supreme Court’s Smith test to determine whether its restrictions were in fact punitive.32×32. Id. at 521–22. Under this frame, courts first ask whether the legislature intended for a law to impose punishment.33×33. See Smith, 538 U.S. at 92–93. Because the plaintiffs did not make this argument,34×34. Vasquez, 895 F.3d at 521. the court moved on to the second inquiry: whether the law was so punitive in effect as to negate the state’s nonpunitive intent.35×35. Id. (citing Smith, 538 U.S. at 92). To assess the law’s effect, the panel considered the five factors from Smith: whether the statute “[(1)] in its necessary operation . . . [would be] regarded in our history and traditions as a punishment[, (2)] imposes an affirmative disability or restraint[, (3)] promotes the traditional aims of punishment[, (4)] has a rational connection to a nonpunitive purpose[, or (5)] is excessive with respect to this purpose.”36×36. Id. (second alteration in original) (quoting Smith, 538 U.S. at 97).
The court first examined whether the residency law resembled traditional methods of punishment. It rejected the plaintiffs’ comparison of the law with the historical punishments of shaming and banishment, observing that the law’s restrictions did not inflict public disgrace, which is the purpose of shaming, and that, unlike banishment, they did not force offenders to permanently or completely leave their communities.37×37. Id. But see People v. Tucker, 879 N.W.2d 906, 921 (Mich. Ct. App. 2015) (finding similar residency restrictions resembled banishment); Corey Rayburn Yung, Banishment by a Thousand Laws: Residency Restrictions on Sex Offenders, 85 Wash. U. L. Rev. 101, 135–37 (2007) (arguing the same). The court additionally noted that, since the restrictions limited only where those subject to them can live, they did not sufficiently resemble the “comprehensive control” of criminal punishments like probation and supervised release.38×38. Vasquez, 895 F.3d at 521.
The court next asked whether the law imposed an affirmative disability or restraint. It conceded that the plaintiffs had difficulty finding compliant housing and acknowledged that the 2008 amendment’s inclusion of day-care homes compounded this difficulty, since “a private residential property can become a day-care home without anyone in the neighborhood noticing.”39×39. Id. at 522. Nevertheless, the court concluded that the restriction failed to meet the standard of functioning as a “paradigmatic” disability or restraint because it did not impose a physical restraint resembling imprisonment.40×40. Id. (quoting Smith, 538 U.S. at 100). But see Doe v. Miller, 405 F.3d 700, 720–21 (8th Cir. 2005) (holding that, while imprisonment is the “paradigmatic” disability or restraint, other restraints can also be considered under this factor and concluding that a residency restriction does impose such a restraint); Commonwealth v. Baker, 295 S.W.3d 437, 445 (Ky. 2009) (same).
To determine whether the statute promoted the traditional aims of punishment, the court assessed whether it was enacted for retributive purposes.41×41. Vasquez, 895 F.3d at 522. In finding retribution to be the only relevant traditional aim of punishment, the court parted ways with at least two circuit courts, which have also considered deterrence in this prong. See, e.g., Does #1–5 v. Snyder, 834 F.3d 696, 704 (6th Cir. 2016); Miller, 405 F.3d at 720. It rejected any argument to this effect, finding that “the obvious aim of the statute [was] to protect children from the danger of recidivism by convicted child sex offenders.”42×42. Vasquez, 895 F.3d at 522. But see Snyder, 834 F.3d at 704 (holding that a law limiting where registrants can live and work is “retributive in that it looks back at the offense (and nothing else) in imposing its restrictions, and it marks registrants as ones who cannot be fully admitted into the community”).
Finally, the court considered whether the statute had a rational connection to a nonpunitive purpose and was excessive with respect to that purpose. Because these two factors “are related,” it analyzed them together.43×43. Vasquez, 895 F.3d at 522. Explaining that the plaintiffs were “required to show that the statute’s ‘nonpunitive purpose is a sham or mere pretext,’”44×44. Id. (quoting Smith, 538 U.S. at 103). the court concluded that they failed to do so. Though it noted the plaintiffs’ assertion that those convicted of sex offenses do not recidivate at higher rates than those convicted of other offenses, it argued that, even if this were true, “similar recidivism rates across different categories of crime would not establish that the nonpunitive aim of this statute — protecting children — is a sham.”45×45. Id.
Taking all the factors together, the court concluded that the amended residency restriction was “neither retroactive nor punitive and thus raise[d] no ex post facto concerns.”46×46. Id. Accordingly, it affirmed the district court’s dismissal of the ex post facto claim.47×47. Id.
The Seventh Circuit’s simultaneous consideration of the last two factors from Smith led it to incorrectly conflate them and to neglect the latter. This conflation matters not just as a doctrinal issue, but also for its implications in future cases involving the regulation of those convicted of sex offenses. If courts do not independently inquire into whether a given sex offender statute is excessive, they will be more likely to uphold nearly all such statutes, even the most far-reaching, without confronting the actual effects of the restrictions they impose.
In Smith, the Supreme Court analyzed the sex offender registration statute at issue by conducting a sequential five-factor analysis.48×48. Smith, 538 U.S. at 97. Unlike Judge Sykes, Justice Kennedy applied each of the factors to determine punitive effect independently, considering the statute under each one before moving to the next.49×49. See id. at 97–105. In addressing the fourth factor, he concluded that the statute had a rational connection to the nonpunitive purpose of public safety.50×50. Id. at 102–03. He noted that a statute need not have a close or perfect fit with its nonpunitive purpose (dismissing the respondents’ contention that restrictions must be narrowly drawn), so long as that purpose is not a “sham or mere pretext.”51×51. Id. at 103 (quoting Kansas v. Hendricks, 521 U.S. 346, 371 (1997) (Kennedy, J., concurring)).
After addressing this argument, the Court moved clearly to the following factor: whether the statute was excessive in relation to its regulatory purpose.52×52. Id. Justice Kennedy considered and rejected the argument that the regulatory scheme was excessive because it “applie[d] to all convicted sex offenders without regard to their future dangerousness.”53×53. Id. He did so not because the legislature had proven that its rationale for imposing restrictions on this class of offenders was not a sham, but because it could conclude, based on a survey of available data, that “a conviction for [any] sex offense provides evidence of substantial risk of recidivism.”54×54. Id. But see Ira Mark Ellman & Tara Ellman, “Frightening and High”: The Supreme Court’s Crucial Mistake About Sex Crime Statistics, 30 Const. Comment. 495, 496–505, 508 (2015) (discussing how Justice Kennedy’s contention that the recidivism rate of those convicted of sex crimes is “frightening and high” is unsupported by any data and is, in fact, disproven by statistical studies). Not only did the Court separate out these two inquiries, but it also held the state to a different standard for each one: where the rational relation test requires only that a legislature’s justification rise above a mere “sham,” the excessiveness test involves a closer, more rigorous look at the state’s empirical justification.
While purporting to apply the multifactor Smith approach, the Vasquez court combined and conflated the final two factors. Noting that “[t]he last two factors in the Smith framework are related,” Judge Sykes merged their consideration, first explaining that “[a]t this step of the analysis, the challenger is required to show that the statute’s ‘nonpunitive purpose is a sham or mere pretext.’”55×55. Vasquez, 895 F.3d at 522 (quoting Smith, 538 U.S. at 103). In so doing, she applied the “sham or mere pretext” test to her analysis of both remaining inquiries: whether the statute had a rational connection to a nonpunitive purpose and whether it was excessive with respect to that purpose. She was then able to reject the plaintiffs’ argument that those convicted of sex offenses do not reoffend more than those convicted of other offenses, which resembled the argument that Justice Kennedy addressed under Smith’s excessiveness analysis, because “similar recidivism rates across different categories of crime” did not necessarily indicate that the nonpunitive purpose “of this statute — protecting children — is a sham.”56×56. Id.
Completing its melding of these two factors, the Seventh Circuit concluded by quoting language from Smith’s excessiveness assessment: “Indeed, Smith holds that states may make ‘reasonable categorical judgments . . . without any corresponding risk assessment.’”57×57. Id. (omission in original) (quoting Smith, 538 U.S. at 103–04). In this final move, the court implied that an inquiry into whether and how categorical judgments about certain types of offenders are made in a statutory scheme should go into the initial determination of whether the legislature’s purpose was a sham. Smith, however, envisions such questions about a statute’s scope and methods as primarily part of an analysis of whether that statute is excessive in relation to its purposes.58×58. See Smith, 538 U.S. at 103–05. The effect of this conflation is to eliminate any independent inquiry into whether a statute, even if rationally related to a nonpunitive end, is excessive in its restrictions.
Eliminating this analysis of a statute’s excessiveness not only constitutes a doctrinal misstep, but also makes it easier for states to pass restrictions on those convicted of sex offenses. In defending restrictions imposed on those convicted of offenses against minors in particular, state actors will always have a compelling nonpunitive purpose to put forward: protecting children from sexual harm.59×59. See, e.g., Doe v. Miller, 405 F.3d 700, 714, 721 (8th Cir. 2005); see also Lee Edelman, No Future: Queer Theory and the Death Drive 2 (2004) (positing that perhaps no value is “so unquestioned, because so obviously unquestionable, as that of the Child whose innocence solicits our defense”). Moreover, courts have generally accepted that measures like residency restrictions have a rational relationship to this purpose, as they might prevent some people targeted from reoffending.60×60. See, e.g., Miller, 405 F.3d at 721–23. But see Does #1–5 v. Snyder, 834 F.3d 696, 704–05 (6th Cir. 2016) (noting the lack of proof that residency restrictions “have any beneficial effect on recidivism rates,” id. at 705, and citing studies concluding the opposite); Brief of Eighteen Scholars as Amici Curiae in Support of Petitioners at 1–26, Vasquez v. Foxx, 139 S. Ct. 797 (2019) (No. 18-386) (noting that, while “[t]he importance of the state’s interest in the safety of children is beyond doubt,” id. at 26, there is no empirical data suggesting that residency restrictions help accomplish that goal). In these cases, a separate analysis of excessiveness can act as an important chance for the court to take a second, closer look at the scope and impact of these statutes, as the Supreme Court did in Smith.61×61. See Smith, 538 U.S. at 103–05; see also Miller, 405 F.3d at 721–23; State v. Pollard, 908 N.E.2d 1145, 1153 (Ind. 2009). Indeed, though the majority in Smith ultimately found that the statute at issue was not excessive and was nonpunitive in effect,62×62. Smith, 538 U.S. at 105–06. Justice Ginsburg in dissent disagreed and explained that “[w]hat ultimately tips the balance for me is the Act’s excessiveness in relation to its nonpunitive purpose.”63×63. Id. at 116 (Ginsburg, J., dissenting).
An excessiveness inquiry might have helped the Vasquez court understand the Illinois statute’s effects on those subject to it. First, though Smith held that the legislature is entitled to make “reasonable categorical judgments that conviction of specified crimes should entail particular regulatory consequences,”64×64. Id. at 103–04 (majority opinion). and though a court might accept the controversial claim that the risk of recidivism among those convicted of sex offenses is “frightening and high,”65×65. Id. at 103 (quoting McKune v. Lile, 536 U.S. 24, 34 (2002) (plurality opinion)). But see Ellman & Ellman, supra note 54, at 496–505, 508. the statute encompasses a particularly broad class. Most notably, those subject to its restrictions include people who have been convicted of crimes that are not necessarily sexually motivated or likely to be associated with high rates of recidivism, including kidnapping and unlawful restraint.66×66. See 720 Ill. Comp. Stat. 5/11-9.3(d)(2.5)(iii) (2017). Both crimes are defined quite broadly and can include, for example, actions taken in the course of an armed robbery, People v. Lee, 876 N.E.2d 671, 674–75, 688 (Ill. App. Ct. 2007), or a carjacking, People v. Fuller, 756 N.E.2d 255, 257 (Ill. App. Ct. 2001). See also 720 Ill. Comp. Stat. 5/10-1 to 5/10-3.1 (defining kidnapping and unlawful restraint). If minors are involved in the commission of these offenses, the residency restriction applies. Id. 5/11-9.3(d)(2.5)(iii). Further, though the plaintiffs were unable to develop this issue through discovery, it is quite likely that the restriction, especially after its 2008 expansion, would prevent those affected from living in large swaths of Chicago67×67. See Brief of Plaintiffs-Appellants Joshua Vasquez and Miguel Cardona, supra note 17, at 21 n.6 (attesting to the difficulty of locating compliant residences under the restrictions); id. at 15 (noting that there are over 2600 licensed day-care providers in Chicago alone, not even considering the other establishments that trigger the residency restrictions). : in other residency restriction cases with more developed factual records, plaintiffs have demonstrated that restrictions like the Illinois one can make it nearly impossible to live in densely populated urban areas,68×68. See, e.g., Commonwealth v. Baker, 295 S.W.3d 437, 447 (Ky. 2009) (noting that a 1000-foot residency restriction “becomes a serious burden” in metropolitan areas); In re Berlin v. Evans, 923 N.Y.S.2d 828, 835 (Sup. Ct. 2011) (finding that a 1000-foot residency restriction “effectively . . . banished [the petitioner] from Manhattan”); see also Beth Schwartzapfel & Emily Kassie, Banished, Marshall Project (Oct. 3, 2018, 7:00 AM), https://www.themarshallproject.org/2018/10/03/banished [https://perma.cc/4H7M-3K66] (documenting registrants in Miami who have been made homeless by residency restrictions). particularly for those who are low income or who have healthcare needs.69×69. See, e.g., In re Arroyo v. Annucci, 85 N.Y.S.3d 700, 705–06 (Sup. Ct. 2018) (detailing the difficulty of finding an affordable, compliant residence in New York City “for [a] terminally ill individual in need of round-the-clock medical care,” id. at 706); Human Rights Watch, No Easy Answers: Sex Offender Laws in the U.S. 102 (2007), https://www.hrw.org/sites/default/files/reports/us0907webwcover.pdf [https://perma.cc/FRG4-TY6W] (noting that residency restrictions are “particularly problematic for registrants who have limited resources”). These features of the statutory scheme at issue in Vasquez could have led the court to conclude that its restrictions were excessive in relation to the risk identified.70×70. See, e.g., Mikaloff v. Walsh, No. 06-CV-96, 2007 U.S. Dist. LEXIS 65076, at *33–34 (N.D. Ohio Sept. 4, 2007) (finding that a residency restriction’s scope was excessive in relation to its purposes); Doe v. Miller, 298 F. Supp. 2d 844, 871 (S.D. Iowa 2004) (same); Baker, 295 S.W.3d at 446–47 (same). But even if a fully engaged excessiveness inquiry does not change a court’s outcome on this factor or on its ultimate ex post facto determination, it is worthwhile for courts to substantively consider the details and scope of statutes — especially when they are explicitly engaged in an analysis of their effects.
In conflating its inquiries into whether the statute had a rational relation to a nonpunitive purpose and whether it was excessive with respect to that purpose, the Seventh Circuit failed to fully assess the regulatory scheme before it. Instead, by asking only whether the nonpunitive purpose was a sham, the court substantially and troublingly deferred to the legislature. In an area where lawmakers are motivated to pass increasingly punitive restrictions on a politically marginalized and socially reviled group,71×71. See, e.g., Human Rights Watch, supra note 69, at 2 (“The reality is that sex offenders are a great political target . . . .” (quoting Ill. State Rep. John Fritchey)). a full judicial inquiry into not only the purposes but also the effects of these laws can and should serve as an important check on legislative overreach, prompting courts to consider how these laws actually operate on those subject to them.72×72. Close judicial attention to laws regulating those convicted of sex offenses is particularly important as legislatures enact statutes whose restrictions go far beyond those at issue in Smith. See Catherine L. Carpenter & Amy E. Beverlin, The Evolution of Unconstitutionality in Sex Offender Registration Laws, 63 Hastings L.J. 1071, 1105–22 (2012) (describing the expansion of sex offender regulations). While the Vasquez court mistakenly conflated two Smith factors, other courts have also upheld statutes through hasty applications of Smith. See Corey Rayburn Yung, The Emerging Criminal War on Sex Offenders, 45 Harv. C.R.-C.L. L. Rev. 435, 459–63 (2010) (“[C]ourts have largely cited Smith [when evaluating newer laws] without looking at the underlying differences in the cases.” Id. at 462.); see also Carpenter & Beverlin, supra, at 1107 (“However ‘tempting’ it is to conclude that Smith controls [for newer laws], it would be a mistake to do so because the statutory landscape has so dramatically altered.” (footnotes omitted)). Without closer scrutiny of those effects, plaintiffs like Vasquez and Cardona will have little hope of being protected or even heard by either the legislature or the courts.