971 F.3d 1174
10th Cir.
2020

DAVID MILLARD; EUGENE KNIGHT; ARTURO VEGA, Plaintiffs - Appellees, v. JOHN CAMPER, in his official capacity as Director of the Colorado Bureau of Investigation, Defendant - Appellant.

No. 17-1333

United States Court of Appeals, Tenth Circuit

August 20, 2020

PUBLISH

UNITED STATES COURT OF APPEALS FOR THE TENTH CIRCUIT

FILED
United States Court of Appeals Tenth Circuit
August 20, 2020
Christopher M. Wolpert Clerk of Court

DAVID MILLARD; EUGENE KNIGHT; ARTURO VEGA,
Plaintiffs - Appellees,
v.
JOHN CAMPER, in his official capacity as Director of the Colorado Bureau of Investigation,
Defendant - Appellant.

MIKE HUNTER, Attorney General of the State of Oklahoma; DEREK SCHMIDT, Attorney General of the State of Kansas; HECTOR H. BALDERAS, Attorney General of the State of New Mexico; SEAN REYES, Attorney General of the State of Utah; PETER K. MICHAEL, Attorney General of the State of Wyoming; 17 SCHOLARS WHO STUDY SEX OFFENSES; NATIONAL ASSOCIATION FOR RATIONAL SEXUAL OFFENSE LAWS,
Amici Curiae.

Appeal from the United States District Court for the District of Colorado
(D.C. No. 1:13-CV-02406-RPM)

Melanie J. Snyder, Chief Deputy Attorney General of Colorado (Frederick R. Yarger, Solicitor General of Colorado; Chris W. Alber, Senior Assistant Attorney General of Colorado; James X. Quinn, First Assistant Attorney General of Colorado; Robert C. Huss, Assistant Attorney General of Colorado; Russell D. Johnson, Assistant Solicitor General of Colorado, with her on the briefs), for Petitioner-Appellant.

Ty Gee, Haddon, Morgan and Foreman P.C., Denver, Colorado (Adam Mueller, Haddon, Morgan and Foreman P.C.; Alison Ruttenberg; Mark Silverstein, American Civil Liberties Union of Colorado; Sara R. Neel, American Civil Liberties Union of Colorado, with him on the brief), for Respondent-Appellee.

Before HARTZ, SEYMOUR, and EID, Circuit Judges.

EID, Circuit Judge.

Plaintiff-Appellees David Millard, Eugene Knight, and Arturo Vega challenge the constitutionality of Colorado‘s Sex Offender Registration Act (CSORA). The district court held CSORA was unconstitutional as applied to the Appellees1 because the statute inflicted cruel and unusual punishment and violated substantive due process guarantees. Additionally, the district court held that the state courts’ application of CSORA‘s deregistration procedures to Vega violated his procedural due process rights. Defendant-Appellant, the State,2 appeals from the entirety of the district court‘s decision. Because the district court‘s ruling contravenes binding Supreme Court and Tenth Circuit precedent, we reverse.

I. BACKGROUND

A. Development of sex offender registration acts and CSORA

States began adopting sex offender registry laws in the early 1990‘s in response to the high-profile sexual assaults and murders of children by individuals with prior sex-offense convictions. See Nichols v. United States, 136 S. Ct. 1113, 1116 (2016); see also 34 U.S.C. § 20901(1)–(17) (listing names). In 1994, Congress “conditioned federal funds on [s]tates’ enacting sex-offender registry laws meeting certain minimum standards.” Id. (discussing the Jacob Wetterling Crimes Against Children and Sexually Violent Offender Registration Act). “By 1996, every State, the District of Columbia, and the Federal Government had enacted some variation of” a sex-offender registry. Smith v. Doe, 538 U.S. 84, 90 (2003).

In 2006, Congress revised its federal sex-offender registry requirements with the adoption of the Sex Offender Registration and Notification Act (SORNA), Pub. L. No. 109-248, §§ 101–155, 120 Stat. 587, codified at 42 U.S.C. § 16901 et seq. (2006 ed.), transferred to 34 U.S.C. § 20901 et seq. (2017 ed.). With SORNA, Congress created a nationwide registry intended to “protect the public from sex offenders.” 34 U.S.C. §§ 20901, 20921 (national registry). SORNA continues to condition federal funding to states’ maintenance of their own sex-offender registries, but also requires states (and registrants) to provide registry information to the federal government. Id. at §§ 20913(c), 20914, 20918. SORNA requires states “make available on the Internet, in a manner that is readily accessible to all jurisdictions and to the public,” pertinent information “about each sex offender in the registry.” Id. at § 20920(a). And, the website must be searchable by zip code or geographic area. Id.

Colorado has maintained some variant of a sex-offender registry since 1991. See 1991 COLO. SESS. LAWS, ch. 69. It is against this legal backdrop that, in 2002, Colorado adopted CSORA. 2002 COLO. SESS. LAWS, ch. 297. CSORA‘s provisions render Colorado compliant with SORNA.3 CSORA has three basic elements: (1) registration by sex offenders with local law enforcement; (2) compilation of a sex-offender registry by the Colorado Bureau of Investigation (CBI); and (3) limited public disclosure of some of the information contained in the registry. The statutory text itself explains that “it [was] not the general assembly‘s intent that the information [contained in the Registry] be used to inflict retribution or additional punishment on any person,” but rather CSORA was intended to address “the public‘s need to adequately protect themselves and their children” from those with prior sexual convictions. COLO. REV. STAT. § 16-22-110(6)(a); see also id. § 112(1).

Registration is required under CSORA if a person is (a) convicted of any of the enumerated thirty-plus misdemeanor or felony offenses, (b) convicted of any offense in which the “underlying factual basis involves” any of those offenses, or (c) released from the custody of the state department of corrections after serving a sentence for any of the offenses or an offense with the requisite “underlying factual basis.” Id. at § 103(1)–(2). Juveniles must also register but can petition for deregistration if they satisfy certain criteria. Id. at § 103(5)(a).

Persons subject to the registration requirement must register with local law enforcement and disclose certain personal information including their name, date of birth, address, place of employment, all names previously used, the identity of any school they are attending, vehicle identification number, and, for certain registrants, email addresses. Id. at § 109(1).4 Each registrant must also provide a current photograph and a set of fingerprints. Id. at § 108(6). Most registrants must register annually, but sexually-violent offenders and those convicted of particularly serious offenses must register quarterly. Id. at § 108(1)(b), (d).

The information obtained pursuant to registration is then compiled by the CBI, the agency that maintains the Colorado Sex Offender Registry (Registry). The Registry is available to state and federal criminal justice agencies, id. at §§ 109, 110, 112, and it includes the offender‘s name, registration status, date of birth, and description of the offender‘s unlawful sexual behavior and crimes, id. at § 110(2).

In addition to compiling the Registry itself, CSORA also requires the CBI make some of the registrants’ information available to the public. Upon request, the CBI must issue a list of persons on the Registry. Id. at § 110(6)(c). That list must include, at minimum, the registrant‘s name and aliases, address, birth date, photograph, and the offense that required him or her to register. Id. at § 110(6)(c), (f). The CBI also maintains a public website searchable by name and geographic area. Id. at § 111(1). This website does not include information about registrants whose offenses were misdemeanors, or information about registrants who committed the triggering offense while they were juveniles. Id. at § 111(1)(a)–(d). However, information about those registrants is available via the Registry list, which as explained above, any person could obtain upon request to the CBI. Id. at § 110(6)(c). Since the creation of the Registry, private, third-party businesses have emerged that republish registrants’ personal information on the internet with no limitation or regulation on republication.

Some registered sex offenders can petition the court to discontinue registration under certain conditions, see id. at § 113(1), while certain categories of sex offenders are ineligible for deregistration and must continue to abide by CSORA‘s registration requirements for the rest of their lives, see id. at § 113(3). Whether a sex offender is eligible for removal from the Registry depends on the underlying offense.5

B. The Appellees

Millard, Knight, and Vega are convicted sex-offenders subject to CSORA‘s registration requirements.6 Together, they brought suit pursuant to 42 U.S.C. § 1983, arguing that CSORA‘s registration requirements constitute cruel and unusual punishment under the Eighth Amendment; violate their Fourteenth Amendment right to privacy; and, in the case of juvenile-offenders like Vega, violate their right to Due Process under the Fourteenth Amendment. Each alleges that registration has been onerous, making it harder to hold a job, find housing, raise his children, or pass a criminal background check. Millard contends that he had to move to a different work location because of his sex-offender status, and that he was forced to move—and had to file hundreds of applications before finally finding a new residence—because of a TV news story focusing on sex offenders in apartment housing, and that he incurred shame from the broad dissemination of his information and loud visits from police. Knight allegedly has experienced shame due to private businesses republishing the registry, is no longer able to enter his children‘s school, and was denied employment at Home Depot after his background check was “red flagged.” Vega, who committed his underlying offense as a juvenile, contends that his placement on the registry prevents him from finding secure employment, and that the state court violated procedural due process when denying his petitions for deregistration.

C. The District Court Order

Following a bench trial, the district court concluded that CSORA (1) constituted cruel and unusual punishment7 in violation of the Eighth Amendment as applied to Appellees, Millard v. Rankin, 265 F. Supp 3d 1211, 1231–32 (D. Colo. 2017), (2) violated the Appellees’ substantive due process rights, id. at 1235; and (3) violated—through the state-court‘s misapplication of the law when denying Vega‘s petition to deregister—Vega‘s procedural due process rights, id. at 1233. The court granted declaratory judgment to Appellees.8

II. STANDARD OF REVIEW

On appeal from a bench trial, “we review the district court‘s factual findings for clear error and its legal conclusions de novo.” Keys Youth Servs., Inc. v. City of Olathe, KS, 248 F.3d 1267, 1274 (10th Cir. 2001). In addition, as Appellees have made an as-applied challenge to CSORA, we are limited to analyzing the contours of their claims “under the particular circumstances” of Millard, Knight, and Vega.9 See United States v. Carel, 668 F.3d 1211, 1217 (10th Cir. 2011) (quotation marks and citation omitted).

III. ANALYSIS

A. Eighth Amendment

A deprivation cannot violate the Eighth Amendment‘s prohibition against “cruel and unusual punishment” unless it first qualifies as “punishment.” See Carney v. Okla. Dept. of Public Safety, 875 F.3d 1347, 1352 (10th Cir. 2017). Whether a law equates to “punishment” depends on whether the legislature‘s intent was to punish; and if not, whether there is the “clearest proof” that any punitive effects of the law negate the legislature‘s civil intent. Smith, 538 U.S. at 92. This court has twice, and the Supreme Court has once, determined that sex-offender registration requirements were not “punishments” because their respective legislatures lacked punitive intent and their application lacked punitive effect. Id. at 96, 105 (finding the Alaska Sex Offender Registration Act nonpunitive); Shaw v. Patton, 823 F.3d 556, 562, 577 (10th Cir. 2016) (finding the application of Oklahoma‘s sex-offender reporting and residency requirements nonpunitive); Femedeer v. Haun, 227 F.3d 1244, 1249, 1253 (10th Cir. 2000) (finding the application of Utah‘s sex-offender registration and notification system, including dissemination of the information on the internet, nonpunitive). Evaluating the Colorado legislature‘s intent and comparing CSORA‘s effects to those deemed nonpunitive by binding precedent in Smith, Shaw, and Femedeer, we conclude that CSORA‘s registration requirements as applied to Appellees do not amount to punishment.

1. Legislative Intent

The Colorado General Assembly did not intend for CSORA to inflict a “punishment.” The legislature expressly indicated through the statutory text that CSORA was not intended to “be used to inflict retribution or additional punishment on any person,” but was rather intended to address “the public‘s need to adequately protect themselves and their children” from those with prior sexual convictions. COLO. REV. STAT. § 16-22-110(6)(a); see also id. § 112(1). We therefore conclude that the Colorado legislature‘s intent behind the registration requirements was civil, and not punitive. See Smith, 538 U.S. at 93 (finding civil intent behind Alaska‘s sex-offense registration requirements where the “statutory text itself” noted that the primary goal of the law was to “protect[] the public from sex offenders“).

2. Effects

Past precedent shows that CSORA‘s effects on Appellees are not punitive enough to negate the legislature‘s civil intent. To determine the punitive nature of CSORA‘s effects, we consider five factors used in Kennedy v. Mendoza-Martinez, 372 U.S. 144, 168 (1963), whether the scheme: resembles traditional forms of punishment, imposes an affirmative disability or restraint, promotes the traditional aims of punishment, has a rational connection to a nonpunitive purpose, or is excessive with respect to this purpose. Smith, 538 U.S. at 97; Shaw, 823 F.3d at 562; Femedeer, 227 F.3d at 1249.10 The Courts in Smith, Shaw, and Femedeer determined that each of the above factors supported the conclusion that sex-offender laws similar to or more restrictive than CSORA were nonpunitive. As explained below, neither the district court nor Appellees distinguish their case from our binding precedent in Smith, Shaw, and Femedeer in any meaningful way with regard to any of the Mendoza-Martinez factors, leading us to conclude that CSORA‘s effects do not constitute punishment.

a. CSORA does not resemble traditional forms of punishment.

For this factor, we examine whether CSORA‘s registration requirements are “regarded in our history and traditions as a punishment.” Smith, 538 U.S. at 97. The district court below concluded the requirements imposed by CSORA resembled three traditional forms of punishment: public shaming, banishment, and parole and probation. Millard, 265 F. Supp 3d at 1226–27. We disagree.

First, CSORA does not resemble public shaming. The district court based its finding on the “indignity” that the Appellees suffered from “loud public visits from the police” and “ubiquitous influence of social media.” Id. at 1226–27. But past precedent shows that this does not amount to public shaming. The State here is not putting Appellees on display but is rather disseminating “accurate information about a criminal record,” which the Smith Court found was a “legitimate governmental objective,” and not “punishment.” 538 U.S. at 98. Any attendant “public shame” or “humiliation“—even when magnified by the “reach of the Internet“—is a “collateral consequence of a valid regulation.” Id. at 99; see also Femedeer, 227 F.3d at 1250 (“Public shaming [and] humiliation . . . involve more than the dissemination of information.” (quoting E.B. v. Verniero, 119 F.3d 1077, 1099–1100 (3d Cir. 1997)).

Second, CSORA does not resemble banishment. The district court based its conclusion on the fact that one of the Appellees was forced to move—and had to file hundreds of applications before finally finding a new residence—“because of a TV news story focusing on sex offenders in apartment housing.” Millard, 265 F. Supp 3d at 1227. But such obstacles are less onerous than Oklahoma‘s residency restrictions that place an outright ban on sex-offenders “living within 2,000 feet of a school, playground, park, or child care center,” which this court in Shaw found did not equate to banishment. 823 F.3d at 559, 568. Moreover, the Appellees’ struggles here did not stem from affirmative state action—such as the residency restrictions in Shaw—but instead from third parties and businesses implementing their own procedures. CSORA mandates the dissemination of information, which does not constitute “banishment.” Smith, 538 U.S. at 98 (“[B]anishment[] involve[s] more than the dissemination of information.“); see also Femedeer, 227 F.3d at 1250.

Last, the application of CSORA to the Appellees does not amount to probation. The district court concluded otherwise, reasoning that CSORA‘s requirements for in-person reporting and disclosure of email addresses and chat-handles provided law enforcement supervisory capabilities akin to probation. Millard, 265 F. Supp 3d at 1228 (citing COLO. REV. STAT. § 16-22-108(2.5)(a)). But in Shaw we concluded that in-person reporting does not amount to probation. 823 F.3d at 566. And our reasoning there applies equally to CSORA‘s email and chat-handle disclosure requirements. Any monitoring capability stemming from these requirements falls short of the “far more active” role law enforcement plays in a probationer‘s life, such as mandating employment, requiring consent before moving or changing jobs, and forbidding drug and alcohol use. Id. at 564–65. Moreover, CSORA‘s mandates are “regulatory requirements separate from [the] underlying sex-offense conviction,” whereas “probation historically involve[s] a ‘deferred sentence’ based on the underlying offense.” Id. at 566.

b. CSORA does not impose affirmative disability or restraint.

For this factor, we must assess whether CSORA “imposes an affirmative disability or restraint.” Smith, 538 U.S. at 97. The district court held that it did because its in-person reporting requirements and use of the Registry restrained the Appellees’ “abilities to live, work, accompany their children to school, and otherwise freely live their lives.” Millard, 265 F. Supp 3d at 1229. We disagree, however, because we held in Shaw that neither the in-person reporting requirements, nor the more onerous residency-restrictions, of Oklahoma‘s sex-offender laws constituted an affirmative disability or restraint. Shaw, 823 F.3d at 569, 571. It is true that Millard‘s employer moved him to a different work location and that Home Depot declined to hire Knight because of his background. But we find these effects to be less harsh than a lifelong bar on work in a particular industry, which the Supreme Court in Hudson v. United States found did not constitute an affirmative disability or restraint. 522 U.S. 93, 104 (1997). We therefore find that CSORA has not affirmatively disabled or restrained the Appellees here.

c. CSORA does not promote the traditional aims of punishment.

For this Mendoza-Martinez factor we assess whether CSORA as applied promotes traditional aims of punishment. Smith, 538 U.S. at 97. The district court held that CSORA appears to promote the traditional aims of retribution and deterrence. Millard, 265 F. Supp 3d at 1229–30. The district court premised this conclusion on CBI Director Rankin‘s testimony that deterrence was one of CSORA‘s purposes, see id., and that CSORA‘s requirements are based purely on statutory offense classifications, not an individual assessment of how likely a registrant is to reoffend, which implies that CSORA is retributive and deterrent in nature. Id. at 1230.

However, both these determinations run afoul of Smith. First, as the Smith Court explained, deterrent purpose alone is not enough to render a regulatory scheme criminal in nature. Smith, 538 U.S. at 102. “Any number of governmental programs might deter crime without imposing punishment. To hold that the mere presence of a deterrent purpose renders such sanctions ‘criminal’ would severely undermine the Government‘s ability to engage in effective regulation.” Id. (quotation marks, alteration, and citations omitted). Second, the Smith Court also rejected the argument that tying the length of the reporting requirement to the nature of the offense, rather than individual risk, renders the registration obligation retributive. Id. (“The broad categories, however, and the corresponding length of the reporting requirement, are reasonably related to the danger of recidivism, and this is consistent with the regulatory objective.“). We therefore reject the district court‘s reasoning and hold that CSORA‘s requirements are consistent with its regulatory objectives.

d. CSORA is rationally related to a nonpunitive purpose.

Here, we evaluate whether CSORA “has a rational connection to a nonpunitive purpose.” Smith, 538 U.S. at 97. The district court found that CSORA is rationally connected to Colorado‘s “avowed regulatory purpose of public safety.” Millard, 265 F. Supp 3d at 1230. We agree. CSORA requires more serious offenders to register more often than others. COLO. REV. STAT. § 16-10-108(1)(b), (d). Its requirements are therefore, like those in Shaw and Smith, “keyed to the seriousness of [the] underlying sex-offense conviction,” and are thus “consistent with the regulatory objective of protecting public safety.” Shaw, 823 F.3d at 572 (quotations omitted); see also Smith, 538 U.S. at 90.

e. CSORA is not excessive in relation to its nonpunitive purpose.

For this last Mendoza-Martinez factor, we assess whether CSORA is “excessive” in relation to its public safety objective. Smith, 538 U.S. at 97. We conclude that it is not. Below, the court held that CSORA‘s very long registration requirements and substantial disclosure of personal information, without any individual risk assessment or opportunity to soften CSORA‘s requirements based on evidence of rehabilitation, were excessive in relation to CSORA‘s supposed public safety objective. Millard, 265 F. Supp 3d at 1230.

Again, we reject the district court‘s conclusions because they run counter to governing precedent. In Smith, the Court rejected the court of appeals’ reasoning that the registration statute was excessive because it (1) applied to all sex offenders without regard to future dangerousness; and (2) placed no limits on the number of persons who had access to the information. 538 U.S. at 103. Instead, the Court held states were permitted to impose even very long reporting requirements based on categorical judgments about specific crimes, and that states were not required to evaluate individual risk to avoid a finding that registration acts were excessive in relation to their regulatory purpose. Id. at 104. Further, if the weekly to quarterly reporting requirement was not excessive in Shaw, 823 F.3d at 576, then neither is the less-demanding quarterly to annual requirement here.

In sum, Supreme Court precedent and our governing case law compel us to reject the district court‘s decision. Upon de novo review of the application of the Mendoza-Martinez factors, we conclude that the Appellees have not presented the clearest proof of punitive effect, and that therefore CSORA is not punitive as applied to Appellees.11 Accordingly, the Eighth Amendment does not bar its application in this case.12

B. Substantive Due Process

We now review Appellees’ claims that CSORA violates their substantive due process rights. The district court found that the application of CSORA violated substantive due process because it gives the public the power to “arbitrarily,” and without notice, “inflict punishments beyond those imposed through the courts.” Millard, 265 F. Supp 3d at 1235. Separately, the Appellees argue that CSORA offends substantive due process because it creates an “irrebuttable presumption” that a registered sex offender will reoffend. We disagree with the district court and the Appellees and find instead that CSORA does not violate substantive due process.

The Supreme Court has held that “the Due Process Clause specially protects . . . fundamental rights and liberties.” Washington v. Glucksburg, 521 U.S. 702, 720–21 (1997). Laws that violate such rights undergo heightened scrutiny and will be deemed invalid unless they are narrowly tailored to a compelling state interest. Id. at 721. But if a law does not violate fundamental rights, it need only be rationally related to a legitimate government interest. Id. at 728.

The Appellees fail to show how CSORA violated their fundamental rights. They cite no case holding that compliance with a sex-offender registration law implicates a “deeply rooted” fundamental right. Rather, all courts to have considered the issue have concluded otherwise. See Litmon v. Harris, 768 F.3d 1237, 1242 (9th Cir. 2014); Does v. Munoz, 507 F.3d 961, 963–65 (6th Cir. 2007); Doe v. Moore, 410 F.3d 1337, 1345–46 (11th Cir. 2005).

We therefore must determine whether CSORA is rationally related to a legitimate government interest. CSORA easily passes this test because, as discussed earlier, there is a rational connection between it and the government‘s interest in public safety.13 Accordingly, we reject the Appellees’ substantive due process claims.14

The district court concluded that CSORA violates substantive due process because of the public‘s allegedly “arbitrary” infliction of punishment. But this would not amount to substantive due process even if the public has indeed acted arbitrarily. The “touchstone of due process is protection of the individual against arbitrary action of government.” County of Sacramento v. Lewis, 523 U.S. 833, 845 (1998) (alterations omitted) (emphasis added). And below, the alleged “arbitrariness” with which the district court took issue related not to government action, but to public action. Millard, 265 F. Supp 3d at 1235.

We similarly reject the Appellees’ arguments that CSORA‘s alleged “irrebuttable presumption” of re-offense offends substantive due process. The Appellees do not connect this argument to any recognized fundamental liberty interest, or otherwise explain why an “irrebuttable presumption” in this context would be unconstitutional. Their briefing suggests it is per se unconstitutional to erect an irrebuttable presumption without a certain “adequacy of ‘fit’ between the classification and the policy that the classification serves.” Aple. Br. at 26 (citing Michael H. v. Gerald D., 491 U.S. 110, 121 (1989)). But Michael H. explains that the problem with a “presumption” is not the presumption itself, but whether the “presumption” implicates a constitutionally-problematic legal classification. Id. at 120–21. The answer to that question, again, depends on the nature of the right at issue: classifications affecting fundamental rights must be narrowly tailored to a compelling interest, but other classifications need not be so narrowly tailored. See Glucksburg, 521 U.S. at 720–21, 728.15 Accordingly, because the Appellees fail to connect CSORA‘s alleged “irrebuttable presumption” of re-offense to a fundamental right, their argument on this score fails.

C. Procedural Due Process

Vega claims that two Colorado magistrate judges violated procedural due process when rejecting his petitions to discontinue his registration obligation because, according to him, the judges misapplied the relevant standard. The district court below agreed with Vega and declared that “[t]he procedures followed by the state in considering his petitions did not comport with basic principles of fundamental fairness.” Millard, 265 F. Supp 3d at 1232. But we vacate this judgment because the district court lacked jurisdiction under the Rooker-Feldman doctrine to hear Vega‘s claims related to the state-court decisions.

The Rooker-Feldman doctrine precludes lower federal courts from exercising appellate jurisdiction over state-court judgments. See Rooker v. Fidelity Trust Co., 263 U.S. 413 (1923); see also Kline v. Biles, 861 F.3d 1177, 1180 (10th Cir. 2017). A plaintiff cannot circumvent Rooker-Feldman by framing his or her challenge to a state-court judgment as a due process claim. See Campbell v. City of Spencer, 682 F.3d 1278, 1284 (10th Cir. 2012) (holding Rooker Feldman barred Fifth Amendment due process claim); see also Castro v. Kondaur Capital Corp., 541 F. App‘x 833, 837 (10th Cir. 2013) (unpublished) (holding due process claim barred by Rooker-Feldman when reviewing that claim would impermissibly involve reexamination of the state-court judgment). Vega‘s procedural due process claim is, at its core, that the state court misapplied CSORA. The district court thus could not have granted relief without reexamining the state-court judgments, and it therefore lacked jurisdiction to decide the issue.

IV. CONCLUSION

For the foregoing reasons, we REVERSE the district court‘s judgment on the Eighth Amendment and substantive due process claims and REMAND for further proceedings consistent with this opinion. Further, we VACATE the district court‘s judgment regarding the procedural due process claim and REMAND with instructions to dismiss that claim for lack of subject matter jurisdiction.

ALLISON H. EID

UNITED STATES CIRCUIT JUDGE

Notes

1
For simplicity and readability, we will refer to Millard, Knight, and Vega as “the Appellees” throughout this opinion, unless discussing one of them individually.
2
The original appellant was Michael Rankin, who was sued in his official capacity as the then-Director of the CBI. The current appellant, John Camper, replaced Rankin as the Director of the CBI in January 2018, and under Fed. R. App. P. 43(c)(2), he is substituted for Rankin as the Appellant in this appeal. For simplicity and readability, we will refer to the Appellant as “the State” throughout this opinion.
3
See Office of Sex Offender Sentencing, Monitoring, Apprehending, Registering, and Tracking, “Substantially Implemented,” https://smart.ojp.gov/sorna/substantially-implemented (last updated May 13, 2020) (noting Colorado has “substantially implemented” SORNA‘s requirements).
4
These requirements are less restrictive than those in some other states, which limit where registrants can live and what employment they can seek. See, e.g., ALA. CODE § 15-20A-11 (prohibition on residence); CAL. PENAL CODE § 290.95 (prohibited from work with minors); GA. CODE ANN. § 42-1-15(b)–(c) (prohibition on residence and employment); KY. REV. STAT. ANN. § 17.545 (prohibition on residence), see id. § 17.546 (prohibition on online or social media interactions with minors); MISS. CODE ANN. § 45-33-24(4a) (prohibition on residence); OHIO REV. CODE ANN. § 2950.034 (prohibition on residence).
5
A registrant who committed a class 1, 2, or 3 felony can petition to deregister twenty years after discharge from incarceration or final release from the jurisdiction of the court, id. at § 113(1)(a), and for lower-level felonies, the registrant may petition for deregistration ten years after discharge or release. Id. at § 113(1)(b). Those who were convicted of their offenses as juveniles can also petition for deregistration. Id.
6
Millard is required to register quarterly for the rest of his life. COLO. REV. STAT. § 16-22-113(3)(b)(I). The district court incorrectly stated that Millard would become eligible to petition for deregistration in 2017. Millard v. Rankin, 265 F. Supp 3d 1211, 1217 (D. Colo. 2017). Millard was accused of forcible sexual assault by his daughter and was convicted pursuant to the then-existing sexual assault provision, see COLO. REV. STAT. § 18-3-403 (1999), which according to CSORA, makes him ineligible for deregistration for the rest of his natural life. See COLO. REV. STAT. § 16-22-113(3)(b)(I) (“The following persons are not eligible for relief . . . any person who is convicted as an adult of . . . sexual assault in the second degree, in violation of section 18-3-403, as it existed prior to July 1, 2000.“). Knight is eligible to petition for deregistration in 2021. Id. at § 113(1)(b). And Vega was eligible for deregistration at the time he completed his juvenile sentence. See id. at § 113(1)(e). He has since twice petitioned for deregistration but was denied both times by the state court. These denials form the basis of his procedural due process claim.
7
The district court concluded “the experience of these plaintiffs . . . [shows] the effect of publication of the information required by registration is to expose the registrants to punishments inflicted not by the state but by their fellow citizens.” Millard, 265 F. Supp 3d at 1226.
8
Although Appellees requested a permanent injunction against enforcement of CSORA, the district court found that Millard, Knight, and Vega had not submitted any evidence to establish the requisite factors other than success on the merits. Accordingly, the district court granted only declaratory relief.
9
However, the district court‘s use of third-parties’ testimony below does not rise to the level of clear error, see Keys Youth Serv., 248 F.3d at 1274, because the district court‘s order shows it considered third-party evidence as reinforcement of the Appellees’ own experiences.
10
There are two additional factors in Mendoza-Martinez—whether the sanction has a scienter requirement, and whether the behavior to which it applies is already a crime. See 372 U.S. at 168. However, as the Smith Court explained, these factors are less relevant to our analysis. 538 U.S. at 105.
11
We recognize that the Colorado Court of Appeals has split amongst itself on the question of whether juveniles convicted of multiple sex offenses can be constitutionally subjected to lifetime registration pursuant to CSORA. Compare People in Interest of C.M.D., 452 P.3d 133, 136–39 (Colo. App. 2018) (holding that CSORA‘s requiring of juveniles with multiple convictions to register for life does not constitute punishment) with People in Interest of T.B., --- P.3d ---, No. 16CA1289, 2019 WL 2528764 at *4–*9 (Colo. App. June 20, 2019) (holding the opposite, and finding that not only does CSORA impose a punishment but it is cruel and unusual), cert. granted, People in Interest of T.B., No. 19-SC-690 (Colo. 2020). We conclude the Colorado Supreme Court‘s impending decision in T.B. is inapplicable to the present matter, as none of the Appellees is a juvenile offender with multiple juvenile sex offenses such that they are subject to lifetime registration. See id. (limiting cert. question to “[w]hether mandatory lifetime sex-offender registration for multiple juvenile offenses is facially cruel and unusual punishment under the Eighth Amendment” (emphasis added)). Millard is subject to lifetime registration, but for an offense he committed as an adult. And, although Vega was a juvenile offender, he had only one sex offense conviction (juvenile or otherwise), so he is eligible to petition for deregistration and is not subject to a lifetime registration requirement.
12
Because we hold that CSORA does not impose “punishment,” we need not analyze whether its requirements are “cruel and unusual.”
13
In the district court‘s words, “[t]here is a rational relationship between the registration requirements and the legislative purpose of giving members of the public the opportunity to protect themselves and their children from sex offenses.” Millard, 265 F. Supp 3d at 1235.
14
CSORA‘s scheme satisfies rational basis even for Vega, who was a juvenile at the time of his crimes. CSORA treats juvenile offenses differently from adult offenses, and Vega has benefitted from those heightened protections. His name does not appear on the state website, and he was eligible to petition for deregistration at the end of his juvenile sentence. And, as the State correctly notes, the out-of-state cases Vega cites imposed much longer registration requirements on juveniles. See Aple. Br. at 30–31 (discussing Interest of C.K., 182 A.3d 917, 934–36 (N.J. 2018) (lifetime registration for juveniles); In re J.B., 107 A.3d 1, 14–20 (Pa. 2014) (25 years mandatory registration); In re C.P., 967 N.E.2d 729, 748 (Ohio 2012) (25 years mandatory registration)).
15
And, in fact, in Michael H., the Court upheld the presumption at issue, concluding there was no “historical tradition” or fundamental right implicated. 491 U.S. at 124.
Read the detailed case summary