Brian Hope v. Commissioner of Indiana Depart

• Court: Court of Appeals for the Seventh Circuit
• Decided: January 6, 2021
• Docket: 19-2523
• Status: Published

Opinion

In the

United States Court of Appeals For the Seventh Circuit ____________________ No. 19-2523 BRIAN HOPE, et al., Plaintiffs-Appellees, v.

COMMISSIONER OF INDIANA DEPARTMENT OF CORRECTION, et al, Defendants-Appellants. ____________________

Appeal from the United States District Court for the Southern District of Indiana, Indianapolis Division. No. 1:16-cv-02865-RLY-TAB — Richard L. Young, Judge. ____________________

ARGUED JANUARY 14, 2020 — DECIDED JANUARY 6, 2021 ____________________

Before ROVNER, WOOD, and ST. EVE, Circuit Judges. ROVNER, Circuit Judge. Sex offender registration and noti- fication laws have a unique place at the intersection of crimi- nal and civil law. These civil laws impose cumbersome and often lifelong burdens on former criminal perpetrators, many of whom have finished all forms of imprisonment and post- imprisonment supervision. For this reason, they are fre- quently challenged as unconstitutional. In this case, the 2 No. 19-2523

plaintiffs have challenged Indiana’s Sex Offender Registra- tion Act (SORA) as it applies to offenders who have relocated to Indiana from other states after the enactment of SORA, and who are forced to register under the law, but would not have been required to do so had they committed their crimes as residents of Indiana prior to the enactment of the relevant portions of SORA and maintained citizenship there. The dis- trict court found the registration requirements to be unconsti- tutional, and we uphold the district court’s finding that this application of SORA violates the plaintiffs’ right to travel. I. Although sex offender registries had been around for some time prior, they proliferated in the early 1990’s due to a few high profile and highly publicized heinous crimes against children by repeat sex offenders. We can assume that more widespread access to the internet in the 1990’s also contrib- uted to the proliferations of these laws. For the first time, an- yone with an internet connection could access the information in these registries from their homes with a few mouse clicks and find out the location of convicted sex offenders in their communities. In 1994, Indiana enacted its own version of a sex offender registry, SORA, also called “Zachary’s Law,” after a 10-year-old boy who was tragically sexually assaulted and murdered by a neighbor with a previous criminal conviction for sexual assault of a child. 1994 Ind. P.L. 11 § 7 (codified as Indiana Code §§ 5-2-12-1 through 5-2-12-13) (current version at Ind. Code §§ 11-8-8-1 through 11-8-8-23). Around the same time, in the federal arena, Congress was enacting sex offender registration and notification laws, cul- minating in 2006 with the federal Sex Offender Registration and Notification Act (SORNA), 34 U.S.C.A. § 20901 et. seq., No. 19-2523 3

which requires states to maintain public registries with spec- ified sex offender information. Indiana has periodically amended its SORA to remain in compliance with changing requirements of the federal SORNA, and, according to Indi- ana’s brief, to target those most likely to recidivate. In 1996, the Indiana General Assembly revised SORA to require regis- tration by one convicted elsewhere of a state offense that is substantially equivalent to an Indiana offense that triggers a duty to register. 1996 Ind. P.L. 33 § 2; see also 2001 Ind. P.L. 238 § 4 (making substantial equivalency provision retrospective).1 And, most relevantly, in 2006, the legislature amended SORA to apply the statute’s requirements to any “person who is re- quired to register as a sex offender in any jurisdiction.” 2006 P.L. 140 § 5(b)(1) (codified at Ind. Code §§ 11-8-8-4.5(b)(1), 11- 8-8-5(b)(1)). Those amendments have greatly expanded both the list of persons required to register and the information those regis- trants are required to provide. The current SORA require- ments are many. A person required to register under SORA must report in person at least once annually to the local sher- iff’s office in the county of residence, and if the registrant is employed or attends school in a different county, the regis- trant must report to the sheriff’s office in each of those coun- ties as well. Sexual offenders who have committed one of nine specified offenses are considered to be “sexually violent pred- ators” and must report to the local sheriff’s office every ninety days. Ind. Code. § 11-8-8-14(b). A person who is homeless or

1During the briefing of this case, this provision was codified at Ind. Code §§ 11-8-8-4.5(a)(22), 11-8-8-5(a)(24). As of July 1, 2020, the “substantial equivalency” provision has been moved to Ind. Code 1-1-2-4(b) and made more generally applicable across Indiana’s Code. 4 No. 19-2523

lives in transitional or temporary housing must appear in per- son at least once every seven days. Id. at § 11-8-8-12(b)(2). Registration requires more than simply appearing at the sheriff’s office. The person registering must be photographed and provide information including their name, date of birth, race, height, weight, hair color, eye color, identifying features such as scars and tattoos, social security number, driver’s license or state identification card number, vehicle description and license plate number of any vehicle the registrant might operate regularly, principal address, name and address of any employer or educational institution, any electronic mail addresses, any instant messaging user names, any social networking website user name and “[a]ny other information required by the [Department of Corrections (DOC)].” Ind. Code § 11-8-8-8(a).2 Most of this information is published on the public registry, although some of the information (such as an individual’s e-mail address) is not available to the public. If any of this information changes, the registrant must go in person to the sheriff’s office, within seventy-two hours, to report it. Ind. Code. § 11-8-8-8(c). That means, for example, if a registrant gets a Pinterest account, that person must report the new account, in person, at the local sheriff’s office, within seventy-two hours. Convicted sex offenders are required to maintain a valid driver’s license or state identification card and are prohibited from seeking a name change. Ind. Code. §§ 11-8-8-15(b), 16. In addition to all of these requirements, a sexually violent predator must inform law enforcement of any absences away

2 This is a simplified list. The full version can be found at Ind. Code § 11-8-8. No. 19-2523 5

from home that are longer than seventy-two hours. Ind. Code § 11-8-8-18.3 And an “offender against children” may not work, volunteer, or reside within 1,000 feet of a school, a youth program center, or a public park. Ind. Code §§ 35-42-4-10, 11.4 A person who is a “serious sex offender” may not enter school property. Ind. Code. § 35-42-4-14(b).5 To verify addresses, a local law enforcement officer must visit a registrant’s home at least once per year, and at least once every ninety days if the offender is a “sexually violent predator.” Ind. Code § 11-8-8-13(a). As of March 16, 2018, there were close to 10,000 persons required to register as sex or violent offenders in Indiana. R. 100-2 at 9. After cataloguing the burdens that we have just enumer- ated, the Indiana Supreme Court concluded that “the Act im- poses significant affirmative obligations and a severe stigma on every person to whom it applies. … [and the] duties im- posed on offenders are significant and intrusive.” Wallace v. State, 905 N.E.2d 371, 379 (Ind. 2009). As a result, the Indiana Supreme Court concluded that the Act had the “effect of add- ing punishment beyond that which could have been imposed when his crime was committed,” and therefore the State could not impose the requirements of SORA on anyone whose of- fense predated the enactment of that statute. Id. at 384. To do so, it held, would violate the ex post facto clause of the Indiana Constitution. Id. As a result, Indiana does not require any

3 A “sexually violent predator” is defined in Ind. Code § 35-38-1-7.5. 4 An “offender against children” is defined in Ind. Code § 35-42-4-11. 5 A “serious sex offender” is defined in Ind. Code § 35-42-4-14(a). 6 No. 19-2523

person to register if the offense occurred prior to SORA— provided that person remains a resident of Indiana. This case is before us now, however, because, despite the Wallace decision, persons with pre-SORA convictions who re- locate to Indiana from another state where registration was required or relocate from Indiana to another state requiring registration and then back again, must register in Indiana, even if Indiana would not have required them to register had they committed their offenses in Indiana and never left. We must pause here, before getting to the State’s reasons for requiring these registrations, in order to untangle a Gor- dian knot in this case. The State has argued in its briefs that there are two situations in which the DOC decides that a per- son is required to register upon moving to Indiana.6 Those cir- cumstances are as follows: (1) The substantial equivalency requirement: If an individual relocates to Indiana after the offense of which that individual was convicted, or its out-of-state-equivalent, became a regis- trable offense, the DOC requires that individual to register based on its determination that, at the time that the individual relocated to Indiana, he was “on notice” that the offense re- quires registration. (R. 100-1 at 24–25, R. 100-2 at 15–16). Indi- ana requires the person to register whether or not that person was obligated to register in the state from which that person came.

6 The DOC and local county sheriff’s offices jointly maintain SORA and the DOC makes a final determination as to who is required to register and for how long each offender must register. Ind. Code, §§ 11-8-2-12.4, 11-8- 2-13(b), 36-2-13-5.5. No. 19-2523 7

(2) The other jurisdiction requirement: If the individual is re- quired to register in another jurisdiction and relocates to In- diana after July 1, 2006, the DOC requires the individual to register pursuant to Indiana Code § 11-8-8-5(b)(1), which de- fines a “sex or violent offender” to include an individual “re- quired to register as a sex or violent offender in any jurisdic- tion.” (R. 100-1 at 24-25, 39; R. 100-2 at 16–17).7 Of course, one can be required to register in another state as a result of one’s employment or school enrollment in that state, even if he does not live there. E.g., 730 ILCS 150/3(a-5). Thus, the State points out that “[a] lifelong Indiana resident who would otherwise fall within the Wallace rule will be required to register if he works in another state that requires him to register;” State Br. at 5. But that particular application of the other jurisdiction requirement is not at issue here: So far as the briefing reveals, none of the six plaintiffs was subject to registration in another jurisdiction as a result of work, study, or some conduct apart from residence in that jurisdiction. In its fact section generally, and in the description of each plaintiff, and throughout its brief, the State maintains that all of the plaintiffs must register because of both of the statutory requirements we have just identified. State’s Brief at 22, 23, 24, 25. As an example, the State asserts in its brief that Gary Snider must register because of both the substantial equiva- lency requirement (the crime he committed in Michigan in 1988 is substantially equivalent to an Indiana registrable of- fense), and because of the other jurisdiction requirement (he

7 July 1, 2006 is the date that the Indiana General Assembly amended SORA to apply the statute’s requirements to any “person who is required to register as a sex offender in any jurisdiction.” See 2006 Ind. P.L. 140 § 5(b)(1) (codified at Ind. Code §§ 11-8-8-4.5(b)(1), 11-8-8-5(b)(1)). 8 No. 19-2523

was required to register in Michigan). State’s Brief at 23–24. Snider committed his crime in 1988, long before Indiana’s SORA was enacted, and moved to Indiana in 2003—three years before Indiana enacted the other jurisdiction require- ment of SORA. Nevertheless, the State’s brief asserts that he was required to register under both provisions. This is but one of several conflicting assertations that the State makes about Snider. It also presents two others. In its representative depo- sition pursuant to Federal Rule of Civil Procedure 30(b)(6), the State asserted that because Snider moved to Indiana be- fore the other jurisdiction requirement was added to SORA, he would only have to register on the basis of the substantial equivalency requirement. R. 100-1 at 25. And at oral argu- ment, the State asserted that Indiana’s ex post facto clause would not allow the state to require registration from some- one whose only basis for that registration was a pre-SORA of- fense that is the substantial equivalent of an Indiana offense made registrable by SORA. Oral argument at 8:57–12:00. To put this all together, the State’s briefs assert that Snider had to register for both reasons. At the 30(b)(6) deposition the State maintained that Snider would not have to register under the other jurisdiction prong because he arrived in Indiana before that requirement was added. And at oral argument, the State asserted that Snider could not be required to register because of the substantial equivalency requirement alone because of the Supreme Court decision in Wallace. We do not what else to call this other than “a mess.” Part of this confusion stems from the odd manner in which Indiana describes the operation of SORA. It refers to the stat- utory requirements of SORA as one aspect of the law, and then distinguishes the statutory law from the rulings by the Indiana Supreme Court invalidating certain applications of No. 19-2523 9

those laws. This, however, is not how we ordinarily describe operative state law. For example, in 2003, the U.S. Supreme Court declared unconstitutional the Texas statute making it a crime for two persons of the same sex to engage in certain in- timate sexual conduct. Lawrence v. Texas, 539 U.S. 558, 123 S. Ct. 2472 (2003). Despite this ruling, the Texas statute that makes it a crime if a person “engages in deviate sexual inter- course with another individual of the same sex” remains on the books in Texas to this day. See Tx. Penal Code § 21.06. Yet no one ought to write a brief which describes same sex behav- ior as illegal in Texas under the statute but allowed by the Su- preme Court’s interpretation of the Constitution. Legisla- tively enacted laws, modified by case law, together as a whole become the law of the land and we do not continue to refer to the statutory law of Texas separately from the law of Texas as limited, clarified, or modified by the judiciary. From this we can make several brief conclusions before continuing on with the facts, with the promise for more anal- ysis later. First, the decision in Wallace prevents the State from requiring new (or returning) residents to Indiana to register under the substantial equivalency prong alone, if their crime occurred before the date that SORA would have required reg- istration for the substantially equivalent crime in Indiana. The State concedes this in its brief, at oral argument, and in a sup- plemental filing. See State’s brief at 21 (“[A]n offender who committed his registrable offense prior to the adoption of SORA and who would not have any registration obligations but for SORA cannot be required to register; under Wallace, the marginal effects of such an application would be puni- tive.); id. at 38 (“[A] pre-SORA offender who moves to Indiana from a State where he was not required to register will also not be required to register in Indiana.”); State’s Reply Brief at 10 No. 19-2523

6 (“[W]hether SORA can be applied to an offender whose criminal conduct predates the statute turns on whether he has already been required to register[;] … these decisions would thus permit applying SORA if he were a lifelong Hoosier whose out-of-state travel triggered another State’s registra- tion requirement, while they would prohibit applying SORA if he were a recent resident whose prior State did not require registration.”) (emphasis removed); Oral argument at 3:28– 3:44 and 8:57–12:00 (asserting that if a pre-SORA offender moved to Indiana from a state where he was not required to register, he would not be required to register in Indiana even if his crime was a substantially equivalent offense); State’s Rule 28(j) letter dated April 13, 2020 at 1–2 (“Under the Indi- ana Supreme Court’s decisions, Indiana’s Ex Post Facto Clause prohibits applying SORA to someone whose offense predates SORA and who is not required to register in another State.”) Second, as we will explain later, Wallace also prevents the State from requiring registration under the other jurisdiction prong alone if the new (or returning) resident relocated to In- diana before 2006, when the other jurisdiction requirement was added to SORA. With this in mind, we can continue with the remaining facts. The plaintiffs maintain that five of them have been re- quired to register as sex offenders in Indiana because of both the substantial equivalency determination and the other juris- diction requirement, and that Snider had to register solely be- cause of the substantial equivalency requirement.8 As we

8 Hope has to register because of the “registrable offense” part of the sub- stantial equivalency/registrable offense requirement. Because he No. 19-2523 11

noted above, we cannot understand why the State maintains that each plaintiff had to register for both reasons under the statute, but at the same time oddly concedes that it cannot re- quire a pre-SORA offender to register on the basis of the sub- stantial equivalency requirement alone. As we shall see, ultimately these distinctions about why an offender was required to register are not relevant to the outcome of this case. Instead, our outcome depends on the fact that two people who committed the same crime at the same time have different registration requirements depend- ing on their history of residency in Indiana. Nevertheless, be- cause the State has created much confusion with its bifurca- tion of the “statutory law” and the “constitutional law,” we will make clear that we are proceeding with our analysis of the case with the understanding that the State cannot apply the substantial equivalency registration requirement to any plaintiff who committed his offense before that offense be- came registrable in Indiana.9 As a matter of Indiana law, it may only require registration of pre-SORA offenders by those who were required to register in another jurisdiction.10 This

committed his crime in Indiana it need not be compared to Indiana crime to establish equivalence. 9 The first iteration of SORA became effective on March 2, 1994. All of the plaintiffs committed their offenses prior to this date, with the exception of Standish, who committed his offense on February 1, 1995. But the offense Standish committed did not become a registrable offense in Indiana until the statute was amended years later. 10 We conclude, as we discuss further below, that this creates an over- looked problem with requiring registration from Snider and Bash, both of whom arrived in Indiana prior to the addition of the other jurisdiction re- quirement. We address this problem later. 12 No. 19-2523

in turn is what gives rise to the plaintiffs’ right-to-travel claim: having relocated from other states that required them to reg- ister, they are burdened with an obligation to register in Indi- ana that would not be imposed on a similarly-situated of- fender who has lived in Indiana continuously since commit- ting his offense. Before we turn to the individual circumstances of the plaintiffs, we can describe the generalized facts that apply to all of them. As a historical matter, it appears that all six of the plaintiffs were required to register in Indiana based on a determination by the DOC and local sheriff’s departments that they had committed a registrable offense or the out-of- state equivalent to such an offense and that they had been required to register in another jurisdiction. Although they committed their offenses before SORA was amended to require registration on these grounds, the State deemed the amendments applicable to the plaintiffs because they had relocated (or returned) to Indiana after SORA was revised to include these registration requirements. Indiana would not have required them to register on these grounds had they been living in the State at the time they committed their offenses and remained there continuously thereafter. The State now concedes that, as a matter of Indiana law (including the Indiana Supreme Court’s decision in Wallace), the plaintiffs can only be compelled to register based on the other jurisdiction requirement—that is, because they were required to register in another state. (We will discuss below why this theory is problematic as it relates to Snider and Bash, who moved to Indiana before the other jurisdiction requirement was enacted). All of the plaintiffs committed their crimes a long time ago—between approximately twenty-five to thirty- five years ago. All have fully served their sentences resulting No. 19-2523 13

from their sex-offense conviction.11 All qualify as an “offender against children,” and as a “serious sex offender.” All have lifelong labels, which are prominently displayed on the publicly available registry. Snider is now in his mid-sixties and committed his crime thirty-two years ago. Hope was nineteen when he committed his crime—twenty-four years ago. Mr. Bash was in his early teens or even younger when, about thirty-five years ago, in the mid-1980’s, he committed his crime. Standish, Rice, and Rush, like Snider, are all over fifty years old, and have families that include children (and in at least one case, grandchildren) of their own. Brian Hope pled guilty to child molestation in 1996 for a crime that occurred in 1993 (twenty-seven years ago). He completed his probation in 2000 and has not been under any form of supervised release since then. In 2004 he left Indiana and relocated to California and then Texas, where he was re- quired to register as a sex offender.12 He returned to Indiana in 2013 to help care for a sick grandfather. Hope is the only plaintiff who committed his offense in Indiana before the

11 Consequently, there is no restriction on their right to travel resulting from the restraints of parole or probation. See Jones v. Helms, 452 U.S. 412, 420–21, 101 S. Ct. 2434, 2441 (1981); Williams v. Wis., 336 F.3d 576, 581 (7th Cir. 2003). 12 There are conflicting explanations in the briefs about why Hope was required to register in Texas. Hope claimed that it was because he was required to register in Indiana. The State asserts that he was required to register in Texas because his offense in Indiana was “substantially similar” to a Texas offense and thus required registration under Texas Penal Code Ann. § 21.11. In any event, the relevant facts here are that Hope is now required to register in Indiana because he had been required to register in Texas. For purposes of the application of Indiana’s SORA, the reasons for the initial registration elsewhere are not relevant. 14 No. 19-2523

enactment of SORA, left, and then returned after SORA. Upon his return, the state required him to register for the remainder of his life as an “offender against children” because he was required to register in Texas. Ind. Code § 11-8-8-5(b)(1). Be- cause he is homeless, every seven days he must walk one to two miles each way to the Sheriff’s office and wait in line to register. The whole process—including travel, wait time, and registration—can take several hours. Hope cannot live within 1,000 feet of a park, daycare, or certain other facilities. On at least one occasion, he had to relocate from a homeless shelter because it was located within 800 feet of a park. It is uncon- troverted that had he remained a resident of Indiana, Wallace would have relieved him of any obligation to register. He un- successfully appealed the determination that he must do so to both the Marion County Sheriff’s Department and the DOC. A Michigan jury convicted Gary Snider in 1994 of criminal sexual conduct in the first degree. Snider continues to deny liability for his offense but stated in his affidavit that it was his recollection that, at trial, the victim did not have a precise memory of when the offense took place but testified that it occurred in the first half of 1988 (thirty-two years ago). R. 100- 4 at 1. He married his wife while in prison, and the day he completed his prison term in 2003, he moved to Indiana where his wife lived and worked—three years before the In- diana legislature added the other jurisdiction requirement to SORA. Prior to the decision in Wallace, Snider was registered as a sex offender. In 2006, he moved away from his wife be- cause their home was located within 1,000 feet of a daycare. In 2010, the Huntington County Sheriff’s Department in- formed him that he was no longer required to register because of the decision in Wallace (his offense predated the enactment of that statute). In 2016, the DOC told him that Wallace no No. 19-2523 15

longer applied and he would have to register as both a “sex- ually violent predator” and an “offender against children” every ninety days for the remainder of his life. He cannot en- ter school property to see any of his five grandchildren or great grandchildren perform in school activities. In 1996, Joseph Standish pleaded no contest to attempted criminal sexual conduct which occurred in Michigan in 1995 (25 years ago). He completed probation in 2001, and, when his wife received a job in Indiana in 2013, he moved with her. In- itially DOC did not require him to register but changed course in 2016. He is now required to register as an “offender against children” and a “sexually violent predator,” and must do so at least every ninety days. Mr. Standish cannot watch his chil- dren participate in school activities and cannot take them to and from school. In 1989, an Illinois court convicted and sentenced Patrick Rice for an aggravated criminal sexual assault that took place that year (31 years ago). Lacking a home or resources after his release from prison in 2017, he relocated to Indiana to live with his sister. Although Illinois required that Rice register only for ten years, Indiana requires him to register for life because he qualifies as a “sexually violent predator.” The registration process for Madison County, where he first settled, required him to pay an initial registration fee of fifty dollars and to make multiple trips within a seventy-two-hour period. Shortly after he was released from prison and arrived in Indiana, he had to make eight to ten trips to the local sheriff’s office—to register initially, to provide a copy of his newly obtained government identification, then his newly obtained social security card, his new telephone number, a new e-mail address, and a Facebook account. He must repeat 16 No. 19-2523

the process any time he changes any of the listed information. Without reliable transportation, he has to rely on other people and their schedules to take him the six or seven miles to register, and he must often wait an hour or more before registering. A few months after registering in Madison County, he had to repeat the process in Delaware County after moving there with a partner. As an “offender against children” and a “sexually violent predator,” he must repeat this process every ninety days for the remainder of his life. In 1990, Adam Bash pleaded guilty but mentally ill to the Kentucky crime of rape in the first degree and sodomy in the first degree for crimes committed in the mid-1980s, when he was somewhere between a pre-teen and an early teen. In 1998, he completed his prison sentence—which he spent mostly in psychiatric or medical facilities—without any required parole or probation. Upon his release, he relocated to Ohio, where he was required to register, before settling in Indiana around 1999 or 2000, about six years before SORA’s other jurisdiction requirement was enacted. Nevertheless, he is required to reg- ister in Indiana annually as an “offender against children.” Bash subsists on social security benefits, and because of his conviction, does not qualify for any public housing assistance. His housing options also have been limited by the prohibition on living within 800 feet of certain facilities. All of this makes it difficult for him to pay the fifty-dollar registration fee and the five-dollar change-of-address fee, the latter of which, de- spite its name, has been imposed when he registered a change in car and haircut. Because of his precarious financial situa- tion, he sometimes has to go on a payment plan. Bash has full legal custody of his young son, but he cannot enter school property for his activities or for parent-teacher conferences. No. 19-2523 17

In 1992, Scott Rush was charged and convicted in a Florida state court of sexual battery of a child less than twelve years old. He completed his sentence in 1995, and his probation in 2005. In 2017 his employer closed its Florida office and offered him a position in Indiana, which required him to relocate. Rush qualifies as a sexually violent predator and an offender against children and must register every ninety days (or sooner if his information changes). He lives approximately six miles from the sheriff’s office where he must register, and the process generally takes more than an hour, but he must take an entire day off of work to complete the process, as his job is not flexible enough to allow him to come and go. Mr. Rush has been affected most significantly from the prohibition on entering school property, as his daughter has a learning disa- bility, and every two to three months he must miss the school meeting convened to discuss her needs and individualized educational program. Hope and Snider filed suit for declaratory and injunctive relief against the Commissioner of the DOC, their respective county prosecutors’ offices and respective county sheriffs challenging the constitutionality of the statute, later adding Standish as a plaintiff. The district court entered a preliminary injunction on April 6, 2017, enjoining Indiana’s enforcement of SORA against all three plaintiffs. A few months after Hope and Snider filed their complaint, Rice, Bash, and Rush filed a similar complaint. By agreement, the cases were consolidated, and the preliminary injunction was extended to the new plaintiffs. On July 9, 2019, the district court issued its Entry on Cross-Motions for Summary Judgment holding that “SORA violates Plaintiffs’ fundamental right to travel, Plaintiffs’ right to equal protection of the laws, and the Constitution’s prohi- bition against retroactive punishment.” R. 118 at 36. 18 No. 19-2523

Indiana’s rule that those moving into the state must register while similarly situated residents do not have to register violates Plaintiffs’ funda- mental right to travel and guarantee to equal protection of the laws. The application of SORA’s requirements retroactively also violates the Constitution’s prohibition against retroac- tive punishment. That means the registration re- quirements as applied here cannot stand. Id. at 2. The State defendants appealed. We now affirm the district court’s finding that application of SORA to this class of offenders violates their fundamental right to travel in that it treats them less favorably than Indiana citizens with com- parable criminal histories who lived in Indiana before the other jurisdiction requirement of SORA was enacted. II. We review the district court’s summary judgment deci- sion de novo. E.g., Johnson v. Enhanced Recovery Co., 961 F.3d 975, 982 (7th Cir. 2020). Because we conclude that the State’s application of SORA to the plaintiffs impermissibly interferes with their right to travel, we do not reach the district court’s alternative finding that it also violates their rights under the ex post facto clause of the United States Constitution. Although all six of the plaintiffs were convicted of sex of- fenses before SORA would have required them to register for those offenses, Indiana nonetheless requires each of them to register based on subsequent amendments to SORA. As we have noted, throughout this litigation, Indiana has repre- sented that each of the plaintiffs is required to register on ei- ther of two independent grounds: (1) he relocated to Indiana No. 19-2523 19

after his offense (or its out-of-state equivalent) became a reg- istrable offense under SORA (the substantial equivalency re- quirement); or (2) he was required to register in his previous state of residence (the other jurisdiction requirement). But the State’s appellate briefs defend the plaintiffs’ registration obli- gation solely on the basis of the other jurisdiction require- ment, and as noted the State’s lawyer conceded at oral argu- ment that Indiana’s ex post facto clause precludes application of the substantial equivalency requirement to the plaintiffs. It is therefore clear that Indiana places no reliance on the sub- stantial equivalency requirement as a basis for demanding that the plaintiffs register as sex offenders. As to the plaintiffs’ claim that Indiana has interfered with their right to travel by requiring them to register, the State’s legal theory is that SORA does not make the sort of distinction between newer and more longstanding citizens that the per- tinent line of Supreme Court right-to-travel cases forbids. What triggers SORA’s application to the plaintiffs, the State emphasizes, is not the timing of their arrival in Indiana but rather the fact that they were subject to a registration require- ment in another jurisdiction. Thus, someone relocating to In- diana today will have no obligation to register there if he was under no such obligation in his former state of residence, whereas a lifelong Indiana citizen will incur an obligation to register in Indiana if he becomes obligated to register in an- other state by virtue of taking a job or enrolling in school in that state. Our dissenting colleague makes essentially the same point: Application of the other jurisdiction requirement turns not on whether or when an offender moved to Indiana from another state but rather on another state’s imposition of a duty to register, period. Indiana is not intentionally treating newly arrived offenders differently and thus burdening their 20 No. 19-2523

right to travel; at most, the burden that an offender incurs on relocating to Indiana is incidental to his interstate travel. Whatever superficial appeal this line of reasoning might have in the abstract, it does not defeat the plaintiff’s right-to- travel claim. Indiana is not contending that the plaintiffs must register because they committed a registrable offense or its equivalent, nor is Indiana relying on some other aspect of the plaintiffs’ conduct in another state signaling a danger that warrants the plaintiffs’ registration in Indiana. It instead relies solely on the fact that another state, in the exercise of its inde- pendent judgment, required each of the plaintiffs to register, although Indiana itself would not have required the plaintiffs to do so in the first instance. Because Indiana is placing exclu- sive reliance upon another state’s decision to require an of- fender to register, it is necessarily (if implicitly) using an of- fender’s travel as the trigger for its own registration require- ment. It is true that there are two types of travel implicated by the other jurisdiction requirement: relocation to Indiana from another state and commuting from Indiana to another state for work or study. All six of the plaintiffs have relocated to Indiana, and it is that particular type of travel, and its unique place in the Supreme Court’s right-to-travel jurisprudence, that is at issue here. Indiana commuters who have picked up registration obligations elsewhere may or may not have their own constitutional claim—theirs is a different form of travel for constitutional purposes—but no such claim has been raised or briefed in this case. For the plaintiffs, all of whom committed sex offenses before those offenses became registra- ble in Indiana, it is the fact of their subsequent relocation that gives rise to a duty to register in Indiana; had they lived in Indiana at the time of their offenses and never left, they would not be required to register today. True enough, their former No. 19-2523 21

states of residence also required them to register. Indiana is choosing to continue (or as to Bash and Snider, resurrect) those obligations. But what matters for purposes of the plain- tiffs’ constitutional claim is that they are now Indiana citizens. And because they are citizens who relocated (traveled) to In- diana from other jurisdictions, they are subject to burdens that Indiana pre-SORA offenders are not by virtue of the Wal- lace decision. Indiana has thus created two classes of other- wise similarly situated citizens based on whether or not they previously lived (or were otherwise present) in a state that re- quired them to register. The distinction is purposeful, as it ex- pressly looks to what obligations have been imposed on a per- son elsewhere to determine what obligations he will now have within the borders of Indiana. This disparate treatment is incompatible with the Supreme Court’s right-to-travel ju- risprudence, as we now explain. Although a right to travel is not expressly mentioned in the Constitution, it is nonetheless firmly embedded in federal jurisprudence. Saenz v. Roe, 526 U.S. 489, 498, 119 S. Ct. 1518, 1524 (1999). See also Attorney General of N.Y. v. Soto-Lopez, 476 U.S. 898, 901, 106 S. Ct. 2317, 2320 (1986) (“Freedom to travel throughout the United States has long been recognized as a basic right under the Constitution.”) (cleaned up); Shapiro v. Thompson, 394 U.S. 618, 629, 89 S. Ct. 1322, 1329 (1969) (“This Court long ago recognized that the nature of our Federal Un- ion and our constitutional concepts of personal liberty unite to require that all citizens be free to travel throughout the length and breadth of our land uninhibited by statutes, rules, or regulations which unreasonably burden this movement.”), overruled in part on other grounds by Edelman v. Jordan, 415 U.S. 651, 671, 94 S. Ct. 1347, 1359–60 (1974). Indeed, the short-lived Articles of Confederation expressly recognized a right of “free 22 No. 19-2523

ingress and regress to and from any other State” and entitled the free inhabitants of each state to “all privileges and immun- ities of free citizens in the several States.” Articles of Confed- eration, art. IV, § 1 (1778). As the cases have defined it, the right to travel encom- passes at least three distinct, but related, components: (1) the right of a citizen of one state to enter and leave another state; (2) the right of a citizen of the first state to be treated as a wel- come visitor rather than an unfriendly alien by the second state; and (3) the right of a traveler who elects to settle in and become a permanent resident of another state to be treated on par with other citizens of that state. Saenz, 526 U.S. at 500, 119 S. Ct. at 1525. The parties agree it is the third right, if any, that is impli- cated here. So far as the plaintiffs in this case are concerned, whether or not SORA imposes a duty to register depends upon whether one settled (or re-settled) in Indiana after the relevant provision of SORA was enacted: A sex offender who lived in Indiana before the other jurisdiction requirement was adopted and has remained a citizen of Indiana since that time, without taking a job or engaging in some other activity in an- other state that triggers a duty to register in that state, is ex- empt from SORA’s other jurisdiction requirement, whereas an individual with the identical criminal history who relo- cated to Indiana after that requirement was adopted in 2006, from a state that compelled him to register there, is subject to SORA’s registration requirement. Historically, there has been some uncertainty as to the con- stitutional underpinning of the right to travel and thus as to the appropriate framework for evaluating claims that a state provision like SORA intrudes upon that right. The right to No. 19-2523 23

travel has been variously ascribed to the Privileges and Im- munities Clause of Article IV of the Constitution (the succes- sor to a kindred provision in the Articles of Confederation), the Commerce Clause, and the Privileges or Immunities Clause of the Fourteenth Amendment. Soto-Lopez, 476 U.S. at 902, 106 S. Ct. at 2320. And, as relevant here, a number of cases evaluating the legitimacy of state statutes according benefits to residents based on the date of their arrival to or the dura- tion of their residence within a state have examined those classifications under the Equal Protection Clause of the Four- teenth Amendment. See, e.g., Hooper v. Bernalillo Cnty. Asses- sor, 472 U.S. 612, 618 & n.6, 623, 105 S. Ct. 2862, 2866 & n.6, 2869 (1985); Zobel v. Williams, 457 U.S. 55, 60–64 & n.6, 102 S. Ct. 2309, 2112–2315 & n.6 (1982); Shapiro, 394 U.S. at 632–33, 89 S. Ct. at 1330 (1969).13 But the Court’s decision in Saenz placed the third compo- nent of the right to travel squarely within the domain of the Privileges or Immunities Clause of the Fourteenth Amend- ment, which provides: All persons born or naturalized in the United States, and subject to the jurisdiction thereof, are citizens of the United States and of the State

13 “Durational residency requirements are those that demand a person re-

side in a state for a given period of time before gaining benefits. Fixed- point residency requirements demand that at a legislatively determined moment (a specific date or event, for example, a veteran’s date of enlist- ment) the applicant was a resident in the state. While an unsuccessful ap- plicant can satisfy a durational residence requirement in the future, an ap- plicant who fails to satisfy a fixed-point residence requirement cannot cure the defect with the passage of time.” Harris v. Hahn, 827 F.3d 359, 362 n.4 (5th Cir. 2016). 24 No. 19-2523

wherein they reside. No State shall make or en- force any law which shall abridge the privileges or immunities of citizens of the United States; …. U.S. Const., amend. XIV, § 1. Saenz, 526 U.S. at 502–03, 119 S. Ct. at 1526. Saenz concerned a California statute which, for a period of twelve months after a new resident’s arrival, lim- ited the maximum welfare benefits available to that resident to the amount he was eligible to receive in the state where he formerly resided (assuming that amount was less than Cali- fornia’s relatively generous benefits). This durational resi- dency provision, the Court observed, implicated “the right of the newly arrived citizen to the same privileges and immuni- ties enjoyed by other citizens of the same State. That right is protected not only by the new arrival’s status as a state citizen, but also by her status as a citizen of the United States.” Id. at 502, 119 S. Ct. at 1526. And it is this latter right that the Four- teenth Amendment’s Privileges or Immunities Clause ex- pressly addresses: Despite fundamentally differing views concern- ing the coverage of the Privileges or Immunities Clause of the Fourteenth Amendment, most no- tably expressed in the Slaughter-House Cases, 16 Wall. 36, 21 L. Ed. 394 (1872), it has always been common ground that this Clause protects the third component of the right to travel. Writing for the majority in the Slaughter-House Cases, Justice Miller explained that one of the privi- leges conferred by this Clause “is that a citizen of the United States can, of his own volition, be- come a citizen of any State of the Union by a No. 19-2523 25

bonâ fide residence therein, with the same rights as other citizens of the State. Id., at 80. Justice Bradley, in dissent, used even stronger lan- guage to make the same point: The states have not now, if they ever had, any power to restrict their citizenship to any classes or persons. A citizen of the United States has a perfect constitutional right to go to and reside in any State he chooses, and to claim citi- zenship therein, and an equality of rights with every other citizen; and the whole power of the nation is pledged to sustain him in that right. He is not bound to cringe to any superior, or to pray for any act of grace, as a means of enjoying all the rights and privileges enjoyed by other citizens. Id., at 112–113. That newly arrived citizens “have two political capacities, one state and one federal,” adds spe- cial force to their claim that they have the same rights as others who share their citizenship. Nei- ther mere rationality nor some intermediate standard should be used to judge the constitu- tionality of a state rule that discriminates against some of its citizens because they have been domiciled in the State for less than a year. The appropriate standard may be more 26 No. 19-2523

categorical than that articulated in Shapiro, … but it is surely no less strict. Saenz, 526 U.S. at 503–04, 119 S. Ct. at 1526–27 (footnote omit- ted). See also A.W. by & through Doe v. Neb., 865 F.3d 1014, 1020 n.3 (8th Cir. 2017); Harris v. Hahn, 827 F.3d 359, 370 (5th Cir. 2016); Selevan v. N.Y. Thruway Auth., 584 F.3d 82, 103 (2d Cir. 2009); Gean v. Hattaway, 330 F.3d 758, 771 (6th Cir. 2003); Rus- sell v. Hug, 275 F.3d 812, 822 (9th Cir. 2002). Insofar as the plaintiffs here are concerned, Indiana’s SORA creates two classes of Indiana citizens, with the 2006 adoption of the other jurisdiction requirement marking the di- viding line between the two classes: those who resided in In- diana prior to the enactment of SORA’s other jurisdiction re- quirement (and remained residents thereafter without incur- ring a registration obligation in any other state), and those who arrived later. The former enjoy the full protection of In- diana’s ex post facto clause as interpreted by Wallace and may not be burdened with the various aspects of the duty to regis- ter that Wallace deemed to be penalties, so long as they do not venture out of the State to engage in activity that might sub- ject them to a registration requirement elsewhere. If, however, they settled in Indiana after the other jurisdiction provision of SORA was enacted, they may be subject to those very penal- ties, regardless of when they committed their offenses. This sets up the very sort of multi-tiered state citizenship that the Supreme Court’s right-to-travel cases prohibit. See Saenz, 526 U.S. at 507, 119 S. Ct. at 1528 (“Neither the duration of respondents’ California residence, nor the identity of their prior States of residence, has any relevance to their need for benefits. Nor do those factors bear any relationship to the State’s interest in making an equitable allocation of the funds No. 19-2523 27

to be distributed among its needy citizens.”); Soto-Lopez, 476 U.S. at 904, 106 S. Ct. at 2322 (civil service employment pref- erence granted only to those veterans who resided in state at time they entered military service) (“The analysis in all of these cases … is informed by the same guiding principle—the right to migrate protects residents of a State from being dis- advantaged, or from being treated differently, simply because of the timing of their migration, from other similarly situated residents.”); Hooper, 472 U.S. at 623, 105 S. Ct. at 2868 (prop- erty tax exemption limited to veterans who resided in state prior to specified date) (“The State may not favor established residents over new residents based on the view that the State may take care of ’its own,’ if such is defined by prior resi- dence. Newcomers, by establishing bona fide residence in the State, become the State’s ’own’ and may not be discriminated against solely on the basis of their arrival in the State after May 8, 1976.”); Zobel, 457 U.S. at 64, 102 S. Ct. at 2314–15 (pub- lic oil dividends distributed to state residents based on the length of their residence) (“If the states can make the amount of a cash dividend depend on length of residence, what would preclude varying university tuition on a sliding scale based on years of residence—or even limiting access to finite public facilities, eligibility for student loans, for civil service jobs, or for government contracts by length of domicile? Could states impose different taxes based on length of residence? Alaska’s reasoning could open the door to state apportionment of other rights, benefits, and services according to length of res- idency. It would permit the states to divide citizens into ex- panding numbers of permanent classes. Such a result would be clearly impermissible.”) (footnotes omitted); Mem. Hosp. v. Maricopa Cnty., 415 U.S. 250, 261–62, 94 S. Ct. 1076, 1084 (1974) (requiring one year’s residence to be eligible for non- 28 No. 19-2523

emergency medical care at public expense) (“Not unlike the admonition of the Bible that, ‘Ye shall have one manner of law, as well for the stranger, as for one of your own country,’ Leviticus 24:22 (King James Version), the right of interstate travel must be seen as insuring new residents the same right to vital government benefits and privileges in the States to which they migrate as are enjoyed by other residents. The State of Arizona’s durational residence requirement for free medical care penalizes indigents for exercising their right to migrate to and settle in that State.”); Shapiro, 394 U.S. at 633, 89 S. Ct. at 1330 (various state provisions requiring one year’s residence to be eligible for welfare assistance) (“We recognize that a State has a valid interest in preserving the fiscal integ- rity of its programs. … But a State may not accomplish such a purpose by invidious distinctions between classes of its citi- zens.”). Just as in those cases, Indiana’s decision to make the ap- plicability of SORA dependent upon the date of a citizens’ ar- rival to the State (before or after the relevant statutory provi- sion took effect) implicates the right to travel by imposing greater burdens on newly arrived residents. In this respect, newer citizens of Indiana—including the plaintiffs—are not accorded the same rights as more longstanding residents who are otherwise similarly situated in terms of their criminal his- tory. This discriminatory classification is a penalty in and of itself and can only survive if it satisfies strict scrutiny. See Saenz, 526 U.S. at 504–05, 119 S. Ct. at 1527. Accordingly, the State must demonstrate that its differential treatment of Indi- ana citizens is necessary to promote a compelling governmen- tal interest. Id. at 499, 119 S. Ct. at 1524 (citing Shapiro, 394 U.S. at 634, 89 S. Ct. at 1331). No. 19-2523 29

The discriminatory application of SORA to newer resi- dents does not satisfy this demanding standard and, indeed, the State’s counsel conceded at oral argument that it cannot do so. Indiana surely has a strong interest in protecting its res- idents from the potential predations of convicted sex offend- ers. We may assume arguendo that SORA’s requirements fur- ther that interest. But singling out only newer citizens with a history of sex offenses to the exclusion of more longstanding citizens with the same criminal history does not further that interest: the distinction is not even rational. Cf. F.C.C. v. Beach Commc’ns, Inc., 508 U.S. 307, 313, 113 S. Ct. 2096, 2101 (1993) (statutory classification that neither proceeds along suspect lines nor infringes on fundamental constitutional rights is re- viewed for rational basis); see Doe v. Penn. Bd. of Prob. & Parole, 513 F.3d 95, 108–112 (3d Cir. 2008) (subjecting an out-of-state sex offender to community notification if he transfers his su- pervision to Pennsylvania, but subjecting Pennsylvania of- fender to community notification only if, following a hearing, he is designated a “sexually violent predator,” lacks rational basis and therefore violates equal protection); Hendricks v. Jones ex rel. Okla. Dep’t of Corr., 349 P.3d 531, 536 (Okla. 2013) (requiring Oklahoma resident convicted of sex offense in an- other jurisdiction to register regardless of offense date, but re- quiring Oklahoma resident convicted of comparable offense in Oklahoma to register only if convicted after enactment of Oklahoma’s SORA, was irrational and violates equal protec- tion); A.C.L.U. of N.M. v. City of Albuquerque, 137 P.3d 1215, 1226–27 (N.M. Ct. App. 2006) (requiring sex offenders visiting city to register depending on whether they committed their offenses out of state or in state was irrational and violated equal protection); see also Williams v. Vt., 472 U.S. 14, 23, 105 S. Ct. 2465, 2472 (1985) (restricting Vermont tax credit for 30 No. 19-2523

taxes paid to another state on automobile purchase in that state solely to Vermont residents who lived in Vermont at time of purchase violates equal protection: “residence at the time of purchase is a wholly arbitrary basis on which to dis- tinguish among present Vermont registrants …”). We are pointed to no evidence indicating, nor does the State suggest, that individuals who began to reside in Indiana after the other jurisdiction provision of SORA was enacted are more likely to re-offend than those who were already residents prior to that time. Below, the State suggested that absent this arrangement, an individual currently living in another state whose sex of- fense predates SORA’s other jurisdiction provision but who is subject to registration in that other state might have an incen- tive to relocate to Indiana in order to evade registration. But we have no reason to suspect that a move to Indiana would be prompted by that incentive as opposed to the prospect of a better job, a wish to be close to family, or Indiana’s lower cost of living, and the State itself has abandoned this argu- ment on appeal. In any case, the aim of deterring in-migration to Indiana from other states is constitutionally impermissible. See Saenz, 526 U.S. at 506, 119 S. Ct. at 1527–28 (citing Shapiro, 394 U.S. at 631, 89 S. Ct. at 1329). We recognize, as the district court did, that the Indiana legislature may have wished to avoid this discriminatory clas- sification and instead apply SORA’s requirements to all sex offenders residing in Indiana, regardless of the date of their convictions, but that it was precluded from doing so by Wal- lace’s holding that imposing a registration requirement on of- fenders whose convictions pre-dated SORA’s regulatory re- quirements violated Indiana’s ex post facto provision. So the distinction may well be driven more by the Indiana Supreme Court’s holding in Wallace than by any legislative judgment No. 19-2523 31

as to the relative risks posed by newer and more longstanding residents. But that merely makes plain the point that the clas- sification deprives newer residents of a valuable right granted by the Indiana Constitution—the right not to be penalized for offenses of which they were convicted before the other juris- diction provision of SORA was enacted. Against all of this, the State as noted argues that the right to travel as recognized in cases like Saenz is, in reality, not bur- dened here, in that the applicability of SORA is tied not to the duration of one’s residency in Indiana but rather to the SORA- like requirements to which arriving citizens were subject in their former states of residence. In other words, from the State’s point of view, what triggers SORA is not the recency of one’s arrival to Indiana but the registration requirement to which the new resident was subject in his former state; and in that respect, Indiana is simply choosing to leave in effect (or reimpose) the very requirement that would have burdened the offender had he never traveled across state lines at all. And if the new arrival was not subject to a registration re- quirement in his former state of residence, Indiana will not impose one. The fact that the other jurisdiction requirement can also apply to a lifelong Indiana resident who becomes subject to a registration requirement in another state by virtue of his work, study, or other activity in that state reinforces the State’s point. But for at least three reasons, the argument is unavailing. First, notwithstanding Indiana’s insistence that its scheme is not tied to the duration of one’s residency, for individuals like the plaintiffs, the fact of one’s relocation from another jurisdiction is necessarily and implicitly an element of the other jurisdiction requirement, as one cannot become subject 32 No. 19-2523

to this provision of SORA without having moved to Indiana from another state (where he was required to register); it is this relocation from one state to another that brings the third component of the right to travel into play. See A.W., 865 F.3d at 1020 n.3 (noting that construing Nebraska’s SORA to compel juvenile to register based on registration obligation imposed by Michigan before he relocated to Nebraska would “raise[ ] troubling implications under the third prong of the right to travel”). That connection is most clearly illustrated by application of the other jurisdiction requirement to Hope: Hope, having committed his offense in Indiana pre-SORA, would not have been required to register but for the fact that he later relocated to Texas and was required to register there; his subsequent decision to return to Indiana was what triggered a duty to register under SORA’s other jurisdiction requirement. Had he never left Indiana, he would not have been required to register. Furthermore, the timing of a person’s relocation to Indiana is a critically important factor in the application of SORA. As we have said, for the Indiana citizen who was living in Indiana before the other jurisdiction requirement was added to SORA in 2006, there will be no obligation to register unless he chooses to leave the State for work or other activity which might trigger a registration requirement elsewhere. But for the individual who moves to Indiana after 2006, the other jurisdiction requirement will come into play and trigger a duty to register so long as he was subject to registration in his former state. In this respect, this case is not unlike Hooper, 472 U.S. 612, 105 S. Ct. 2862, in which a fixed-point residency requirement made one’s eligibility for a veteran’s tax exemption turn upon whether he was a resident of New Mexico as of a particular date. The Supreme Court held unequivocally that the State could not No. 19-2523 33

discriminate against its own citizens based on the timing of their arrival in the State. Id. at 623, 105 S. Ct. at 2868.14 Here, relocation to Indiana after 2006 will not always trigger a requirement to register under SORA: If one was not required to register in his former home state, the move by itself will not require registration in Indiana. At the same time, relocation to Indiana from a jurisdiction where one was required to register is not the only way in which the other jurisdiction requirement is triggered: work, study, or other activity engaged in by an Indiana resident (however long-tenured) in another jurisdiction that requires him to register there will trigger a duty to register in Indiana. Even so, as to relocating citizens who arrive in Indiana after 2006, the other jurisdiction requirement operates to create two classes of otherwise similarly-situated residents, one of which must register and one of which need not. Cf. Saenz, 526 U.S. at 497 & n.8, 119 S. Ct. at 1523 & n.8 (because California public benefit levels were sixth highest in nation, its one-year cap on benefits for new residents would not adversely affect all persons relocating to California but only those arriving from one of 44 states or

14 By contrast, in Sklar v. Byrne, 727 F.2d 633 (7th Cir. 1984), the City of Chicago had banned the possession of handguns beginning on April 10, 1982, while grandfathering the rights of city residents who had registered their handguns before that date. The plaintiff, who had moved to Chicago from a suburb shortly after the handgun ordinance took effect and thus could not lawfully possess a gun, argued that the ordinance disadvan- taged new residents of the city and in that way interfered with his right to travel. We held to the contrary, noting that the ordinance did not “single out new residents of Chicago for discriminatory treatment.” Id. at 638. Ra- ther, any Chicago resident, new or longstanding, who did not possess a registered handgun before the ordinance took effect would be unable to do so thereafter. Id. Any impact on the travel rights of new Chicago resi- dents was “only indirect.” Id. 34 No. 19-2523

District of Columbia that had lower benefit levels). Creating such classes of citizens based on the fact and timing of their relocation is directly at odds with the Fourteenth Amendment’s Privileges or Immunities Clause. See id. at 504– 07, 119 S. Ct. at 1527–28. Second, to the extent the State’s theory presumes that the other jurisdiction requirement applies to plaintiffs Bash and Snider, who relocated to Indiana before that requirement was added to SORA in 2006, a few additional words are in order. The State’s counsel has insisted that the other jurisdiction re- quirement applies to them as it does to the other plaintiffs be- cause Snider and Bash were required to register in their for- mer states of residence (Ohio in Bash’s case, Michigan in Snider’s). This is obviously consistent with the State’s broader point that the timing of one’s arrival in Indiana does not mat- ter. Our dissenting colleague shares this understanding. But we cannot see any possible reason why the Indiana Supreme Court’s Wallace decision permits this application given the timing of Bash’s and Snider’s arrival in Indiana. When Bash and Snider arrived in Indiana pre-2006, the other jurisdiction requirement was not yet in effect and, of course, that is the only provision on which the State now relies to justify their obligation to register. So, at the time they relocated to Indiana, they were in the same position as a lifelong resident of Indi- ana with a similar criminal history: there was no provision re- quiring them to register (at least not one the State is willing to rely upon now, or one that the Indiana Supreme Court has not found to be unconstitutional).15 Whatever registration

15 As a factual matter, we know that Snider was nonetheless required to register when he moved to Indiana in 2003, presumably on the basis that his offense was the substantial equivalent of one that SORA deems No. 19-2523 35

requirements to which Bash and Snider had been subject else- where came to an end upon their becoming citizens of Indi- ana.16 Re-imposing a duty to register in 2006, when SORA was amended to include the other jurisdiction requirement, would not constitute a mere continuation of the registration obliga- tions to which these men were subject in their former states, which the Indiana Supreme Court has permitted; it would in- stead constitute a resurrection of an expired obligation and, to that extent, a material and detrimental change in their obli- gations as Indiana citizens. Compare Wallace, 905 N.E.2d at 384 (“[SORA] violates the prohibition on ex post facto laws contained in the Indiana Constitution because it imposes bur- dens that have the effect of adding punishment beyond that which could have been imposed when [Wallace’s] crime was committed”), with Tyson v. State, 51 N.E.3d 88, 96 (Ind. 2016) (“[U]nlike Wallace, where the offender had no obligation to register anywhere before the Act was passed, Tyson was re- quired to register in Texas years before our statutory defini- tion was amended to include him …; the challenged

registrable. But that was before the Wallace decision in 2009 made clear that this was not permissible under Indiana’s ex post facto clause. After Wallace was decided, Indiana freed Snider from any registration obligation until 2016, when, upon inquiry from Michigan (Snider’s former State of residence), Indiana again required him to register—this time, apparently, on the basis of the other jurisdiction requirement. 16 New York’s SORA has been interpreted to require an offender’s contin-

uing registration in that state notwithstanding his relocation to another state. Doe v. O’Donnell, 924 N.Y.S.2d 684, 686–87 (App. Div. 2011). It is likely an outlier in that regard, however. See Samantha R. Millar, Note, Doe v. O’Donnell and New York’s Sex Offender Registration Act: The Problem of Continued Registration under SORA After Leaving the State, 38 CARDOZO L. REV. 337, 359–61 (2016) (contrasting New York’s law with those of Mich- igan, California, and New Jersey in this respect). 36 No. 19-2523

amendments merely lengthened that requirement. We simply cannot say that transferring the obligation upon moving is any more punitive than lengthening it to potentially last a life- time.”) (emphasis in original), State v. Zerbe, 50 N.E.3d 368, 371 (Ind. 2016) (“the significant responsibilities with respect to Zerbe’s registration are merely maintained across state lines, to be fulfilled where he currently lives and works”), and Ammons v. State, 50 N.E.3d 143, 145 (Ind. 2016) (per curiam) (“Because Ammons was already under an obligation to regis- ter [in Iowa] and [SORA’s other jurisdiction provisions] do not impose any additional punishment on him, we find no ex post facto violation.”). Why the State believes it is lawful to effect such a material change in their obligations years after they became Indiana citizens is a puzzle, and one it never ex- plains: If Indiana’s ex post facto clause would forbid the appli- cation of the substantial equivalency requirement to any of the plaintiffs, as the State has conceded it would, why would it not also preclude the application of the other jurisdiction requirement to Snider and Bash, who relocated to Indiana be- fore there was any such requirement? The district court made no findings as to how SORA’s other jurisdiction requirement operates, in light of Wallace, vis-à-vis persons who arrived in Indiana before the requirement’s enactment (so far as we can discern, it was not asked to do so). But we can see no rationale for allowing the other jurisdiction prong to be applied to per- sons who arrived before the enactment of that provision and who were, in the years between their arrival and the enact- ment, free and clear of any lawful registration obligation. Cer- tainly, no Indiana case cited by the State or the dissent con- dones a revival, as opposed to a continuation across state lines, of a duty to register. Ultimately, we need not reduce this conclusion about Indiana law to a holding, as, at the end of No. 19-2523 37

the day, our conclusion about the right to travel precludes the State from imposing a registration requirement on any of these plaintiffs. It is worth noting, however, that even without our consideration of federal constitutional law, it seems that under the Indiana Supreme Court’s holding in Wallace, nei- ther Snider nor Bash (both of who arrived in Indiana prior to the enactment of the other jurisdiction requirement) should ever have been subject to a registration requirement in Indi- ana. Even if the State and the dissent are correct that, as a mat- ter of state law, the other jurisdiction requirement is fully ret- rospective and can properly apply to individuals like Bash and Snider, there remains a dichotomy among Indiana resi- dents based on the date of their arrival in Indiana. One who was a resident of Indiana before SORA required registration for his offense and remains so thereafter is not subject to a duty to register, period (Wallace leaves no doubt in that regard at all), whereas one who arrived in Indiana later may be sub- ject to registration pursuant to the other jurisdiction require- ment. In that respect, the timing of one’s relocation to Indiana still matters. Third, although the applicability of SORA as relevant here depends on whether or not a new citizen was subject to com- parable requirements elsewhere, the State’s theory that it is merely recognizing and choosing to continue a burden im- posed by another jurisdiction cannot somehow obviate the ef- fect that its actions have on the right to travel. Indiana may not be burdening newly-arrived sex offenders vis-à-vis the re- quirements that their former states of residence imposed on 38 No. 19-2523

them, as our dissenting colleague emphasizes,17 but Indiana certainly is treating them less favorably as compared with sex offenders who lived in the State before SORA’s other jurisdic- tion requirement was enacted; and the relative burden Indi- ana imposes on new arrivals is necessarily one that penalizes the exercise of one’s constitutional right to relocate to another state. Indiana’s statutory scheme is no different in kind from California’s effort to temporarily cap a new citizen’s welfare benefits at the amount of assistance she received in her former state of residence (assuming that amount was lower than what California would otherwise provide). Nominally, that cap did not penalize and therefore did not dis-incentivize an indigent person’s decision to relocate to California, because she would not receive less than what she had in her former home state; in that sense, there was no direct burden imposed on the exercise of one’s right to travel from state to state. 526 U.S. at 504, 119 S. Ct. at 1527. But that was “beside the point” as far as the Supreme Court was concerned. Ibid.

17 It bears noting that in at least some cases, SORA’s other jurisdiction re- quirement increases the burden upon a newly arriving sex offender as compared with what would have been required of him in his former home State. Recall that Illinois required Rice to register only for a period of ten years following his release from prison. But because Rice qualifies as a “sexually violent predator” under SORA, Indiana requires him to register for life (although he does have the ability to petition for removal of this designation). But the Indiana Supreme Court has deemed such marginal effects on an offender’s obligations to be insufficient to trigger the state’s ex post facto clause. See Jensen v. State, 905 N.E.2d 384 (Ind. 2009); Lemmon v. Harris, 949 N.E.2d 803 (Ind. 2011); but cf. Gonzalez v. State, 980 N.E.2d 312 (Ind. 2013) (increase of registration obligation from ten years to life violates ex post facto clause where limited opportunity to file petition for removal did not permit offender to argue he was rehabilitated and no longer posed a threat to the public). No. 19-2523 39

Were we concerned solely with actual deter- rence to migration, we might be persuaded that a partial withholding of benefits constitutes a lesser incursion on the right to travel than an outright denial of all benefits. See Dunn v. Blum- stein, 405 U.S. 330, 339, 92 S. Ct. 995, 31 L.Ed.2d 274 (1972). But since the right to travel embraces the citizen’s right to be treated equally in her new State of residence, the discriminatory clas- sification is itself a penalty. Saenz, 526 U.S. at 504–05, 119 S. Ct. at 1527. That is precisely the problem here: As to relocating individuals, Indiana has established a two-tiered system of regulating offenders that is tied in the first instance to when the individual became a citi- zen of Indiana. For constitutional purposes, the relevant com- parison is not how Indiana treats an offender versus how his former state of residence treated him, but rather how Indiana distinguishes among its own citizens based on whether they arrived pre- or post-enactment of SORA’s other jurisdiction provision. Indiana grants the former the full protection of its ex post facto clause but deprives newer arrivals of the same protection. See Saenz, 526 U.S. at 502, 119 S. Ct. at 1526 (third aspect of the right to travel encompasses “the right of the newly arrived citizen to the same privileges and immunities enjoyed by other citizens of the same State”); Soto-Lopez, 476 U.S. at 904, 106 S. Ct. at 2322 (“the right to migrate protects residents of a State from being disadvantaged, or from being treated differently, simply because of the timing of their mi- gration, from other similarly situated residents”). And, unlike the California scheme at issue in Saenz, the burden that one incurs by arriving in Indiana after the enactment of SORA is not temporary, insofar as the plaintiffs here are concerned, 40 No. 19-2523

but permanent. See Hooper, 472 U.S. at 623, 105 S. Ct. at 2869 (“the Constitution will not tolerate a state benefit program that ‘creates fixed, permanent distinctions … between … clas- ses of concededly bona fide residents, based on how long they have been in the State”) (quoting Zobel, 457 U.S. at 59, 102 S. Ct. at 2312). Indeed, on what we might call the State’s “lesser of two state citizenships” theory, Indiana would be free to create a boundless array of classes among its citizens tied to the greater regulatory burdens imposed by their for- mer home states (theoretically subjecting them to differential treatment on anything from tax rates to minimum drinking ages). The Supreme Court’s right-to-travel jurisprudence plainly forecloses such scenarios. See Saenz, 526 U.S. at 507, 119 S. Ct. at 1528; Zobel, 457 U.S. at 64, 102 S. Ct. at 2314–15. To return to a key point of the dissent: No, the overlap be- tween the set of offenders whom SORA burdens with a regis- tration requirement and the set of offenders who relocate to Indiana is not complete. Some number of relocating individ- uals will not be affected by the other jurisdiction requirement because their former domiciles did not require them to regis- ter. And some number of offenders will be required to register not because they are relocating from other states where they registered but because they have commuted from Indiana for work or study to other states that mandated their registration. The latter group, as we have noted, has still engaged in inter- state travel, but not the sort of travel for relocation to another state that Saenz and the other residency cases address. Mem- bers of that group may have their own constitutional claim, but it is not one that is presented here. But the fact that relo- cation from one state to another is not the exclusive means of triggering the other jurisdiction requirement and does not in- variably result in a registration obligation in Indiana does not No. 19-2523 41

detract from the inescapable point that for the six plaintiffs here, relocating to Indiana from other states has yielded them a set of rights and obligations that is concretely different from, and more burdensome than, the rights and obligations of of- fenders who arrived in Indiana before they did. Our dissenting colleague cites Connelly v. Steel Valley Sch. Dist., 706 F.3d 209 (3d Cir. 2013), in an effort to demonstrate that it is not the plaintiffs’ travel that explains their differen- tial treatment by Indiana. But key distinctions between Con- nelly and this case actually demonstrate the opposite. In Connelly, a Pennsylvania school district gave less credit to its teachers for prior out-of-state teaching experience than it did for in-state experience in establishing starting salaries. The plaintiff, who had lived and taught in Maryland before relocating to Pennsylvania, argued that he was being treated differently based on his former domicile in Maryland, in vio- lation of his right to travel. In rejecting that claim, the court emphasized that the school district was not treating its teacher-citizens differently based on where they came from. Rather, what mattered was where they gained their prior ex- perience. As a result, a former Maryland citizen whose prior teaching experience was in neighboring Pennsylvania (the two states share a border along the Mason-Dixon line) would receive full credit for his experience, whereas a lifelong Penn- sylvania citizen who previously taught in Maryland would receive only partial credit. 706 F.3d at 214–15. “[O]nly the teacher’s lack of Pennsylvania teaching experience—not his residency—would adversely affect his starting pay.” Id. at 215. Any burden this scheme imposed on interstate travel was, at most, “incidental.” Id. The court went on to hold that the school district had a rational basis for paying its teachers 42 No. 19-2523

differently based on the extent of their teaching experience in Pennsylvania: Teachers with experience working at Pennsyl- vania schools have a greater familiarity with the state’s edu- cational policies, procedures, and regulations and are likely to have a better grasp of what teaching methods are likely to be successful in achieving the state’s educational goals. Id. at 216–17. Our colleague analogizes the registration obligation that an offender has borne in another state to the teaching experi- ence that the plaintiff in Connelly acquired in another state: both are historical facts, on their face unconnected to an indi- vidual’s travel, that may legitimately inform a state’s judg- ment as to how the individual should be treated as a newly- arrived citizen. The difference is that when other states required the plaintiffs to register, they did so based on the very same criminal history that Indiana itself would not treat as sufficient to require registration. Individual states are, of course, free to reach different conclusions about what offenses require registration—that is a feature of our federalist system of government. And just as states may compile their own lists of registrable offenses, they may (and do) reach different conclusions as to whether the ex post facto provisions in their own constitutions permit the retrospective imposition of a duty to register. Indiana has conceded that its own ex post facto provision, as construed by Wallace, prohibits the application of the Indiana SORA’s substantial equivalence requirement to the plaintiffs: they were all convicted before their offenses were identified as registrable offenses by the Indiana Legislature. Had they been Indiana citizens at the time of No. 19-2523 43

those offenses, and remained in Indiana thereafter, they could not have been required to register. So unlike the school district in Connelly, which relied on the mise en scène of one’s prior experience and the perceived qualitative differences between in-state and out-of-state teaching experience, Indiana is not relying on something that any of the plaintiffs did in another state that distinguishes him from an otherwise similarly-situated Indiana sex offender— e.g., commit another offense, violate the terms of his proba- tion, or fail to comply with his registration and reporting ob- ligations. No plaintiff did anything outside of Indiana that would have triggered an obligation to register under SORA had he done it in Indiana. Indiana is relying solely on another state’s judgment that registration was required in that other state, so long as the offender was present in that state. Upon his relocation to Indiana, the State requires each plaintiff to carry that obligation with him, notwithstanding the fact that Indiana itself would not have imposed that obligation in the first instance. A simplified hypothetical helps to make clear why it is the offender’s relocation to Indiana from another state that is the real trigger for the mandate that he register in Indiana. Con- sider two offenders, A and B, who are similar in all respects but for the fact that A lives in Indiana and B lives in Illinois. In 1993, both commit the same sex offense, are convicted in their respective states, and commence six-year prison terms. In 1994, both states enact laws requiring lifetime registration for the sex offense that A and B committed; but the Indiana legislature makes its registration obligation prospective only, whereas Illinois makes the obligation fully retrospective. Thus, when A is released from prison in 2000, he is not 44 No. 19-2523

obliged to register in Indiana, but B must register in Illinois. In 2006, Indiana adopts a requirement obliging an Indiana resident who has been required to register in another jurisdic- tion to register in Indiana. In 2008, B moves to Indiana. Rely- ing on the other jurisdiction provision, Indiana demands that he register. Compare A and B, who are now both Indiana cit- izens. There is no difference in their criminal histories: They committed the same offense in the same year (and in all other respects are similar) so the danger that they pose to Indiana citizens is the same. But Indiana, in the exercise of its judg- ment, has not deemed the 1993 offense to be registrable, so A has never been required to register. Illinois required B to reg- ister for life, but now that he has moved to Indiana, that obli- gation is no longer operative. As a newly arrived citizen of Indiana, what distinguishes B from A? The answer is plain: B relocated from a state that required him to register. His travel is the one and only cause of his duty to register in Indiana, and in that regard, he is being treated differently from A, a lifelong resident of Indiana. The point is more clearly made if we assume that both A and B were Indiana citizens from the beginning and commit- ted the same sex offense in Indiana in 1993 and that B in 2008 relocated briefly to (and established residency in) Illinois, which required him to register, before returning to Indiana the following year. Indiana, relying on the other jurisdiction provision, now requires B to register in Indiana as well. Has B’s criminal history changed? No. Did he do anything that materially changed the risk that he might pose to his fellow Indiana citizens? No. All that occurred was his move to an- other state that required him to register before he returned to and re-established residency in Indiana. No. 19-2523 45

The plaintiffs no doubt belong to a relatively small class: They all committed their crimes before those offenses became registrable in Indiana (twenty-five or more years ago) and be- fore Indiana adopted the other jurisdiction requirement (four- teen years ago). With the passage of time, this class will dis- appear, and any individual who commits a sex offense will have been on constructive notice that registration is an obli- gation that he will have to shoulder. But however small in number the plaintiffs may be, Indi- ana has assigned them to a class of citizenship that is inferior to that enjoyed by other, similarly situated Indianans, and for the plaintiffs, it is their relocation from other states that has resulted in that second-class status. Indiana, as a matter of its own statutes and judicial precedents, would not have re- quired the plaintiffs to register had they lived in Indiana prior to 2006, when the other jurisdiction requirement was enacted. Only their travel from states that did require them to register has triggered this burden. The Supreme Court’s right-to- travel jurisprudence instructs that this two-tiered model of state citizenship is not permissible under the Fourteenth Amendment’s Privileges or Immunities Clause. A sex of- fender who has paid his debt to society has the right to relo- cate to Indiana like any other individual and be treated on equal terms with other similarly-situated citizens. These plaintiffs have been denied that equitable treatment. Indiana has, in effect, told the plaintiffs, “You are not from here. In- stead of applying our rules to you, we will apply your former state’s rules.” In this way, the plaintiffs remain outsiders in Indiana’s regulatory framework. Indiana nonetheless argues that because it is not denying newer arrivals a public benefit, a tax exemption, or the right 46 No. 19-2523

to vote,18 the Supreme Court’s right-to-travel cases are inap- plicable. It may be true as a factual matter that the Court’s du- rational and fixed-point residency cases have not addressed the right to travel beyond these sorts of factual contexts.19 But to return to first principles, the right at issue here is the right

18 See, e.g., Dunn v. Blumstein, 405 U.S. 330, 92 S. Ct. 995 (1972) (state laws requiring prospective voter to have been resident for one year in state and three months in county penalize individuals who have migrated to state in order to establish new residence during qualifying period, thereby in- terfering with right to travel and violating equal protection). 19 It is worth noting, however, that the Court’s decision in Edwards v. Cal., 314 U.S. 160, 62 S. Ct. 164 (1941) invalidated a California statute that made it a criminal offense to bring a non-resident indigent person into the state. The statute was born of fears triggered by the massive influx of migrants from other states as a result of the Dust Bowl and the Great Depression and the “problems of health, morals, and especially finance” attributed to that influx. Id. at 173, 62 S. C.t at 167. The Court pointed out that among the opportunities this exclusionary statute deprived indigent persons was “the opportunity to exert political pressure upon the California legislature in order to obtain a change in policy.” Id.at 174, 62 S. Ct. at 167. Because the statute was aimed at excluding indigents from the state, it arguably implicated all three components of the right to travel; and, indeed, the Court’s majority concluded that the statute interfered with interstate com- merce and as such was inconsistent with the Commerce Clause. Id. at 172– 77, 62 S. Ct. at 166–68. But the concurrences invoked the Fourteenth Amendment’s Privileges or Immunities Clause, making clear that among the rights implicated by the statute was a right of national citizenship that precludes a state from creating different classes among its residents with different rights. See id. at 181, 62 S. Ct. at 170 (Douglas, J., concurring) (state’s restriction of free movement “would permit those who were stig- matized by a State as indigents, paupers, or vagabonds to be relegated to an inferior class of citizenship”); id. at 184, 62 S. Ct. at 172 (Jackson, J., con- curring) (“We should say now, and in no uncertain terms, that a man’s mere property status, without more, cannot be used by a state to test, qual- ify, or limit his rights as a citizen of the United States.”). No. 19-2523 47

of a newly arrived citizen of Indiana to the same privileges and immunities enjoyed by other citizens of the state. See, e.g., Saenz, 526 U.S. at 502–05, 119 S. Ct. at 1526–27. Nothing in the Supreme Court’s jurisprudence concerning this right is tied to the nature of the particular benefit at issue; the right is one to citizenship on the same terms as other residents of the state. See ibid.; Soto Lopez, 476 U.S. at 904, 106 S. Ct. at 2322; Hooper, 472 U.S. at 623, 105 S. Ct. at 2868–69. SORA imposes a signifi- cant and lasting burden on a class of citizens who are other- wise similarly situated to those whose Indiana citizenship pre-dates the statute’s other jurisdiction requirement. As we have said many times now, those persons who were citizens of Indiana and were convicted of sex offenses before SORA’s other jurisdiction requirement was enacted enjoy the full pro- tection of the State’s ex post facto clause as construed in Wal- lace, and they are exempt from the various registration and reporting requirements that SORA imposes on such offend- ers. But those persons who were convicted of sex offenses be- fore the other jurisdiction requirement was enacted but have since relocated to Indiana are deemed subject to SORA’s reg- istration and other requirements—provisions that Wallace deemed punitive—so long as their former states imposed comparable requirements on them. In no sense are these two classes of Indiana citizens being treated equally: one is af- forded a valuable, constitutionally mandated protection against the after-the-fact imposition of what the Indiana Su- preme Court has labeled penalties, and one is not. Finally, Indiana posits that any understanding that a state cannot classify its citizens differently depending on whether they have engaged in interstate travel is inconsistent with the many federal criminal laws that apply precisely on that basis. This is a misguided argument. First, the sort of interstate 48 No. 19-2523

travel to which Indiana is referring implicates a different com- ponent of the constitutional right to travel than the right, upon relocating from one state to another, to be treated on equal terms with other citizens in one’s new home state. The latter component is the only one at issue here. Second, federal criminal law does not create separate classes of citizens within a state who enjoy lesser or greater rights depending upon when they became citizens of that state. It simply employs the jurisdictional “hook” of conduct within interstate commerce to regulate, on equal terms, all who move or act across state lines. This has nothing to do with the classes Indiana has es- tablished within its own citizenry. III. The other jurisdiction requirement of Indiana’s SORA im- poses a duty to register and its attendant burdens upon a re- locating citizen that it would not impose upon a lifelong Indi- ana resident. The Privileges or Immunities Clause of the Four- teenth Amendment prohibits this differential treatment. We affirm the district court’s judgment on this basis, without reaching the separate question of whether application of the other jurisdiction requirement also violates the ex post facto clause of the U.S. Constitution. AFFIRMED No. 19-2523 49

ST. EVE, Circuit Judge, dissenting. I disagree with the ma- jority’s conclusion that the Indiana Supreme Court’s interpre- tation of the Indiana Constitution’s Ex Post Facto Clause as applied to Indiana’s Sex Offender Registration Act (SORA) implicates plaintiffs’ right to travel under the Privileges or Im- munities Clause of the Fourteenth Amendment of the United States Constitution. I write separately to explain my reason- ing. I. A. Addressing the plaintiffs’ federal claims requires a deep dive into Indiana law and the reasons for Indiana’s complex rules surrounding SORA. The majority starts off on the wrong foot by misunderstanding those rules. Like other states, Indiana requires persons convicted of sex offenses to register as sex offenders. Ind. Code § 11-8-8-1 to -23. Under SORA, a person must register if he (1) commit- ted a registrable offense under Indiana law, (2) committed a crime under the laws of another jurisdiction that is substan- tially equivalent to Indiana’s registrable offenses, Ind. Code § 1-1-2-4(b)(3) (formerly Ind. Code § 11-8-8-5(a)(24)), or (3) is required to register in any other jurisdiction, Ind. Code § 11- 8-8-5(b)(1). As a statutory matter, SORA is fully retrospective and does not depend on when someone was convicted of an of- fense. And as a federal constitutional matter, this retrospec- tive application is not inherently suspect under the Federal Ex Post Facto Clause. See Smith v. Doe, 538 U.S. 84, 90 (2003). The Indiana Supreme Court, however, has interpreted its state constitution’s Ex Post Facto Clause such that persons who 50 No. 19-2523

would be required to register as a statutory matter are not re- quired to register as a constitutional matter. This began with Wallace v. State, 905 N.E.2d 371 (Ind. 2009), in which the court, applying its own version of the U.S. Supreme Court’s “intent- effects” test, held that SORA had a punitive effect on those who had been charged, convicted, and served their sentences before SORA was enacted. Id. at 379, 384. Even after Wallace, however, not all applications of SORA to prior convictions offend the Indiana Constitution. On the same day as Wallace, the Indiana Supreme Court decided Jen- sen v. State, 905 N.E.2d 384 (Ind. 2009). Jensen was convicted of a sex offense in 2000, and at the time, SORA required him to register as a sex offender for ten years. Id. at 389. Before Jensen’s ten years were up, the Indiana General Assembly amended SORA in 2006 to mandate that he now register for life. He argued that this extension violated the state’s Ex Post Facto Clause, but the Indiana Supreme Court disagreed. Un- like Wallace, who had no obligations before SORA was amended to cover him, the “‘broad and sweeping’ disclosure requirements were in place and applied to Jensen at the time of his guilty plea in January 2000. Nothing in that regard was changed by the 2006 amendments.” Id. at 394. The marginal effect of increasing only the length of an existing registration obligation did not rise to the level of “punishment” such that it violated the Indiana constitution. Id. at 391–93. After Jensen, the Indiana Supreme Court continued to fo- cus on the marginal effects of SORA and its amendments. In State v. Pollard, 908 N.E.2d 1145 (Ind. 2009), it said that a new residency restriction was “adding punishment.” Id. at 1154. The court’s decision in Lemmon v. Harris, 949 N.E.2d 803 (Ind. 2011), though, concluded that an amendment that reclassified No. 19-2523 51

someone from a sex offender to a “sexually violent predator” was not punitive because, just like for Jensen, it amounted only to an extension of pre-existing obligations and was not “any more punitive.” Id. at 810–11, 813 n.19. Up to this point, however, each case had asked whether the Indiana SORA had a marginal punitive effective com- pared to those requirements already imposed by Indiana law. In 2016, the Indiana Supreme Court encountered three cases challenging SORA’s effect on those who had been required to register under another state’s laws. Applying the same marginal-effects test, the Indiana Su- preme Court concluded that the effect of maintaining an out- of-state registration in Indiana was not punitive, regardless of when or where the registrable crime had been committed. First, in Tyson v. State, 51 N.E.3d 88 (Ind. 2016), the court up- held registration for a man obligated to register under Texas law at the time of his conviction, before Indiana’s SORA cov- ered his offense. Id. at 92. The court concluded that the mar- ginal effect of “maintaining a registry requirement across state lines does not amount to a punitive burden” in violation of the state constitution. Id. at 90. The court extended this reasoning in State v. Zerbe, 50 N.E.3d 368 (Ind. 2016). Zerbe was convicted in Michigan in 1992, before either Michigan or Indiana had enacted sex of- fender registration laws. Id. at 369. Zerbe was nevertheless re- quired to register under Michigan law upon release from prison because Michigan did not share Indiana’s stricter Ex Post Facto Clause. Id. at 371. This twist changed nothing; the marginal effect of maintaining that registration was not puni- tive. Id. at 370–71. As the court clarified, “it is not Zerbe’s crime that triggers his obligation to register as a sex offender in 52 No. 19-2523

Indiana; rather, it is his Michigan registry requirement that does so.” Id. at 370 (emphasis in original). The trilogy finished with Ammons v. State, 50 N.E.3d 143 (Ind. 2016) (per curiam). Am- mons had been convicted in Indiana before SORA, but he moved to Iowa, which obligated him to register for his Indi- ana crime. Id. When he moved back to Indiana, the Indiana Supreme Court confirmed that, just like for Tyson and Zerbe, maintaining Ammons’s Iowa registration for his Indiana crime did not amount to “additional punishment.” Id. at 145. In sum, the question under SORA and Indiana’s Ex Post Facto Clause is always whether SORA’s marginal effect is pu- nitive. Maintaining, extending, or modifying a duty under SORA generally is not punitive, but imposing a new duty is. It is immaterial to the analysis whether Indiana law is main- taining, extending, or modifying its own duties or those of an- other state. Likewise, it is immaterial where or when the con- viction occurred, as long as some state imposed a lawful reg- istration obligation on the offender and SORA does not so sig- nificantly alter that obligation to result in added punishment. B. Plaintiffs argue that the Indiana Supreme Court’s marginal-effects rule violates their right to travel under the Federal Constitution. While the majority aptly summarizes the development of the law regarding this right, it overreads the right to travel as articulated by the Supreme Court. The U.S. Supreme Court has identified three components to the right to travel: (1) “the right of a citizen of one State to enter and to leave another State,” (2) “the right to be treated as a welcome visitor rather than an unfriendly alien when temporarily present in the second State,” and (3) “for those No. 19-2523 53

travelers who elect to become permanent residents, the right to be treated like the other citizens of that State.” Saenz v. Roe, 526 U.S. 489, 500 (1999). Everyone agrees that only the third facet of the right is at issue here. That aspect is derived from the Privileges or Immunities Clause of the Fourteenth Amendment. U.S. Const. amend. XIV, § 2; Saenz, 526 U.S. at 502–03. In Saenz, the plaintiffs challenged the constitutionality of a California statute that limited new residents to only the wel- fare benefits to which they would have been entitled in their prior state of residence. 526 U.S. at 492. The Supreme Court held that this rule violated the third aspect of the right to travel. The Court was not concerned with whether California was trying to penalize or deter travel or even if it was suc- ceeding. Id. at 504. Instead, the Court found that “the right to travel embraces the citizen’s right to be treated equally in her new State of residence” and that “the discriminatory classifi- cation is itself a penalty.” Id. at 505. In addressing this discrim- ination, the Court applied strict scrutiny, which California’s law failed. Id. at 504–05. The duration of a citizen’s residency and the location of his or her prior residence had no relevance to the citizen’s welfare needs, and the bare desire to reduce the state’s budget was not compelling enough to justify a complex layered hierarchy among bona fide California resi- dents. Id. at 507. Saenz broadly stands for the proposition that durational residency requirements violate the right to travel unless they pass strict scrutiny. The full scope of that right, however, is uncertain. There have been no Supreme Court decisions inter- preting the right to travel after Saenz. Decisions before it held other durational residency requirements unlawful but did so 54 No. 19-2523

under the Equal Protection Clause. See Mem'l Hosp. v. Mari- copa Cty., 415 U.S. 250, 251, 261–62 (1974) (striking down a state law requiring an indigent person to be a county resident for one year to receive free medical care); Dunn v. Blumstein, 405 U.S. 330, 334–35, 360 (1972) (holding unlawful a state law permitting only residents who have lived in state for one year to vote); Shapiro v. Thompson, 394 U.S. 618, 622 (1969) (invali- dating statutes that deny welfare assistance to individuals during their first year of residency). But see Sosna v. Iowa, 419 U.S. 393, 396, 409 (1975) (upholding an Iowa law requiring resident to live in state for one year to obtain a divorce de- cree). I agree with the majority, however, that the right to travel should be understood to go beyond prohibiting only durational residency requirements that place a waiting period on benefits. It seems unlikely that a permanent distinction between bona fide residents would be any more lawful than a temporary one. The Supreme Court’s cases illustrate this point, though a majority of the Court has yet to endorse it. In Zobel v. Williams, 457 U.S. 55 (1982), Alaska implemented a natural resource dividend statute that created “fixed, permanent distinctions between an ever-increasing number of perpetual classes of concededly bona fide residents, based on how long they have been in the State.” Id. at 59. The Supreme Court held that this scheme was improper even under rational-basis review. Id. at 64. The Court did the same thing with a New Mexico tax exemption for Vietnam veterans who were state residents before a specific date. Hooper v. Bernalillo Cty. Assessor, 472 U.S. 612, 624 (1985). Although the Supreme Court did not directly hold that the laws at issue in Zobel and Hooper implicated the right to travel, No. 19-2523 55

a plurality of the Court later concluded that the right was re- ally driving those decisions. See Att’y Gen. of N.Y. v. Soto- Lopez, 476 U.S. 898, 907–08 (1986) (plurality opinion). In the plurality’s view, “a permanent deprivation of a significant benefit, based only on the fact of nonresidence at a past point in time, clearly operates to penalize appellees for exercising their right to migrate” and thus mandated strict scrutiny. Id. at 909. All of the Supreme Court’s decisions in this area have something in common. Each involved a rule that explicitly discriminated between old and new residents. As the Court noted, the challenged classifications in Saenz were “defined entirely by (a) the period of residency in California and (b) the location of the prior residences of the [plaintiffs].” 526 U.S. at 505 (emphasis added). Likewise, in Soto-Lopez, the plurality emphasized that New York was depriving the plaintiffs “of a significant benefit, based only on the fact of nonresidence at a past point in time.” 476 U.S. at 909 (emphasis added). In each case, there was a direct causal connection between a person’s status as a new resident and the deprivation of a benefit. In legal parlance, each was a “disparate treatment” claim. II. The question in this case is whether Indiana’s registration requirement, as applied through the marginal-effects test, vi- olates the right to travel identified by the Supreme Court. In other words, does the marginal-effects test treat bona fide res- idents differently based on when they became residents. I part ways with the majority because I conclude it does not. Neither SORA nor Indiana’s Ex Post Facto Clause dis- criminates based on residency. Neither even mentions 56 No. 19-2523

residency. As a statutory matter, SORA obligates all people— both old and new residents—to register based on prior con- victions. Indiana’s Ex Post Facto Clause then relieves a subset of those who must register from that statutory obligation. Who receives the clause’s benefits, though, does not depend on when one became an Indiana resident but on whether one is subject to an existing registration requirement. That re- quirement can come from Indiana or from another state. The twist in this case is that for those like the plaintiffs, convicted before Indiana’s SORA covered their crimes, such a registra- tion obligation must come from elsewhere. For the majority, that fact is determinative. The majority offers two theories to support its view that Indiana violated the plaintiffs’ right to travel. The primary argument is that the other-jurisdiction provision in SORA “creates two classes of Indiana citizens”: those who were Indiana residents prior to the provision’s enactment on July 1, 2006, and those who moved to Indiana after that date. Because the provision applies only to newer Indiana residents who moved to the state after July 1, 2006, the argument goes, the provision impermissibly classifies Indiana residents based on the length of their residency in the state and generally treats newer residents worse than long-term Indianans. I disagree because the underlying premise to this conclusion is incorrect—the other-jurisdiction provision does apply retroactively to offenders who became Indiana residents prior to July 1, 2006. The majority’s conclusion to the contrary is rooted in its narrow reading of Indiana caselaw applying the state’s Ex Post Facto Clause to SORA. It suggests that any gap in time between the placement of an initial registration requirement No. 19-2523 57

and the later imposition of a related requirement transforms Indiana’s subsequent requirement into a “material change” in obligations rather than a “mere continuation” of them, failing the marginal-effects test. Under this logic, offenders who relocated to Indiana prior to July 1, 2006, and who were not required to register in Indiana until 2006 would be exempt from a later registration requirement under Wallace. That interpretation misapprehends Wallace’s more recent progeny: Tyson, Zerbe, and Ammons. While those cases dealt with plaintiffs who moved to Indiana after it enacted SORA’s other-jurisdiction provision in 2006, the Indiana Supreme Court did not base its decisions on that factor. Rather, it held, across three distinct factual patterns, that SORA’s retroactive application does not violate the state’s Ex Post Facto Clause as long as the offender is “already required to register in another jurisdiction.” Zerbe, 50 N.E.3d at 369–70. That condition is satisfied here. Because SORA’s other-jurisdiction provision may apply equally to Indianans who became residents prior to July 1, 2006, as well as those who move to the state after that date, the majority’s argument that SORA classifies Indiana residents by date of residency is unpersuasive.1 The majority’s second, more implicit argument is that as a practical effect of Indiana’s SORA, out-of-state residency is a determinative factor in the plaintiffs’ case and that of other offenders like them. Undoubtedly having a registration obli- gation in another state is correlated with changing one’s state

1 I likewise depart from the majority’s suggestion that Bash and Snider

cannot be required to register under Indiana law; that conclusion depends upon an interpretation of Indiana law that bars retroactive application of SORA’s other-jurisdiction provision, which I reject. 58 No. 19-2523

of residence, as we can see not only from the six plaintiffs here but also from Tyson, Zerbe, and Ammons, all of whom had to register after moving to Indiana. But the correlation is imper- fect. Some lifelong Indiana residents who committed crimes before SORA might well have a registration obligation based on their employment or schooling in an adjacent state. See, e.g., 730 ILCS 150/3(a-5) (requiring out-of-state students or employees to register in Illinois). On the flipside, some new Indiana residents who committed their crimes elsewhere might have no registration requirement in their prior state be- cause of state-law protection against retroactivity. See, e.g., Doe v. State, 189 P.3d 999, 1004 (Alaska 2008) (holding that the state’s Ex Post Facto Clause prohibits retroactive application of SORA). While prior, out-of-state residency is often an ele- ment in the application of SORA’s registration requirements, residency is not the trigger for the other-jurisdiction provi- sion. The majority acknowledges this but suggests that the fact that some new residents are not adversely affected by SORA’s requirements does not immunize the law from unconstitu- tionality. For support, the majority points to Saenz. There, the Supreme Court struck down the law, even though some of the new residents experienced more favorable welfare benefits than long-term Californians. Saenz, 526 U.S. at 497. What mat- tered, the Court concluded, was that California explicitly based its provision of benefits to new Californians—favorable or not—on the duration of their residence in California. Id. at 497, 505. The implication of the majority’s citation to Saenz is that SORA likewise cannot be saved by the fact that some new Indianans may not be subject to the registration requirements while some lifelong Indianans may be covered. That is a false equivalence. California employed an express, durational- No. 19-2523 59

residency classification; it applied to all persons who had re- sided in California for less than a year and happened to pro- duce a favorable effect for some of them. In contrast, SORA by its terms does not base its application on any length of res- idency in Indiana; new Indiana residents who arrive in Indi- ana without any prior registration requirements do not expe- rience a favorable effect under SORA—the law simply does not apply to them at all. In sum, Saenz involved a discrimina- tory test that some new residents passed, while SORA in- volves a non-discriminatory test that some new residents fail. Prior, out-of-state residency represents neither causation nor perfect correlation for the application of SORA’s registration requirements, and there is no evidence that anyone in Indiana intended to deter travel through the other- jurisdiction provision. The result? A disparate-treatment claim under the right to travel necessarily must fail. All that is left is a disparate-impact claim—an argument that, as a practical matter, more new residents than old residents must register under the law. The Supreme Court, though, has never extended the right to travel this far. Cf. Washington v. Davis, 426 U.S. 229, 239 (1976) (holding that only disparate treatment or discriminatory purpose violates the Equal Protection Clause). Nor, to my knowledge, has any other court of appeals before today. The Third Circuit has actively refused to take this step, and I would follow its lead. In Connelly v. Steel Valley Sch. Dist., 706 F.3d 209 (3d Cir. 2013), a Pennsylvania school district set its teachers’ salaries based on years of teaching experience but gave full credit for years teaching in the district, partial credit for years teaching in Pennsylvania, and reduced credit for years teaching elsewhere. Id. at 211–12. A teacher who taught 60 No. 19-2523

nine years in Maryland and received one year of credit argued that the school district’s salary scheme violated his right to travel. Id. at 213. The Third Circuit recognized that the district was not discriminating based on duration of residency but on location of teaching experience. Id. at 214. A lifelong Pennsyl- vania resident who taught across the border in Maryland would have received the same treatment as a similar Mary- land resident who moved to Pennsylvania. Id. at 214–15. In the Third Circuit’s view, “[t]he right to travel simply is not implicated when there is no discrimination based on the du- ration of one's residency." Id. at 215. Because SORA likewise does not discriminate based on the duration of one’s resi- dency but rather on the existence of a registration obligation, I would conclude that it does not implicate the right to travel or merit strict scrutiny. There are good reasons for limiting the right to travel to actual discrimination, as we recognized more than a decade before Saenz. In Sklar v. Byrne, 727 F.2d 633 (7th Cir. 1984), we considered a right-to-travel objection to an ordinance banning unregistered handguns in the City of Chicago. Because one needed to be a Chicago resident to register a handgun for law- ful possession—and Chicago stopped new registrations after 1982—Sklar argued that the ordinance violated the right to travel. Id. We recognized then that this could not be how the right to travel functions, for applying strict scrutiny “based merely on a showing that newer residents would not benefit” would make huge swaths of the law vulnerable. Id. at 639. Like in Sklar, the plaintiffs here want to apply strict scrutiny on the showing that they, as new residents, are “merely one group among several who do not benefit” from the protec- tions of Indiana’s Ex Post Facto Clause. Id. at 639. We refused the invitation in 1984, and I would refuse it again now. No. 19-2523 61

The Privileges or Immunities Clause of the Fourteenth Amendment simply does not prohibit a state from inci- dentally burdening travel to or from the state. It guarantees only “the right of the newly arrived citizen to the same privi- leges and immunities enjoyed by other citizens of the same State.” Saenz, 526 U.S. at 502. Because both old and new Indi- ana residents are treated equally under SORA and Indiana’s Ex Post Facto Clause, I respectfully dissent from the major- ity’s holding that either law violates the right to travel. III. Because Indiana’s law does not implicate a fundamental right, it is subject to rational basis review. To survive this level of scrutiny, the Supreme Court has required that there be a rational basis for the classification. See FCC v. Beach Commc’ns Inc., 508 U.S. 307, 313 (1993) (stating that a statutory classifi- cation will survive rational-basis scrutiny “if there is any rea- sonably conceivable state of facts that could provide a rational basis for the classification”). Because the district court did not undertake a rational-basis review, I would remand this case to the district court to determine whether this level of scrutiny has been met.