AOYAGI, P. J.

Argued and submitted December 13, 2023

Board of Parole and Post-Prison Supervision

Anna Pervukhin Sammons argued the cause and fled the briefs for petitioner.

Benjamin Gutman, Solicitor General, argued the cause for respondent. Also on the brief was Ellen F. Rosenblum, Attorney General.

Before Aoyagi, Presiding Judge, Joyce, Judge, and Jacquot, Judge.

AOYAGI, P. J.

Petitioner seeks judicial review of a September 2022 order of the Board of Parole and Post-Prison Supervision. In that order, the board set petitioner's sex offender notification level (SONL) at Level 3 (High), based on the board's application of current OAR 255-085-0005 and OAR 255-085-0020. Petitioner challenges the order on several legal grounds, but we need only consider one, because it is dispositive. We recently held in Thomsen v. Board of Parole, 333 Or.App. 703, 704-05,P.3d(2024), that the foregoing rules are invalid because they conflict with the enabling statute, specifically by requiring the board to classify some registrants based on their historical recidivism risk rather than their recidivism risk at the time of classification. We reverse and remand on the basis addressed in Thomsen, without reaching petitioner's other arguments.

BACKGROUND

In 2004, petitioner was convicted of sex offenses that he committed when he was 22 years old. He received a probationary sentence and was required to register as a sex offender. Petitioner has not been convicted of any new sex offenses since 2004, although he has twice been convicted of failure to register.

Oregon's sex offender registry is discussed at some length in Sohappy v. Board of Parole, 329 Or.App. 28, 540 P.3d 568 (2023). In short, as relevant here, ORS 163A.100 requires the board to classify registered sex offenders into three risk levels based on the risk of reoffending that each person" 'presents,'" which determines the" 'range of notification'" that the person "'requires.'" Id. at 40 (quoting ORS 163A.100). The board has decided to use the Static-99R as its actuarial risk assessment instrument for adult males. Id. at 41. The Static-99R entails scoring a person on 10 factual items, resulting in a total score that provides a statistically valid estimate of the person's relative risk of committing a new sex offense as of the date of his release from supervision. See id. at 41-45 (describing the Static-99R). A person's Static-99R score validly predicts the person's risk of sexually reoffending for approximately two years from the release date. Id. at 42. Beyond those two years, if the person has not committed and continues not to commit any new sex offenses, the person's risk of reoffending declines in a predictable way, according to research related to the Static-99R. Id. at 42-43 ("[A] person's Static-99R score remains the same over time (absent a new index offense), while the risk of sexually reoffending changes over time, predictably declining for those who remain sex-offense-free in the community").

In this case, the board classified petitioner for the first time in its September 2022 order. Petitioner received a score of "6" on the Static-99R, representing the risk of sexually reoffending that he presented in 2004, when he was released into the community on his index offenses, through approximately 2006. Applying OAR 255-085-0020, the board relied entirely on petitioner's Static-99R score of "6" to set his SONL at Level 3 (High), without considering petitioner's conduct in the 18 years since his release or its effect on his present risk of sexually reoffending. Petitioner unsuccessfully challenged the board's approach in the board proceeding and now challenges it on judicial review.

ANALYSIS

We have issued a number of decisions in the last year regarding the board's rules on SONL classification. In Sohappy, 329 Or.App. at 28, and Watson v. Board of Parole, 329 Or.App. 13, 540 P.3d 20 (2023), we held that the board had misinterpreted a previous version of OAR 255-085-0020 when it classified registrants based on the risk of sexual reoffending that they presented years in the past, rather than the risk that they present at the time of classification. Most recently, in Thomsen, we held that the current versions of OAR 255-085-0005 and 255-085-0020 are invalid because "the board's methodology for initial classification of an individual's SONL does not measure what the legislature has charged it with measuring under ORS 163A.100 (i.e., present risk), but instead measures 'the registrant's risk as it was at the time of their release from their index offense.'" Thomsen, 333 Or.App. at 715 (quoting OAR 255-085-0005).

The board applied the current versions of OAR 255-085-0005 and OAR 255-085-0020 in this case. As such, this appeal is controlled by Thomsen, and we could simply reverse and remand without further discussion. The board's arguments have evolved over time, however, and we feel it prudent to address here a newer version of an argument that we first addressed in Watson and that the board makes again in this case in a somewhat different way. See Watson, 329 Or.App. at 24-25 (describing a board argument regarding the 2013 legislature's intent as to existing registrants). The thrust of the argument is that, notwithstanding the text of ORS 163A.100 itself, the larger statutory scheme of which ORS 163A.100 is a part evinces a legislative intent to classify all registrants based on their risk of reoffending at the time of release, not at the time of classification.

The board's argument focuses on the disparate treatment of two groups of people. The first group consists of registrants who are classified and given an SONL immediately upon their release into the community. The second group consists of registrants who are classified and given an SONL after years of living in the community. Imagine that the board classifies people in both groups in 2022. A person in the first group who scores "6" on the Static-99R will be initially classified in Level 3, based solely on that score, and consequently will never be eligible for reclassification below Level 2 or removal from the registry. See ORS 163A.125(3) (b) (so providing for registrants initially classified in Level 3). Meanwhile, a person in the second group who scores "6" on the Static-99R but has lived in the community for years without sexually reoffending will be initially classified in Level 2 or 1, based on the combination of his score and his offense-free time in the community, and consequently will be eligible for reclassification to Level 1 or removal from the registry. See ORS 163A.125(2)(c), 163A.125(1)(a) (so providing as to Level 2 and 1 offenders).

The board views the foregoing disparity as unfair. It argues that the legislature would have wanted everyone with the same Static-99R score to receive the same initial SONL classification, regardless of when initial classification occurs.

There is undoubtedly some tension in the statutory scheme, as we have acknowledged in prior cases. See, e.g., Watson, 329 Or.App. at 25. On the one hand, it requires initial classification based on the recidivism risk that each registrant "presents," while on the other hand it requires some registrants to remain on the registry for life regardless of changes in their actual risk profile. That tension may be due to the fact that, "when it created the three-tiered SONL system in 2013, the legislature likely was unaware of the significance of sex-offense-free time to sexual recidivism risk and likely was not thinking about that issue with respect to existing registrants." Sohappy, 329 Or.App. at 47 n 10; see also id. at 43 n 8 (noting research advancements since 2013). In any event, only the legislature can resolve it. Until and unless the legislature revisits the issue, there will tension within the statutory scheme regardless of our holdings in these cases-it is just a matter of which part of the statutory scheme we view as taking precedence with respect to the issue before us.

In our view, the statutory text and the purpose of the registry are the most persuasive indication of the legislative intent regarding initial classification. The legislature chose to create a classification system based on the recidivism risk that a person "presents." ORS 163A.100. That legislative choice is consistent with the fact that the registry serves a regulatory, rather than punitive, purpose. The stated purpose of the registry is "to assist law enforcement agencies in preventing future sex offenses." ORS 163A.045(1). In responding to constitutional challenges, "the Supreme Court has held that the registry is regulatory in nature, not punitive, in part precisely because '[t]he operation of the law conforms to the legislature's declared purpose[.]'" Sohappy, 329 Or.App. at 42 (quoting State v. MacNab, 334 Or. 469, 480, 51 P.3d 1249 (2002)); see also, e.g., State v. Benson, 313 Or.App. 748, 770, 495 P.3d 717, rev den, 369 Or. 69 (2021) (explaining that a registry requirement was not punitive because it "exists as part of a noncriminal regulatory framework, is individually regulatory in nature, and serves the noncriminal purpose of effectuating Oregon's sex offender registration system").

If the registry were punitive in nature, we might be more persuaded by the board's argument that the legislature intended (or would have intended) for all registrants who score similarly on the Static-99R to receive the same initial classification, regardless of their actual recidivism risk at the time of classification. Such an approach is at odds, however, with a regulatory system that is not supposed to be punitive but, rather, is supposed to protect the public by assisting law enforcement to identify the past offenders most likely to reoffend so that available resources may be directed toward them. See Sohappy, 329 Or.App. at 37-38 (discussing concerns about the registry from law enforcement and other stakeholders that led to the three-tiered system adopted in 2013).

We therefore reject the board's argument regarding disparate treatment of different groups of offenders and reaffirm our decisions in Sohappy, Watson, and Thomsen. Applying Thomsen, we reverse and remand the board's order in this case.

Reversed and remanded.