508 P.3d 1099
Cal.
2022

Conservatorship of the Person and Estate of ERIC B.

S261812

IN THE SUPREME COURT OF CALIFORNIA

April 28, 2022

PUBLIC GUARDIAN OF CONTRA COSTA COUNTY, as Conservator, etc., Petitioner and Respondent, v. ERIC B., Objector and Appellant.

First Appellate District, Division Five A157280; Contra Costa County Superior Court P18-01826

Justice Corrigan authored the opinion of the Court, in which Chief Justice Cantil-Sakauye and Justices Liu, Kruger, Groban, Jenkins, and Moore* concurred.

* Associate Justice of the Court of Appeal, Fourth Appellate District, Division Three, assigned by the Chief Justice pursuant to article VI, section 6 of the California Constitution.

Justice Kruger filed a concurring opinion, in which Justices Liu and Groban concurred.

Opinion of the Court by Corrigan, J.

The Lanterman-Petris-Short (LPS) Act authorizes one-year conservatorships for those gravely disabled by a mental disorder or chronic alcoholism. (Welf. & Inst. Code, § 5350.) Conservatorship proceedings are civil in nature, so the constitutional protections afforded criminal defendants do not directly apply. However, the Legislature has extended many of the same rights by statute to the commitment of persons found not guilty of crimes by reason of insanity (NGI‘s). (Pen. Code, § 1026.5, subd. (b)(7).) Among those is the right not to give compelled testimony at trial. (See Hudec v. Superior Court (2015) 60 Cal.4th 815, 826 (Hudec).) The question here is whether those facing conservatorship due to an inability to care for themselves should enjoy the same protection. We conclude that, for purposes of the right against compelled testimony, the groups are sufficiently similar that equal protection principles require the government to justify its disparate treatment of these proposed conservatees. However, because it is undisputed any error here was harmless, we need not decide what level of scrutiny is appropriate or whether the disparate treatment of conservatees can be constitutionally justified. We affirm the judgment.

I. BACKGROUND

The Contra Costa County Public Guardian (Public Guardian) petitioned for an LPS conservatorship on the ground that appellant Eric B. was gravely disabled. Appellant requested a jury trial on the petition and objected to giving compelled testimony, based on the holding in Hudec, supra, 60 Cal.4th 815. The court overruled the objection.

Psychiatrist Michael Levin, M.D., testified that appellant has chronic schizophrenia. Treatment included three medications, one of which required weekly white blood cell monitoring. Appellant‘s minimal insight about his illness made it difficult for him to cooperate with treatment. When not housed in a treatment facility, he had failed to take his medication, which aggravated his symptoms. Levin considered appellant gravely disabled and doubted he could provide for his basic needs without a conservatorship.

Therapist James Grey became appellаnt‘s case manager at the Concord Adult Mental Health Clinic in 2016, after paranoid behaviors put appellant‘s subsidized housing at risk. Appellant had tried to change door locks and damaged his apartment searching for monitoring devices. Although Grey arranged transportation for clinic appointments, appellant was usually unwilling to go. According to Grey, appellant displayed the paranoia, guardedness, and agitation typical of schizophrenia, and his cooperation with treatment was “very inconsistent.” Appellant had full bottles of medication that were months old and other psychiatric prescriptions went unfilled. The county had been serving as appellant‘s money manager, providing him an allowance, but he often failed to cash these checks. Appellant was treated as a psychiatric inpatient when a temporary conservatorship was ordered but was later released against Grey‘s advice. Within a week, he was admitted to an emergency psychiatric facility and was eventually transferred to his current placement. Appellant remained guarded and paranoid, with an extremely flat affect and disorganized thoughts. He sometimes believed his mother was not actually his mother and that others posed a threat to him. He had significant difficulty complying with treatment and medications and was generally unable to meet his needs for food and clothing without support.

Called to the stand by the Public Guardian, appellant testified that he lived in a board and care facility and was previously in an intensive treatment unit. After multiple questions about where he had lived, appellant remarked, “I didn‘t know[,] T-Con had to deal with being here and being there. It has nothing to do with each other.” He knew that Grey believed he should be moved from a temporary to a full conservatorship. Asked what he wanted to happen, appellant gave a rambling and partially incoherent response, asserting he might not need a conservatorship because, though he had a mental health disorder, he did not always need medications for it.1 He said he was told he had attention deficit disorder as a child. “I just had a learning disability. They didn‘t say anything about anxiety disorders or any manic problem or anything else like that.” He could name two of his medications but did not understand why he was taking them. He believed, “[T]here‘s just a basic medication standard issue in a given area. And they hand you medication.” Apparently referring to his inpatient admission, he said: “I was admitted out of unbreeching contract. There‘s something just going on.” Asked to clarify this statement, he responded, “This is penetrating. That‘s what I mean. We‘ll pass on this.” He acknowledged that he was “sort of still dependent” on his current program. He had no plans for where he would live or how he would support himself if released from the conservatorship. He thought he might get a job but acknowledged he had not worked since 2011. He said he would take his medications but when asked how he would pay for food responded, “Pay for food? Rely on the conservatorship.”

The jury found appellant gravely disabled. The court appointed the Public Guardian as conservator, ordered that appellant continue in his current placement, and restricted his ability to possess firearms and refuse treatment. On appeal, appellant challenged the order compelling his testimony. He argued that because the right to silence is statutorily provided in NGI extension proceedings, equal protection required that the same right should apply in the LPS context. The Court of Appeal held that LPS conservatees are similarly situated with NGI‘s for this purpose but ruled the error in compelling his testimony was harmless. Because the Court of Appeal expressly disagreed with the contrary holding in Conservatorship of Bryan S. (2019) 42 Cal.App.5th 190 (Bryan S.), we granted review to resolve the conflict.2

II. DISCUSSION

A. Overview of Relevant Civil Commitment Schemes

“California has no fewer than nine involuntary commitment procedures that may apply to persons who have various mental problems, and who pose a threat to their own welfare or to the safety of others. Some of these laws . . . operate in a manner largely independent of the criminal justice system. (See [Welf. & Inst. Code, § 4825 [developmentally disabled persons . . .], 5000 et seq. [mentally ill persons under the LPS Act].) Others apply depending on whether a criminal prosecution has occurred.” (People v. Barrett (2012) 54 Cal.4th 1081, 1093 (Barrett).) We discuss only the most pertinent commitment schemes here.

1. Extended Commitments Connected to a Criminal Case

NGI Commitments “A person found not guilty of a felony by reason of insanity may be committed to a state hospital for a period no longer than the maximum prison sentence for” the offense. (Hudec, supra, 60 Cal.4th at p. 818; Pen. Code, § 1026.5, subd. (a).) Thereafter, the district attorney may petition to extend the NGI commitment by two years if the person “represents a substantial danger of physical harm to others” because of “a mental disease, defect, or disorder.” (Pen. Code, § 1026.5, subd. (b)(1).) The respondent has a statutory right to representation by counsel and a jury trial. (Id., subd. (b)(3)-(4).) As discussed further below (see post, at pp. 13-15), statutes also require that NGI extension hearings comply with certain federal and state constitutional guarantees applicable in criminal proceedings. (Pen. Code, § 1026.5, subd. (b)(7).) The commitment can be renewed for two-yeаr periods without limitation, subject to the same procedural requirements. (Id., subd. (b)(10).) Although provided for by the Penal Code, NGI extension trials are considered “essentially civil in nature, rather than criminal, because they are directed at confinement for treatment rather than punishment.” (Hudec, at p. 819.) NGI‘s are typically confined in state hospital facilities. (See Pen. Code, § 1026, subd. (a).)

Other Criminally Based Commitments The Penal Code also provides for the involuntary civil commitment of violent offenders with mental health disorders (see Pen. Code, § 2960 et seq.) (OMHD‘s)3 and those convicted of sexually violent offenses (see Welf. & Inst. Code, § 6600 et seq.) (SVP‘s). In these instances, the person has been convicted of serious crimes and incarcerated. The civil commitment proceedings may be brought once the term of incarceration has ended. (Pen. Code, §§ 2970, subd. (b), 2972, subd. (c); Welf. & Inst. Code, §§ 6601-6603.) In both cases, the statutes provide for renewable terms of commitment, as well as the rights to counsel, jury trial, proof beyond a reasonable doubt, and a unanimous verdict. (Pen. Code, § 2972, subds. (a)(1)-(2), (e); Welf. & Inst. Code, §§ 6603, subd. (a), 6604.)4 As does appellant, we focus our analysis primarily on the comparison between LPS Act commitments and those under the NGI scheme.

2. LPS Act Commitments

The Legislature has also enacted a civil commitment scheme for involuntary mental health treatment without an underlying criminal offense. The LPS Act authorizes short-term involuntary detentions (see Welf. & Inst. Code, §§ 5150, 5250) and one-year conservatorships for those who are gravely disabled due to a mental health disorder or chronic alcoholism (see id., § 5350).

When a treatment professional determines a person is gravely disabled and unwilling or unable to accept treatment voluntarily, the county‘s public guardian may petition to establish a conservatorship. (Welf. & Inst. Code, § 5352; see Conservatorship of K.P., supra, 11 Cal.5th at pp. 708-709.) If the matter proceeds to trial and the person is found gravely disabled, the court appoints a conservator (Welf. & Inst. Code, § 5350), imposes “disabilities” as needed (id., § 5357), and determines an appropriate treatment placement (id., § 5358). (See Conservatorship of K.P., at pp. 709-710.) A conservatorship terminates after one year but may be extended for additional one-year terms upon petition. (Welf. & Inst. Code, § 5361.)

The LPS Act provides for two types of conservatorships. The first and most common is for those who are unable to meet their own needs for food, clothing, or shelter due to a mental health disorder. (Welf. & Inst. Code, § 5008, subd. (h)(1)(A).) This type, which we refer to as a traditional conservatorship, is the kind at issue here. Those subject to a traditional conservatorship have a right to be treated in “the least restrictive alternative placement” (id., § 5358, subd. (a)(1)(A)), with first priority given to their home or that of a relative (see id., subd. (c)(1)). However, a significant number of these conservatees are placed in locked facilities, including state hospitals. For example, as of February 2019, about 63 percent of LPS conservatees in the City and County of San Francisco were placed in locked facilities. (City and County of S.F., Budget and Legis. Analyst‘s Office, Policy Analysis Report: Review of Lanterman-Petris-Short (LPS) Conservatorships in San Francisco (Nov. 12, 2019) p. A-11 (San Francisco Analyst‘s Report).) As of November 2019, LPS conservatees made up approximately 11 percent of the population in state hospital facilities, with the remainder composed of individuals whose commitments arose from the criminal justice system. (Cal. State Auditor, Rep. No. 2019-119 (July 2020) Lanterman-Petris-Short Act: California Has Not Ensured That Individuals With Serious Mental Illnesses Receive Adequate Ongoing Care, p. 25 (State Auditor‘s Report).)5

A second type of LPS conservatorship, not at issue here, may be imposed when a person has been ruled incompetent to stand trial for a criminal accusation (see Pen. Code, § 1370) yet still “represents a substantial danger of physical harm to others by reason of a mental disease, defect, or disorder” (Welf. & Inst. Code, § 5008, (h)(1)(B)(iv)). This kind of commitment is commonly referred to as a “Murphy conservatorship,” after the legislator who sponsored the amendment adding this ground to the LPS Act. (Jackson v. Superior Court (2017) 4 Cal.5th 96, 102; People v. Karriker (2007) 149 Cal.App.4th 763, 775.) Criminal defendants ruled incompetent for trial are initially committed under Penal Code section 1370. If they do not regain competence within the statutory period, or if there is no substantial likelihood competence will be regained, the court will order the public guardian to initiate LPS proceedings. (Pen. Code, § 1370, subd. (c)(2); see Jackson, at p. 102.) A Murphy conservatorship may be imposed only if the person has been charged with a violent felony, a formal finding of probable cause supports the charge, a mental health disorder prevents the person from understanding the proceedings, and the person poses a substantial danger of physical harm to others. (Welf. & Inst. Code, § 5008, subd. (h)(1)(B).)6

LPS conservatees have the right to a jury trial to determine whether they are gravely disabled, as that condition is statutorily defined. (Conservatorship of K.P., supra, 11 Cal.5th at p. 709; see Welf. & Inst. Code, § 5350, subd. (d)(1).) They enjoy the right to counsel and a unanimous verdict based on proof beyond a reasonable doubt. We extended these trial rights to the LPS context in Conservatorship of Roulet (1979) 23 Cal.3d 219, 235 (Roulet), reasoning that “commitment to a mental hospital, despite its civil label, threatens a person‘s liberty and dignity on as massive a scale as that traditionally associated with criminal prosecutions.” (Id. at p. 223; see also Addington v. Texas (1979) 441 U.S. 418, 425.) “At the same time, a civil commitment proceeding is not a criminal proceeding, even though it is often collateral to a criminal trial.” (Blackburn, supra, 61 Cal.4th at p. 1119.) Thus, although some constitutional protections have been extended from the criminal context based on due process concerns, “we have also found various constitutional protections inapplicable.” (Id. at p. 1120.) For example, Conservatorship of Susan T. (1994) 8 Cal.4th 1005, 1015 (Susan T.) held that thе exclusionary rule does not apply in conservatorship proceedings because the purpose of an LPS commitment is treatment, not punishment. For similar reasons, we concluded conservatees have no constitutional right to the appellate review procedures of Anders v. California (1967) 386 U.S. 738 and People v. Wende (1979) 25 Cal.3d 436. (Conservatorship of Ben C. (2007) 40 Cal.4th 529, 538-540, 543 (Ben C.).)

B. No Constitutional Right Against Compelled ‍​​‌​​​‌​​​‌​‌‌‌‌​‌​‌‌‌​​​​‌‌​‌‌‌​​‌‌‌​‌​​​​‌​​‌​‍Testimony in Civil Commitment Proceedings

As a matter of constitutional protection, criminal defendants cannot be compelled to testify against themselves. (U.S. Const., 5th Amend.; Cal. Const., art. I, § 15.)7 Furthermore, witnesses in both criminal and civil proceedings have the right to refuse to answer any question that might tend to incriminate them. (Evid. Code, § 940.)8

The constitutional right against compelled testimony has not been extended to civil commitment proceedings, however. Citing the “predominantly civil character of the proceedings,” this court in Cramer v. Tyars (1979) 23 Cal.3d 131, 137 (Cramer) did not extend the right to individuals who faced confinement under former statutes governing the commitment of developmentally disabled persons. (See Welf. & Inst. Code, former § 6500 et seq.) We declined to analogize the proceedings to criminal prosecutions because the statutory scheme served only the purposes of “custodial care, diagnosis, treatment, and protection,” and the resulting commitment could not be deemed punishment. (Cramer, at p. 137.) We further reasoned that the individual‘s testimony would provide the best evidence of whether commitment was necessary: “Reason and common sense suggest that it is appropriate under such circumstances that a jury be permitted fully to observe the person sought to be committed, and to hear him speak and respond in order that it may make an informed judgment as to the level of his mental and intellectual functioning. The receipt of such evidence may be analogized to the disclosure of physical as opposed to testimonial evidence and may in fact be the most reliable proof and probative indicator of the person‘s present mental condition.” (Id. at p. 139.) Later decisions extended Cramer‘s holding to conservatorship trials (Conservatorship of Baber (1984) 153 Cal.App.3d 542, 550 (Baber)) and LPS proceedings for the confinement of imminently dangerous persons9 (Conservatorship of Bones (1987) 189 Cal.App.3d 1010, 1015-1016).

Further, the constitutional right against compelled testimony does not apply in commitment proceedings that arise in connection with criminal charges. In Allen v. Illinois (1986) 478 U.S. 364, 373-374, the high court held that the federal privilege against self-incrimination did not apply in proceedings under Illinois‘s Sexually Dangerous Persons Act because the commitments were essentially civil in nature. California courts extended Allen‘s holding in the SVP (People v. Leonard (2000) 78 Cal.App.4th 776, 792-793) and OMHD commitment contexts. (People v. Clark (2000) 82 Cal.App.4th 1072, 1081-1082; People v. Merfeld (1997) 57 Cal.App.4th 1440, 1446.) These courts reasoned that the proceedings were designed only to determine the subjects’ status, including the potential for danger and need of mental health treаtment, and that their testimony offered reliable evidence on these issues. (See Clark, at p. 1082; Leonard, at pp. 792-793.)

In recognition of this precedent, appellant does not claim he is entitled to refuse to testify as a matter of constitutional right. (See Hudec, supra, 60 Cal.4th at p. 819McKee, supra, 47 Cal.4th at p. 1207 C. Statutory Right Against Compelled Testimony in Commitment Proceedings Connected to a Criminal Case The statutory right against compelled testimony in an NGI extension proceeding is found in Penal Code section 1026.5, subdivision (b)(7). The history of its enactment is informative. Before 1978, criminal defendants who successfully asserted an insanity defense were most often committed to a state hospital or other facility indefinitely and could be released only if they proved their sanity had been restored. (Pen. Code, former §§ 1026, 1026a; see In re Moye (1978) 22 Cal.3d 457, 461 (Moye).)10 The NGI commitment scheme was substantially altered thereafter in response to a series of decisions from this court. In companion cases dealing with the since-repealed Mentally Disordered Sex Offender (MDSO) law (Welf. & Inst. Code, former § 6300 et seq.), People v. Burnick (1975) 14 Cal.3d 310, 318 held that due process required the offender‘s status to be proven beyond a reasonable doubt. People v. Feagley (1975) 14 Cal.3d 338, 349-352, 375-376 recognized the right to a unanimous jury verdict and disapproved indefinite commitments. In 1977, the Legislature amended the former MDSO statutes to codify these holdings. (See Moye, supra, 22 Cal.3d at p. 464.) The revised statutes provided for renewable annual commitments once the maximum allowable incarceration term had expired. (Welf. & Inst. Code, former §§ 6316.1, 6316.2, subds. (a), (h).) The statutes аlso provided for counsel, discovery, and a jury trial. (Welf. & Inst. Code, former § 6316.2, subds. (d), (e); see Hudec, supra, 60 Cal.4th at p. 821.) One provision gave MDSO‘s the constitutional rights applicable in criminal trials. (Welf. & Inst. Code, former § 6316.2, subd. (e).) The following year, Moye concluded equal protection principles required that initial NGI commitments likewise be limited to the maximum term applicable to the underlying criminal offense. (Moye, at p. 467.) As with the MDSO decisions, the Legislature codified the Moye holding. (See Sen. Com. on Judiciary, Analysis of Sen. Bill No. 1022 (1979-1980 Reg. Sess.) as amended Apr. 30, 1979, p. 2; Hudec, supra, 60 Cal.4th at p. 821.) Penal Code, section 1026.5, enacted in 1979, limits initial NGI commitments to the longest available term of imprisonment for the underlying offense. The commitment may be extended by renewable two-year terms if a “mental disease, defect, or disorder” renders the person a substantial risk of physical harm to others. (Pen. Code, § 1026.5, subd. (b)(1); see id., subd. (b)(8), (10).) Mirroring the former MDSO statutes, Penal Code section 1026.5 provides for counsel, discovery, and jury trial rights. (Id., subd. (b)(3), (4).) Significantly, the statute also declares: “The person shall be entitled to the rights guaranteed under the federal and State Constitutions for criminal proceedings. All proceedings shall be in accordance with applicable constitutional guarantees.” (Id., subd. (b)(7).) In quasi-civil commitment trials, the statute effectively confers many of the rights available by constitutional mandate in criminal proceedings.11 Hudec, supra, 60 Cal.4th 815 considered the scope of this statutory language. Appellant Hudec acknowledged that the trial to extend his NGI commitment was civil in nature, and thus he had no constitutional right to refuse to testify. (Id. at p. 819.) Nevertheless, he argued Penal Code section 1026.5, subdivision (b)(7) granted him this statutory right. (Hudec, at pp. 819-820.) We agreed. (Id. at p. 826.) Although not evеry constitutional right from the criminal context can be sensibly imported into civil proceedings, Hudec concluded no inconsistency or absurdity would result from recognizing a right against compelled testimony in NGI commitment extension trials. (Id. at p. 829People v. Haynie (2004) 116 Cal.App.4th 1224, 1227), NGI commitments could be extended even if the respondent declined to testify. (Hudec, at p. 829.) Hudec acknowledged that recognizing this right would sometimes exclude relevant evidence and that the ability to hear and observe the person‘s testimony can assist the fact finder‘s assessment of mental state. (See id. at pp. 829-830.) However, “[g]ranting that trial accuracy considerations arguably support compelling a committee‘s testimony,” the court concluded, “other considerations,” such as fairness, “militat[ed] against such compulsion.” (Id. at p. 830.)12 After Hudec, a number of Court of Appeal decisions considered whether equal protection required extending the statutory right against compelled testimony to offenders facing postconviction treatment under other commitment schemes. These courts uniformly extended the right in SVP and OMHD contexts. (See People v. Flint (2018) 22 Cal.App.5th 983, 989 (Flint) [SVP]; People v. Alsafar (2017) 8 Cal.App.5th 880, 882-883 [OMHD]; People v. Field (2016) 1 Cal.App.5th 174, 193-194 [SVP]; People v. Dunley (2016) 247 Cal.App.4th 1438, 1450 [OMHD]; People v. Landau (2016) 246 Cal.App.4th 850, 865 [SVP]; People v. Curlee (2015) 237 Cal.App.4th 709, 720 (Curlee) [SVP].) While recognizing differences between the statutory schemes, these courts concluded the differences were not dispositive. (See, e.g., Dunley, at pp. 1449-1450.) Individuals in all three groups had committed criminal acts; all had been diagnosed with mental health disorders that made them potеntially dangerous to others; and all were subject to commitment in a state facility for involuntary treatment. (See Curlee, at p. 720.) Further, the purpose of commitment in all three statutory schemes was the same: “To protect the public from those who have committed criminal acts and have mental disorders and to provide mental health treatment for the disorders. (See Pen. Code, § 1026.5, subd. (b); McKee[], supra, 47 Cal.4th at pp. 1203, 1207; Moye, supra, 22 Cal.3d at p. 466.)” (Curlee, at p. 720.) D. Extending the Statutory Right Against Compelled Testimony to LPS Commitment Proceedings The LPS Act does not include a statutory right against compelled testimony, nor does it contain the broad mention of rights set out in Penal Code section 1026.5, subdivision (b)(7). Nevertheless, appellant argues equal protection demands that the same right to refuse testimony applies. “Because of the fundamental interests at stake, equal protection principles are often invoked in civil commitment cases to ensure that the statutory scheme applicable to a particular class of persons has not treated them unfairly in comparison with other groups with similar characteristics.” (Barrett, supra, 54 Cal.4th at p. 1107.) An equal protection analysis has two steps. “‘The first prerequisite . . . is a showing that the state has adopted a classification that affects two or more similarly situated groups in an unequal manner.’ [Citations.] This initial inquiry is not whether persons are similarly situated for all purposes, but ‘whether they are similarly situated for purposes of the law challenged.‘” (McKee, supra, 47 Cal.4th at p. 1202, some italics added.) If the groups are similarly situated, the next question is whether the disparate treatment can be justified by a constitutionally sufficient state interest. (See id. at pp. 1207-1209; Moye, supra, 22 Cal.3d at pp. 465-466.) 1. The Similarly Situated Prong Three lower court decisions have addressed whether traditional LPS conservatees are similarly situated with individuals facing an extended NGI commitment. Bryan S., supra, 42 Cal.App.5th at pages 196-197 concluded they are not, because a conservatorship may be imposed without any connection to a crime or any showing of danger to others, and conservatees may be placed in nonhospital settings.13 The Court of Appeal decisions here, Conservatorship of E.B. (2020) 45 Cal.App.5th 986 (E.B.), and in Conservatorship of J.Y. (2020) 49 Cal.App.5th 220 disagreed with Bryan S. They concluded traditional LPS conservatees are similarly situated with those facing an NGI commitment extension because both are subject to involuntary confinement that could be extended indefinitely, and both are committed for the dual purposes of mental health treatment and public protection. (See J.Y., at pp. 229–231; E.B., at pp. 993–994.) We agree with these latter cases that the groups are similarly situated for purposes of the right not to give compelled testimony.14 An equal protection analysis typically focuses on the practical consequences of a challenged law to the groups in question. In McKee, for example, we conсluded SVP‘s and OMHD‘S were similarly situated with regard to certain procedural rights because, despite their differences in other respects, both had “the same interest at stake — the loss of liberty through involuntary civil commitment.” (McKee, supra, 47 Cal.4th at p. 1204.) Here, too, the most striking and decisive similarity between the groups is the potential loss of liberty both face in the proceedings at issue. Like NGI‘s, LPS conservatees are subject to physical confinement and the loss of many personal rights. (See Ben C., supra, 40 Cal.4th at p. 540; Roulet, supra, 23 Cal.3d at p. 223.) Although traditional conservatees are entitled to be placed in the least restrictive suitable setting (Welf. & Inst. Code, § 5358, subds. (a), (c)), the LPS statutes authorize confinement in a residential facility or hospital when appropriate (see Welf. & Inst. Code, § 5358, subd. (a)(2)). Here, the Public Guardian‘s petition for conservatorship requested authority to seek this most restrictive placement for appellant. As noted, institutional placements for LPS conservatees ‍​​‌​​​‌​​​‌​‌‌‌‌​‌​‌‌‌​​​​‌‌​‌‌‌​​‌‌‌​‌​​​​‌​​‌​‍are fairly common; so much so that in July 2020 the state auditor criticized the long wait times LPS conservatees had to endure before state hospital admission. (State Auditor‘s Report, supra, at pp. 22–26.) Although LPS conservatees occupied around 11 percent of state hospital beds in 2019, the auditor reported that 200 more were waiting for admission and, as a result, receiving lower levels of care than they needed. (Id. at p. 25.) The Public Guardian concedes that LPS conservatees are frequently confined in locked facilities but argues the prevalence of such commitments is “not surprising” given that conservatorships are only ordered for individuals who are unable to care for themselves. The parties do not dispute that there may be good reasons for such confinements, or that they may be necessary to provide the care and treatment a conservatee requires. Both traditional LPS conservatorships and those relating to criminal proceedings share the goal of treatment, not punishment. Nonetheless, it cannot be denied that “civil commitment for any purpose constitutes a significant deprivation of liberty . . . .” (Addington v. Texas, 441 U.S. at p. 425; see Blackburn, supra, 61 Cal.4th at p. 1119.) “In addition to physical restraint, ‘[t]he gravely disabled person for whom a conservatorship has been established faces the loss of many other liberties . . . .‘” (Ben C., supra, 40 Cal.4th at p. 540.) Apart from their possible confinement, conservatees may lose the rights to drive, vote, enter contracts, and make decisions about their treatment. (See Welf. & Inst. Code, § 5357.) In light of the potential for such a significant loss of liberty, conservatorship cases are governed by many of the same procеdural protections that apply in criminal trials. (See Welf. & Inst. Code, § 5350, subd. (d)(1); Ben C., at p. 541; but see Ben C., at p. 538 [recognizing “that the analogy between criminal proceedings and proceedings under the LPS Act is imperfect at best” and that “not all of the safeguards required in the former are appropriate to the latter“]; Susan T., supra, 8 Cal.4th at p. 1015 [holding the exclusionary rule does not apply in conservatorship proceedings].) Moreover, a year-long conservatorship may be extended through the filing of successive petitions. (Welf. & Inst. Code, § 5361.) As a result, the LPS statutes can “assure in many cases an unbroken and indefinite period of state-sanctioned confinement.” (Roulet, supra, 23 Cal.3d at p. 224.) In San Francisco, for example, almost 38 percent of LPS conservatorships, excluding Murphy conservatorships, had been extended for 10 years or more as of December 2018. (San Francisco Analyst‘s Report, supra, at p. A-9.) An additional 23 percent had been extended from five to 10 years. (Ibid.) Thus, in practice, traditional LPS conservatorships can impose substantially the same restraint on liberty as involuntary commitments connected to criminal proceedings. To be sure, traditional LPS conservatees differ in certain respects from civilly committed NGI‘s. The latter are adjudged to have committed a criminal actus reus but are found nоt guilty because their insanity negates the required mens rea. (See Moye, supra, 22 Cal.3d at p. 466.) While those confined as an SVP or OMHD have been convicted of crimes, most conservatorships are not based on criminal allegations. LPS conservatees are ordinarily imposed solely because a mental illness prevents the conservatee from providing for basic survival needs. (See Welf. & Inst. Code, §§ 5008, subd. (h)(1)(A), 5350.) For these individuals, “[t]he commitment is not initiated in response, or necessarily related, to any criminal acts ....” (Susan T., supra, 8 Cal.4th at p. 1015.) Murphy conservatorships bear a much closer resemblance to NGI commitments in this regard. Murphy conservatees have been charged with serious felonies involving actual or threatened physical harm (Welf. & Inst. Code, § 5008, subd. (h)(1)(B)(i)–(ii)), and, unlike the traditional LPS conservatees at issue in this case, their dangerousness to others is assessed in determining whether a conservatorship is necessary (see id., subd. (h)(1)(B)(iv)). Murphy conservatorships are comparatively rare, however, accounting for only around 2 percent of all LPS conservatorships in San Francisco, for example. (See San Francisco Analyst‘s Report, supra, at p. A-11.) It is “incontrovertible” that conservatees “do not share identical characteristics” with civilly committed NGI‘s. (McKee, supra, 47 Cal.4th at p. 1203.) But these differences are not dispositive of whеther the groups are similarly situated with respect to the testimonial privilege. (See ibid.) In this part of an equal protection analysis, the question “‘is not whether persons are similarly situated for all purposes, but “whether they are similarly situated for purposes of the law challenged.“‘” (People v. Valencia (2017) 3 Cal.5th 347, 376, italics added.) “In other words, we ask at the threshold whether two classes that are different in some respects are sufficiently similar with respect to the laws in question to require the government to justify its differential treatment of these classes under those laws.” (McKee, at p. 1202.)15 In some cases, we have concluded traditional LPS conservatees were not sufficiently similar to other groups in regard to a challenged law. For example, in Cooley v. Superior Court (2002) 29 Cal.4th 228, 253–254, we concluded individuals facing an SVP probable cause hearing were not similarly situated with those seeking habeas review of a short-term detention under the LPS Act because the purposes served by the standard of proof at the LPS hearing had no rational application in the SVP context. Here, however, we reach a different conclusion. In rejecting the same equal protection challenge raised here, the Bryan S. court considered the purpose served by the testimonial privilege. It reached back to Cramer, supra, 23 Cal.3d 131, where we held the constitutional privilege does not apply in civil commitment proceedings. Cramer explained that “the historic purpose of the privilege against being called as a witness has been to assure that the criminal justice system remains accusatorial, not inquisitorial. [Citations.] The extension of the privilege to an area outside the criminal justice system . . . would contravene both the language and purpose of the privilege.” (Id. at pp. 137–138; see Bryan S., supra, 42 Cal.App.5th at p. 197.) After Cramer was decided, however, the Legislature chose to extend the privilege beyond the criminal justice system by enacting Penal Code section 1026.5, subdivision (b)(7). We observed in Hudec that ”Cramer‘s constitutional reasoning ha[d] no bearing on the interpretation of” Penal Code section 1026.5, subdivision (b)(7). (Hudec, supra, 60 Cal.4th at p. 830.) It is likewise inapt to the equal protection challenge here. The issue is not whether traditional LPS conservatees are similar to criminal defendants, but whether they are similar to NGI‘s. Like these conservatees, NGI‘s no longer stand accused of crimes. And, like conservatorships, NGI extension proceedings are civil in nature and examine only whether the statutory grounds for commitment have been met. (See Hudec, at p. 819.) The more precise similarity question, then, is what purpose does the testimonial privilege serve in civil commitment proceedings? Hudec offers one answer. Hudec acknowledged that testimony from those facing commitment may be particularly helpful in determining their mental condition but noted that “other considerations” might weigh against compelling their testimony, “notably ‘our sense of fair play which dictates “a fair state-individual balance by requiring the government . . . in its contest with the individual to shoulder the entire load.“’ (Murphy v. Waterfront Comm‘n. (1964) 378 U.S. 52, 55.)” (Hudec, supra, 60 Cal.4th at p. 830.) “The right to not be compelled to testify against oneself is clearly and relevantly implicated when a person is called by the state to testify in a proceeding to [commit or] recommit him or her even if what is said on the witness stand is not per se incriminating.” (People v. Haynie, supra, 116 Cal.App.4th at p. 1230.) The privilege‘s role in enforcing fair play, and ensuring the government meets its burden, is not unique to the criminal context. Like NGI‘s, traditional LPS conservatees also face the prospect of extended involuntary confinement and the loss of other liberties. In reaching a different conclusion, the trial court here cited the importance of allowing the trier of fact to observe the “physical and mental characteristics” of the proposed conservatee. Compelled testimony from the conservatee may well assist the fact finder and contribute to more accurate verdicts in conservatorship trials. (See Cramer, supra, 23 Cal.3d at p. 139; Baber, supra, 153 Cal.App.3d at p. 550.)16 It might also be argued that the predicates for traditional LPS and NGI commitments are significantly different. Most of those for whom an LPS conservatorship is sought will not have been subject to a criminal adjudication or any showing that they pose a danger to others. As a result, they will not have undergone the kinds of extended restraints on liberty and resultant therapeutic and rehabilitative efforts extended to NGI, SVP, and OMHD individuals. While we acknowledge these differences and note that they may bear on whether the disparate treatment of traditional LPS conservatees and NGI‘s is constitutionally justified, they are not sufficient to undermine the two groups’ similarity for purposes of the testimonial privilege. Accordingly, despite their differences, we conclude NGI‘s and traditional LPS conservatees “are sufficiently similar to bring into play equal protection principles that require a court to determine ‘“whether distinctions between the two groups justify the unequal treatment.“’ (People v. Hofsheier (2006) 37 Cal.4th 1185, 1200.)” (In re Marriage Cases (2008) 43 Cal.4th 757, 832, fn. 54.) Conservatorship of Bryan S., supra, 42 Cal.App.5th 190 is disapproved to the extent it conflicts with the views expressed herein. 2. Justification for Disparate Treatment The next step of an equal protection analysis asks whether the disparate treatment of two similarly situated groups is justified by a constitutionally sufficient state interest. (See McKee, supra, 47 Cal.4th at pp. 1207–1208.) Varying levels of judicial scrutiny apply depending on the type of claim. “[M]ost legislation is tested only to determine if the challenged classification bears a rational relationship to a legitimate state purpose.” (People v. Hofsheier, supra, 37 Cal.4th at p. 1200.) However, differences “in statutes that involve suspect classifications or touch upon fundamental interests are subject to strict scrutiny, and can be sustained only if they are necessary to achieve a compelling state interest.” (Ibid.) Decisions from the Courts of Appeal have reached differing conclusions about the level of scrutiny appropriate for assessing claims of disparate treatment in civil commitments. (Compare Flint, supra, 22 Cal.App.5th at pp. 992–993 [strict scrutiny] with People v. Nolasco (2021) 67 Cal.App.5th 209, 225 [rational basis].) Because the courts below did not reach this prong of the equal protection analysis, arguments have not been well developed here concerning the proper degree of scrutiny or whether the government can demonstrate a sufficient justification for granting the testimonial privilege to NGI‘s but not traditional LPS conservatees. Ordinarily, we would remand to the trial court for a hearing at which the Public Guardian would have an opportunity to show why the differential treatment is constitutionally justified. (See McKee, supra, 47 Cal.4th at pp. 1207–1209; see also Curlee, supra, 237 Cal.App.4th at p. 722.) However, the Court of Appeal determined the error in this case was harmless under either the state (People v. Watson, supra, 46 Cal.2d at p. 836) or federal (Chapman v. California, 386 U.S. at p. 24) standard for harmless errоr. The court observed that, apart from appellant‘s testimony, “two other witnesses who were familiar with appellant . . . painted a vivid picture of someone who was unable to care for himself left to his own devices due to his mental illness.” (E.B., supra, 45 Cal.App.5th at p. 999.) Appellant does not challenge that conclusion. Accordingly, although we have concluded traditional LPS conservatees are similarly situated with NGI‘s for purposes of the right against compelled testimony, a remand is not appropriate here. Whether the government can justify its differential treatment of traditional conservatees with regard to this right must await decision in another case. III. DISPOSITION We affirm the judgment of the Court of Appeal. CORRIGAN, J. We Concur: CANTIL-SAKAUYE, C. J. LIU, J. KRUGER, J. GROBAN, J. JENKINS, J. MOORE, J.* * Associate Justice of the Court of Appeal, Fourth Appellate District, Division Three, assigned by the Chief Justice pursuant to article VI, section 6 of the California Constitution. Concurring Opinion by Justice Kruger This case involves a federal equal protection challenge to the statutory procedures for establishing conservatorships for persons with grave disabilities. Eric B., a potential conservatee, argues the statute is unconstitutional ‍​​‌​​​‌​​​‌​‌‌‌‌​‌​‌‌‌​​​​‌‌​‌‌‌​​‌‌‌​‌​​​​‌​​‌​‍because it contains no right to refuse to testify akin to the statutory right enjoyed by NGI‘s (that is, persons found not guilty of a crime by reason of insanity) in commitment extension proceedings. (Compare Welf. & Inst. Code, § 5350, subd. (d)(1), (2) with Pen. Code, § 1026.5, subd. (b)(7).) But the question now before this cоurt is not the ultimate question whether this difference in treatment is constitutional. Rather, the sole question before us concerns a threshold inquiry: Whether potential conservatees are sufficiently similarly situated to NGI‘s, for purposes of the challenged law, to warrant further inquiry into whether the differential treatment violates equal protection. The court answers yes. (Maj. opn., ante, at pp. 1, 25.) I agree with this limited holding and have signed the court‘s opinion. I write separately, however, to suggest that this threshold inquiry doesn‘t serve much purpose. Worse, it risks harm. The simple fact that a law differently benefits or burdens two identifiable groups is — or at least ought to be — sufficient reason for us to examine whether the difference in treatment is consistent with equal protection. To the extent our cases have taken a different approach, it is probably time to reevaluate. I. In answering the question before us, the court‘s opinion describes a two-step approach for analyzing equal protection challenges. “‘“‘The first prerequisite . . . is a showing that the state has adopted a classification that affects two or more similarly situated groups in an unequal manner.’ [Citations.] This initial inquiry is not whether persons are similarly situated for all purposes, but ‘whether they are similarly situated for purposes of the law challenged.‘“‘” (People v. McKee (2010) 47 Cal.4th 1172, 1202, some italics added.) If the groups are similarly situated, the next question is whether the disparate treatment can be justified by a constitutionally sufficient state interest. (See id. at pp. 1207–1209; In re Moye (1978) 22 Cal.3d 457, 465–466.)” (Maj. opn., ante, at p. 18.) In other words: (1) Are the parties sufficiently similarly situated to call for further inquiry? If no, the analysis is done. But (2) if yes, can the challenged disparity be justified? At the second step, we employ the familiar tiered system of scrutiny to determine the amount of justification required. We apply the most lenient standard — so-called rational basis review — to most forms of differential treatment; we apply more searching scrutiny to, and thus require greater justification for, differential treatment that either infringes on a fundamental right or is based on a suspect or quasi-suspect classification, such as race or sex. (People v. Chatman (2018) 4 Cal.5th 277, 288–289.) This is the approach set out in many — though not all — of our recent equal protection cases. Both parties assume it applies here, as did the Court of Appeal in this case, and as have many other California courts addressing similar questions. Whether the approach makes sense is a different matter. A. This two-step approach is not how equal protection analysis was always done in California. This court did often observe that equal рrotection requires like treatment for those “similarly situated with respect to the legitimate purpose of the law.” (Purdy & Fitzpatrick v. State of California (1969) 71 Cal.2d 566, 578.) But we did not initially use this general observation about the concept of equal protection as a springboard for engaging in a threshold inquiry into whether two groups are similarly situated. We instead described the relevant constitutional inquiry solely in terms of whether the challenged difference in treatment was justified under the applicable standard of scrutiny. (Id. at pp. 578–579; see, e.g., In re Antazo (1970) 3 Cal.3d 100, 110–111.) The two-step approach appears to have emerged from two cases decided in the late 1970‘s, both concerning challenges to statutes governing the treatment of juveniles. In the first case, In re Roger S. (1977) 19 Cal.3d 921 (Roger S.), a minor objected to involuntary admission to a state mental hospital on the application of a parent. He argued that he was denied equal protection because his admission was not conditioned on a finding that he was gravely disabled or a danger to himself or others, as it would have been for an adult or a minor ward of the court. This court rejected the argument. “[T]he Constitution,” we observed, “does not require things which are different in fact or opinion to be treated in law as though they were the same.” (Id. at p. 934, quoting Tigner v. Texas (1940) 310 U.S. 141, 147.) Given the differences between thе liberty interests of children and adults, we concluded that minors “are not ‘similarly situated’ with adults for purposes of equal protection analysis.” (Roger S., at p. 934.) We also found minors like Roger S. dissimilar from court wards, explaining that courts have options for the psychiatric treatment of nondangerous minors that parents may not. The difference in the standards for the involuntary confinement of the two groups, we held, “does not in our view deny equal protection to either class.” (Id. at p. 935, citing, inter alia, Reed v. Reed (1971) 404 U.S. 71, 75–76.) In the second case, In re Eric J. (1979) 25 Cal.3d 522 (Eric J.), this court considered a juvenile‘s equal protection challenge to laws extending more favorable sentencing treatment to an adult convicted of a crime warranting imprisonment than to juveniles subject to confinement for committing the same crime. Rejecting the claim, the court cited Roger S., supra, 19 Cal.3d at page 934 for the proposition that the “first prerequisite to a meritorious claim under the equal protection clause is a showing that the state has adopted a classification that affects two or more similarly situated groups in an unequal manner.” (Eric J., at p. 530; see also id. at p. 530, fn. 1 [quoting, as Roger S. had, Tigner v. Texas, supra, 310 U.S. at p. 147 for the proposition that “‘[t]he Constitution does not require things which are different in fact or opinion to be treated in law as though they were the same‘“].) We went on to conclude that “because minors and adults are not ‘similarly situated’ with respect to their interest in liberty,” аnd because the two groups “are not confined for the same purposes,” the difference in treatment did not violate equal protection. (Eric J., at p. 533.) The two-step framework the court applies today traces back to this particular gloss on the United States Supreme Court‘s admonition that equal protection “does not require things which are different in fact or opinion to be treated in law as though they were the same.” (Tigner v. Texas, supra, 310 U.S. at p. 147.) Of course, it is not clear that either Roger S. or Eric J. in fact applied anything like the two-step framework; both cases undertook what was essentially a one-step, holistic inquiry into whether the challenged differential treatment violated equal protection. Roger S. looked for support to Reed v. Reed, supra, 404 U.S. 71, a high court decision that had evaluated an equal protection challenge to a sex-based classification by asking whether the classification was justified in view of the state‘s interests (Roger S., supra, 19 Cal.3d at p. 935); Eric J., in turn, looked to Roger S. And notwithstanding the language in Eric J. suggesting the existence of a preliminary “similarly situated” step as a “first prerequisite” to further inquiry (Eric J., supra, 25 Cal.3d at p. 530, italics omitted), the cases were not initially understood as establishing a two-step framework. In a case decided not long after Eric J., this court considered an equal protection challenge to a decision limiting a school district election to a certain group of district residents, while excluding a second group. “The first step in evaluating this contention,” we explained, “is to determine the applicable level of judicial review,” rational basis or heightened scrutiny. (Fullerton Joint Union High School Dist. v. State Bd. of Education (1982) 32 Cal.3d 779, 798 (Fullerton).) We dismissed the notion that Eric J. required a different order of operations: “Some decisions speak of an initial constitutional inquiry to determine whether the groups affected are similarly situated with respect to the purpose of the legislation or other state action. (See, e.g., In re Eric J.[, supra,] 25 Cal.3d [at p.] 531.) To ask whether two groups are similarly situated in this context, however, is the same as asking whether the distinction between them can be justified under the appropriate test of equal protection. Obvious dissimilarities between groups will not justify a classification which fails strict scrutiny (if that test is applicable) or lacks a rational relationship to the legislative purpose.” (Fullerton, at p. 798, fn. 19; accord, People v. Allen (1986) 42 Cal.3d 1222, 1295 (lead opn.).) As time went on, however, the language of Eric J. took precedence over its limiting treatment in Fullerton. Courts repeatedly invoked Eric J.‘s “first prerequisite” language and rejected equal protection claims on the basis that the two groups treated differently were insufficiently similar to one another. (See People v. Williams (1988) 45 Cal.3d 1268, 1330 [“persons convicted under the death penalty law are manifestly not similarly situated to persons convicted under the Determinate Sentencing Act and accordingly cannot assert a meritorious claim to the ‘benefits’ of the act under the equal protection clause“], citing Eric J., supra, 25 Cal.3d at p. 530; People v. Andrews (1989) 49 Cal.3d 200, 223 [citing Eric J. for the proposition that “the first prerequisite to [an equal protection] claim is a showing that ‘the state has adopted a classification that affects two or more similarly situated groups in an unequal manner‘” and rejecting equal protection claim]; Coleman v. Department of Personnel Administration (1991) 52 Cal.3d 1102, 1125 [same]; People v. Massie (1998) 19 Cal.4th 550, 571 [same]; Manduley v. Superior Court (2002) 27 Cal.4th 537, 568–571 [citing Eric J. and rejecting claim on ground the defendant had not shown unequal treatment of similarly situated groups]; People v. Wutzke (2002) 28 Cal.4th 923, 943–944 [same]; Cooley v. Superior Court (2002) 29 Cal.4th 228, 253–254 [same].) The language of Eric J. was repeated from case to case. Eventually, shorn of context, the language morphed and hardened to become the first step of the formal two-step inquiry the court‘s opinion recites today. (See, e.g., People v. Hofsheier (2006) 37 Cal.4th 1185, 1199–1200 [detailed analysis of the similarly situated requirement as a threshold matter independent of subsequent inquiry into justification]; People v. McKee, supra, 47 Cal.4th at p. 1202 [treating the similarly situated inquiry as a necessary “threshold” question]; id. at pp. 1202–1209 [deciding only that question and remanding for further proceedings on the separate question of justification].) Indeed, the court stopped citing Eric J. itself, simply asserting as a settled matter that the “‘initial inquiry in any equal protection analysis is whether persons are “similarly situated for purposes of the law challenged.“‘” (In re Lemanuel C. (2007) 41 Cal.4th 33, 47.) And in some cases, the court has concluded they are not — a conclusion that has simply еnded the equal protection analysis, without review of the challenged governmental action under any level of scrutiny. (See, e.g., People v. Lewis (2004) 33 Cal.4th 214, 231; Conservatorship of Ben C. (2007) 40 Cal.4th 529, 543; People v. Salazar (2016) 63 Cal.4th 214, 227; People v. Valencia (2017) 3 Cal.5th 347, 376.) B. Although the threshold similarly situated test nominally has its roots in United States Supreme Court case law, the high court itself has neither required nor applied any similar gatekeeping test. Rather, in cases involving challenges to discrimination between identifiable groups, the court proceeds directly to the justification step: It identifies the appropriate level of scrutiny for a particular challenged distinction and then examines whether the actual or potential justification for that differentiation is sufficient, without separately analyzing whether the groups receiving differential treatment are otherwise similarly situated. (See, e.g., Grutter v. Bollinger (2003) 539 U.S. 306, 326–343 [determining appropriate level of scrutiny (strict) and moving directly to a consideration of the adequacy of the proffered justification]; United States v. Virginia (1996) 518 U.S. 515, 531–534 [same, applying intermediate scrutiny]; Cleburne v. Cleburne Living Center, Inc. (1985) 473 U.S. 432, 439–450 [same, applying rational basis scrutiny].) The high court‘s cases do make clear that a similarly situated inquiry has a useful role to play in other kinds of cases — particularly cases involving so-called “‘class of one‘” equal protection claims, “where the plaintiff alleges that she has been intentionally treated differently from others similarly situated and that therе is no rational basis for the difference in treatment.” (Village of Willowbrook v. Olech (2000) 528 U.S. 562, 564.) In such cases, where a plaintiff does not allege that she has been treated differently because of “membership in a class or group” (ibid.), a similarly situated inquiry helps identify whether the plaintiff has suffered differential treatment that warrants scrutiny under the equal protection clause. (See also Engquist v. Oregon Dept. of Agriculture (2008) 553 U.S. 591, 601–602 [discussing “class-of-one” claims under Olech].) But in a case like the one before us, as in many others, the law clearly treats Eric B. differently from others because of the group — that is, potential conservatees — to which he belongs. The critical question is whether that group-based difference in treatment comports with equal protection principles. In comparable cases, the high court has proceeded directly to this critical question, without first attempting to gauge the degree of similarity between the groups, as California courts have done. We are, of course, not bound to follow where the United States Supreme Court leads in matters of state constitutional law. So if the two-step framework articulated in our cases had developed as an explication of unique state constitutional principles, there would be no need to concern ourselves with whether it comports with United Stаtes Supreme Court guidance. But in elaborating a two-step approach, we‘ve never invoked any special features of the state Constitution‘s equal protection provision. (Cal. Const., art. I, § 7, subd. (a).) To the contrary, when urged to use that provision to articulate a unique set of state law specific principles, we‘ve declined. (Manduley v. Superior Court, supra, 27 Cal.4th at p. 572 [rejecting petitioners’ invitation to rely on state constitutional principles and “deem[ing]” the “analysis of petitioners’ equal protection claim under the Fourteenth Amendment to the United States Constitution also applicable to their equal protection claim made pursuant to provisions in the California Constitution“]; see, e.g., Johnson v. Department of Justice (2015) 60 Cal.4th 871, 881 [accepting “the high court‘s analysis of federal . . . equal protection principles [as] persuasive for purposes of the state Constitution“].) It is true that while the United States Supreme Court has not used the same two-step approach to analyze federal equal protection issues, it also has never formally repudiated any such approach.1 But if we choose to chart a different path, we at least ought to be clear that that‘s what we‘re doing. Instead, our cases appear to assume the United States Supreme Court has pointed us in the direction of the two-step framework. It has not. C. Even in this court, this two-step approach is not always how the equal protection analysis is done — which is to say, we are not always rigid or consistent in our application of the two-step framework. In a number of cases, we have analyzed equal protection questions much as Fullerton had once instructed and as the United States Supreme Court does regularly: We have begun by asking not whether two groups are similarly situated but what level of scrutiny should apply. (See, e.g., Hernandez v. City of Hanford (2007) 41 Cal.4th 279, 298 [“we begin with the question of the appropriate equal protection standard applicable in this case“]; Kasler v. Lockyer (2000) 23 Cal.4th 472, 480 [“we must address plaintiffs’ equal protection challenge on the merits, and the threshold question we confront is which standard of review applies“].) This line of ‍​​‌​​​‌​​​‌​‌‌‌‌​‌​‌‌‌​​​​‌‌​‌‌‌​​‌‌‌​‌​​​​‌​​‌​‍cases has tackled equal protection questions without requiring the plaintiff to show, at the first step, that other groups are similarly situated. (See, e.g., People v. Turnage (2012) 55 Cal.4th 62, 74–75; California Grocers Assn. v. City of Los Angeles (2011) 52 Cal.4th 177, 208–211; Warden v. State Bar (1999) 21 Cal.4th 628, 640–651.) If we have sometimes done without the two-step approach, the question arises whether we might always do without, or whether instead the approach offers some useful assistance to courts evaluating equal protection challenges like this one. But on a brief review of the cases decided under this apрroach, its utility seems doubtful. The basic reason is the one Fullerton identified decades ago: At least as our cases have described the approach, it is not clear how the threshold similarly situated inquiry differs in any material way from the ultimate question in a group-based discrimination case, except that it offers substantially less guidance about how to answer. That two groups are similarly situated, or are not similarly situated, with respect to the purposes of a law is a conclusion one can only reach after considering the law‘s aims and how the differential treatment relates to those aims. Even then, the issue remains: How similarly situated, precisely, relative to which aims? These are questions courts already explore at the justification step, using the tiers of scrutiny to guide their answers. It is unclear what purpose is served by asking the same questions, in a substantially more general way, as part of a separate threshold step of the analysis. Our cases have not, of course, treated the two prongs of the analysis as merely duplicative or interchangeable. But we have also failed to explain in any meaningful way how the two prongs should differ from one another. This has led to some oddities. Take Johnson v. Department of Justice, supra, 60 Cal.4th 871, which overruled an earlier decision finding an equal protection violation in the statutory requirement that those convicted of oral copulation with a minor, but not those convicted of intercourse with the same, register as sex offenders. (See People v. Hofsheier, supra, 37 Cal.4th 1185.) Hofsheier found the groups similarly situated and then concluded no rational basis existed for treating them unequally. Johnson purported to accept the similarly situated half of Hofsheier‘s analysis, but then concluded that a rational basis existed for differential treatment because of relevant differences between the groups. (Johnson, at pp. 882, 884–887.) In other words, the groups were not similarly situated with respect to the purposes of the law after all. A reader might be forgiven for experiencing a sense of whiplash. (See also, e.g., In re C.B. (2018) 6 Cal.5th 118, 134 [in the span of a few paragraphs, assuming that two groups were similarly situated with respect to the purposes of a voter initiative and then explaining how “voters rationally could differentiate” between them because of an interest in cost savings].) Employing a framework that contains a potentially duplicative step carries more risks than just the possibility of wasted effort or seeming inconsistencies in the analysis. By adding a step not directly focused on the ultimate question of justification, we run the risk of mistakenly cutting off potentially meritorious equal protection claims. Interposing an unnecessary gatekeeping inquiry always raises the possibility that the gate will sometimes slam shut, when the gate shouldn‘t have been there in the first place. At the very least, the two-step framework creates unnecessary confusion. Because it is a requirement of our own creation, the threshold similarly situated inquiry comes with no clear high court guidelines as to its proper application. Nor have we offered much guidance ourselves. This case illustrates the kinds of unresolved questions that courts still confront, decades after the inquiry first emerged in the case law. To decide whether two groups are similarly situated with respect to the purpose of a given law, one must define what that purpose is. But how does one do so when the law‘s purpose involves a balance of considerations (as laws generally do)? Here, the court‘s opinion says one possible purpose for conferring a privilege against testifying on NGI committees is a sense of fair play that outweighs the interest in accurate determinations. (Maj. opn., ante, at p. 24, citing Hudec v. Superior Court (2015) 60 Cal.4th 815, 830.) The court then assesses whether Lanterman-Petris-Short committees are similarly situated for purposes of the fair-play interest (maj. opn., ante, at p. 24), without considering whether they are also similarly situated with respect to the countervailing interest in accurate determinations. Should the inquiry consider one, or the other, or both? It seems impossible to say without knowing what the similarly situated test is meant to achieve. The case law yields no clear answers. The way the court‘s opinion tackles the inquiry is by no means wrong; the point is only that the inquiry itself injects unnecessary uncertainty into the law. That uncertainty might be worth clearing up if the similarly situated test added sufficient value. I doubt that it does. II. All that said, this is not the case in which to reexamine our equal protection framework. The parties have not raised any question about that framework here; instead, in reliance on our current case law, they have focused entirely on the proper application of the similarly situated step some cases have told them is necessary. The Court of Appeal decision likewise focused only on that step, and then, without resolving whether any different treatment would have been justified, found any potential constitutional error harmless under the circumstances of the case. And — as we already knew when we granted review — this case is moot, so it does not make sense to press the issue further. Finally, I agree that the choice of framework would not be outcome-determinative in any event: Given our conclusion that potential conservatees and NGI‘s are sufficiently similarly situated to warrant further scrutiny, if this case were to proceed, the government would be required to come forward with a sufficient justification, just as it would if we were to proceed directly to the justification inquiry. For all these reasons, in today‘s case it makes little difference that we have occupied ourselves with a threshold inquiry into whether two groups are similarly situated. So long as we continue to employ this framework, that is presumably how it should be; the threshold similarly situated test should not cut off inquiry into the core question, whether an admitted difference in treatment of two groups is justified under the law. But going forward, it is unclear why we should hold on to a legal test that serves so little purpose. In an appropriate future case, we ought to consider whether it is time to let the similarly situated test go. KRUGER, J. We Concur: LIU, J. GROBAN, J. See next page for addresses and telephone numbers for counsel who argued in Supreme Court. Name of Opinion Conservatorship of Eric B. Procedural Posture (see XX below) Original Appeal Original Proceeding Review Granted (published) XX 45 Cal.App.5th 986 Review Granted (unpublished) Rehearing Granted Opinion No. S261812 Date Filed: April 28, 2022 Court: Superior County: Contra Costa Judge: Susanne M. Fenstermacher Counsel: Jeremy T. Price, under appointment by the Supreme Court, for Objector and Appellant. Kim Pederson and Anne Hadreas for Disability Rights California, California Association of Mental Health Patients’ Rights Advocates, California Public Defenders Association, American Civil Liberties Union, American Civil Liberties Union of Northern California, Disability Rights Education аnd Defense Fund, Law Foundation of Silicon Valley and Mental Health Advocacy Services as Amici Curiae on behalf of Objector and Appellant. Sharon L. Anderson and Mary Ann McNett Mason, County Counsel, Steven Rettig, Assistant County Counsel, and Patrick L. Hurley, Deputy County Counsel, for Petitioner and Respondent. Jennifer B. Henning for the California State Association of Counties and California State Association of Public Administrators, Public Guardians, and Public Conservators as Amici Curiae on behalf of Petitioner and Respondent. Counsel who argued in Supreme Court (not intended for publication with opinion): Jeremy T. Price First District Appellate Project 475 14th Street #650 Oakland, CA 94612 (415) 495-3119 Patrick L. Hurley Deputy County Counsel 1025 Escobar Street, 3rd Floor Martinez, CA 94553 (925) 655-2251

Notes

1
He stated: “Oh, I even kind of have really spoken not too clearly about this. But I‘m more towards the neutrality and leaving enough area of a cushion that I could have — so I could leave the temporary conservatorship because maybe it‘s that I don‘t need it. And I know I have a mental health — mental health. [¶] . . . [¶] I know what it is. I live with it. I take medications for it. When I know I don‘t need medications, I don‘t need medications. [¶] But if you will there‘s always a little strike pad here that we can always roughly just braze and find out my history find out my — and my future means too. I‘m trying to save this for myself.” A handful of other jurisdictions have also sometimes applied some version of a threshold similarly situated inquiry. (See, e.g., Morrison v. Garraghty (4th Cir. 2001) 239 F.3d 648, 654; Rodriguez v. Lamer (11th Cir. 1995) 60 F.3d 745, 749; T.M. v. State (Fla.Ct.App. 1997) 689 So.2d 443, 444–445; Miami County Bd. v. Kanza Rail-Trails (2011) 292 Kan. 285, 315–316 [255 P.3d 1186, 1207]; DuPont v. Commissioner of Correction (2007) 448 Mass. 389, 399–400, 403, fn. 24 [861 N.E.2d 744, 752–753, 754–755, fn. 24]; Vison Net, Inc. v. Dept. of Revenue (2019) 397 Mont. 118, 124–125 [447 P.3d 1034, 1038]; cf. Jackson v. Raffensperger (2020) 308 Ga. 736, 741 [843 S.E.2d 576, 581] [applying threshold similarly situated inquiry as matter of state constitutional ‍​​‌​​​‌​​​‌​‌‌‌‌​‌​‌‌‌​​​​‌‌​‌‌‌​​‌‌‌​‌​​​​‌​​‌​‍law].) That inquiry has not escaped criticism elsewhere. (See, e.g., State v. Kelsey (2015) 51 Kan.App.2d 819, 830 [356 P.3d 414, 421] (conc. opn. of Atcheson, J.) [noting that in Kansas — much as in California — a “potentially dispositive threshold test has crept fog-like into our cases on little cat feet. It hasn‘t a basis in generally accepted equal protection jurisprudence, and akin to a morning fog, it obscures the landscape to no particularly useful ends and conceivably dangerous ones“].)
2
The Public Guardian represents that the conservatorship at issue here terminated on June 16, 2020, rendering the appeal moot. The problem frequently arises in this area of law given the short duration of conservatorships. (See Conservatorship of John L. (2010) 48 Cal.4th 131, 142 fn. 2.) Because the case raises important issues capable of repetition but likely to evade review, we exercise our discretion to decide this otherwise moot appeal. (See Conservatorship of K.P. (2021) 11 Cal.5th 695, 705, fn. 3.)
3
Such prisoners were previously described as mentally disordered offenders, or MDO‘s. (See, e.g., People v. Blackburn (2015) 61 Cal.4th 1113, 1116 (Blackburn).) The Legislature recently changed this terminology to “offender with a mental health disorder.” (Pen. Code, § 2962, subd. (d)(3); Stats. 2019, ch. 9, § 7.) In accordance with this change, we now refer to extension proceedings under Penal Code section 2962 as OMHD commitments.
4
The original SVP statutes provided for renewable two-year commitments. (See People v. McKee (2010) 47 Cal.4th 1172, 1185 (McKee).) Now, however, SVP‘s are committed for an indeterminate period (Welf. & Inst. Code, § 6604) but may petition for discharge if they are no longer “a danger to the health and safety of others and . . . not likely to engage in sexually violent criminal behavior” (id., § 6605, subd. (a)(2); see id., §§ 6608-6609).
5
We granted judicial notice of the San Francisco Analyst‘s Report and State Auditor‘s Report at the request of amici curiae Disability Rights California, et al.
6
Many of the statistics cited throughout this opinion do not differentiate between traditional and Murphy conservatees. However, it appears that Murphy conservatees make up a very small proportion of the total number. (See, e.g., San Francisco Analyst‘s Report, supra, at p. A-11.)
7
The Fifth Amendment privilege against self-incrimination is, of course, broader than the right not to testify against oneself in a criminal proceeding. (See, e.g., Miranda v. Arizona (1966) 384 U.S. 436, 467.) Here, however, we are concerned only with the right against giving compelled testimony at a commitment trial. We need not and do not decide whether any other aspect of the privilege applies outside the context of a criminal prosecution.
8
Other privileges are set out in the Evidence Code and relate to a variety of circumstances. (See, e.g., Evid. Code, §§ 954 [attorney-client privilege], 980 [marital communications], 1014 [psychotherapist-patient privilege], 1033-1034 [clergy and penitent privileges].) None of these Evidence Code privileges is implicated in this appeal.
9
In addition to short-term holds for intensive treatment and one-year conservatorships, the LPS Act provides for commitments up to 180 days for individuals who present a substantial risk of physical harm to others as a result of a mental health disorder. (Welf. & Inst. Code, § 5300.)
10
Indefinite commitments for outpatient treatment could also be ordered under certain circumstances. (See Pen. Code, former § 1026.1; Moye, supra, 22 Cal.3d at p. 461.)
11
The distinction primarily impacts the applicable standard of review. Constitutional errors require reversal if thеre is a reasonable possibility they affected the verdict (Chapman v. California (1967) 386 U.S. 18, 23-24), whereas state law errors require reversal only if it is reasonably probable a different result would have been reached absent the error (People v. Watson (1956) 46 Cal.2d 818, 837).
12
Hudec discussed varying approaches taken in the Courts of Appeal grappling with just how broadly Penal Code section 1026.5, subdivision (b)(7) should be interpreted to sweep. It rejected cases employing an overly narrow interpretation but acknowledged that an application leading to absurd consequences could not have been what the Legislature intended. (See Hudec, supra, 60 Cal.4th at pp. 826-830.) Hudec did not attempt to plumb the depths of the question, limiting its analysis to the right against compelled testimony. We do the same.
13
Although conservatorship proceedings were initiated after Bryan S. was found incompetent to stand trial, it appears that a traditional conservatorship was ultimately imposed because the trial court ruled “that Bryan was gravely disabled as a result of a mental disorder and was currently unable to provide for food, clothing, or shelter.” (Bryan S., supra, 42 Cal.App.5th at p. 194, italics added.)
14
We consider only the first rationale articulated by E.B. and J.Y., recognizing that the traditional conservatorships under consideration here are ordinarily imposed for the protection of the conservatee, not the public.
15
Because an equal protection analysis considers whether groups are similarly situated with respect to a particular law, cases cited by the Public Guardian holding that conservatees or NGI‘s are not similarly situated with other civilly committed groups for purposes other than the testimonial privilege shed little light on the issue here.
16
Of course, even if it is ultimately determined that equal protection requires extending the statutory right against compelled testimony to LPS conservatorship trials, a question we do not reach here, recognition of that right would not preclude testimony from other competent witnesses or the admission of relevant documents bearing on grave disability.
Read the detailed case summary