People v. Hardin

Full title: THE PEOPLE, Plaintiff and Respondent, v. TONY HARDIN, Defendant and…

Court: Supreme Court of California

Date published: Mar 4, 2024

318 Cal. Rptr. 3d 513 (Cal. 2024)

543 P.3d 960

Opinion

S277487

03-04-2024

The PEOPLE, Plaintiff and Respondent, v. Tony HARDIN, Defendant and Appellant.

William L. Heyman, under appointment by the Court of Appeal; Munger, Tolles & Olson, William D. Temko, Sara A. McDermott, Adeel Mohammadi, Los Angeles; USC Post-Conviction Justice Project, Heidi Rummel, Michael Parente, Los Angeles, and Danielle A. Wilkins for Defendant and Appellant. Complex Appellate Litigation Group and Greg Wolff for Human Rights Watch, State Senator Loni Hancock (Ret.), the Anti-Recidivism Coalition, the LWOP Alliance Group at Calipatria State Prison and the National Life Without Parole Leadership Council as Amici Curiae on behalf of Defendant and Appellant. Law Office of B.C. McComas, Brian C. McComas, San Francisco; and Eric Weaver, Albany, for the Santa Clara County Independent Defense Counsel Office as Amicus Curiae on behalf of Defendant and Appellant. Cooley, Kathleen R. Hartnett, Darina Shtrakhman, Prianka Misra, Ariana E. Bustos, Adam S. Gershenson, Matt K. Nguyen; and Marsha L. Levick for Neuroscience, Psychology and Juvenile Justice Scholars, Juvenile Law Center, the American Academy of Pediatric Neuropsychology, the Pacific Juvenile Defender Center and the Sentencing Project as Amici Curiae on behalf of Defendant and Appellant. Kim Saltz; Avram Frey; Summer Lacey, Los Angeles; and Diana Garrido for The ACLU, The ACLU of Northern California, The ACLU of Southern California, The California Public Defenders Association and The Contra Costa Public Defender Office as Amici Curiae on behalf of Defendant and Appellant. Law Office of Michael Laurence and Michael Laurence for Catherine M. Grosso as Amicus Curiae on behalf of Defendant and Appellant. Rob Bonta, Attorney General, Michael J. Mongan, State Solicitor General, Lance E. Winters, Chief Assistant Attorney General, Susan Sullivan Pithey, Assistant Attorney General, Helen H. Hong, Deputy State Solicitor General, Noah P. Hill, Idan Ivri and Nima Razfar, Deputy Attorneys General, for Plaintiff and Respondent. Jason Anderson, District Attorney (San Bernardino), and Brent J. Schultze, Deputy District Attorney, for the Distinct Attorney of San Bernardino County as Amicus Curiae on behalf of Plaintiff and Respondent. Kent S. Scheidegger and Kymberlee C. Stapleton for Criminal Justice Legal Foundation as Amicus Curiae on behalf of Plaintiff and Respondent. Jeffrey F. Rosen, District Attorney (Santa Clara), and David R. Boyd, Deputy District Attorney, for the District Attorney of Santa Clara County as Amicus Curiae on behalf of Plaintiff and Respondent. Gibson, Dunn & Crutcher, Eric D. Vandevelde, Los Angeles, Jamila D. MacEbong, Patrick J. Fuster, Los Angeles, Benjamin W. Holston, Jenna Bernard, Maya M. Halthore and Allison P. Miller for Prosecutors Alliance of California as Amicus Curiae.

Opinion of the Court by Kruger, J.

West Code

Negative Treatment Reconsidered

Cal. Penal Code § 3051(h)

Second Appellate District, Division Seven, B315434, Los Angeles County Superior Court, A893110, Juan Carlos Dominguez, Judge

William L. Heyman, under appointment by the Court of Appeal; Munger, Tolles & Olson, William D. Temko, Sara A. McDermott, Adeel Mohammadi, Los Angeles; USC Post-Conviction Justice Project, Heidi Rummel, Michael Parente, Los Angeles, and Danielle A. Wilkins for Defendant and Appellant.

Complex Appellate Litigation Group and Greg Wolff for Human Rights Watch, State Senator Loni Hancock (Ret.), the Anti-Recidivism Coalition, the LWOP Alliance Group at Calipatria State Prison and the National Life Without Parole Leadership Council as Amici Curiae on behalf of Defendant and Appellant.

Law Office of B.C. McComas, Brian C. McComas, San Francisco; and Eric Weaver, Albany, for the Santa Clara County Independent Defense Counsel Office as Amicus Curiae on behalf of Defendant and Appellant.

Cooley, Kathleen R. Hartnett, Darina Shtrakhman, Prianka Misra, Ariana E. Bustos, Adam S. Gershenson, Matt K. Nguyen; and Marsha L. Levick for Neuroscience, Psychology and Juvenile Justice Scholars, Juvenile Law Center, the American Academy of Pediatric Neuropsychology, the Pacific Juvenile Defender Center and the Sentencing Project as Amici Curiae on behalf of Defendant and Appellant.

Kim Saltz; Avram Frey; Summer Lacey, Los Angeles; and Diana Garrido for The ACLU, The ACLU of Northern California, The ACLU of Southern California, The California Public Defenders Association and The Contra Costa Public Defender Office as Amici Curiae on behalf of Defendant and Appellant.

Law Office of Michael Laurence and Michael Laurence for Catherine M. Grosso as Amicus Curiae on behalf of Defendant and Appellant. Rob Bonta, Attorney General, Michael J. Mongan, State Solicitor General, Lance E. Winters, Chief Assistant Attorney General, Susan Sullivan Pithey, Assistant Attorney General, Helen H. Hong, Deputy State Solicitor General, Noah P. Hill, Idan Ivri and Nima Razfar, Deputy Attorneys General, for Plaintiff and Respondent.

Jason Anderson, District Attorney (San Bernardino), and Brent J. Schultze, Deputy District Attorney, for the Distinct Attorney of San Bernardino County as Amicus Curiae on behalf of Plaintiff and Respondent.

Kent S. Scheidegger and Kymberlee C. Stapleton for Criminal Justice Legal Foundation as Amicus Curiae on behalf of Plaintiff and Respondent.

Jeffrey F. Rosen, District Attorney (Santa Clara), and David R. Boyd, Deputy District Attorney, for the District Attorney of Santa Clara County as Amicus Curiae on behalf of Plaintiff and Respondent.

Gibson, Dunn & Crutcher, Eric D. Vandevelde, Los Angeles, Jamila D. MacEbong, Patrick J. Fuster, Los Angeles, Benjamin W. Holston, Jenna Bernard, Maya M. Halthore and Allison P. Miller for Prosecutors Alliance of California as Amicus Curiae.

Opinion of the Court by Kruger, J.

California’s youth offender parole statute offers opportunities for early release to certain persons who are incarcerated for crimes they committed at a young age. (Pen. Code, §§ 3051, 4801.) When it was first enacted in 2013, the statute applied only to individuals who committed their crimes before the age of 18; the purpose of the statute was to align California law with then-recent court decisions identifying Eighth Amendment limitations on life without parole sentences for juvenile offenders. In more recent years, however, the Legislature has expanded the statute to include certain young adult offenders as well. Under the current version of the statute, most persons incarcerated for a crime committed between ages 18 and 25 are entitled to a parole hearing during the 15th, 20th, or 25th year of their incarceration, (Pen. Code, § 3051, subd. (b).) But not all youthful offenders are eligible for parole hearings. The statute excludes, among others, offenders who are serving sentences of life in prison without the possibility of parole for a crime committed after the age of 18. (Id., subd. (h).)

Appellant Tony Hardin is currently serving a life without parole sentence for a special circumstance murder he committed at age 25. He contends that the youth offender parole statute violates the Fourteenth Amendment’s equal protection guarantee by irrationally discriminating against young adult offenders sentenced to life without parole — including, in particular, those sentenced to life without parole for special circumstance murder. Agreeing with Hardin and disagreeing with other appellate decisions to address the issue, the Court of Appeal held the life without parole exclusion invalid for lack of a rational basis.

We now reverse. The standard we apply here, rational basis review, is necessarily deferential. The law recognizes that "[i]t is both the prerogative and the duty of the Legislature to define degrees of culpability and punishment, and to distinguish between crimes in this regard." (People v. Turnage (2012) 55 Cal.4th 62, 74, 144 Cal. Rptr.3d 489, 281 P.3d 464.) Respect for the Legislature’s proper role — and ours — means that we may not strike down its enactment under a rational basis standard unless the challengers demonstrate that "there is no ‘rational relationship between the disparity of treatment and some legitimate governmental purpose.’ " (Ibid.) Without foreclosing the possibility of other as-applied challenges to the statute, we conclude that Hardin has not demonstrated that Penal Code section 3051’s exclusion of young adult offenders sentenced to life without parole is constitutionally invalid under a rational basis standard, either on its face or as applied to Hardin and other individuals who are serving life without parole sentences for special circumstance murder. Under California law, special circumstance murder is a uniquely serious offense, punishable only by death or life without possibility of parole. When it was considering whether to expand the youth offender parole system to include not only juvenile offenders but also certain young adults, the Legislature could rationally balance the seriousness of the offender’s crimes against the capacity of all young adults for growth, and determine that young adults who have committed certain very serious crimes should remain ineligible for release from prison. Hardin has not demonstrated that the Legislature acted irrationally in declining to grant the possibility of parole to young adult offenders convicted of special circumstance murder, even as it has granted youth offender hearings to young adults convicted of other offenses.

This conclusion does not turn on this court’s judgments about what constitutes sound sentencing policy. It turns on the deference we owe to the policy choices made through the democratic process by the people of California and their elected representatives. The legislative branch may continue to consider the appropriate reach of the youth offender parole statute in light of the recognized capacity of young persons for growth and change. Hardin has not, however, established that the legislative policy choices reflected in current law are irrational and therefore impermissible as a matter of equal protection.

I.

In 1989, Hardin robbed and killed an elderly neighbor. Hardin was then 25 years old. A jury convicted Hardin of first degree murder, among other offenses. The jury also found true a special circumstance allegation that Hardin murdered the victim during the commission of a robbery. Hardin’s conviction for first degree murder with special circumstances carried a mandatory sentence of either death or life in prison without the possibility of parole. (Pen. Code, § 190.2, subd. (a); id., subd. (a)(17)(A).) Although the prosecution had sought the death penalty, the penalty phase jury declined to return a death verdict. The trial court imposed a sentence of life in prison without parole for the murder and stayed the sentences for the other convictions.

Decades later, Hardin filed a postjudgment motion to develop and preserve evidence for later use in a youth offender parole hearing under Penal Code section 3051 (section 3051). (See People v. Franklin (2016) 63 Cal.4th 261, 283–284, 202 Cal.Rptr.3d 496, 370 P.3d 1053 (Franklin) [an offender who will later become eligible for a youth offender parole hearing is entitled to an interim court proceeding to develop and preserve evidence of youth-related characteristics and circumstances at the time of the offense]; In re Cook (2019) 7 Cal.5th 439, 458–459, 247 Cal.Rptr.3d 669, 441 P.3d 912 [an offender whose sentence is otherwise final may obtain a Franklin hearing by filing a postjudgment motion in superior court].) In his motion, Hardin acknowledged that, as an offender sentenced to life without parole for a crime committed as a young adult, he is not eligible for a youth offender parole hearing. (§ 3051, subd. (h).) He contended, however, that his exclusion violates the Equal Protection Clause of the Fourteenth Amendment to the federal Constitution. The superior court rejected the contention and denied Hardin’s motion. The Court of Appeal, however, reversed. (People v. Hardin (2022) 84 Cal.App.5th 273, 291, 300 Cal. Rptr.3d 243 (Hardin).)

On appeal, Hardin raised two equal protection arguments. He first argued that section 3051 violates equal protection by excluding young adult offenders sentenced to life without parole while including juvenile offenders (that is, offenders younger than 18 at the time of the offense) sentenced to life without parole. The Court of Appeal rejected this argument. It explained that the Legislature had a rational basis for distinguishing between juvenile offenders and young adult offenders, since a unique set of constitutional rules restricts sentencing children to life without parole. (Hardin, supra, 84 Cal.App.5th at pp. 285–286, 300 Cal.Rptr.3d 243, citing, inter alia, Miller v. Alabama (2012) 567 U.S. 460, 132 S.Ct. 2455, 183 L.Ed.2d 407 (Miller).) Hardin does not challenge the Court of Appeal’s conclusion on this point.

Hardin next argued that section 3051 violates equal protection by treating young adult offenders sentenced to life without parole for special circumstance murder differently from other young adult offenders serving parole-eligible life sentences for other crimes. On this point, the Court of Appeal agreed with Hardin. (Hardin, supra, 84 Cal.App.5th at p. 291, 300 Cal. Rptr.3d 243.)

Employing the two-step equal protection analysis prescribed by our cases (see, e.g., Conservatorship of Eric B. (2022) 12 Cal.5th 1085, 1102, 293 Cal.Rptr.3d 93, 508 P.3d 1099 (Eric B.)), the Court of Appeal began by considering whether, in light of the purposes of the challenged law, young adult offenders convicted of special circumstance murder and sentenced to life without parole are similarly situated to all other young offenders. The court answered yes. It explained that the Legislature’s stated purpose in enacting section 3051 was to permit "a determination whether a person who committed a serious or violent crime between the age of 18 and 25 has sufficiently matured and outgrown the youthful impulses that led to the commission of the offense." (Hardin, supra, 84 Cal.App.5th at p. 287, 300 Cal.Rptr.3d 243.) The court concluded that all young offenders are similarly situated from this standpoint, since a person’s potential for increased maturity and growth is not crime-specific. (Ibid.)

Turning to the next step of the analysis, the basis for the disparate treatment of similarly situated groups, the court concluded there was no rational basis for section 3051 to distinguish between young adult offenders convicted of special circumstance murder and sentenced to life without parole and other young adult offenders. The court again adverted to the stated purpose of section 3051: "[I]f, as the Legislature stated, the goal of section 3051 was … to permit youth offenders a meaningful opportunity for parole if they demonstrate increased maturity and impulse control, then for that purpose there is no plausible basis for distinguishing between same-age offenders based solely on the crime they committed." (Hardin, supra, 84 Cal.App.5th at p. 288, 300 Cal.Rptr.3d 243; see id. at pp. 278–279, 300 Cal.Rptr.3d 243.)

The Court of Appeal acknowledged other appellate cases had reached a different conclusion. In those cases, the courts reasoned that the Legislature, in determining which young adult offenders should be afforded opportunities for early release, permissibly decided to take into account the seriousness of the offender’s crime and rationally decided to exclude those who had committed crimes sufficiently serious to warrant a sentence of life without parole. (Hardin, supra, 84 Cal.App.5th pp. 288–289, 300 Cal.Rptr.3d 243 [citing cases].) But the court in this case rejected this "superficially plausible justification" as "belied by the statutory provisions that allow [a youth offender parole] hearing for individuals who have committed multiple violent crimes (albeit not special circumstance murder) and were sentenced to a technically parole-eligible indeterminate state prison term that is the functional equivalent of life without parole." (Id. at p. 289, 300 Cal.Rptr.3d 243.) The court also deemed "illusory" any differences between the culpability of individuals convicted of first degree murder without special circumstances and first degree murder with special circumstances. (Id. at p. 290, 300 Cal.Rptr.3d 243.) The court relied for this conclusion on a law review article finding that, as a result of the expansion of the special circumstance statute over time, at least one special circumstance could have been alleged in 95 percent of first degree murder cases. (Id. at p. 290 & fn. 11, 300 Cal.Rptr.3d 243 [citing Com. on Revision of the Pen. Code, Annual Report and Recommendations (2021) p. 51, in turn citing Baldus et al., Furman at 45: Constitutional Challenges from California’s Failure to (Again) Narrow Death Eligibility (2019) 16 J. Empirical Legal Studies 693].) Ultimately, finding no rational basis for the challenged life without parole exclusion, the court concluded that "the disparate treatment of offenders like Hardin cannot stand." (Hardin, at p. 291, 300 Cal.Rptr.3d 243.)

We granted review to resolve the conflict between the Court of Appeal’s decision in this case and the decisions of the other appellate courts to address the issue.^{1 1}

1 The Court of Appeal in this case was the first to conclude that section 3051’s exclusion of young adults sentenced to life without parole violated equal protection. Before Hardin, several published appellate opinions had reached the opposite conclusion. (See In re Williams (2020) 57 Cal.App.5th 427, 271 Cal. Rptr.3d 453; People v. Sands (2021) 70 Cal. App.5th 193, 285 Cal.Rptr.3d 238; People v. Morales (2021) 67 Cal.App.5th 326, 282 Cal. Rptr.3d 151; People v. Jackson (2021) 61 Cal. App.5th 189, 279 Cal.Rptr.3d 396; People v. Acosta (2021) 60 Cal.App.5th 769, 275 Cal. Rptr.3d 110; People v. Montano (2022) 80 Cal.App.5th 82, 295 Cal.Rptr.3d 437.) More appellate decisions have done so since Hardin. (People v. Ngo (2023) 89 Cal.App.5th 116, 305 Cal.Rptr.3d 630, review granted May 17, 2023, S279458; People v. Bolanos (2023) 87 Cal.App.5th 1069, 304 Cal.Rptr.3d 237, review granted Apr. 12, 2023, S278803 [distinguishing Hardin on the ground that it involved a murder conviction, as opposed to a sex offense conviction carrying a life without parole sentence under the One Strike law, Pen. Code, § 667.61].)

1 All further unspecified statutory references are to the Penal Code.

II.

A.

Section 3051 provides that, at a time designated in the statute, the Board of Parole Hearings must hold a parole hearing "for the purpose of reviewing the parole suitability of any prisoner who was 25 years of age or younger … at the time of the controlling offense." (§ 3051, subd. (a)(1); id., subd. (d).) How much time must pass before an eligible youth offender receives a parole hearing depends on the length of the original sentence for the " ‘[c]ontrolling offense,’ " a term defined to mean "the offense or enhancement for which any sentencing court imposed the longest term of imprisonment." (Id., subd. (a)(2)(B).) An offender sentenced to a determinate term becomes eligible for parole after 15 years (id., subd. (b)(1)); an offender sentenced to an indeterminate life term of fewer than 25 years to life becomes eligible after 20 years (id., subd. (b)(2)); and an offender sentenced to an indetermi- nate life term of 25 years to life, or an offender sentenced to life without parole for a crime committed before the age of 18, becomes eligible after 25 years (id., subd. (b)(3), (4)).

Certain persons are, however, categorically ineligible for youth offender parole hearings, including offenders sentenced for multiple violent or serious felonies under the "Three Strikes" law (Pen. Code, §§ 667, subds. (b)–(i), 1170.12); offenders sentenced for sex offenses under the One Strike law (id., § 667.61); and offenders who, "subsequent to attaining 26 years of age, commit[ ] an additional crime for which malice aforethought is a necessary element of the crime or for which the individual is sentenced to life in prison." (§ 3051, subd. (h).) The statute also excludes those who, like Hardin, are sentenced to life without parole for a controlling offense committed after reaching the age of 18. (Ibid.) In Hardin’s case, as in most of the appellate cases addressing the issue, the offense is first degree murder with one or more special circumstances. (Pen. Code, § 190.2.)

B.

The Legislature first created this system of youth offender parole hearings in 2013, following a series of court decisions identifying Eighth Amendment limits on the sentencing of juvenile offenders. (Stats. 2013, ch. 312, § 1; see generally Franklin, supra, 63 Cal.4th at p. 277, 202 Cal.Rptr.3d 496, 370 P.3d 1053.) In Roper v. Simmons (2005) 543 U.S. 551, 125 S.Ct. 1183, 161 L.Ed.2d 1 (Roper), the high court held that the Eighth Amendment forbids imposing the death penalty for crimes committed before age 18, given the diminished culpability of juveniles relative to adult offenders. (Roper, at p. 575, 125 S.Ct. 1183.) Five years later, the high court held in Graham v. Florida (2010) 560 U.S. 48, 130 S.Ct. 2011, 176 L.Ed.2d 825 (Graham) that the Eighth Amendment also forbids life without parole sentences for nonhomicide crimes committed before age 18. (Graham, at p. 82, 130 S.Ct. 2011.) Finally, in Miller, supra, 567 U.S. 460, 132 S.Ct. 2455, the high court held that the Eighth Amendment forbids mandatory life without parole sentences for homicides committed before the age of 18. (Miller, at pp. 479–480, 132 S.Ct. 2455; see id. at pp. 477–478, 489, 132 S.Ct. 2455.)

In each case, the high court explained why juvenile offenders are "constitutionally different" from adult offenders for purposes of criminal sentencing. (Miller, supra, 567 U.S. at p. 471, 132 S.Ct. 2455.) Relying "not only on common sense — on what ‘any parent knows’ — but on science and social science," the court identified three primary differences between juveniles and adults. (Ibid.) First, the "hallmark features" of youth — "among them, immaturity, impetuosity, and failure to appreciate risks and consequences" — both diminish a child’s moral culpability and increase the chances that the child’s moral shortcomings will be reformed with age. (Id. at p. 477, 132 S.Ct. 2455; see id. at p. 472, 132 S.Ct. 2455.) Second, children " ‘are more vulnerable … to negative influences and outside pressures,’ including from their family and peers; they have limited ‘contro[l] over their own environment’ and lack the ability to extricate themselves from horrific, crime-producing settings." (Id. at p. 471, 132 S.Ct. 2455, quoting Roper, supra, 543 U.S. at p. 569, 125 S.Ct. 1183.) And finally, compared to an adult, a juvenile’s character is "not as ‘well formed’ … his traits are ‘less fixed’ " and thus "his actions less likely to be ‘evidence of irretrievabl[e] deprav[ity].’ " (Miller, at p. 471, 132 S.Ct. 2455, quoting Roper, at p. 570, 125 S.Ct. 1183.) In Graham and Roper, the court held that these features of youth categorically preclude a death sentence, or a sentence of life without parole for a nonhomicide offense. But in ruling out life without parole sentences for nonhomicide offenses committed by juveniles, the court in Graham "took care" to distinguish homicide offenses, which raise different considerations as a matter of "both moral culpability and consequential harm." (Miller, supra, 567 U.S. at p. 473, 132 S.Ct. 2455.) When confronted with the issue in Miller, the court did not categorically rule out life without parole sentences for juvenile offenders, instead concluding that before a court may impose such a sentence, "a judge or jury must have the opportunity to consider mitigating circumstances," including the hallmark features of youth and their relation to the offense. (Id. at p. 489, 132 S.Ct. 2455.) The court further observed that, in light of "children’s diminished culpability and heightened capacity for change, we think appropriate occasions for sentencing juveniles to th[e] harshest possible penalty [of life without parole] will be uncommon. That is especially so because of the great difficulty we noted in Roper and Graham of distinguishing at this early age between ‘the juvenile offender whose crime reflects unfortunate yet transient immaturity, and the rare juvenile offender whose crime reflects irreparable corruption.’ " (Id. at pp. 479–480, 132 S.Ct. 2455, quoting Graham, supra, 560 U.S. at p. 68, 130 S.Ct. 2011.)

Not long after the high court issued its decision in Miller, this court clarified in People v. Caballero (2012) 55 Cal.4th 262, 268, 145 Cal.Rptr.3d 286, 282 P.3d 291 (Caballero) that Graham's prohibition on life without parole sentences for juvenile nonhomicide offenders applies to a term-of-years sentence that is "the functional equivalent of a life without parole sentence" — there, a sentence of 110 years. (Ibid.) Without dictating "a precise time-frame" for holding parole hearings for juvenile offenders who had received actual or de facto life sentences for nonhomicide crimes, this court explained that, under Graham, "a state must provide a juvenile offender ‘with some realistic opportunity to obtain release’ from prison during his or her expected lifetime." (Id. at pp. 269, 268, 145 Cal.Rptr.3d 286, 282 P.3d 291.)

The Legislature enacted section 3051 to bring California juvenile sentencing law into line with Graham, Miller, and Caballero. (Stats. 2013, ch. 312, § 1; see Franklin, supra, 63 Cal.4th at p. 268, 202 Cal.Rptr.3d 496, 370 P.3d 1053; id. at pp. 278–280, 202 Cal.Rptr.3d 496, 370 P.3d 1053 [holding that the youth offender parole statute remedied any Eighth Amendment defects in the sentences of juvenile offenders].) In language echoing the holdings of these cases, section 3051 provided for youth offender parole hearings at which the Board of Parole Hearings must provide "a meaningful opportunity" for release (§ 3051, subd. (e)), giving "great weight to the diminished culpability of youth as compared to adults, the hallmark features of youth, and any subsequent growth and increased maturity" (Pen. Code, § 4801, subd. (c)).

As initially enacted, section 3051 provided youth offender parole hearings only for juvenile offenders incarcerated for crimes committed before the age of 18. (Former § 3051, subd. (a)(1), added by Stats. 2013, ch. 312, § 4.) But it did not include all juvenile offenders; the statute excluded several categories of individuals, including juvenile offenders sentenced to life without possibility of parole. (Former § 3051, subd. (h), added by Stats. 2013, ch. 312, § 4.) A different statute, enacted not long before section 3051, had created an alternative mechanism for relief that, with some exceptions, permitted juvenile offenders sen- fenced to life without parole to petition for recall of sentence and resentencing to a term that included an opportunity for parole. (Stats. 2012, ch. 828, adding Pen. Code, § 1170, subd. (d).)

Since the youth offender parole statute was first enacted, the Legislature has expanded it in two primary respects. The first area of change concerns juvenile offenders sentenced to life without possibility of parole. In 2017, this court concluded the recall and resentencing scheme did not provide an adequate remedy for juvenile offenders who had been sentenced to life without parole terms without adequate consideration of the youth-related factors set out in Miller. (In re Kirchner (2017) 2 Cal.5th 1040, 1043, 216 Cal.Rptr.3d 876, 393 P.3d 364 (Kirchner).) That same year, the Legislature expanded section 3051 to include juvenile offenders sentenced to life without parole, making them eligible for youth offender parole hearings after their 25th year of incarceration. (Stats. 2017, ch. 684, § 1.5, adding § 3051, subd. (b)(4); see Assem. Com. on Public Safety, Analysis of Sen. Bill No. 394 (2017-2018 Reg. Sess.) as amended May 26, 2017, p. 1.)

The second area of change concerns the statute’s application to older offenders. In 2015, the Legislature raised the age of eligibility for youth offender parole hearings to include most young adults incarcerated for offenses committed before the age of 23. (Stats. 2015, ch. 471, § 1.) In expanding section 3051 beyond the constitutional minimum age of 18 set out in Graham and Miller, the Legislature considered scientific evidence that neurological development, particularly in areas of the brain relevant to judgment and decisionmaking, continues beyond adolescence and into the mid-20’s. (See Sen. Com. on Public Safety, Rep. on Sen. Bill No. 261 (2015–2016 Reg. Sess.) Apr. 28, 2015, p. 3.) In 2017, motivated by these same considerations, the Legislature once again raised the age cut-off for section 3051 parole hearings, this time to age 25. (Stats. 2017, ch. 675, § 1; see Assem. Com. on Public Safety, Analysis of Assem. Bill No. 1308 (2017–2018 Reg. Sess.) as amended Mar. 30, 2017, p. 2.)

The expansion to young adults did not, however, include all persons who committed crimes between the age of 18 and 25: The Legislature carried forward preexisting exclusions, including the exclusion for those sentenced to life in prison without the possibility of parole. (See Assem. Com. on Public Safety, Analysis of Assem. Bill No. 1308, supra, as amended Mar. 30, 2017, p. 2.) Similarly, when it expanded the youth offender parole system to include juvenile offenders sentenced to life without parole, the Legislature preserved the life without parole exclusion for youthful offenders who committed their controlling offense after the age of 18. (See Assem. Com. on Public Safety, Analysis of Sen. Bill No. 394, supra, as amended May 26, 2017, p. 1.)

Hardin challenges the statute’s exclusion of young adult offenders sentenced to life without parole as violative of equal protection. As noted, in the trial court, Hardin challenged the statute’s disparate treatment of juvenile and young adult offenders sentenced to life without possibility of parole. But the Court of Appeal in this case held, and he does not dispute, that the Legislature acted reasonably in distinguishing between offenses committed before and after the age of 18 because the Eighth Amendment (and the law more generally) makes the same distinction. (Hardin, supra, 84 Cal.App.5th at pp. 285–286, 300 Cal.Rptr.3d 243 [noting that age 18 generally marks the difference between childhood and adulthood].)

As the case comes to us, the parties agree that the Legislature was not constitutionally obligated to expand youth of- fender parole opportunities to young adults over the age of 18. Hardin argues, however, that once the Legislature decided to expand such opportunities to young adults, it could not rationally treat those sentenced to life without parole differently from those convicted of other serious crimes and serving lengthy parole-eligible sentences. Once the Legislature decided to include one class of young adult offenders, it was obligated to include both.

Hardin effectively challenges the life without parole exclusion on its face, in all of its applications. He also challenges the exclusion more specifically as it applies to young adult offenders who are, like him, serving life without parole sentences following convictions for first degree murder with one or more special circumstances.

III.

[1, 2] The Equal Protection Clause of the Fourteenth Amendment to the United States Constitution provides that no state may "deny to any person within its jurisdiction the equal protection of the laws."^{2 2} (U.S. Const., 14th Amend.) This provision is "essentially a direction that all persons similarly situated should be treated alike." (Cleburne v. Cleburne Living Center, Inc. (1985) 473 U.S. 432, 439, 105 S.Ct. 3249, 87 L.Ed.2d 313 (Cleburne).) "At core, the requirement of equal protection ensures that the government does not treat a group of people unequally without some justification." (People v. Chatman (2018) 4 Cal.5th 277, 288, 228 Cal.Rptr.3d 379, 410 P.3d 9 (Chatman).)

2 The California Constitution also guarantees equal protection of the law. (Cal. Const., art. I, § 7, subd. (a).) Hardin does not raise any arguments specific to the California Constitution, however, and we see " ‘no reason to suppose’ that federal equal protection analysis would yield a result different from what would emerge from analysis of the state Constitution." (Chatman, supra, 4 Cal.5th at p, 288, 228 Cal.Rptr.3d 379, 410 P.3d 9.)

2 The majority declines to consider the significance of the racially disparate impact of the LWOP exclusion by faulting Hardin for not having raised a claim that heightened scrutiny applies on that basis. (Maj. opn., supra, 318 Cal.Rptr.3d at p. 539, fn. 8, 543 P.3d at pp. 981–982, fn. 8.) However, amici raised that argument and, following our order directing him to do so, the Attorney General filed an answer addressing that argument. Thus, due consideration of the racially disparate impact of the LWOP exclusion is warranted.

[3, 4] The degree of justification required to satisfy equal protection depends on the type of unequal treatment at issue. Courts apply heightened scrutiny when a challenged statute or other regulation involves a suspect classification such as race, or a fundamental right such as the right to vote, and accordingly will demand greater justification for the differential treatment. (E.g., Chatman, supra, 4 Cal.5th at p. 288, 228 Cal.Rptr.3d 379, 410 P.3d 9; Massachusetts Bd. of Retirement v. Murgia (1976) 427 U.S. 307, 312, 96 S.Ct. 2562, 49 L.Ed.2d 520.) But when a statute involves neither a suspect classification nor a fundamental right, the "general rule is that legislation is presumed to be valid and will be sustained if the classification drawn by the statute is rationally related to a legitimate state interest." (Cleburne, supra, 473 U.S. at p. 440, 105 S.Ct. 3249; see Chatman, at pp. 288-289, 228 Cal.Rptr.3d 379, 410 P.3d 9.) A court applying this standard finds "a denial of equal protection only if there is no rational relationship between a disparity in treatment and some legitimate government purpose." (Chatman, at pp. 288–289, 228 Cal.Rptr.3d 379, 410 P.3d 9.)

Here, both sides agree that rational basis review applies; Hardin makes no argument that this case involves a suspect classification or a fundamental right. (See Chatman, supra, 4 Cal.5th at pp. 282, 287, 228 Cal.Rptr.3d 379, 410 P.3d 9 [rational basis review applied to evaluate constitutionality of law prescribing different collateral consequences for different types of criminal convictions]; People v. Wilkinson (2004) 33 Cal.4th 821, 838, 16 Cal.Rptr.3d 420, 94 P.3d 55l (Wilkinson) [A defendant " ‘does not have a fundamental interest in a specific term of imprisonment’ "].)

In the past, our cases have set out a two-part inquiry to evaluate equal protec- tion claims. "We first ask whether the state adopted a classification affecting two or more groups that are similarly situated in an unequal manner. [Citation.] If we deem the groups at issue similarly situated in all material respects, we consider whether the challenged classification" is adequately justified. (Chatman, supra, 4 Cal.5th at p. 289, 228 Cal.Rptr.3d 379, 410 P.3d 9.) In a case, like this one, subject to rational basis review, the question is "whether the challenged classification ultimately bears a rational relationship to a legitimate state purpose." (Ibid.)

The Courts of Appeal that have addressed the issue presented here concerning the life without parole exclusion have fractured over the proper analysis of the threshold "similarly situated" inquiry. At this first step of the two-part equal protection inquiry, the reviewing court asks "not whether persons are similarly situated for all purposes, but ‘whether they are similarly situated for purposes of the law challenged.’ " (Cooley v. Superior Court (2002) 29 Cal.4th 228, 253, 127 Cal.Rptr.2d 177, 57 P.3d 654, quoting People v. Gibson (1988) 204 Cal.App.3d 1425, 1438, 252 Cal. Rptr. 56.) If the challenging party fails to satisfy this threshold " ‘similarly situated’ " inquiry, the equal protection analysis is at an end. (Cooley, at p. 254, 127 Cal.Rptr.2d 177, 57 P.3d 654.)

The Court of Appeal in this case held that offenders serving life without parole sentences are, for purposes of the youth offender parole statute, similarly situated to offenders serving parole-eligible life terms for offenses committed at the same age. It then went on to hold that the statute’s disparate treatment of the two groups is not adequately justified. (Hardin, supra, 84 Cal.App.5th at pp. 287–288, 290, 300 Cal.Rptr.3d 243.) Several other courts have likewise concluded that the groups are similarly situated for purposes of the challenged law, but that the difference in treatment is justified. A still larger group of courts have concluded that the groups are not similarly situated for purposes of the law, while citing essentially the same reasons other courts have cited at the justification step of the inquiry. And the largest group of courts have avoided the question by assuming without deciding that the two groups are similarly situated and proceeding to hold that the difference in treatment is justified under rational basis review.

Despite this state of uncertainty, the Attorney General asks us to join the group of courts that have avoided the issue by assuming without deciding that a young adult offender serving a parole eligible life sentence is similarly situated to an individual serving a sentence of life without parole for an offense committed at a similar age. The Attorney General thus would have us proceed directly to the operative question, which is whether the disparate treatment has a rational basis.

We have taken this assume-without-deciding approach to the "similarly situated" inquiry in other recent equal protection cases and could do the same here. (Chatman, supra, 4 Cal.5th at p. 290, 228 Cal. Rptr.3d 379, 410 P.3d 9 [moving to the second step of the equal protection analysis without deciding the first, "similarly situated" step]; Johnson v. Department of Justice (2015) 60 Cal.4th 871, 882, 183 Cal.Rptr.3d 96, 341 P.3d 1075 (Johnson).) But to do so would simply perpetuate the uncertainty that has led courts to so many different conclusions about how the "similarly situated" test ought to apply, and that has so often led both this court and the Courts of Appeal to avoid the test altogether.

There is a reason for this uncertainty. As we recognized decades ago, in cases involving challenges to statutes like section 3051, subdivision (h) that facially distinguish between identifiable groups or classes of individuals, "[t]o ask whether two groups are similarly situated in this context," given the interests underlying the law challenged, is essentially "the same as asking whether the distinction between them can be justified under the appropriate test of equal protection." (Fullerton, Joint Union High School Dist. v. State Bd. of Education (1982) 32 Cal.3d 779, 798, fn. 19, 187 Cal.Rptr. 398, 654 P.2d 168 (plur. opn.).) This is because one can only reach the conclusion that two groups are similarly situated with respect to the purposes of a particular law after considering the law’s aims and how the differential treatment relates to those aims. But the first, "similarly situated" step of the analysis provides substantially less guidance about how this inquiry is to proceed: "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." (Eric B., supra, 12 Cal.5th at p. 1115, 293 Cal.Rptr.3d 93, 508 P.3d 1099 (conc. opn. of Kruger, J.).) In the context of challenges like this one, the similarly situated test serves no real purpose. At best it duplicates the justification inquiry prescribed at the second step of the analysis; at worst it creates an unnecessary threshold obstacle to the adjudication of potentially meritorious constitutional challenges; and in all events it injects unnecessary uncertainty into the analysis, particularly in the situations in which the challenged law reflects multiple, sometimes competing aims.

Our cases purported to derive the threshold "similarly situated" test from United States Supreme Court guidance, but the high court itself has not employed any similar threshold test in equal protection cases involving challenges to facial legal classifications. (See, e.g., Cleburne, supra, 473 U.S. at pp. 439–450, 105 S.Ct. 3249.) Even when this court first began to speak in terms of a "similarly situated" test, it did not initially understand this to mean that courts must always engage in that inquiry as a separate analytical step. (See In re Roger S. (1977) 19 Cal.3d 921, 141 Cal.Rptr. 298, 569 P.2d 1286; In re Eric J. (1979) 25 Cal.Sd 522, 159 Cal.Rptr. 317, 601 P.2d 549.) Rather, courts reciting the rules of these cases over time came to lay out a two-step analysis, even though no court ever identified precisely what independent function the first step is supposed to serve. Unsurprisingly, then, courts did not apply it consistently, often adopting an approach of assuming-without-deciding that the groups or classes facing disparate treatment are similarly situated, or skipping the inquiry altogether, to reach the critical question of whether the justification for the alleged disparate treatment is adequate. (See, e.g., Chatman, supra, 4 Cal.5th at p. 290, 228 Cal.Rptr.3d 379, 410 P.3d 9; Johnson, supra, 60 Cal.4th at p. 882, 183 Cal.Rptr.3d 96, 341 P.3d 1075; Hernandez v. City of Hanford (2007) 41 Cal.4th 279, 299, 59 Cal.Rptr.3d 442, 159 P.3d 33 (Hernandez); People v. Floyd (2003) 31 Cal.4th 179, 190, 1 Cal.Rptr.3d 885, 72 P.3d 820.)

After directing the parties and inviting amici curiae to address this issue, none has identified any substantive reason why we should continue to prescribe a two-step analysis in cases like this one, in which the only real question is whether a facial difference in treatment is adequately justified by the purposes the law was meant to serve. The primary concern raised by the Attorney General relates to stare decisis — the idea that once an issue is decided, it should ordinarily remain decided.

[5, 6] Stare decisis plays a vitally important role in our work as a common law court; the policy of adherence to precedent ensures the certainty, stability, and predictability on which the rule of law depends. But stare decisis concerns have no real place here. The doctrine "does not ‘ "shield court-created error from correction" ’ " but "permits us ‘to reconsider, and ultimately to depart from, our own prior precedent in an appropriate case.' " (People v. Mendoza (2000) 23 Cal.4th 896, 924, 98 Cal.Rptr.2d 431, 4 P.3d 265.) Here, none of the factors we have identified as relevant to the question of adherence to precedent — including "the age of the precedent, the nature and extent of public and private reliance on it, and its consistency or inconsistency with other related rules of law" (Trope v. Katz (1995) 11 Cal.4th 274, 288, 45 Cal.Rptr.2d 241, 902 P.2d 259) — suggests we are bound to preserve an analytical framework that has generated uncertainty and confusion, with no discernible effect on the actual outcomes of cases.

[7, 8] For these reasons, we now hold that, when plaintiffs challenge laws drawing distinctions between identifiable groups or classes of persons, on the basis that the distinctions drawn are inconsistent with equal protection, courts no longer need to ask at the threshold whether the two groups are similarly situated for purposes of the law in question. The only pertinent inquiry is whether the challenged difference in treatment is adequately justified under the applicable standard of review. The burden is on the party challenging the law to show that it is not.

To be clear, we cast no doubt on the utility of "similarly situated" inquiries in other contexts. In cases that do not involve challenges to classifications appearing on the face of the law, to ask whether a person has been treated differently from another person similarly situated is typically how we determine whether a person has been treated differently on the basis of group membership or another actionable basis. We do not call into question the established role the similarly situated inquiry plays in, for instance, cases involving claims of group-based discrimination against individuals, in which plaintiffs bear the burden of showing disparate treatment along class lines, or so-called "class of one" cases that do not allege differential treatment on the basis of class membership. (See, e.g., United States v. Armstrong (1996) 517 U.S. 456, 465–467, 116 S.Ct. 1480, 134 L.Ed.2d 687; Village of Willowbrook v. Olech (2000) 528 U.S. 562, 564, 120 S.Ct. 1073, 145 L.Ed.2d 1060.)

Nor, in dispensing with the threshold "similarly situated" test in equal protection challenges like this one, do we call into question any of this court’s precedent that purported to dispose of an equal protection challenge upon deciding that the challenged disparate treatment did not involve groups that were similarly situated for purposes of the law in question. As we have explained, the conclusion in each of those cases could just as well have been cast as a conclusion about whether the difference in treatment was adequately justified under the applicable standard of review. (See, e.g., People v. Salazar (2016) 63 Cal.4th 214, 227, 202 Cal.Rptr.3d 638, 371 P.3d 161 [noting individuals who commit a capital crime after being convicted of a juvenile murder in superior court are not similarly situated to those whose prior murder was adjudicated in juvenile court, because the Legislature may fairly distinguish these groups based on culpability]; People v. Johnson (1992) 3 Cal.4th 1183, 1242–1243, 14 Cal.Rptr.2d 702, 842 P.2d 1 [noting capital defendants are not similarly situated to those subject to ordinary sentencing enhancements because of the aggravating circumstances surrounding the capital offense].) Having thus clarified the governing analytical framework, we turn to the central inquiry in this case: whether there is a rational basis justifying section 3051’s disparate treatment of individuals who, like Hardin, are serving sentences of life without parole for special circumstance murder.

IV.

A.

[9] Rational basis review "sets a high bar" for litigants challenging legislative enactments. (Chatman, supra, 4 Cal.5th at p. 289, 228 Cal.Rptr.3d 379, 410 P.3d 9.) The reasons for this lie at the heart of our democratic system of governance. "Coupled with a rebuttable presumption that legislation is constitutional, [rational basis review] helps ensure that democratically enacted laws are not invalidated merely based on a court’s cursory conclusion that a statute’s tradeoffs seem unwise or unfair." (Ibid.)

[10–14] Under this deferential standard, we presume that a given statutory classification is valid "until the challenger shows that no rational basis for the unequal treatment is reasonably conceivable." (Chatman, supra, 4 Cal.5th at p. 289, 228 Cal.Rptr.3d 379, 410 P.3d 9.) The underlying rationale for a statutory classification need not have been "ever actually articulated" by lawmakers, nor "be empirically substantiated." (People v. Turnage, supra, 55 Cal.4th at pp. 74, 75, 144 Cal. Rptr.3d 489, 281 P.3d 464 (Turnage).) Evaluating potential justifications for disparate treatment, a court reviewing a statute under this standard must "treat the statute’s potential logic and assumptions far more permissively than with other standards of constitutional or regulatory review." (Chatman, at p. 294, 228 Cal. Rptr.3d 379, 410 P.3d 9.) "If a plausible basis exists for the disparity, courts may not second-guess its ‘ "wisdom, fairness, or logic." ’ " (Johnson, supra, 60 Cal.4th at p. 881, 183 Cal.Rptr.3d 96, 341 P.3d 1075.) "[T]he logic behind a potential justification need [not] be persuasive or sensible — rather than simply rational." (Chatman, at p. 289, 228 Cal.Rptr.3d 379, 410 P.3d 9.)^{3 3}

3 The high court has on occasion applied a more searching form of rational basis review that looks to the Legislature’s actual motivations in enacting a statute rather than hypothesized ones. (See, e.g., U.S. Dept. of Agriculture v. Moreno (1973) 413 U.S. 528, 535–538, 93 S.Ct. 2821, 37 L.Ed.2d 782.) The high court has generally reserved this form of review for cases in which the sole motivation underlying the enactment is baseless prejudice against a politically unpopular group. (See, e.g., ibid.; Cleburne, supra, 473 U.S. at pp. 448–450, 105 S.Ct. 3249.) Those are not the circumstances we confront here, and no party argues otherwise.
Justice Liu lays out an argument for reconsidering rational basis review under our state equal protection guarantee to require a focus on the Legislature’s actual, rather than hypothesized, reasons for the challenged classification. (Dis. opn. of Liu, J., post, 318 Cal. Rptr.3d at p. 551, 543 P.3d at p. 992.) We note, however, that our analysis focuses on the apparent motivations underlying the challenged classification, as revealed in the statutory text and history; we do not endeavor to exhaustively catalog all conceivable concerns that might be hypothesized in support of the challenged distinction.

3 AU Internet citations in this opinion are archived by year, docket number, and case name at http://www.courts.ca.gov/38324.htm>.

B.

Hardin’s central argument is that section 3051’s exclusion of offenders sentenced to life without possibility of parole has no rational basis because it is inconsistent with what he understands to be the "sole" purpose behind the statute: to create "a meaningful opportunity for release for youthful offenders, who were 25 or younger at the time of their crimes, through demonstrated growth and rehabil- itation." Pointing to the high court’s reasoning concerning juvenile offenders in Miller, supra, 567 U.S. 460, 132 S.Ct. 2455, and the scientific research that prompted the Legislature to expand section 3051 to young adults, Hardin contends that all youthful offenders, by virtue of their age and the limitations associated with still-developing judgment and impulse control, possess the same characteristics that prompted the enactment and expansion of section 3051, including diminished culpability and the potential for change.

[15, 16] Hardin acknowledges the core of the counterargument. "It is both the prerogative and the duty of the Legislature to define degrees of culpability and punishment, and to distinguish between crimes in this regard." (Turnage, supra, 55 Cal.4th at p. 74, 144 Cal.Rptr.3d 489, 281 P.3d 464.) Life without parole is the most severe sentence of imprisonment in California law, applicable only in cases of special circumstance murder and a small number of other offenses the law regards as particularly serious.^{4 4} By excluding persons sentenced to life without parole from youth offender parole proceedings, the Legislature exercised its prerogative to define degrees of culpability and punishment by leaving in place longstanding judgments about the seriousness of these crimes and, relatedly, the punishment for them.

4 These offenses include certain aggravated sex offenses against minors (Pen. Code, § 667.61, subds. (j)(1), (l)); kidnapping for ransom resulting in death or bodily harm or exposure to a substantial likelihood of death (id., § 209, subd. (a)); certain felonies inflicting great bodily injury that are committed by a "habitual offender" (id., § 667.7, subd. (a)); hate crime first degree murder (id., § 190.03, subd. (a)); willful and malicious ignition of an explosive device causing death (id., § 18755, subd. (a)); and intentional train wrecking (id., §§218, 219).

4 Recently, the Massachusetts Supreme Court relied on similar brain science pertaining to 18 to 20 year olds in holding LWOP sentences for that age group violated the Massachusetts State Constitution’s prohibition against cruel and unusual punishment. (Commonwealth v. Mattis (2024) 493 Mass. 216, 224 N.E.3d 410.)

Hardin asserts, however, that the seriousness of the offenses "provides no basis for their exclusion because the purpose of the statute was ameliorative, not punitive." The Court of Appeal made a similar point: "[I]f, as the Legislature stated, the goal of section 3051 was to apply the Miller youth-related mitigating factors to young adults up to the age of 26 in light of neuroscience research that demonstrated the human brain continues to develop into a person’s mid-20’s, and thus to permit youth offenders a meaningful opportunity for parole if they demonstrate increased maturity and impulse control, then for that purpose there is no plausible basis for distinguishing between same-age offenders based solely on the crime they committed." (Hardin, supra, 84 Cal.App.5th at p. 288, 300 Cal.Rptr.3d 243.)

This argument rests on the premise that "there was only a single purpose underlying" section 3051. (Hernandez, supra, 41 Cal.4th at p. 300, 59 Cal.Rptr.3d 442, 159 P.3d 33.) But as we explained in Hernandez, legislation does not always — or even often — work this way. Legislation is frequently the "‘product of multiple and somewhat inconsistent purposes that led to certain compromises.’ " (Id. at p. 301, 59 Cal.Rptr.3d 442, 159 P.3d 33, quoting U. S. Railroad Retirement Rd. v. Fritz (1980) 449 U.S. 166, 181, 101 S.Ct. 453, 66 L.Ed.2d 368 (conc. opn. of Stevens, J.).) This is only to be expected, for "[d]eciding what competing values will or will not be sacrificed to the achievement of a particular objective is the very essence of legislative choice." (Rodriguez v. United States (1987) 480 U.S. 522, 526, 107 S.Ct. 1391, 94 L.Ed.2d 533.) "Past cases establish that the equal protection clause does not preclude a … legislative measure that is aimed at achieving multiple objectives, even when such objectives in some respects may be in tension or conflict." (Her- nandez, at p. 300, 59 Cal.Rptr.3d 442, 159 P.3d 33.)

[17] Section 3051 is such a measure. No one doubts that the Legislature’s primary purpose in expanding section 3051 to include young adult offenders was to give these young persons the opportunity to obtain release based on demonstrated growth and rehabilitation. Even though the Eighth Amendment requires that this opportunity be afforded only to persons who committed their crimes as juveniles, the Legislature determined that comparable opportunities should be available to some older offenders as well. But the structure and history of the expansion make clear that the Legislature sought to balance this primary objective with other, sometimes competing, concerns, including concerns about culpability and the appropriate level of punishment for certain very serious crimes.

This balancing has been evident throughout the history of the youth offender parole statute. Even as initially drafted, the statute did not categorically extend youth offender parole hearings to all persons below the age of 18, but instead distinguished between offenders based on the crimes they committed. (Stats, 2013, ch. 312, § 1; former § 3051, added by Stats. 2013, ch. 312, § 4; cf. Sen. Com. on Appropriations, Analysis of Sen. Bill No. 394 (2017–2018 Reg. Sess.) Apr. 17, 2017, p. 2 ["[The bill that created section 3051] established a parole process for persons sentenced to prison for certain crimes committed before attaining 18 years of age" (italics added)].) Through multiple rounds of statutory amendments gradually expanding the statute, the Legislature retained crime-based distinctions, and the legislative history accompanying the amendments confirms that these were deliberate choices. (See, e.g., Assem. Com. on Appropriations, Analysis of Assem. Bill No. 1308 (2017–2018 Reg. Sess.) as amended Mar. 30, 2017, p. 2 ["Some offenders are not eligible [for parole hearings] based on the crime that was committed, or actions taken by the inmate after the age of 23" (italics added)]; Sen. Com. on Public Safety, Rep. on Sen. Bill No. 394 (2017–2018 Reg. Sess.) Mar. 21, 2017, p. 4 ["This bill would apply the youth offender parole process to juveniles sentenced to [life without parole]. … [¶] The bill makes clear that … the provisions applying to juvenile [life without parole] apply only to those sentenced before the age of 18" and thus exclude individuals sentenced to life without parole for crimes committed after the age of 18].)

The end result is that under the youth offender parole statute as enacted and since amended, the nature of the sentence received for a particular crime — what the statute terms the "controlling offense" — sometimes determines whether an individual is eligible for a youth offender parole hearing in the first instance. And for those who are eligible, the nature of the sentence determines when they will receive such a hearing: whether after 15, 20, or 25 years. In other words, in designing section 3051, the Legislature consciously drew lines that altered the parole component of offenders’ sentences based not only on the age of the offender (and thus the offender’s amenability to rehabilitation) but also on the offense and sentence imposed. The lines the Legislature drew necessarily reflect a set of legislative judgments about the nature of punishment that is appropriate for the crime.

It may be true, as Hardin argues, that these crime-based categories are not rationally related to the Legislature’s purpose of expanding opportunities for early release based on the attributes of youth since, as Miller explained, the attributes of youth are not "crime-specific." (Miller, su- pra, 567 U.S. at p. 473, 132 S.Ct. 2455.) No doubt the Legislature — which consciously enacted section 3051 in language that borrowed from Miller and other Eighth Amendment juvenile sentencing cases — was aware of this point. The Legislature nonetheless crafted a statutory scheme that assigns significance to the nature of underlying offenses and accompanying sentences. The most natural conclusion to draw from this is not, as Hardin would have it, that the Legislature enacted a statute at odds with its own rehabilitative ends, but instead that the Legislature — as legislatures often do — was attempting to pursue other " ‘(perhaps even contrary) ends as well.’ " (Hernandez, supra, 41 Cal.4th at p. 301, 59 Cal.Rptr.3d 442, 159 P.3d 33, quoting Fitzgerald v. Racing Assn. of Central Iowa (2003) 539 U.S. 103, 108, 123 S.Ct. 2156, 156 L.Ed.2d 97.)

The statutory framework indicates that the Legislature aimed to increase opportunities for meaningful release for young adult offenders, while taking into account the appropriate punishment for the underlying crimes, depending on their severity. These are essentially the same considerations involved whenever the Legislature exercises its responsibility "for determining which class of crimes deserves certain punishments and which crimes should be distinguished from others." (Wilkinson, supra, 33 Cal.4th at p. 840, 16 Cal.Rptr.3d 420, 94 P.3d 551.) They are also not dissimilar from the considerations that prompted the high court to distinguish, for Eighth Amendment purposes, between sentencing juveniles for homicide offenses and sentencing juveniles for nonhomicide offenses. (Miller, supra, 567 U.S. at p. 473, 132 S.Ct. 2455 [based on considerations of "both moral culpability and consequential harm," juvenile homicide offenders, unlike juvenile nonhomicide offenders, may be sentenced to life without possibility of parole, but only after individualized sentencing that gives appropriate consideration to the mitigating attributes of youth].) Much as the high court invoked culpability-related concerns to distinguish among crimes in that context, it is reasonable to infer that the Legislature considered such concerns in this one.

Hardin argues that the Legislature’s decision to adopt a parole process indicates it was unconcerned with culpability and instead had only rehabilitation in mind. If the Legislature had been concerned with calibrating the appropriate sentence for particular crimes, Hardin reasons, the Legislature could have instead enacted a statute providing for the recall of sentence and resentencing, as it had done in Penal Code section 1170, subdivision (d) — the predecessor statute to section 3051 discussed in Kirchner, supra, 2 Cal.5th at pages 1049–1050, 216 Cal.Rptr.3d 876, 393 P.3d 364 — and as it has done in other recently enacted ameliorative statutes (e.g., Pen. Code, § 1172.6; see People v. Lewis (2021) 11 Cal.5th 952, 959–960, 281 Cal.Rptr.3d 521, 491 P.3d 309). In Hardin’s view it is "telling" that the Legislature instead enacted a parole process, since "California’s parole process explicitly measures rehabilitation. … To the extent the crime of commitment can be taken into consideration at all, it is only for purposes of determining the present level of risk."

What Hardin says is true of the task of the Parole Board at a parole hearing. (§ 3051, subd. (d), citing Pen. Code, § 3041; see Pen. Code, § 3041, subd. (b)(1) ["The panel … shall grant parole … unless it determines that the gravity of … current or past convicted offense or offenses … is such that consideration of the public safety requires a more lengthy period of incarceration"].) But the Legislature has a different role, which is to determine not only whether an incarcerated individual may be suitable for release on parole, but when and whether it is appropriate to afford that individual the opportunity to demonstrate suitability for release. Parole eligibility is frequently an important component of the sentence prescribed for a crime, and so the Legislature frequently considers multiple sentencing objectives — including both the prospects for rehabilitation and the degree of culpability demonstrated by the crime — in determining when, and if, a particular category of offenders will become eligible for a parole hearing.

Hardin also argues that section 3051’s focus on the "controlling offense" — that is, the single "offense or enhancement for which any sentencing court imposed the longest term of imprisonment" (§ 3051, subd. (a)(2)(B)) — is indicative of the Legislature’s rehabilitative concerns rather than concerns with appropriate punishment. Hardin points out, for example, that the statute sets a 25-year eligibility date for all youthful offenders who have received a sentence of 25 years to life for any one offense or enhancement — even if another individual with the same youth offender parole eligibility date may be serving a much longer aggregate sentence on account of other crimes; and even if the offender’s "controlling offense" is merely an enhancement, rather than a substantive crime. Hardin argues that the Legislature that enacted section 3051 therefore must not have been concerned with the relatively greater culpability of the individual with the longer aggregate sentence, or of the substantive crime to which the longer enhancement was attached.

To be sure, the statute’s "controlling offense" framework does rely on a certain amount of generalization about the relationship between the lengthiest individual sentence the offender has received and the culpability of the underlying criminal conduct. But " ‘[w]hen conducting rational basis review, we must accept any gross generalizations and rough accommodations that the Legislature seems to have made.’ " (Turnage, supra, 55 Cal.4th at p. 77, 144 Cal.Rptr.3d 489, 281 P.3d 464.) Hardin’s argument presumes there is only one way to evaluate culpability for these purposes — by focusing on the offender’s entire criminal history rather than examining an individual offense, or by focusing on substantive crimes and ignoring the role of sentence enhancements. But these are not the only possible ways to evaluate culpability. That the Legislature may have prescribed a measurement of culpability different from Hardin’s does not mean the Legislature was not attempting to measure culpability at all. While section 3051 is not, in terms, a statute prescribing sentences for particular crimes, it does "set[ ] the consequences of criminal offenses." (Johnson, supra, 60 Cal.4th at p. 887, 183 Cal. Rptr.3d 96, 341 P.3d 1075.) It is reasonable to infer that in setting those consequences through operation of the youth offender parole system, the Legislature balanced multiple considerations, including both concerns about increasing opportunities for release for young adults able to show growth and maturity and concerns about calibrating the level of punishment appropriate for certain serious criminal offenses.

Hardin also suggests that, by enacting a system of single-offense-based staggered eligibility terms and exclusions, the Legislature was attempting to capture the moment when, based on the sentence received for a single offense or enhancement, "a person might be first expected to demonstrate meaningful rehabilitation." Hardin provides no logical or evidentiary support for this view. It is unclear how the Legislature could have determined that 15 years marks the relevant line of maturation for an offender who received a determinate sentence for a controlling offense; 20 years marks the maturation line for an offender sentenced to a life term of less than 25 years to life; and so on. But more fundamentally, this is not an either/or matter. Parole eligibility dates are an important component of the sentences prescribed for crimes. As such, they presumptively reflect the full range of usual penological considerations, including rehabilitative and retributive purposes. Even assuming the staggered parole eligibility terms reflect some set of legislative judgments about when an offender is most likely to be rehabilitated, the critical point is that they also necessarily reflect a judgment about the degree to which the youth offender parole statute should reduce potential punishment. Concerns about both appropriate punishment and rehabilitation underlie this provision, just as the same balance of penological considerations underlie the other provisions of the statute. These are unquestionably legitimate purposes. (E.g., Wilkinson, supra, 33 Cal.4th at p. 840, 16 Cal.Rptr.3d 420, 94 P.3d 551.) The exclusion that Hardin challenges may or may not be rationally related to those purposes — we will turn to that question below — but the exclusion is not invalid simply because it reflects interests on the other side of a legislative balance.

Finally, Hardin argues that the other exclusions from youth offender parole eligibility set forth in section 3051, subdivision (h) "further undermine the rationality of the statute." The only question before us here concerns the constitutionality of the exclusion of youthful offenders sentenced to life without parole, and there is no occasion for us to pass judgment on the validity of any other exclusion. It suffices to observe, however, that nothing in the other exclusions undermines the conclusion that the Legislature that crafted the youth offender parole statute was attempting to balance multiple penological considerations in addition to rehabilitation. Whether or not each of the other exclusions is adequately justified in light of those considerations is beyond the scope of our inquiry in this case.^{5 5}

5 We do not, for instance, decide the issue presented in People v. Williams (2020) 47 Cal.App.5th 475, 261 Cal.Rptr.3d 70, review granted July 22, 2020, S262229, in which the Court of Appeal held that section 3051, subdivision (h) violates equal protection principles by excluding youthful offenders convicted and sentenced for aggravated sex crimes under the One Strike law (Pen. Code § 667.61) from youth offender parole consideration.

5 The majority declines to engage with this uncontroverted research on the grounds that it was not subject to adversarial testing in the trial court (see maj. opn., ante, 318 Cal. Rptr.3d at p. 536, 543 P.3d at p. 979) and that, even if taken at face value, the study does not "say[] nor suggest[] that California’s special circumstance law is categorically invalid." (Maj. opn., ante, 318 Cal.Rptr.3d at p. 536, 543 P.3d at p. 979.) But, contrary to the majority’s suggestion, our hands are not tied. We can do as we have previously done and remand the matter to the trial court for further factual development (see dis. opn. of Liu, J., ante, 318 Cal.Rptr.3d at pp. 561–563, 543 P.3d at pp. 1001–1002) and then decide whether there is a rational basis for the differential treatment of youthful offenders.

C.

[18] Hardin argues that even if the life without parole exclusion reflects culpability-related concerns, it nonetheless fails rational basis review because there is no reasonable basis to conclude that young adult offenders sentenced to life without parole are more culpable or less deserving of the opportunity for release than other young adult offenders. Hardin’s arguments focus specifically on individuals who, like him, received life without parole sentences following convictions for special circumstance murder. The Legislature, he argues, "would have had no rational basis to distinguish between youthful offenders sentenced to life without parole for special circumstance murder and youthful offenders sentenced either to the functional equivalent of life without parole or to indeterminate life terms for first degree murder. That is because, from a culpability standpoint, these groups cannot rationally be distinguished." Hardin, however, fails to demonstrate that the life without parole exclusion is irrational, and therefore unconstitutional, as applied to individuals sentenced for special circumstance murder.

In California, a conviction for first degree murder generally results in a life sentence with parole eligibility after 25 years. (See Pen. Code, § 190, subd. (a); id., §§ 190.1-190.5.) Penal Code section 190.2 (section 190.2) lists special circumstances that, under California law, mark a first degree murder particularly egregious and thus render the perpetrator eligible for the death penalty, consistent with Eighth Amendment requirements. (People v. Bacigalupo (1993) 6 Cal.4th 457, 467-468, 24 Cal.Rptr.2d 808, 862 P.2d 808 [the "special circumstances" statute performs the constitutionally required function of " ‘narrowing’ " the "class of murderers eligible for the death penalty"].) If a defendant is convicted of first degree murder with a special circumstance under section 190.2, there are only two possible sentences: death or life without the possibility of parole. (Id., subd. (a).)

To understand the function of special circumstances in California’s capital sentencing law is to understand why Hardin faces a particularly difficult task in establishing that the Legislature’s decision to exclude offenders convicted of special circumstance murder from the youth offender parole system is "so devoid of even minimal rationality that it is unconstitutional as a matter of equal protection." (Chatman, supra, 4 Cal.5th at p. 289, 228 Cal.Rptr.3d 379, 410 P.3d 9.) The core of Hardin’s argument is that the Legislature could not rationally conclude that a conviction for special circumstance murder is a reliable indication of the seriousness of an offense or the culpability of the offender, such that it could rationally decide to exclude the offender from receiving the youth offender parole consideration to which other young adults are statutorily entitled. In making this argument, Hardin does not focus on any single special circumstance or any particular factual scenarios; his argument is a categorical one, aimed at special circumstance murder in general. This argument about the relative insignificance of special circumstance murder, as a category, is inconsistent with what are by now legions of decisions holding that special circumstance murder is sufficiently serious and morally culpable as to justify imposing the most severe sanctions available under the law, up to and including death.

In the Eighth Amendment context, this court has consistently rejected arguments that section 190.2’s potential coverage is too broad to perforai its constitutionally required function of identifying those convicted of murders whose crimes are sufficiently egregious to warrant the law’s most severe penalty.^{6 6} We have explained why various challenged provisions of section 190.2 adequately separate the most egregious first degree murders — those deserving of the most severe punishment available — from the rest. (See, e.g., People v. Anderson (1987) 43 Cal.3d 1104, 1147, 240 Cal.Rptr. 585, 742 P.2d 1306 ["[B]y making the felony murderer but not the simple murderer death-eligible, a death penalty law furnishes the ‘meaningful basis [required by the Eighth Amendment] for distinguishing the few cases in which [the death penalty] is imposed from the many cases in which it is not’ "].)

6 See, e.g., People v. Wilson (2023) 14 Cal.5th 839, 865-866, 309 Cal.Rptr.3d 211, 530 P.3d 323; People v. Thomas (2023) 14 Cal.5th 327, 408, 304 Cal.Rptr.3d 1, 523 P.3d 323; People v. Ramirez (2022) 13 Cal.5th 997, 1160, 298 Cal.Rptr.3d 1, 515 P.3d 1085; People v. Parker (2022) 13 Cal.5th I, 89, 293 Cal.Rptr.3d 813, 510 P.3d 404; People v. Wright (2021) 12 Cal.5th 419, 455–456, 287 Cal.Rptr.3d 510, 500 P.3d 267; People v. Scully (2021) 11 Cal.5th 542, 610, 278 Cal.Rptr.3d 792, 486 P.3d 1029; People v. Schultz (2020) 10 Cal.5th 623, 682, 271 Cal.Rptr.3d 387, 475 P.3d 1073; People v. Frederickson (2020) 8 Cal.5th 963, 1026, 258 Cal.Rptr.3d 114, 457 P.3d 1; People v. Capers (2019) 7 Cal.5th 989, 1012–1013, 251 Cal.Rptr.3d 80, 446 P.3d 726; People v. Brooks (2017) 3 Cal.5th 1, 114–115, 219 Cal. Rptr.3d 331, 396 P.3d 480; People v. Johnson (2016) 62 Cal.4th 600, 654-655, 197 Cal. Rptr.3d 461, 364 P.3d 359.

6 It would be useful to compare racial disparities among youthful offenders sentenced to LWOP with youthful offenders sentenced to an indeterminate life term for first degree murder. The parties, however, have not provided any such data.

[19, 20] Here, for example, Hardin was convicted of murdering his victim in the course of robbing her. We have explained why the law treats robbery-murder as more culpable than simple murder. The special circumstance is limited to those defendants who commit "a ‘willful, deliberate and premeditated’ murder ‘during the commission’ of a robbery or other listed felony" rather than "when the defendant’s intent is not to steal but to kill and the robbery is merely incidental to the murder." (People v. Green (1980) 27 Cal.3d 1, 61, 164 Cal.Rptr. 1, 609 P.2d 468.) The law treats as particularly egregious a murder "in cold blood in order to advance an independent felonious purpose, e.g., who carried out an execution-style slaying of the victim of or witness to a holdup, a kidnaping, or a rape." (Ibid.) "[T]he purpose of this special circumstance is to make eligible for the most severe punishment those defendants who escalate a serious felony into a murder, thereby attempting to deter such escalation." (People v. Mora and Rangel (2018) 5 Cal.5th 442, 520, 235 Cal. Rptr.3d 92, 420 P.3d 902 (cone. & dis. opn. of Liu, J.).)

Given this body of case law, it is difficult to see how the Legislature that enacted section 3051 could have acted irrationally in singling out special circumstance murder as a particularly culpable offense. In concluding otherwise, the Court of Appeal in this case pointed to a law review article’s finding that, because of the expansion of the special circumstances over the years, at least one special circumstance could be alleged in many if not most first degree murder cases, "leaving the decision whether a life without parole sentence may be imposed to the discretion of local prosecutors, rather than a matter of statewide policy." (Hardin, supra, 84 Cal.App.5th at p. 290, 300 Cal.Rptr.3d 243; id. at p. 290, fn. 11, 300 Cal.Rptr.3d 243, citing Com. on Revision of the Pen. Code, supra, Annual Report and Recommendations, p. 51, in turn citing Baldus et al., Furman at. 45; Constitutional Challenges from California’s Failure to (Again) Narrow Death Eligibility, supra., 16 J. Empirical Legal Studies at pp. 713–714 (Baldus study).) Hardin now invokes the same article in support of his challenge to section 3051’s disparate treatment of individuals sentenced for special circumstance murder.

Hardin’s argument is not that prosecutorial discretion itself offends equal protection. (See People v. Keenan (1988) 46 Cal.3d 478, 505, 506, 250 Cal.Rptr. 550, 758 P.2d 1081 [rejecting the argument that "prosecutorial discretion to select those eligible cases in which the death penalty will actually be sought" in and of itself "offend[s] principles of equal protection," and explaining that "[m]any circumstances may affect the litigation of a case chargeable under the death penalty law. These include factual nuances, strength of evidence, and, in particular, the broad discretion to showleniency"]; see also United. States v. Batchelder (1979) 442 U.S. 114, 125, 99 S.Ct. 2198, 60 L.Ed.2d 755 [rejecting an equal protection challenge to "the discretion a prosecutor exercises when deciding whether to charge under one of two statutes"].) Nor does he bring or develop a claim that prosecutorial discretion has been exercised in an arbitrary or invidious manner. Rather, in light of the findings of the cited law review article, Hardin "challenges the Legislature’s ability to rely on a distinction between two groups — youthful offenders convicted of special circumstance murders and youthful offenders convicted of first degree murders — that collapses on further scrutiny."

We have previously considered a similar argument raised in the Eighth Amend- ment context. In People v. Frye (1998) 18 Cal.4th 894, 1028-1029, 77 Cal.Rptr.2d 25, 959 P.2d 183 (Frye), a capital defendant relied on "a statistical analysis based on an examination of published appeals from murder convictions for the years 1988-1992" that showed "virtually all first degree murders are death eligible." The defendant in that case attributed this result to "the broad interpretation of the lying-in-wait special circumstance and the expansive sweep of the felony-murder special circumstance." (Id. at p. 1029, 77 Cal. Rptr.2d 25, 959 P.2d 183.) We rejected the argument, citing case law upholding the validity of both the lying-in-wait special circumstance and the felony-murder special circumstance in cases in which the defendant did not harbor an intent to kill but was instead a major participant in a felony who acted with reckless indifference to human life. (Ibid., citing, inter alia, People v. Morales (1989) 48 Cal.3d 527, 557-558, 257 Cal.Rptr. 64, 770 P.2d 244 & People v. Marshall (1990) 50 Cal. 3d 907, 946, 269 Cal.Rptr. 269, 790 P.2d 676.)

Our treatment of the issue in Frye was admittedly terse, and it relied on a different study than the one on which Hardin now relies. But based on the arguments and evidence that have been presented to us here, we have no adequate basis to fault the Legislature for distinguishing, as a categorical matter, between a conviction for special circumstance murder and a conviction for a different homicide offense, as the law has long done.

At the outset, we note that the Baldus study on which Hardin relies is not part of the record in this case, having been first raised not by the parties but by the Court of Appeal in its opinion. (See Hardin, supra, 84 Cal.App.5th at p. 290, 300 Cal. Rptr.3d 243.) The study’s findings were not litigated in the trial court, so they have never been the subject of any sort of adversarial testing that would afford us insight into either the methodology employed or the ultimate accuracy or significance of the results. To strike down an act of the Legislature as irrational based on a set of untested empirical findings would be antithetical to multiple settled principles of judicial review.

Even if we were to take the study’s findings at face value, however, they do not support Hardin’s claim that it is, as a categorical matter, irrational to treat individuals convicted of first degree special circumstance murder differently from individuals convicted of first degree murder without special circumstances. The study neither says nor suggests that California’s special circumstance law is categorically invalid. Rather, as the Court of Appeal noted in its opinion, the study appears to suggest that certain special circumstances, added through various amendments after the initial enactment of section 190.2, have led to the results found in the study.^{7 7} (See Hardin, supra, 84 Cal.App.5th at p. 290, 300 Cal.Rptr.3d 243.) But Hardin makes no challenge specific to any particular special circumstance or special circumstances added or changed by postenactment amendments. (The special circumstance finding at issue in Hardin’s own case is based on a provision of the law that dates back to the initial enactment of section 190.2. (See Stats. 1973, ch. 719, § 5, pp. 1299-1300.)) While we do not foreclose the possibility of other challenges to the distinctions drawn by the special circumstances statute based on a more robust record or a more focused as-applied inquiry, Hardin has not carried his burden to demonstrate that legislative reliance on the special circumstance murder statute in section 3051, subdivision (h) is categorically irrational.

7 As Justice Liu notes, the Baldus study also reports that robbery-murder is factually present in a majority of special circumstance murder cases. (Dis. opn. of Liu, J., post, 318 Cal.Rptr.3d at pp. 559-560, 543 P.3d at p. 999.) But standing alone, that finding has no clear relevance; a special circumstance is not legally invalid simply because it may be the most frequently recurring form of special circumstance murder.
Justice Liu also invokes a different study, cited in the Baldus study but not raised by either party to this case, in support of the view that " 'the felony murder special circumstances alone defeat any possibility of genuine narrowing.’ " (Dis. opn. of Liu, J., post, 318 Cal.Rptr.3d at p. 560, 543 P.3d at p. 999.) Particularly without any adversarial testing or argument concerning the relationship between this limited set of empirical findings and the Eighth Amendment's narrowing requirement, we have no adequate basis for drawing this sweeping conclusion, which would call into question a substantial body of precedent of both this court and of the United States Supreme Court. (See Pulley v. Harris (1984) 465 U.S. 37, 51, fn. 13, 53, 104 S.Ct. 871, 79 L.Ed.2d 29 [upholding the 1978 version of the special circumstance murder statute]; Frye, supra, 18 Cal.4th at pp. 1028–1029, 77 Cal.Rptr.2d 25, 959 P.2d 183.)

7 Notably, the Human Rights Watch submitted its report to the Legislature in support of Semite Bill No. 394 (2017–2018 Reg. Sess.), which expanded the youth offender parole eligibility scheme to include youthful offenders who were younger than 18 at the time of the offense and were sentenced to LWOP.

Hardin next argues that an individual who commits special circumstance murder may not actually be more culpable than an offender who commits a string of other violent crimes. Agreeing with Hardin, the Court of Appeal raised for comparison two hypothetical offenders who would be eligible for a section 3051 parole hearing: (1) "a 20 year old who shot and killed his victim one day, committed a robbery the next, and was sentenced to an indeterminate term of 50 years to life"; and (2) an individual "who committed multiple violent crimes … and received a parole-eligible indeterminate life term that far exceeded his or her life expectancy." (Hardin, supra, 84 Cal.App.5th at p. 289, 300 Cal. Rptr.3d 243.) In the court’s view, these crimes "cannot rationally" be considered less severe than "[t]he crime of a 20-year-old offender who shot and killed his victim while attempting to commit robbery and was sentenced to life without parole." (Ibid.) Yet section 3051 would deny a parole hearing to that offender. The court concluded that "[b]y defining the youth parole eligible date in terms of a single ‘controlling offense,’ rather than by the offender’s aggregate sentence, the Legislature has eschewed any attempt to assess the offenders’ [sic] overall culpability, let alone his or her amenability to growth." (Hardin, at p. 289, 300 Cal.Rptr.3d 243.)

That view again rests on the assumption that the Legislature is required to evaluate culpability in a particular way — a way that would, essentially, regard special circumstance murder as similar in culpability to a string of other violent crimes that leads to technically parole-eligible sentences. But the Legislature that enacted section 3051 was not obligated to see things this way. Indeed, the law in general does not see things this way: In the criminal law, there is no violent crime or set of violent crimes considered more serious, or that trigger more severe punishment, than special circumstance murder. We thus cannot say that the decision to deny a parole hearing to an offender convicted of special circumstance murder is irrational, even if it is possible that in certain cases some might consider an individual offender convicted of multiple violent crimes more culpable, in a holistic sense, than an individual convicted of special circumstance murder. (Turnage, supra, 55 Cal.4th at pp. 77-78, 144 Cal.Rptr.3d 489, 281 P.3d 464 ["When conducting rational basis review … [a] plausible reason for distinguishing between [two groups of individuals] need not exist in every scenario in which the statutes might apply"].)

Hardin notes that we have described an aggregate sentence that fixes parole eligibility outside of an offender’s life expectan- cy as the "functional equivalent of a life without parole sentence." (Caballero, supra, 55 Cal.4th at p. 268, 145 CaI.Rptr.3d 286, 282 P.3d 291.) But we have employed that description in the context of identifying the category of juvenile offenders to whom the Eighth Amendment limitations on life without parole sentences apply; for that purpose, what matters is only whether the sentence, by its nature, forecloses any realistic chance for a juvenile offender to rejoin society. (See People v. Contreras (2018) 4 Cal.5th 349, 368, 229 Cal.Rptr.3d 249, 411 P.3d 445.) We have not held that a lengthy term-of-years sentence is necessarily equivalent to a life without parole sentence for all purposes. Nor, more specifically, have we suggested that a set of crimes punishable by a lengthy term-of-years sentence is necessarily more culpable, or equivalent in culpability, to a single crime for which the law prescribes a sentence of life without parole. It was not irrational for the Legislature to exclude from youth offender parole eligibility those young adults who have committed special circumstance murder, an offense deemed sufficiently culpable that it merits society’s most stringent sanctions.

V.

In holding that Hardin has not demonstrated that the exclusion of offenders who are serving sentences of life in prison without the possibility of parole for a crime committed after the age of 18 from youth offender parole eligibility is irrational, we pass no judgment on the validity of any of the other exclusions set forth in section 3051, subdivision (h). Nor do we resolve here the constitutionality of section 3051, subdivision (h) as it might arise in other as-applied challenges based on particular special circumstances or the factual circumstances of individual cases.

[21] We emphasize, finally, that the question before us concerns only the constitutional permissibility of the lines the Legislature has drawn. It is not for us to pass judgment on the wisdom or desirability of its policy choices. (Chatman, supra, 4 Cal.5th at p. 297, 228 Cal.Rptr.3d 379, 410 P.3d 9.) Recognizing this, every published Court of Appeal decision other than the decision in this case has upheld the life without parole exclusion against equal protection challenge. At the same time, several opinions have taken the additional step of calling on the Legislature to give further careful consideration to the issue. (See, e.g., In re Murray (2021) 68 Cal. App.5th 456, 464, 283 Cal.Rptr.3d 601; People v. Marales, supra, 67 Cal.App.5th at p. 349, 282 Cal.Rptr.3d 151; People v. Jackson, supra., 61 Cal.App.5th at p. 202, 279 Cal.Rptr.3d 396 (conc. stmt. of Liu, J.) review den. June 9, 2021, S267812; id. at pp. 201-202, 279 Cal.Rptr.3d 396 (conc. opn. of Dato, J.); People v. Acosta, supra, 60 Cal.App.5th at p. 781, 275 Cal.Rptr.3d 110; People v. Montelongo (2020) 55 Cal. App.5th 1016, 1041a, 269 Cal.Rptr.3d 883 (conc. stmt. of Liu, J.) review den. Jan. 27, 2021, S265597; id. at pp. 1035-1036, 269 Cal.Rptr.3d 883 (conc. opn. of Segal, J.); In re Jones (2019) 42 Cal.App.5th 477, 486–487, 255 Cal.Rptr.3d 571 (conc. opn. of Pollak, J.).)

That so many judges across the state have taken this step reflects the significance of this issue. Special circumstance murder is an unquestionably grave offense, one that exacts an unimaginable toll on the lives of victims and those the victims leave behind. But we also know that young people — even young people who have committed grave offenses — are capable of significant, sometimes transformative, change over the course of their lifetimes. To extinguish any hope of release, particularly for an individual just past the cusp of adulthood, is a form of retribution that exacts its own price — one borne not just by the individuals involved, but by their families, by their communities, and by society as a whole.^{8 8}

8 In addition to pointing to neuroscience research showing that all youthful offenders, irrespective of their offense, bear the mitigating attributes of adolescent cognitive development and are capable of reform, various amici curiae also caution against legislative reliance on the special circumstance law given the geographic, temporal, and racial disparities in its application. Justice Evans’s dissent, too, argues that the exclusion of offenders sentenced to life without parole perpetuates racial disparities, and that this bias "should inform this court’s mode of deference." (Dis. opn. of Evans, J., post, 318 Cal. Rptr.3d at p. 567, 543 P.3d at p. 1005.) Hardin himself, however, has never argued that heightened scrutiny should apply to the facially neutral section 3051, subdivision (h), nor has he brought a constitutional claim based on the unequal or invidious enforcement of the special circumstance law. We do not here address how claims concerning racial disparities might be raised or addressed in a different case, whether under the Equal Protection Clause or under the California Racial Justice Act of 2020 (Pen. Code, § 745).
Some amici curiae on the other side of the issue argue that, if we were to find an equal protection violation in section 3051, subdivision (h), the only possible remedy would be to deny youth offender parole hearings to all young adult offenders; we could not instead order that treatment be equalized by granting youth offender parole hearings to young adults convicted of special circumstance murder. These amici curiae argue that because the current version of section 190.2 was enacted by voter initiative (Prop. 7, as approved by voters, Gen. Elec. (Nov. 7, 1978) § 6), extending parole eligibility to youthful offenders sentenced to special circumstance murder would constitute an impermissible amendment by the Legislature. In response, Hardin contends that the penalty scheme set forth in section 190.2 was first enacted by the Legislature, so the Legislature remains free to amend the penalties available for special circumstance murder. We have no occasion to reach this issue, since Hardin has not established that section 3051’s exclusion of young adult offenders sentenced to life without parole violates equal protection.

8 During the era of the superpredator myth, LWOP sentences swelled in the United States — increasing by over 400 percent between 1992 and 2016. (Seeds, Life Sentences and Perpetual Confinement (2021) Annual Review of Criminology, at p. 288 https://www.annualreviews.org/doi/pdf/10.1146/annurev-criminol-061020-022154> [as of Mar. 4, 2024]; see Annual Report, supra, at p. 0.) "[T]he overwhelming majority of JLWOP sentences were imposed in the mid-1990s … pursuant to policies adopted at the height of fear over the myth of the superpredator. … A handful of jurisdictions [including California] are responsible for imposing two-thirds of all JLWOP sentences." (Mills et al., supra, 65 Am. U. L.Rev. at pp. 560, 563.) Notably, prior to this proliferation in the use of the sentence, LWOP operated as a sentence wherein a governor (rather than a parole board) was tasked with granting parole, whereas it is now treated as " 'the other death penalty’ " — a death-in-prison sentence. (Seeds, Life Sentences and Perpetual Confinement, supra, Annual Review of Criminology at p. 302.)

[22] We acknowledge our dissenting colleagues’ view that, in light of these overarching concerns, the Legislature should have made a different choice. But for us to hold that the Legislature was constitutionally compelled to do so would require us to set aside multiple settled rules of constitutional adjudication. As this court has repeatedly explained, the purpose of these rules is to ensure that courts act as courts, and allow for the development of policy through the democratic process without putting the Legislature to unwarranted all-or-nothing choices. "While this court will not condone unconstitutional variances in the statutory consequences of our criminal laws," rational basis review requires us to extend substantial respect to the Legislature’s judgments, for " ‘ " ‘[o]nly by faithful adherence to this guiding principle of judicial review of legislation is it possible to preserve to the legislative branch its rightful independence and its ability to function.’ " ’ " (Johnson, supra, 60 Cal.4th at p. 889, 183 Cal.Rptr.3d 96, 341 P.3d 1075.)

[23] Our legislative bodies may continue to consider the issue and how to balance concerns about the severity of certain crimes with the overarching concera that prompted enactment of the youth offender parole hearing system and its eventual expansion to young adult offenders — that is, the recognition of the potential of young persons for growth and change. We are, however, mindful that the issue in this case arises in the first instance because the Legislature chose to expand opportunities for early parole consideration to many categories of young adult offenders, even though it was under no constitutional compulsion to do so. We are also mindful that the legislative branch is entitled to proceed incrementally, so long as it proceeds ra- tionally, in "walking [the] tightrope" of the political process. (Kasler v. Lockyer (2000) 23 Cal.4th 472, 487, 97 Cal.Rptr.2d 334, 2 P.3d 581.) Our task is limited to determining whether Hardin has shown that the Legislature’s decision to expand youth offender parole hearings to most young adult offenders, while excluding Hardin and others similarly situated, violates equal protection under a rational basis standard. For reasons explained above, we cannot so conclude.

VI.

We reverse the judgment of the Court of Appeal.

We Concur:

GUERRERO, C. J.

CORRIGAN, J.

GROBAN, J.

JENKINS, J.

Dissenting Opinion by Justice Liu

In a series of statutes over the past decade, the Legislature has established a parole eligibility process that provides young people who have committed serious crimes "the opportunity to obtain release when he or she has shown that he or she has been rehabilitated and gained maturity." (Stats. 2013, ch. 312, § 1.) Although the initial version of the parole scheme applied to persons serving sentences for crimes committed before age 18, the Legislature soon expanded eligibility by increasing the age cutoff, first to 23 and then to 26. In these enactments, the Legislature repeatedly recognized that "youthfulness both lessens a juvenile’s moral culpability and enhances the prospect that, as a youth matures into an adult and neurological development occurs, these individuals can become contributing members of society" (Stats. 2013, ch. 312, § 1), and that brain development affecting judgment and decisionmaking "continues beyond adolescence and into the mid-20’s" (maj. opn., ante, 318 Cal.Rptr.3d at p. 523, 543 P.3d at p. 968 [citing legislative history]).

Parole eligibility is now available to young adult offenders serving sentences for crimes committed before age 26, but with exceptions. (Pen. Code, § 3051; undesignated citations are to the Penal Code.) In 1989, at age 25, appellant Tony Hardin killed his elderly neighbor in the course of robbing her, and he was convicted of special-circumstance murder and sentenced to life imprisonment without the possibility of parole (LWOP). The parole eligibility scheme from its inception has excluded young offenders sentenced to LWOP. That exclusion has been lifted for juvenile offenders (§ 3051, subd. (b)(4)), but it still applies to individuals like Hardin who committed their crimes between the ages of 18 and 25 (id., subd. (h)), even though it is undisputed that the Legislature’s concerns about youth offenders’ diminished culpability and capacity for rehabilitation are not "crime-specific." (Miller v. Alabama (2012) 567 U.S. 460, 473, 132 S.Ct. 2455, 183 L.Ed.2d 407 (Miller); see maj. opn., ante, 318 Cal.Rptr.3d at pp. 530–531, 543 P.3d at pp. 974–975.) Hardin says this exclusion violates equal protection of the laws, and he is right.

Today’s opinion rationalizes the exclusion by imputing to the Legislature a purpose — calibrating "culpability and the appropriate level of punishment for certain very serious crimes" (maj. opn., ante, 318 Cal.Rptr.3d at p. 530, 543 P.3d at p. 974) — that is nowhere stated in the statute or its legislative history. It then posits that special-circumstance murder is generally distinguishable from simple first degree murder in terms of culpability (id. at pp. 533–538, 543 P.3d at pp. 977–981) de- spite strong evidence to the contrary. According to the court, nothing more is required under rational basis review.

Although I agree that rational basis review applies to Hardin’s claim, I disagree with how the court has applied it here. Today’s opinion ignores the considerable variation and nuance in our case law applying rational basis review and undertakes the sort of lax analysis that has become typical " ‘[i]n areas of social and economic policy.’ " (Warden v. State Bar (1999) 21 Cal.4th 628, 644, 88 Cal.Rptr.2d 283, 982 P.2d 154 (Warden).) But the issue in this case is a far cry from, say, whether the State Bar may exempt retired judges from continuing education requirements applicable to other licensed attorneys. (Id. at p. 633, 88 Cal.Rptr.2d 283, 982 P.2d 154.) Hardin, who is Black, is challenging a law that spells the difference between dying in prison and having a chance to earn freedom. The law targets a class of offenders who are overwhelmingly Black or Hispanic, and whose crimes — no less than the crimes of other youth offenders — reflect the "transient rashness, proclivity for risk, and inability to assess consequences" that are characteristic of young minds still undergoing neurological development. (Miller, supra, 567 U.S. at p. 472, 132 S.Ct. 2455.) In light of today’s decision, nearly 3,000 inmates continue to be denied any chance to demonstrate — as no doubt many could — that as mature adults they are more than the worst thing they ever did in their youth.

We have applied rational basis review more rigorously in cases with lower stakes. Rational basis review "require[s] the court to conduct ‘a serious and genuine judicial inquiry into the correspondence between the classification and the legislative goals.’ " (Newland v. Board of Governors (1977) 19 Cal.3d 705, 711, 139 Cal.Rptr. 620, 566 P.2d 254 (Newland).) Here, such inquiry reveals that the exclusion of young offenders convicted of special-circumstance murder is irrational when measured against the Legislature’s stated purpose for establishing and expanding youth offender parole eligibility. And even if we were to impute a purpose of excluding young offenders who have committed the most serious crimes, the exclusion of those convicted of special-circumstance murder does not withstand scrutiny. That is because, as the Court of Appeal found, they are not meaningfully distinguishable from young offenders convicted of simple first degree murder, a group that is parole eligible under the statute.

Today’s opinion concludes by echoing judges throughout the state who have urged the Legislature to reconsider the statute. (Maj. opn., ante, 318 Cal.Rptr.3d at pp. 538-539, 543 P.3d at p. 981.) One can hope the Legislature will take up the invitation, but that is no salve for what should have happened here. It is indeed imperative that "courts act as courts" (maj. opn., ante, 318 Cal.Rptr.3d at p. 539, 543 P.3d at p. 982), and in our system of government, courts are the ultimate guarantor of constitutional rights against arbitrariness or excesses of majoritarian rule. Although courts owe deference to the democratic process, deference is not abdication. Upon a serious and genuine judicial inquiry, it is evident that the exclusion of persons convicted of special-circumstance murder from youth offender parole eligibility does not meet the basic test of rationality. I respectfully dissent.

I.

While I agree that rational basis review is the appropriate equal protection standard in this case, today’s opinion largely ignores the way this standard has been articulated and applied in our case law. One feature that distinguishes our equal protection doctrine from its federal counterpart is that the standards of review under our doctrine are limited to two: rational basis review and strict scrutiny. (See In re Marriage Cases (2008) 43 Cal.4th 757, 832, 76 Cd.Rptr.3d 683, 183 P.3d 384.) Unlike the federal courts, we have declined to adopt intermediate scrutiny as a third standard of review. (See Hawkins v. Superior Court (1978) 22 Cal.3d 584, 595–603, 150 Cal.Rptr. 435, 586 P.2d 916 (conc. opn. of Mosk, J.); id. at pp. 607-610, 150 Cal. Rptr. 435, 586 P.2d 916 (conc. opn. of Bird, C. J.).) This means that rational basis review, in our doctrine, covers a wide range of eases and must be applied with nuance and sensitivity if we are to avoid the "rigidity of [a] two-tiered framework" that "applies either a standard that is virtually always met [rational basis] or one that is almost never satisfied [strict scrutiny]." (Id. at p. 598, 150 Cal.Rptr. 435, 586 P.2d 916 (conc. opn. of Mosk, J.).) In the pages that follow, I discuss the rational basis standard in depth. I regret the length of this discussion, but patient readers will understand why careful attention to our case law is essential to proper resolution of Hardin’s equal protection claim.

A.

In Brown v. Merlo (1973) 8 Cal.3d 855, 106 Cal.Rptr. 388, 506 P.2d 212 (Brown), we applied rational basis review and struck down an automobile guest statute that "deprive[d] an injured automobile guest of any recovery for the careless driving of his host unless the injury results from the driver’s willful misconduct or intoxication." (Id. at pp. 858-859, 106 Cal.Rptr. 388, 506 P.2d 212, citing Veh. Code, former § 17158.) Our opinion examined the two rationales traditionally offered for the statute — protecting hospitality and preventing collusive lawsuits — and rejected both with extensive analysis. As to protecting hospitality, the court found this rationale underinclusive in that it "provides no explanation for the statute’s differential treatment of automobile guests as distinguished from other guests, or indeed, all other recipients of hospitality." (Brown, at p. 864, 106 Cal.Rptr. 388, 506 P.2d 212.) Further, we said that any interest in protecting drivers from claims by "ungrateful" guests had been undermined by the advent of widespread liability insurance. (Id. at p. 868, 106 Cal.Rptr. 388, 506 P.2d 212; see id. at p. 869, 106 Cal.Rptr. 388, 506 P.2d 212 ["a classification which once was rational because of a given set of circumstances may lose its rationality if the relevant factual premise is totally altered"].) As to preventing collusive lawsuits, the court explained that "it is unreasonable to eliminate causes of action of an entire class of persons simply because some undefined portion of the designated class may file fraudulent lawsuits." (Id. at p. 875, 106 Cal.Rptr. 388, 506 P.2d 212.) "[B]y broadly prohibiting all automobile guests from instituting causes of action for negligence because a small segment of that class may file collusive suits, the guest statute presents a classic case of an impermissibly overinclusive classification scheme …. " (Id. at p. 876, 106 Cal.Rptr. 388, 506 P.2d 212.)

The court in Brown did not rationalize the statute’s underinclusivity by saying that a legislature "may take one step at a time, addressing itself to the phase of the problem which seems most acute to the legislative mind." (Williamson v. Lee, Optical of Oklahoma, Inc. (1955) 348 U.S. 483, 489, 75 S.Ct. 461, 99 L.Ed. 563 (Lee Optical).) Nor did Brown rationalize the statute’s overinclusivity by saying that a classification does not fail rational basis review "simply because [it] ‘is not made with mathematical nicety or because in practice it results in some inequality,’ " or that practical problems of government " ‘may justify, if they do not require, rough accommodations — illogical, it may be, and unscientific.’ " (Dandridge v. Williams (1970) 397 U.S. 471, 485, 90 S.Ct. 1153, 25 L.Ed.2d 491.) In fact, Brown began its discussion of the rational basis standard by observing that a classification " ‘must rest upon some ground of difference having a fair and substantial relation to the object of the legislation,’ " (Brown, supra, 8 Cal.3d at p. 861, 106 Cal.Rptr. 388, 506 P.2d 212, italics omitted, quoting Reed v. Reed (1971) 404 U.S. 71, 76, 92 S.Ct. 251, 30 L.Ed.2d 225 (Reed).) Although Reed presaged the development of intermediate scrutiny under federal law (see Craig v. Boren (1976) 429 U.S. 190, 197–199, 204, 97 S.Ct. 451, 50 L.Ed.2d 397; Frontiera v. Richardson (1973) 411 U.S. 677, 682–684, 690–691, 93 S.Ct. 1764, 36 L.Ed.2d 583 (plur. opn.)), Brown assimilated it into our explication of rational basis review.

The next year, this court in D’Amico v. Board of Medical Examiners (1974) 11 Cal.3d 1, 112 Cal.Rptr. 786, 520 P.2d 10 (DAmico) invalidated statutes that barred persons with osteopathic training (holders of O.D. degrees) from obtaining a physician’s license available to persons with allopathic training (holders of M.D. degrees). (Id. at p. 23, 112 Cal.Rptr. 786, 520 P.2d 10.) We began by explaining that rational basis review, not strict scrutiny, applies to occupational licensing laws. (Id. at pp. 16–18, 112 Cal.Rptr. 786, 520 P.2d 10.) We then noted the Attorney General’s admissions "(1) that osteopathy, like allopathy, is a complete school of medicine and surgery whose practitioners successfully engage in the full range of activities commonly thought of as constituting medical science …, and (2) that there exists in the state examining and licensing boards the technical capacity to screen osteopathic applicants for licensure, as allopathic applicants are now screened, so as to insure that the people of the state will be protected from incompetent and unqualified practitioners." (Id. at p. 23, 112 Cal.Rptr. 786, 520 P.2d 10.) "This showing," we said, "demonstrates beyond peradventure of a doubt that there exists no rational relationship between the protection of the public health and the exclusion from licensure of all medical practitioners who … hold D.O. rather than M.D. degrees." (Ibid.) We further said that in light of the admissions above, the same result would obtain even if "evidence might show differences of emphasis and quality between osteopathic training and allopathic training." (Id. at p. 24, 112 Cal.Rptr. 786, 520 P.2d 10.) We did not posit that the Legislature could proceed "one step at a time" in protecting public health (Lee Optical, supra, 348 U.S. at p. 489, 75 S.Ct. 461) or that the additional cost of screening osteopathic applicants for licensure could justify the exclusion (cf. Reed, supra, 404 U.S. at p. 76, 92 S.Ct. 251 [rejecting administrative efficiency as a valid rationale for an otherwise "arbitrary legislative choice"]).

Three years later, in Newland, supra, 19 Cal.3d 705, 139 Cal.Rptr. 620, 566 P.2d 254, we applied rational basis review and invalidated a statute that barred persons with a misdemeanor conviction, but not persons with a felony conviction, from eligibility for a teaching credential. (Id. at p. 707, 139 Cal.Rptr. 620, 566 P.2d 254.) The differential treatment turned on the fact that one of the statutory eligibility requirements was a certificate of rehabilitation, which was available to felons but not misdemeanants. (Ibid.) We speculated that the certificate requirement "may simply be a case of legislative oversight — a failure to realize that this requirement would block any relief to a misdemeanant." (Id. at p. 712, 139 Cal.Rptr. 620, 566 P.2d 254.) But we did not rest our reasoning on that ground. Instead, we said "our inquiry must begin with an identification of the purpose of [the statute] so that we may determine whether the statutory classification … rationally relates to that purpose." (Id. at p. 711, 139 Cal.Rptr. 620, 566 P.2d 254.) We determined that the credentialing statute’s purpose was "to protect the students, faculty and others who might be harmed by the employment of an unfit teacher." (Id. at pp. 711–712, 139 Cal.Rptr. 620, 566 P.2d 254.) We then explained: "This statutory discrimination against misdemeanants can claim no rational relationship to the protective purpose of [the statute]. … The Legislature could not possibly or sensibly have concluded that misdemeanants, as opposed to felons, constitute a class of particularly incorrigible offenders who are beyond hope of rehabilitation." (Id. at p. 712, 139 Cal.Rptr. 620, 566 P.2d 254.)

In applying rational basis review, Newland hewed to the statute’s clear purpose and evaluated the classification against that purpose. We did not posit any competing purposes, though it would have been easy to do so: The state could have had an interest in minimizing the costs associated with determining which persons with criminal history have been rehabilitated and are thus fit to be a teacher. Whereas an existing mechanism (a certificate of rehabilitation) simplified that determination for persons with a felony conviction, no such mechanism existed for persons with a misdemeanor conviction, a far larger group. Educational institutions, if they wished to screen such applicants, would have needed to incur the burden of conducting their own fitness hearings, as Newland acknowledged. (Newland, supra, 19 Cal.3d at p. 714, fn. 11, 139 Cal.Rptr. 620, 566 P.2d 254.) Had we taken the view that " ‘[i]f a plausible basis exists for the disparity, courts may not second-guess its " ‘wisdom, fairness, or logic’ " ’ " (maj. opn., ante, 318 Cal.Rptr.3d at p. 528, 543 P.3d at p. 973), Newland would have come out the other way. But we did not deploy such reasoning. After canvassing various formulations of the rational basis standard, we said that "[a]ll of the formulas require the court to conduct ‘a serious and genuine judicial inquiry into the correspondence between the classification and the legislative goals’ " and that such inquiry demonstrated the unconstitutionality of the classification at issue. (Newland, at p. 711, 139 Cal.Rptr. 620, 566 P.2d 254, italics added.)

The following year, we applied the inquiry as stated in Neuland to invalidate a Vehicle Code provision barring passengers who own the car in which they were injured from suing the negligent driver: "[H]aving conducted a ‘serious and genuine judicial inquiry into the correspondence between the [statutory] classification and the legislative goals’ [citation to Newland], we are convinced that the disparate treatment accorded by the statute is not rationally related to a realistically conceivable legislative purpose." (Cooper v. Bray (1978) 21 Cal.3d 841, 855, 148 Cal.Rptr. 148, 582 P.2d 604 (Cooper).) From "the origin and legislative history of the provision," we found it "rather clear" that "the provision was not intended to impose special burdens on owner-passengers but rather proposed to place such owner-passengers on an equal plane with most other injured automobile passengers" at a time when the Vehicle Code also barred recovery by nonpaying automobile guests. (Id. at p. 848, 148 Cal.Rptr. 148, 582 P.2d 604.) Because the court in Brown had since invalidated the automobile guest statute, the bar on recovery by owner-passengers no longer " ‘further[ed] the legislative purpose of according owner-passengers the same treatment as such guests, but rather defeat[ed] that purpose by singling out owner-passengers for differential treatment from all other automobile accident victims.’ " (Id. at p. 851, 148 Cal.Rptr. 148, 582 P.2d 604.)

In evaluating the classification against the statute’s actual purpose, we declined to impute other purposes to the Legislature, such as an "interest in promoting automobile safety by encouraging the careful selection and supervision of permissive drivers by car owners." (Cooper, supra, 21 Cal.3d at p. 851, 148 Cal.Rptr. 148, 582 P.2d 604.) We found the statute overinclusive with regard to such an interest because it "bar[red] recovery by all ownerpassengers, including the most careful owners who selected the most cautious drivers and who scrupulously supervised the driving." (Id. at p. 852, 148 Cal.Rptr. 148, 582 P.2d 604.) We also explained that if the Legislature had intended "to encourage care in the selection and supervision of drivers," then the statutory exceptions allowing recovery by owner-passengers for injury caused by a driver’s intoxication or willful misconduct "obviously make[ ] no sense." (Ibid.) We were unmoved by the dissent’s argument that "the Legislature, pursuing the clearly legitimate goal of achieving a fair distribution of liability for damage caused by unreasonable conduct," could have reasonably "weighed the conflicting interests of driver and owner and concluded that the driver should be protected, given the owner’s selection of, and supervision over, the driver," even if this "reasoning was unwise, or … the purpose of the Legislature could have been better furthered by another means." (Id. at pp. 857-858, 148 Cal.Rptr. 148, 582 P.2d 604 (dis. opn. of Richardson, J.).)

Our approach in Cooper, Newland, and earlier cases was consonant with a contemporaneous high court case, United States Department of Agriculture v. Moreno (1973) 413 U.S. 528, 93 S.Ct. 2821, 37 L.Ed.2d 782 (Moreno), which invalidated a statute excluding households "containing an individual who is unrelated to any other member of the household" from food stamp eligibility. (Id. at p. 529, 93 S.Ct. 2821.) Applying rational basis review, the high court quoted the Food Stamp Act’s stated purpose " ‘[t]o alleviate … hunger and malnutrition’ " among " ‘low-income households’ " and concluded that "[t]he challenged statutory classification … is clearly irrelevant to the stated purposes of the Act." (Id. at pp. 533-534, 93 S.Ct. 2821.) The court then considered whether "Congress might rationally" have had an "interest in minimizing fraud in the administration of the food stamp program." (Id. at p. 535, 93 S.Ct. 2821.) It rejected this rationale on the grounds that the statute already contained other antifraud provisions (id. at pp. 536-537, 93 S.Ct. 2821) and that the exclusion of unrelated households "in practical operation" did not target "persons who are ‘likely to abuse the program’ " (id. at p. 538, 93 S.Ct. 2821). The court did not posit that Congress could have desired a belt-and-suspenders approach to combating fraud. Nor did it accept the generalization that limiting food stamps to related households "provides a guarantee … that the household exists for some purpose other than to collect federal food stamps" (id. at p. 546, 93 S.Ct. 2821 (dis. opn. of Rehnquist, J.), citing evidence to the contrary (id. at pp. 537–538, 93 S.Ct. 2821 (maj. opn.)). Like Cooper, Moreno evaluated the classification against the stated legislative purpose and declined to impute other purposes, and the high court did not defer to plausible yet unsubstantiated generalizations, even while acknowledging that rational basis review "does not require that every classification be drawn with precise ‘ "mathematical nicety." ’ " (Moreno, at p. 538, 93 S.Ct. 2821.)

We continued to apply this mode of analysis in Hays v. Wood (1979) 25 Cal.3d 772, 160 Cal.Rptr. 102, 603 P.2d 19 (Hays), where we invalidated a voter-enacted disclosure law that required public officials who were lawyers or brokers to disclose any source of payments equal to or greater than $1,000, but which required filers with other business interests to disclose only sources of payments equal to or greater than $10,000. (Id. at p. 795, 160 Cal.Rptr. 102, 603 P.2d 19.) We again applied Newland’s formulation of the rational basis inquiry (Hays, at p. 787, 160 Cal.Rptr. 102, 603 P.2d 19) and focused on the legislative purpose stated in "the Act itself," i.e., "insuring disclosure of income which may be materially affected by the official actions of the filing public official" (id. at p. 788, 160 Cal.Rptr. 102, 603 P.2d 19). We recognized that the potential for conflict of interest is a function of an official's "actual profits" derived from business dealings (ibid.) and that providers of professional services have "substantially greater" profit margins than business entities that make or sell goods (id. at p. 789, 160 Cal.Rptr. 102, 603 P.2d 19). But this distinction did not justify "special treatment" of lawyers and brokers as compared to other professionals with comparable profit margins. (Ibid.) We said this " ‘underinclusiv[ity]’ " could not be justified on the ground that "a legislative body … need not attack all phases [of a problem] at once." (Id. at p. 790, 160 Cal. Rptr. 102, 603 P.2d 19.) "[W]hen the legislative body proposes to address an area of concern in less than comprehensive fashion by ‘striking the evil where it is felt most’ [citation], its decision as to where to ‘strike’ must have a rational basis in light of the legislative objectives." (Id. at p. 791, 160 Cal.Rptr. 102, 603 P.2d 19.)

We then proceeded to reject four possible bases for distinguishing lawyers from "all others similarly situated in terms of profit margin." (Hays, supra, 25 Cal.3d at p. 792, 160 Cal.Rptr. 102, 603 P.2d 19.) It was argued that lawyers are more likely to have potential conflicts because they often represent private interests in dealings with government; we said other professionals may have a higher volume of clients, making potential conflicts more frequent. (Id. at pp. 792–793, 160 Cal.Rptr. 102, 603 P.2d 19.) It was argued that "the unique nature" of the lawyer-client relationship, including "habits of loyalty," make lawyers more prone to conflict; we said the professional relationships of physicians and psychotherapists are at least as "personal and intense." (Id. at p. 793, 160 Cal.Rptr. 102, 603 P.2d 19.) It was argued that "the customary practice of many lawyers of accepting retainers" can serve as "a unique device for channeling money in payment for public favors"; we said a "disguised payment for political favor" can occur "in any number of ways." (Id. at pp. 793, 794, 160 Cal.Rptr. 102, 603 P.2d 19.) And it was argued that the public may perceive a lawyer, " ‘more so than … persons in other professions,’ " as promoting client interests when serving as a public official; we said this was a "curious assertion" that provided no basis for "significantly different standards of disclosure for members of different professions." (Id. at pp. 794, 795, 160 Cal.Rptr. 102, 603 P.2d 19.) We thus rejected a series of unsubstantiated assertions en route to holding that the classification "fails to exhibit any fair and reasonable relationship to the stated legislative objectives." (Id. at p. 795, 160 Cal.Rptr. 102, 603 P.2d 19.)

Another case in this line was United States Steel Corp. v. Public Utilities Commission (1981) 29 Cal.3d 603, 175 Cal. Rptr. 169, 629 P.2d 1381 (U.S. Steel), which involved a challenge to a Public Utilities Commission order exempting commodities carried by private vessels (as opposed to common carriers) from intrastate minimum shipping rates. The effect of this order was to make foreign steel cheaper to transport compared to domestic steel. (Id. at p. 607, 175 Cal.Rptr. 169, 629 P.2d 1381.) We annulled the order on the ground that the Commission had adopted it without having satisfied its statutory duty to "assess the economic impact of its action," including whether the exemption would drive shippers out of business and cost jobs. (Id. at p. 610, 175 Cal.Rptr. 169, 629 P.2d 1381.) "To guide the commission in further proceedings" (ibid.), we went on to discuss the requirements of equal protection in this context. We again quoted Newland's formulation of the rational basis inquiry and observed that "[t]he aim of minimum rate regulation is to preclude destructive rate practices and to provide for movement at the lowest rates compatible with the maintenance of adequate transportation service. [Citations.] Rates below the minimum do not serve that aim absent some showing of a difference in cost in hauling private-vessel steel as compared with domestic steel, or of a difference regarding destructive rate practices. There is no showing here." (U.S. Steel, at p. 612, 175 Cal.Rptr. 169, 629 P.2d 1381.) A further argument for the exemption was that the "difficulty in determining whether imported steel has arrived via common carrier or private vessel" would burden "truckers in determining the appropriate rate as well as on the commission in enforcing minimum rates." (Id. at p. 613, 175 Cal.Rptr. 169, 629 P.2d 1381.) We said this concern was plausible, but "the commission’s finding as to ‘difficulty’ seems inadequately supported by the record," and the equal protection issue could not be settled "[w]ithout a more complete record." (Id. at p. 614, 175 Cal.Rptr. 169, 629 P.2d 1381.) In sum, we again declined to accept plausible yet unsubstantiated assertions under rational basis review.

B.

A few years after U.S. Steel, we decided a series of cases rejecting equal protection challenges to various provisions of the Medical Injury Compensation Reform Act of 1975 (MIGRA). (See American Bank & Trust Co. v. Community Hospital (1984) 36 Cal.3d 359, 370-374, 204 Cal.Rptr. 671, 683 P.2d 670 (American Bank); Barme v. Wood (1984) 37 Cal.3d 174, 181–182, 207 Cal.Rptr. 816, 689 P.2d 446 (Barme); Roa v. Lodi Medical Group, Inc. (1985) 37 Cal.3d 920, 930-931, 211 Cal.Rptr. 77, 695 P.2d 164 (Roa); Fein v. Permanente Medical Group (1985) 38 Cal.3d 137, 161–164, 211 Cal.Rptr. 368, 695 P.2d 665 (Fein).) Our language in those cases featured more deferential formulations of rational basis review, (See, e.g., American Bank, at p. 371, 204 Cal.Rptr. 671, 683 P.2d 670 ["the equal protection clause does not prohibit a Legislature from implementing a reform measure ‘one step at a time’ [citation], or prevent it ‘from striking the evil where it is felt most’ "]; id. at p. 374, 204 Cal.Rptr. 671, 683 P.2d 670 ["the constitutionality of a measure under the equal protection clause does not depend on a court’s assessment of the empirical success or failure of the measure’s provisions"].) But we observed that "our application of equal protection principles in [the MIGRA cases] is not inconsistent with the principles enunciated in [Brown and Cooper] or like cases. As Cooper explains, … what is required is that the court ‘conduct "a serious and genuine judicial inquiry into the correspondence between the classification and the legislative goals." ’ (21 Cal.3d at p. 848, 148 Cal.Rptr. 148, 582 P.2d 604 [quoting Newland v. Board of Governors (1977) 19 Cal.3d 705, 711, 139 Cal.Rptr. 620, 566 P.2d 254, italics added in Cooper].) We have conducted such an inquiry in all of these cases (Fein, at p. 163, 211 Cal.Rptr. 368, 695 P.2d 665.)

Fein is illustrative. The plaintiff argued that MICRA’s $250,000 cap on noneconomie damages violates equal protection be- cause it "discriminates between medical malpractice victims and other tort victims" and because it "discriminates within the class of medical malpractice victims, denying a ‘complete’ recovery of damages only to those malpractice plaintiffs with noneconomic damages exceeding $250,000." (Fein, supra, 38 Cal.3d at pp. 161-162, 211 Cal. Rptr. 368, 695 P.2d 665.) As to the first contention, we cited our earlier cases that had extensively examined the legislative history of MIGRA showing that the Legislature, with ample basis, had targeted medical malpractice cases for reform because of "an insurance ‘crisis’ in that particular area." (Fein, at p. 162, 211 Cal. Rptr. 368, 695 P.2d 665, citing American Bank, Barme, and Roa.) As to the second contention, we said "the Legislature clearly had a reasonable basis" for seeking cost savings "only by limiting the recovery of noneconomic damage." (Fein, at p. 162, 211 Cal.Rptr. 368, 695 P.2d 665.) While acknowledging other plausible means of distributing cost savings across malpractice plaintiffs, we explained that the size and unpredictability of noneconomic damages awards rationally justified the Legislature’s approach. (Id. at pp. 162–163, 211 Cal.Rptr. 368, 695 P.2d 665.) We noted that "the unpredictability of the size of large noneconomic damage awards" was "[o]ne of the problems identified in the legislative hearings" (id. at p. 163, 211 Cal.Rptr. 368, 695 P.2d 665), and we cited legal scholarship and an American Bar Association report to show that the issue was one on which "reasonable persons can certainly disagree" (id. at p. 160, 211 Cal. Rptr. 368, 695 P.2d 665; see id. at pp. 159–160 & fns. 16–17, 211 Cal.Rptr. 368, 695 P.2d 665). Fein did not rely on imputed legislative purposes or unsubstantiated assertions to uphold the challenged provision.

A subsequent case, Warden, supra, 21 Cal.4th 628, 88 Cal.Rptr.2d 283, 982 P.2d 154, marks perhaps our most deferential application of rational basis review. We rejected an equal protection challenge to an exemption for retired judges, elected officials, and law professors from continuing education requirements that are generally applicable to practicing attorneys. (Id. at p. 634, 88 Cal.Rptr.2d 283, 982 P.2d 154.) We said "it would not have been irrational to conclude that the attorneys in each of the exempted categories, as a general matter, are less likely than other attorneys to represent clients on a full-time basis, thus rendering the need for a continuing education requirement less vital," and that "in view of their particular professional roles and experience, the attorneys in each of the exempt classes (again, as a general matter) are less likely than lawyers in general to need continuing education courses in order to be familiar with recent legal developments or to remain competent practitioners." (Id. at pp. 645–646, 88 Cal.Rptr.2d 283, 982 P.2d 154.)

The Court of Appeal had observed that there was " ‘no support’ in the legislative history … to indicate that these were the actual explanations of the rationale or motivation for the adoption of the exemptions." (Warden, supra, 21 Cal.4th at pp. 649–650, 88 Cal.Rptr.2d 283, 982 P.2d 154.) Citing federal case law, we said that "when there is a reasonably conceivable justification for a classification, ‘[i]t is … "constitutionally irrelevant whether [the] reasoning in fact underlay the legislative decision." ’ " (Id. at p. 650, 88 Cal.Rptr.2d 283, 982 P.2d 154, quoting United States Railroad Retirement Board. v. Fritz (1980) 449 U.S. 166, 179, 101 S.Ct. 453, 66 L.Ed.2d 368.) We also said, citing federal case law, that " ‘a legislative choice … may be based on rational speculation unsupported. by evidence or empirical data’ " (Warden, at p. 650, 88 Cal.Rptr.2d 283, 982 P.2d 154, italics added, quoting Feder- al Communications Commission v. Beach Communications, Inc. (1993) 508 U.S. 307, 315, 113 S.Ct. 2096, 124 L.Ed.2d 211 (Beach Communications) and that " ‘reform may take one step at a time’ " (Warden, at p. 645, 88 Cal.Rptr.2d 283, 982 P.2d 154, quoting Lee Optical, supra, 348 U.S. at p. 489, 75 S.Ct. 461).

As two dissenting Justices observed, Warden relied heavily on federal authority in elaborating a highly deferential rational basis test without grappling with the fact that whereas it is one of three levels of scrutiny in federal equal protection doctrine, our own case law "has not slavishly followed decisions of the federal high court" and has never adopted intermediate scrutiny. (Warden, supra, 21 Cal.4th at pp. 652-653, 88 Cal.Rptr.2d 283, 982 P.2d 154 (dis. opn. of Kennard, J.); id. at p. 661, 88 Cal.Rptr.2d 283, 982 P.2d 154 (dis. opn. of Brown, J.) ["Our state equal protection jurisprudence grew out of a recognition of the inadequacy of federal standards."].) Justice Brown noted that our decision in Hays had "expressly rejected" Lee Optical in saying that " ‘the legislative body, when it chooses to address a particular area of concern in less than comprehensive fashion by merely "striking the evil where it is felt most" [citation] may not do so wholly at its whim.’ [Citation.] Rather ‘its decision as to where to "strike" must have a rational basis in light of the legislative objectives.’ " (Warden, at p. 664, 88 Cal.Rptr.2d 283, 982 P.2d 154 (dis. opn. of Brown, J.), quoting Hays, supra, 25 Cal.3d at pp. 790, 791, 160 Cal.Rptr. 102, 603 P.2d 19.) "[O]ur state Constitution insists on greater precision …. Rather than merely ‘rubberstamping’ the legislative categories at issue here, we should be engaging in ‘"a serious and genuine judicial inquiry into the correspondence between the classification and the legislative goals" ’ (Newland v. Board of Governors, supra, 19 Cal.3d at p. 711, 139 Cal.Rptr. 620, 566 P.2d 254), and, more particularly, we should be asking whether the legislative classifications substantially advance the legislative purposes without being ‘grossly overinclusive’ or ‘underinclusive.’ (Brown v. Merlo, supra, 8 Cal.3d at p. 877 & fn. 17 [106 Cal.Rptr. 388, 506 P.2d 212].)" (Warden, at pp. 664-665, 88 Cal. Rptr.2d 283, 982 P.2d 154 (dis. opn. of Brown, J.).)

Warden made clear that the standards it elaborated apply " ‘[i]n areas of social and economic policy’ " (Warden, supra, 21 Cal.4th at p. 644, 88 Cal.Rptr.2d 283, 982 P.2d 154, quoting Beach Communications, supra, 508 U.S. at p. 313, 113 S.Ct. 2096), and issues such as continuing legal education requirements or the definition of a "cable system" (Beach Communications, at pp. 310-311, 113 S.Ct. 2096) are paradigmatic examples. Warden’s deferential language has seeped into our case law addressing equal protection challenges to criminal statutes, with no examination of how our doctrine has evolved differently from its federal counterpart. (See People v. Turnage (2012) 55 Cal.4th 62, 75, 79, 144 Cal.Rptr.3d 489, 281 P.3d 464 (Turnage); Johnson v. Department of Justice (2015) 60 Cal.4th 871, 887, 183 Cal.Rptr.3d 96, 341 P.3d 1075 (Johnson); People v. Chatman (2018) 4 Cal.5th 277, 289, 228 Cal. Rptr.3d 379, 410 P.3d 9 (Chatman).) But even in those cases, we have not actually employed the full extent of deference that Warden’s language contemplates.

In Turnage, we upheld a statute allowing felony treatment of placing a false bomb without proof of causing sustained fear, even though a separate statute requires proof of sustained fear for felony treatment of placing a false weapon of mass destruction (WMD). (Turnage, supra, 55 Cal.4th at pp. 67–68, 144 Cal. Rptr.3d 489, 281 P.3d 464.) We said the differential treatment was rational because "[i]t is conceivable from a legislative per- spective" that false WMDs, unlike false bombs, "would not necessarily be recognized or cause fear, even where it is detected and was intended to do so." (Id. at p. 68, 144 Cal.Rptr.3d 489, 281 P.3d 464.) But this rationale was not merely "conceivable"; the history of the false WMD statute implied that the Legislature had actually considered it. Extensively citing a Senate committee report, we observed that "[t]he new false WMD statute was said to be inspired by the false bomb statute" (id. at p. 79, 144 Cal.Rptr.3d 489, 281 P.3d 464) and that "in acknowledging the similarity between the ‘wobbler’ provisions of [the false WMD statute] and [the false bomb statute], the Legislature implied that it was aware of the substance of the latter statute, that proof of sustained fear was not required in felony false bomb cases, and that both felonies nonetheless involved the same level of ‘violent’ fear. … In other words, a showing of sustained fear for felonies under the false WMD statute was necessary to reflect the same level of violent fear that the Legislature assumed was present in false bomb cases …. " (Id. at p. 80, 144 Cal.Rptr.3d 489, 281 P.3d 464.) "[T]he Senate Report implicitly shows a rational connection between the disparate role of sustained fear in the false bomb and false WMD statutes, and the purpose such disparity was apparently meant to serve." (Id. at pp. 79-80, 144 Cal.Rptr.3d 489, 281 P.3d 464.)

In Johnson, we upheld a statutory scheme allowing discretionary sex offender registration for persons convicted of unlawful sexual intercourse with a minor, while imposing mandatory registration for persons convicted of crimes involving other sexual acts with a minor, including nonforcible oral copulation. (Johnson, supra, 60 Cal.4th at pp. 874-875, 183 Cal.Rptr.3d 96, 341 P.3d 1075.) We said that "intercourse is unique in its potential to result in pregnancy and parenthood. Given that unique potential, legislative concerns regarding teen pregnancy and the support of children conceived as a result of unlawful sexual intercourse provide more than just a plausible basis for" the disparate treatment. (Id. at p. 875, 183 Cal.Rptr.3d 96, 341 P.3d 1075.) This rationale was not hypothetical. We examined the history of the statute against unlawful sexual intercourse and observed that the Legislature, by separating this offense from the general rape statute, "sought to eliminate … the social stigma associated with the rape label so that offenders could more readily obtain employment and support children conceived as a result of such intercourse. [Citations to legislative history.] This history confirms that the potential for pregnancy and parenthood has, in fact, influenced legislative decisionmaking regarding unlawful intercourse with minors." (Id. at p. 885, 183 Cal.Rptr.3d 96, 341 P.3d 1075; see ibid. [citing subsequent legislative concerns about teen pregnancy and birth rates resulting from unlawful sexual activity between adult males and teenage girls, and implications for public welfare and health care expenses].) We said this actual legislative concern provided a rational basis for allowing trial courts not to order registration where it "might cause economic or other hardship to a child born to the minor victim and the adult offender," while requiring registration for persons convicted of other unlawful sexual activity with minors. (Id. at p. 886, 183 Cal.Rptr.3d 96, 341 P.3d 1075.)

Similarly, in Chatman, we upheld a statute barring former probationers but not former prisoners from eligibility for a certificate of rehabilitation in certain circumstances. (Chatman, supra, 4 Cal.5th at pp. 282–283, 228 Cal.Rptr.3d 379, 410 P.3d 9.) Our analysis focused on the state’s interest in avoiding the costs associated with extending certificates of rehabilitation to for- mer probationers, a group much larger than former prisoners. (Id. at pp. 291-292, 228 Cal.Rptr.3d 379, 410 P.3d 9.) Although we said the rationale for a legislative classification need not have been articulated by lawmakers and does not need to be empirically substantiated (id. at p. 289, 228 Cal.Rptr.3d 379, 410 P.3d 9), in fact we observed that legislative history "provide[s] at least some indication that this cost concern figured in legislative deliberations" (id. at p. 292, 228 Cal.Rptr.3d 379, 410 P.3d 9), and we went on to cite "data" that substantiated the cost concern (id. at p. 293, 228 Cal.Rptr.3d 379, 410 P.3d 9). In other words, Chatman addressed what appeared to be the actual rationale for the classification and found it had some empirical support. (See also People v. Wilkinson (2004) 33 Cal.4th 821, 834, 839, 16 Cal. Rptr.3d 420, 94 P.3d 551 [legislative history showed rationale for statute allowing potentially harsher treatment of battery on a custodial officer without injury than such battery with injury].)

C.

In sum, this court’s articulation and application of rational basis review has not marched in lock step with federal authority. Our approach is deferential but far from toothless; our case law, though not entirely uniform, reveals several recurring themes: In conducting "‘a serious and genuine judicial inquiry into the correspondence between the classification and the legislative goals’" (Newland, supra., 19 Cal.3d at p. 711, 139 Cal.Rptr. 620, 566 P.2d 254), we have focused on actual legislative purposes instead of imputing hypothetical ones, and we have looked for empirical support instead of relying on conjecture or unsubstantiated assertions. Although we "do not require absolute precision in the designation of classifications," we also "do not tolerate classifications which are so grossly overinclusive as to defy notions of fairness or reasonableness." (Brown, supra, 8 Cal.3d at p. 877, 106 Cal.Rptr. 388, 506 P.2d 212.) And while the Legislature may proceed incrementally, we have said it must do so rationally in light of the legislative objectives and "not … wholly at its whim." (Hays, supra, 25 Cal.3d at p. 790, 160 Cal.Rptr. 102, 603 P.2d 19.)

Today’s opinion ignores this case law and claims that I am offering "an argument for reconsidering rational basis review." (Maj. opn., ante, 318 Cal.Rptr.3d at p. 528, fn. 3, 543 P.3d at p. 973, fn. 3.) To the contrary, it is the court’s refusal to consider, not reconsider, our precedent that is the problem here. The principles above have sturdy foundations in what our cases say and, perhaps more importantly, in what they actually do. Although Warden took a more deferential approach, today’s opinion does not cite Warden for an obvious reason: We are not dealing with a matter at all similar to who is or isn’t subject to continuing legal education requirements. The case before us concerns which young offenders will be condemned to die in prison and which will have a meaningful chance to earn release. Instead of measuring the challenged classification against the Legislature’s stated purpose in enacting the youth offender parole statute, today’s opinion upholds the exclusion of young offenders convicted of special circumstance murder by imputing a different purpose that is nowhere mentioned in the statute’s text or legislative history, and that even on its own terms does not provide a rational basis for treating the excluded group differently from young offenders convicted of first degree murder. The court’s reasoning cannot sustain the result here.

II.

The provision at issue — section 3051, subdivision (h) — is part of the youth offender parole statute enacted by the Legislature in 2013. I begin with some background on the statute and then explain its constitutional infirmity.

A.

The Legislature made clear the purpose of section 3051 in the statute itself. Its opening provision, as originally enacted, says in full: "The Legislature recognizes that youthfulness both lessens a juvenile’s moral culpability and enhances the prospect that, as a youth matures into an adult and neurological development occurs, these individuals can become contributing members of society. The purpose of this act is to establish a parole eligibility mechanism that provides a person serving a sentence for crimes that he or she committed as a juvenile the opportunity to obtain release when he or she has shown that he or she has been rehabilitated and gained maturity, in accordance with the decision of the California Supreme Court in People v. Caballero (2012) 55 Cal.4th 262, 145 Cal.Rptr.3d 286, 282 P.3d 291 and the decisions of the United States Supreme Court in Graham v. Florida (2010) 560 U.S. 48, 130 S.Ct. 2011, 176 L.Ed.2d 825 [(Graham)], and Miller v. Alabama (2012) . Nothing in this act is intended to undermine the California Supreme Court’s holdings in In re Shaputis (2011) 53 Cal.4th 192, 134 Cal.Rptr.3d 86, 265 P.3d 253, In re Lawrence (2008) 44 Cal.4th 1181, 82 Cal.Rptr.3d 169, 190 P.3d 535, and subsequent cases. It is the intent of the Legislature to create a process by which growth and maturity of youthful offenders can be assessed and a meaningful opportunity for release established." (Stats. 2013, ch. 312, § 1 [Sen. Bill No. 260].) The only references to culpability in the statute are the Legislature’s recognition of the diminished culpability of youth (ibid.) and the mandate that any psychological evaluations or risk assessments used by the parole board "shall take into consideration the diminished culpability of youth as compared to that of adults" (§ 3051, subd. (f)(1)).

Subsequently, in light of more "[r]ecent scientific evidence on adolescent and young adult development and neuroscience show[ing] that certain areas of the brain — particularly those affecting judgment and decision-making — do not fully develop until the early- to mid-20s" (Sen. Com. on Public Safety, Rep. on Sen. Bill No. 261 (2015-2016 Reg. Sess.) Apr. 28, 2015, p. 3), the Legislature amended the parole eligibility scheme — first, to include youth offenders who committed crimes before the age of 23 (Stat. 2015, ch. 471, § 1, Sen. Bill No. 261 (2015-2016 Reg. Sess.) § 1), and then, to include offenders who committed crimes before age 26 (Stat. 2017, ch. 675, § 1, Assem. Bill No. 1308 (2017-2018 Reg. Sess.) § 1). A committee report on the latter bill states: "The rationale, as expressed by the author and supporters of this bill, is that research shows that cognitive brain development continues into the early 20s or later. The parts of the brain that are still developing during this process affect judgment and decision-making, and are highly relevant to criminal behavior and culpability. (See Johnson, et al., Adolescent Maturity and the Brain: The Promise and Pitfalls of Neuroscience Research in Adolescent Health Policy, Journal of Adolescent Health (Sept. 2009); National Institute of Mental Health, The Teen Brain: Still Under Construction (2011).) ‘The development and maturation of the prefrontal cortex occurs primarily during adolescence and is fully accomplished at the age of 25 years. The development of the prefrontal cortex is very important for complex behavioral performance, as this region of the brain helps accomplish executive brain functions.’ [Citation to Arain et al., Maturation of the Adolescent Brain (2013) 9 Neuropsychiatric Disease & Treatment 449.]" (Sen. Rules Com., Off. of Sen. Floor Analyses, Analysis of Assem. Bill No. 1308 (2017-2018 Reg. Sess.) Sept. 4, 2017, pp. 4–5.)

As today’s opinion acknowledges, "No one doubts that the Legislature’s primary purpose in expanding section 3051 to include young adult offenders was to give these young persons the opportunity to obtain release based on demonstrated growth and rehabilitation." (Maj. opn., ante, 318 Cal.Rptr.3d at p. 530, 543 P.3d at p. 974.) "The Legislature enacted section 3051 to bring California juvenile sentencing law into line with Graham, Miller, and Caballero.’’ (Id. at p. 522, 543 P.3d at p. 968.) Relying "not only on common sense — on what ‘any parent knows’ — but on science and social science," those cases explain that juveniles have an underdeveloped sense of responsibility, vulnerability to negative influences and outside pressures, lack of control over their own environment, and transitory traits that are not fixed but developing — all of which mitigate their culpability and point to their capacity for rehabilitation. (Miller, supra, 567 U.S. at p. 471, 132 S.Ct. 2455.) These attributes of youth also diminish the traditional penological justifications of retribution, deterrence, and incapacitation. (See id. at p. 472, 132 S.Ct. 2455; Graham, supra., 560 U.S. at pp. 71-73, 130 S.Ct. 2011.) As indicated in the legislative history, the Legislature had these concerns not only about juveniles but also about young adults under the age of 26.

Further, there is no dispute that the scientific evidence cited by the Legislature applies to young offenders across the board. Nothing about the "distinctive (and transitory) mental traits and environmental vulnerabilities" of youth offenders "is crime-specific" (Miller, supra, 567 U.S. at p. 473, 132 S.Ct. 2455), and the Legislature nowhere suggested that young offenders who commit certain crimes, including crimes punishable by LWOP, are immune to those vulnerabilities or incapable of reform. There is also no dispute that providing young offenders with a meaningful opportunity for release in light of their capacity for change is the only purpose stated by the Legislature in creating and expanding the parole scheme. No other purpose is stated in the statute or legislative history.

The implications for equal protection analysis are straightforward, as the Court of Appeal discerned: "[I]f, as the Legislature stated, the goal of section 3051 was to apply the Miller youth-related mitigating factors to young adults up to the age of 26 in light of neuroscience research that demonstrated the human brain continues to develop into a person’s mid-20’s, and thus to permit youth offenders a meaningful opportunity for parole if they demonstrate increased maturity and impulse control, then for that purpose there is no plausible basis for distinguishing between same-age offenders based solely on the crime they committed." (People v. Hardin (2022) 84 Cal.App.5th 273, 288, 300 Cal.Rptr.3d 243 (Hardin).) Just as there was no rational basis for excluding persons with misdemeanor convictions from eligibility for a teaching credential in Newland, or for excluding osteopathic applicants from eligibility for medical licensure in D’Amico, or for imposing more onerous disclosure rules on lawyers than on other professionals in Hays, or for excluding unrelated households from food stamp eligibility in Moreno, there is no rational basis here for excluding inmates convicted of special-circumstance murder from youth offender parole eligibility. In each of the cited cases, the court evaluated the classification against "the primary purpose" (maj. opn., ante, 318 Cal.Rptr.3d at p. 530, 543 P.3d at p. 974) of the statute and found the classi- fication inconsistent with that purpose. Those decisions did not impute additional, unstated purposes that might have justified the challenged classifications.

Nor did those decisions rely on the notion that legislative bodies may proceed incrementally or train their attention wherever they feel it is needed most. In Hays, we made clear that when a legislature "chooses to address a particular area of concern in less than comprehensive fashion," "its decision as to where to ‘strike’ must have a rational basis in light of the legislative objectives." (Hays, supra, 25 Cal.3d at p. 791, 160 Cal.Rptr. 102, 603 P.2d 19.) To be sure, equal protection doctrine gives legislators ample leeway to avoid "all-or-nothing choices." (Maj. opn., ante, 318 Cal.Rptr.3d at p. 539, 543 P.3d at p. 982.) But if the rational basis standard could be met by observing that the challenged legislation reflects a political compromise, then virtually no enacted policy would ever fail rational basis review. The Court of Appeal here rejected the argument that the exclusion should be upheld "on the general principle that, when addressing a problem, the Legislature may choose to proceed incrementally," aptly noting that " ‘the fact that a line has to be drawn somewhere does not justify its being drawn anywhere.’ " (Hardin, supra, 84 Cal.App.5th at pp. 290, 291, 300 Cal. Rptr.3d 243.)

B.

In reaching today’s holding, the court says the Legislature had a second purpose when it expanded youth offender parole eligibility to include young adult offenders: "The structure and history of the [parole eligibility] expansion make clear that the Legislature sought to balance [its] primary objective with … concerns about culpability and the appropriate level of punishment for certain very serious crimes." (Maj. opn., ante, 318 Cal.Rptr.3d at p. 530, 543 P.3d at p. 974.) Presumably the court points to "structure and history" because the text of the original statute focuses solely on the diminished culpability of youth and their capacity for rehabilitation, and nothing in the text of the original statute or subsequent expansions indicates concerns about culpability or appropriate punishment for serious crimes. But neither structure nor history helps the court’s argument either.

As for history, the court says "the legislative history accompanying the amendments [expanding parole eligibility] confirms that [crime-based distinctions] were deliberate choices." (Maj, opn., ante, 318 Cal.Rptr.3d at p. 530, 543 P.3d at p. 974.) That is true, but I see no probative value in quotations from legislative history observing that the Legislature through multiple rounds of statutory amendments retained distinctions based on the crime committed. (Id. at pp. 530–531, 543 P.3d at pp. 974–975.) Those quotations simply describe what the statute does; they shed no light on the Legislature's rationale for the distinctions it drew. (Cf. Cooper, supra, 21 Cal.3d at p. 854, 148 Cal.Rptr. 148, 582 P.2d 604 ["a suggested legislative purpose" of protecting negligent drivers from liability to owner-passengers "does no more than restate the terms of the statute itself and [does not] indicate[ ] the general goal which the Legislature ostensibly intended to promote in providing such ‘protection’ at owner-passengers’ expense"].)

As for structure, the court notes that the statute provides for parole eligibility during a youth offender’s 15th, 20th, or 25th year of incarceration depending on the controlling offense. (Maj. opn., ante, 318 Cal.Rptr.3d at pp. 530-531, 543 P.3d at pp. 974–975; see § 3051, subd. (b).) Based on the fact that "the Legislature consciously drew lines that altered the parole component of offenders’ sentences based not only on the age of the offender (and thus the offender’s amenability to rehabilitation) but also on the offense and sentence imposed," the court infers that those lines "necessarily reflect a set of legislative judgments about the nature of punishment that is appropriate for the crime." (Maj. opn., ante, 318 Cal.Rptr.3d at p. 530, 543 P.3d at p. 975.) As the Court of Appeal explained, this "superficially plausible" inference is implausible when one considers the role of the controlling offense in the parole scheme. (Hardin, supra, 84 Cal. App.5th at p. 289, 300 Cal.Rptr.3d 243.)

The statute defines " ‘controlling offense’ " as "the offense or enhancement for which any sentencing court imposed the longest term of imprisonment." (§ 3051, subd. (a)(2)(B).) The timing of parole eligibility is keyed to a person’s controlling offense, not to the number of offenses or the aggregate sentence. (Id., subd. (b).) This means that a young adult offender serving an aggregate term of 85 years plus 189 years to life for two first degree murders, six attempted premeditated murders, various gang and firearm enhancements, and other crimes (People v. Harris (Dec. 5, 2019, B2886ll) [nonpub. opn.]) is eligible for parole on the same timeline as a young adult offender serving 25 years to life for a single count of first degree murder (§ 190, subd. (a)), It means that a young adult offender serving 80 years to life for two gang-related attempted premeditated murders with firearm enhancements (People v. Itehua (June 30, 2016, B265575) [nonpub. opn.]) is eligible for parole on the same timeline as a young adult offender convicted of a single attempted murder with a 25-year firearm enhancement (§ 12022.53, subd. (d)). And it means a young adult offender sentenced to more than 100 years to life for six attempted premeditated murders, mayhem, gang enhancements, and multiple firearm offenses and enhancements (People v. Jimenez (Apr. 20, 2007, B192157) [nonpub. opn.]) is eligible for parole on the same timeline as a young adult offender convicted of a gang-related felony in which an accomplice shot and injured a victim (§ 12022.53, subd. (e)(1)). What legislature "concern[ed] about culpability and the appropriate level of punishment for certain very serious crimes" (maj. opn., ante, 318 Cal.Rptr.3d at p. 530, 543 P.3d at p. 974) would write a statute with these results?

Today's opinion says "the statute’s ‘controlling offense' framework does rely on a certain amount of generalization about the relationship between the lengthiest individual sentence the offender has received and the culpability of the underlying criminal conduct." (Maj. opn., ante, 318 Cal.Rptr.3d at p. 532, 543 P.3d at p. 976.) I suppose "a certain amount of generalization" is in the eye of the beholder, but consider: There are literally thousands of sentences encompassed by the provision establishing parole eligibility in the 25th year of incarceration for young adult offenders whose controlling offense carries a term of 25 years to life, (§ 3051, subd. (b)(3).) The examples above are easily multiplied. (See, e.g., People v. Sepulveda (2020) 47 Cal.App.5th 291, 295–297, 260 Cal.Rptr.3d 703 [90 years to life for first degree murder, three counts of attempted premeditated murder, drive-by shooting, and firearm and gang enhancements, committed before age 22]; People v. Jones (Dec. 30, 2020, E073115), 2020 WL 7768718 [nonpub. opn.] [50 years to life sentence for single count of first degree murder with firearm and gang enhancements, committed at age 18]; People v. Kennedy (Jan. 15, 2020, B26460l), 2020 WL 218756 [nonpub. opn.] ["sentence of life, plus 173 years and eight months," for second degree murder, four counts of attempted premeditated murder, shooting at an occupied vehicle, and gang and firearm enhancements, committed at age 22]; People v. Windfield (Jan. 15, 2020, E055062) opn. ordered nonpub. Apr. 22, 2020, S260848 [90 years to life for first degree murder, attempted premeditated murder, assault with a semiautomatic weapon, and gang and firearm enhancements, committed before age 23]; People v. Rakisits (Apr. 12, 2018, B280133), 2018 WL 1755531 [nonpub. opn.] [40 years to life for single count of second degree murder with firearm enhancement, committed at age 18].) To say that "the culpability of the underlying criminal conduct" in all of these cases is " ‘ "rough[ly]" ’ " the same goes well beyond " ‘ "any gross generalizations." ’ " (Maj. opn., ante, 318 Cal.Rptr.3d at p. 532, 543 P.3d at p. 976.) It is simply irrational.

The Legislature presumably knew that the cases covered by section 3501, subdivision (b)(3) span a vast range of culpability based on the type and number of crimes committed and enhancements charged, especially in light of the 25 years-to-life firearm enhancement (§ 12022.53, subds. (d), (e)(1)), which applies to a wide range of crimes. (See In re Greg F. (2012) 55 Cal.4th 393, 407, 146 Cal.Rptr.3d 272, 283 P.3d 1160 ["The Legislature is presumed to be aware of all laws in existence when it passes or amends a statute."]; cf. Assem. Com. on Public Safety, Analysis of Assem. Bill No. 1308 (2017-2018 Reg. Sess.) as amended Mar. 30, 2017, p. 4 [noting that the parole board "held 2,519 youth offender hearings" during the first three years of section 3051’s implementation].) Yet the Legislature chose to establish a uniform rule of parole eligibility in the 25th year of incarceration for this large and varied group of young offenders. As Hardin contends, that rule is untethered to relative culpability. Today’s opinion does not explain why the Legislature would have fixated on relative culpability in its treatment of young adult offenders serving LWOP sentences when it was clearly indifferent to the relative culpability of young adult offenders serving non-LWOP sentences ranging from a single term of 25 years to life up to a wide array of de facto LWOP sentences. (See Cooper, supra, 21 Cal.3d at p. 852, 148 Cal.Rptr. 148, 582 P.2d 604 [rejecting putative rationale for a classification because it did not cohere with other parts of the statute]; Moreno, supra, 413 U.S. at pp. 536-537, 93 S.Ct. 2821 [same].)

The court says considerations of "the appropriate punishment for the underlying crimes, depending on their severity," are "not dissimilar from the considerations that prompted the high court [in Miller] to distinguish, for Eighth Amendment purposes, between sentencing juveniles for homicide offenses and sentencing juveniles for nonhomicide offenses." (Maj. opn., ante, 318 Cal.Rptr.3d at p. 531, 543 P.3d at p. 975.) But the line drawn here between LWOP and non-LWOP sentences does not track the line between homicide and nonhomicide offenses. And more fundamentally, the import of Miller is that juvenile sentencing even for the most severe crimes must be bounded by what is known about young offenders’ capacity for change. Miller held that life without parole may be imposed on juvenile homicide offenders only on " ‘rare’ " occasions based on an individualized sentencing determination that "take[s] into account how children are different, and how those differences counsel against irrevocably sentencing them to a lifetime in prison." (Miller, supra, 567 U.S. at pp. 479, 480, 132 S.Ct. 2455.) The Legislature, aware of Miller, determined that young offenders’ transient traits and capacity for change extend through age 25. Denying parole eligibility to a subset of young offenders on the ground that their culpability categorically trumps their potential for rehabilitation is plainly "dissimilar" (maj. opn., ante, 318 Cal.Rptr.3d at p. 531, 543 P.3d at p. 975) from Miller’s emphasis on individualized sentencing determinations to account for youth-related vulnerabilities.

At bottom, the court says that assessing culpability on the basis of aggregate sentences or underlying criminal conduct is "not the only possible way[] to evaluate culpability. That the Legislature may have prescribed a measurement of culpability different from Hardin’s does not mean the Legislature was not attempting to measure culpability at all." (Maj. opn., ante, 318 Cal.Rptr.3d at p. 532, 543 P.3d at p. 976.) But it is not Hardin’s measurement of culpability that the "controlling offense" framework displaces. It is the Legislature’s own measurements of culpability written into the Penal Code, including prescribed sentences for certain crimes and enhancements, as well as requirements for consecutive sentencing. (E.g., §§ 190, subd. (a), 186.22, subd. (b)(1), 12022.53, subds. (b)–(e).) Notably, the court contends that the Legislature leaned into its own measurement of culpability in the special circumstances law (§ 190.2) when it excluded persons like Hardin from parole eligibility. (Maj. opn., ante, 318 Cal.Rptr.3d at p. 537, 543 P.3d at p. 980 [there is no crime that is "more serious, or that trigger[s] more severe punishment, than special circumstance murder"]; see id. at pp. 533-534, 543 P.3d at pp. 977-978.) And yet, the court would have us believe that the Legislature elsewhere in section 3051 simply ditched the multitude of culpability measurements it has made throughout the remainder of the Penal Code in favor of a three-tiered measurement of culpability based only on an offender’s crime or enhancement with the longest sentence. (See id. at pp. 532–533, 543 P.3d at p. 976.)

There is a far simpler explanation: In enacting and expanding section 3051, the Legislature was not in the business of measuring culpability at all, apart from recognizing the diminished culpability of young offenders across the board. The statute’s text and history make clear that the Legislature’s purpose was to recognize the developmental vulnerabilities of young offenders and provide them a chance to earn release by demonstrating growth and change. As Hardin says, the parole scheme simply reflects the Legislature’s general calibration as to how much time is needed for rehabilitation based on a young person’s most serious offense, keeping in mind that eligibility for parole does not mean release. The tiered scheme embodies a legislative judgment that while the attributes of youth are not crime-specific, young people who commit more serious offenses generally require more time for rehabilitation, while young people who commit less serious offenses require less.

This view readily explains the flattening of myriad gradations of crimes and underlying conduct into three tiers of parole eligibility. Within a given tier, committing more crimes typically results in greater harm and culpability, but in view of the science on which the Legislature relied, it does not typically indicate less potential for growth and rehabilitation. Moreover, the 15-, 20-, and 25-year benchmarks are not random numbers; they track the steeply declining risk of offending as people mature beyond early adulthood. (See Lofstrom et al., Pub. Policy Institute of Cal., Are Younger Generations Committing Less Crime? (2023) p. 8, figure 1 [California age-crime curves showing that violent felony arrest rates decline significantly as people mature into their 30s and 40s].) This commonsense understanding of the eligibility framework, unlike the court’s culpability rationale, is tethered to the Legislature’s stated purpose of providing young offenders with an opportunity for release in light of their diminished culpability and capacity for change. Finally, the court says "[t]he most natural" inference is not that the Legislature "enacted a statute at odds with its own rehabilitative ends" but that it was "attempting to pursue other ‘ "(perhaps even contrary) ends as well." ’ " (Maj. opn., ante, 318 Cal.Rptr.3d at p. 531, 543 P.3d at p. 975.) Similar reasoning could have been deployed in Newland, D’Amico, Cooper, Moreno, and other cases. But Newland and D'Amico did not infer a cost-saving purpose for the exclusions at issue, Cooper did not accept an owner-negligence rationale for barring suits by owner-passengers against permissive drivers, and Moreno did not infer a fraud prevention purpose for excluding unrelated households from food stamps. In each case, the court measured the classification against the express or primary purpose of the statute and did not hesitate to find the "statute at odds with its own … ends." (Maj. opn., ante, 318 Cal.Rptr.3d at p. 531, 543 P.3d at p. 975.) The court simply ignores this case law in claiming its view of the statute is "[t]he most natural." (Ibid.)

It is easy to posit that the exclusion here reflects a legislative assessment of culpability and proper punishment; the court even says this is "necessarily" what the Legislature thought. (Maj. opn., ante, 318 Cal.Rptr.3d at pp. 530–531, 533, 543 P.3d at pp. 974–975, 976–977.) But there is not a single mention of such an assessment in the repeated consideration of this legislation. In contrast to the clear statements of rehabilitative aims, nothing in the legislative record states that young adult offenders serving LWOP are categorically more culpable than their parole-eligible peers serving de facto LWOP or other lengthy sentences for very serious crimes, It is not hard to imagine that such a claim, had it been asserted, would have invited skepticism (see post, 318 Cal.Rptr.3d at pp. 558–561, 543 P.3d at pp. 998–1001) as well as heightened attention to the racial skew of the affected group and its traceability to a history of racially inflected tough-on-crime policies. (See dis. opn. of Evans, J., post, 318 Cal.Rptr.3d at p. 574, 543 P.3d at p. 1011 ["The LWOP exclusion perpetuates extreme racial disparities in our criminal and juvenile justice systems."]; Com. on Revision of the Pen. Code, Annual Report and Recommendations (2021), at pp. 51, figure 24, 53 [among California LWOP inmates who were under age 26 at the time of the offense, 86 percent are people of color and 76 percent are Black or Latinx]; cf. Hetey & Eberhardt, Racial Disparities in Incarceration Increase Acceptance of Punitive Policies (2014) 25 Psychological Science 1949 [field study showing that awareness of extreme racial disparities in prison population made voters more accepting of punitive policies and less likely to support reform].)

The Legislature did not say why it excluded persons like Hardin, and we should not paper over this lacuna by imputing a purpose that the Legislature never had. As in past cases, we should take the Legislature’s actual statement of purpose at its word. Doing so, I would hold that the exclusion "fails to exhibit any fair and reasonable relationship to the stated legislative objective." (Hays, supra, 25 Cal.3d at p. 795, 160 Cal.Rptr. 102, 603 P.2d 19.)

III.

Even if we were to assume that the Legislature had concerns about culpability and appropriate punishment when it excluded young adults convicted of special circumstance murder from parole eligibility, we must still inquire whether such concerns provide a reasonable basis to distinguish the excluded group from others who are eligible for youth offender parole.

A.

As the Court of Appeal explained, the rationality of the exclusion "is belied by the statutory provisions that allow [a parole] hearing for individuals who have committed multiple violent crimes (albeit not special circumstance murder) and were sentenced to a technically parole-eligible indeterminate state prison term that is the functional equivalent of life without parole. (Cf. People v. Caballero, supra, 55 Cal.4th at p. 268, 145 Cal.Rptr.3d 286, 282 P.3d 291 [sentence of 110 years to life for three counts of attempted premeditated murder with firearm-use and criminal street gang enhancements ‘amounts to the functional equivalent of a life without parole sentence’]; [citation].) The crime of a 20-year-old offender who shot and killed his victim while attempting to commit robbery and was sentenced to life without parole (see § 190.2, subd. (a)(17)(A)) cannot rationally be considered more severe than those of a 20 year old who shot and killed his victim one day, committed a robbery the next, and was sentenced to an indeterminate term of 50 years to life (see §§ 190, subd. (a), 12022.53, subd. (d)), or who committed multiple violent crimes, like Caballero, and received a parole-eligible indeterminate life term that far exceeded his or her life expectancy." (Hardin, supra, 84 Cal. App.5th at p. 289, 300 Cal.Rptr.3d 243.) As suggested by the many cases cited above, persons sentenced to de facto LWOP but eligible for a youth offender parole hearing are "far from anomalous." (Id. at p. 289, fn. 10, 300 Cal.Rptr.3d 243.)

But there is more for us to consider than a litany of examples. As the Court of Appeal observed, the 2021 Annual Report and Recommendations (2021 Report) of the Committee on Revision of the Penal Code, a statutorily created committee of the California Law Revision Commission (Gov. Code, §§ 8280 et seq.), cited recent research showing that "special circumstance allegations could have been charged in 95 percent of all first degree murder convictions, leaving the decision whether a life without parole sentence may be imposed to the discretion of local prosecutors, rather than a matter of statewide policy. (2021 Report, at p. 51.)" (Hardin, supra, 84 Cal.App.5th at p. 290, 300 Cal. Rptr.3d 243, fns. omitted.)

The 95 percent figure comes from a study of over 27,000 California murder and manslaughter convictions between 1978 and 2002, led by the late Professor David Baldus. The findings, based on a representative sample of 1,900 cases, are comprehensively reported with methodological details in a pair of recent articles. (See Baldus et al., Furman at 45: Constitutional Challenges from California's Failure to (Again) Narrow Death Eligibility (2019) 16 J. Empirical Legal Studies 693 (Baldus study); id, at pp. 713, 714, table 2 [reporting the 95 percent figure]; Grosso et al., Death by Stereotype: Race, Ethnicity, and California's Failure to Implement Furman's Narrowing Requirement (2019) 66 UCLA L.Rev. 1394.) One of the authors, Professor Catherine Grosso, further reports in an amicus brief that among persons under the age pf 26 who were convicted of first degree murder, 98 percent could have been charged with one or more special circumstances, and that in Los Angeles County, the figure is 99 percent. (See also Baldus study, at pp, 719–723 [California has the highest percentage of homicide cases that are death eligible among all states].)

The court’s only direct response to these data is a brief comment that "the study appears to suggest that certain specific special circumstances, added through various amendments after the initial enactment of section 190.2, have led to the results found in the study," and that Hardin is in no position to challenge "[t]he special circumstance finding at issue in [his] own case [i.e., robbery-murder]" because it "is based on a provision of the law that dates back to the initial enactment of section 190.2" in 1973. (Maj. opn., ante, 318 Cal.Rptr.3d at pp. 536-537, 543 P.3d at pp. 979-980.) But this does not accurately characterize the Baldus study or its relevance to Hardin’s case. It ignores the fact that the Baldus study suggests the high rate of factually death-eligible cases among first degree murders is significantly attributable to the felony murder special circumstance, including robbery murder, that has existed since early iterations of section 190.2. (See Baldus study, supra, 16 J. Empirical Legal Studies at p. 729, fn. 122 [robbery felony-murder special circumstance accounted for 55 percent of factually death-eligible homicide cases during a period when that special circumstance required proof of intent to kill].) It also ignores the Baldus study’s citation to an earlier study of several hundred California murder convictions with an appellate decision between 1988 and 1992, which found that 84 percent of first degree murder cases were death-eligible under the statute. (Id. at p. 704, citing Shatz & Rivkind, The California Death Penalty Scheme: Requiem for Furman? (1997) 72 N.YU. L.Rev. 1283, 1332.) That earlier study, on the very page cited by the Baldus study, states: "The majority of first degree murders are felony murders, and felony murders are virtually all special circumstances murders. Thus, the felony murder special circumstances alone defeat any possibility of genuine narrowing." (Shatz & Rivkind, at p. 1332, fn. omitted.)

To be sure, charging decisions that result in a special circumstance conviction may attempt to target the most severe and culpable conduct among factually eligible cases. But it is undeniable that charging decisions are also influenced by many factors unrelated to offense severity and culpability, including the policies and priorities of the district attorney, the attitudes and demographies of the jury pool, the strength of the evidence, and available resources. As noted in an amicus brief filed here by the Prosecutors Alliance of California in support of neither party, "charging decisions have created racial, geographic, and temporal disparities between offenders who were charged, with special-circumstance murder and those who were charged with murder without a special circumstance." Layered on top of well-known racial and geographic disparities are "the vicissitudes of charging decisions (and politics) over time," the brief observes. Such temporal disparities "mean that someone was more likely to be convicted of special-circumstance murder" during the tough-on-crime era of the past than during more recent years "for effectively the same crime."

The court is correct that "Hardin does not argue that prosecutorial discretion itself offends equal protection." (Maj. opn., ante, 318 Cal.Rptr.3d at p. 535, 543 P.3d at p. 979.) But that is not the issue. The issue, on the court’s own view of legislative purpose, is whether it is rational to exclude young offenders convicted of special circumstance murder from parole eligibility on the ground that they are more culpable or have committed more severe offenses than their peers convicted of simple first degree murder. In other words, is special circumstance murder in actuality "a uniquely serious offense" (id. at p. 518, 543 P.3d at p. 964) for purposes of youth offender parole eligibility? No one disputes that "[s]pecial circumstance murder is an unquestionably grave offense, one that exacts an unimaginable toll on the lives of victims and those the victims leave behind." (Id. at p. 538, 543 P.3d at p. 981.) But does that distinguish special circumstance murder from all or any first degree murders in this context?

The panoply of charging factors unrelated to culpability, along with the sheer number of factually eligible cases among first degree murders, casts considerable doubt on the proposition. As Professor Grosso observes in her amicus briefing here, a special circumstance conviction cannot reasonably serve as a proxy for severity of the crime "when virtually all of the youthful first-degree offenders who are eligible for parole consideration committed crimes that qualify for sentencing under California’s special circumstances statute." (Accord, Hardin, supra, 84 Cal.App.5th at pp. 289–290, 300 Cal.Rptr.3d 243 ["[W]ith respect to first degree murder, any purported legislatively recognized distinction in culpability between individuals serving a parole-eligible indeterminate life sentence and those sentenced to life without parole is illusory."].)

B.

Instead of grappling with these data, today’s opinion says the Baldus study is "not part of the record in this case" and "ha[s] never been the subject of any sort of adversarial testing." (Maj. opn., ante, 318 Cal.Rptr.3d at p. 536, 543 P.3d at p. 979.) But the study’s findings are a prominent feature of the Court of Appeal’s reasoning as well as Hardin’s arguments and substantial amicus briefing in this court. Indeed, before we set this matter for oral argument, we took the affinnative step of directing, not inviting, the Attorney General to file an answer to the amicus briefing in this case. In his answer, the Attorney General did not take issue with the study’s findings or methodology.

Among the hundreds of pages of briefing in this case, only one amicus brief, filed by the San Bernardino County District Attorney, questions the reliability of the Baldus study. That brief asserts that the study’s data were "produced by inexperienced law students and recent graduates reading probation reports," that probation reports may omit information important to charging decisions, and that the study yielded the "strange" finding that there is a higher rate of factually death-eligible cases among voluntary manslaughters (47 percent) than among second degree murders (38 percent), though the reported rate among first degree murders is far higher (95 percent). But the amicus brief does not engage with the coding protocol described at length in the peer-reviewed study. The protocol sets forth a series of detailed rules for characterizing cases, acknowledges and addresses the limitations of probation reports, includes methods for curing insufficient information in a probation report, and notes that ultimately 11 percent of cases had insufficient information to permit characterization. (See Baldus study, supra, 16 J. Empirical Legal Studies at pp. 708–713 & fns. 81, 83.) Neither the court nor any party or amicus has identified any specific shortcoming of the research protocol.

While acknowledging the study’s findings, the Attorney General argues that "rational basis review does not require mathematical precision or a perfect fit." But no one is insisting on mathematical precision or a perfect fit. The issue is whether there is any reasonable relation between the classification drawn and the purported purpose of calibrating youth offender parole eligibility to offense severity and culpability. The Baldus study and related findings in Professor Grosso’s amicus brief are clearly relevant to that issue, and these findings have been brought to our attention in the tradition of a "Brandeis brief." (See Muller v. Oregon (1908) 208 U.S. 412, 419 & fn. t, 28 S.Ct. 324, 52 L,Ed. 551 [discussing the 113-page brief by then-attorney Louis Brandeis that documented social science research on the negative effects of long working hours on women’s well-being].) "[W]hen a question of fact is debated and debatable, and the extent to which a special constitutional limitation goes is affected by the truth in respect to that fact," courts "take judicial cognizance of all matters of general knowledge." (Id. at pp. 420–421, 28 S.Ct. 324.) This is a familiar practice in constitutional adjudication. (See, e.g., Graham, supra., 560 U.S. at p. 68, 130 S.Ct. 2011 ["As petitioner’s amici point out, developments in psychology and brain science continue to show fundamental differences between juvenile and adult minds."]; People v. Contreras (2018) 4 Cal.5th 349, 362–363, 229 Cal.Rptr.3d 249, 411 P.3d 445 [citing empirical research outside the record in rejecting an actuarial approach to defining life expectancy of juvenile offenders]; Johnson, supra, 60 Cal.4th at pp. 883–884, 183 Cal.Rptr.3d 96, 341 P.3d 1075 [citing empirical research outside of the record in determining that there is "more than a speculative possibility that sexual predators are more successful in manipulating minors to engage in oral copulation, as opposed to sexual intercourse"].) Does the court believe these decisions, in relying on "untested empirical findings," violate "multiple settled principles of judicial review"? (Maj. opn., ante, 318 Cal.Rptr.3d at p. 536, 543 P.3d at p. 979.)

If the concern is that the Baldus study has not been subject to adversarial testing, then the prudent course is to remand this case for factual development in the trial court. In D’Amico, we were careful to ensure adequate presentation of " ‘constitutional facts’ bearing upon the validity of the [statute excluding osteopaths from medical licensure] under the equal protection clause." (D’Amico, supra, 11 Cal.3d at p. 16, 112 Cal.Rptr. 786, 520 P.2d 10; see id. at pp. 9-10, 112 Cal.Rptr. 786, 520 P.2d 10 [observing that the Court of Appeal had initially declined to decide the equal protection issue and remanded the matter to the trial court for development of relevant facts].) In U.S. Steel, we similarly indicated that the equal protection issue could not be settled "[w]ithout a more complete record"; we said we lacked an adequate factual basis for evaluating how burdensome it would be to trackers and the PUC to determine for rate-setting purposes whether foreign steel is transported by common carrier or private vessel. (U.S. Steel, supra, 29 Cal.3d at p. 614, 175 Cal.Rptr. 169, 629 P.2d 1381.) And the Massachusetts high court, in a recent decision holding LWOP unconstitutional for 18- to 20-year-olds, noted that it had earlier "remanded the defendant’s case to the Superior Court for ‘development of the record with regard to research on brain development after the age of seventeen [in order to] allow us to come to an informed decision as to the constitutionality of sentencing young adults to [LWOP].’ " (Commonwealth v. Mattis (Mass. 2024) 493 Mass. 216, 224 N.E.3d 410, 416, fn. omitted.)

To the extent the court has similar concerns in this case, we should take a similar course. We should not dodge a key component of the equal protection claim on the ground that it was "not litigated in the trial court" (maj. opn., ante, 318 Cal. Rptr.3d at p. 536, 543 P.3d at p. 979) and then proceed to establish a precedent rejecting the equal protection claim. If "untested empirical findings" are the concern (ibid.), then why not remand this matter for factual development in light of the highly relevant information that has been brought to our attention? The court gives no reason. This head-in-the-sand approach — opting for less rather than more information in deciding a major constitutional question — hardly seems like a sound way to exercise judicial review.

C.

Ultimately, the core of the court’s reasoning is that our case law has held in the Eighth Amendment context that "section 190.2 adequately separates the most egregious first degree murders — those deserving of the most severe punishment available — from the rest" and that "[g]iven this body of case law, it is difficult to see how the Legislature that enacted section 3051 could have acted irrationally in singling out special circumstance murder as a particularly culpable offense." (Maj. opn., ante, 318 Cal.Rptr.3d at pp. 534, 535, 543 P.3d at pp. 978, 978.)

It is true that our capital cases have repeatedly upheld the special circumstances statute against claims that it does not properly serve the narrowing function required by the Eighth Amendment for imposition of the death penalty. But what the court leaves unsaid is that none of the cases in its lengthy string cites illuminates the underlying rationale of the Eighth Amendment holding; the cases simply refuse to revisit precedent or summarily reject the claim without analysis. (See maj. opn., ante, 318 Cal.Rptr.3d at pp. 534–535 & fn. 6, 543 P.3d at p. 978 & fn. 6 [citing 12 cases that contain no substantive analysis of the issue].)

Today’s decision cites People v. Green (1980) 27 Cal.3d 1, 61, 164 Cal.Rptr. 1, 609 P.2d 468, which "explained why the law treats robbery-murder as more culpable than simple murder." (Maj. opn., ante, 318 Cal.Rptr.3d at p. 535, 543 P.3d at p. 978.) But that case said nothing about whether the robbery-murder special circumstance, in operation, actually distinguishes crimes that are more culpable than simple murder offenses. The court also cites People v. Bacigalupo (1993) 6 Cal.4th 457, 24 Cal. Rptr.2d 808, 862 P.2d 808 (Bacigalupo) in asserting that special circumstances "mark a first degree murder [as] particularly egregious." (Maj. opn., ante, 318 Cal. Rptr.3d at p. 534, 543 P.3d at p. 977.) We said in that case: "In California, the special circumstances serve to ‘ "guide" ’ and ‘"channel"’ jury discretion ‘by strictly confining the class of offenders eligible for the death penalty.’ [Citation.] As the criteria in the California capital scheme that define the class of murders for which death is a potential penalty, the special circumstances set forth in section 190.2 must comport with Eighth Amendment requirements by providing not only clear and objective standards for channeling jury discretion, but also detailed and specific guidance, thus making the process for imposing a death sentence ‘ "rationally reviewable." ’ [Citation.] [¶] Under our death penalty law, therefore, the section 190.2 ‘special circumstances’ perform the … constitutionally required ‘narrowing’ function" of "circumscrib[ing] the class of murderers eligible for the death penalty." (Bacigalupo, at pp. 467–468, 24 Cal.Rptr.2d 808, 862 P.2d 808.) That was the extent of our analysis.

Bacigalupo described the intended function of special circumstances but undertook no empirical inquiry or factual analysis of the statute’s actual operation. That is because the main issue in Bacigalupo was not whether section 190.2 adequately performs the constitutionally required narrowing function. It was whether the sentencing considerations at the penalty phase (§ 190.3) must satisfy the Eighth Amendment’s narrowing principle. (Bacigalupo, supra, 6 Cal.4th at pp. 462–463, 24 Cal. Rptr.2d 808, 862 P.2d 808.) On that issue, we said that "when a capital punishment statute adequately narrows the class of death-eligible murderers, the Eighth Amendment does not require a further round of ‘narrowing’ at the sentence selection stage." (Id. at p. 475, 24 Cal.Rptr.2d 808, 862 P.2d 808.) Bacigalupo did not address whether special circumstances actually fulfill their intended narrowing function. Today’s opinion cites only one case that examined data on the operation of the special circumstances statute, People v. Frye (1998) 18 Cal.4th 894, 1028–1029, 77 Cal.Rptr.2d 25, 959 P.2d 183 (Frye), and acknowledges that "[o]ur treatment of the issue in Frye was admittedly terse, and it relied on a different study than the one on which Hardin now relies." (Maj. opn., ante, 318 Cal.Rptr.3d at p. 536, 543 P.3d at p. 979.) Given these qualifiers, it is not clear what persuasive value the court thinks Frye has here. The defendant in Frye argued that "virtually all first degree murders are death eligible" based on "a statistical analysis … of published appeals from murder convictions for the years 1988-1992." (Frye, at pp. 1028, 1029, 77 Cal.Rptr.2d 25, 959 P.2d 183; cf. Shatz & Rivkind, supra, 72 N.Y.U. L.Rev. at pp. 1326–1335.) Our opinion in Frye did not discuss or even mention the actual data. Instead, we rejected the claim in three short sentences: "Defendant’s argument notwithstanding, the special circumstances ‘are not overinclusive by their number or terms.’ ([People v. Arias (1996) 13 Cal.4th 92, 187, 51 Cal.Rptr.2d 770, 913 P.2d 980].) Nor have they been construed in an overly expansive manner. (Ibid.; see also People v. Morales (1989) 48 Cal.3d 527, 557–558, 257 Cal.Rptr. 64, 770 P.2d 244 [lying-inwait]; People v. Marshall (1990) 50 Cal.3d 907, 946, 269 Cal.Rptr. 269, 790 P.2d 676 [felony murder]; People v. Anderson (1987) 43 Cal.3d 1104, 1147, 240 Cal.Rptr. 585, 742 P.2d 1306 [felony murder].) Defendant’s statistics do not persuade us to reconsider the validity of these decisions." (Frye, at p. 1029, 77 Cal.Rptr.2d 25, 959 P.2d 183.) The cases cited in Frye shed no further light on the issue, and today’s opinion cites nothing else.

What we have, then, is a body of case law that has never grappled with empirical findings regarding the actual operation of special circumstances, much less with findings based on data as comprehensive as those in the Baldus study. On this point, the court does not and cannot disagree. Instead, today’s decision, like the cases it cites, simply piles citation upon citation. But when one follows the trail of citations in search of a foundational rationale or analysis, one comes up empty. "Truly, this is ‘turtles all the way down.’ " (Rapanos v. United States (2006) 547 U.S. 715, 754, 126 S.Ct. 2208, 165 L.Ed.2d 159 (maj. opn. of Scalia, J.).)

I suppose the Legislature could have posited that special circumstance murder is categorically worse than other murders based on the mere fact that this court has said it again and again. But the fact that our cases have said it does not make it so. Under rational basis review, " ‘the constitutionality of a statute predicated upon the existence of a particular state of facts may be challenged by showing to the court that those facts have ceased to exist.’ " (Brown, supra, 8 Cal.3d at p. 869, 106 Cal.Rptr. 388, 506 P.2d 212.) We have been shown information that forcefully challenges the rationality of including youthful offenders convicted of simple first degree murder in the parole scheme while excluding youthful offenders convicted of special circumstance murder. This information calls for serious engagement and analysis, not repeated citation to ipse dixit in our case law.

D.

Toward the end of today’s opinion, the court says it does not decide "the constitutionality of section 3051, subdivision (h) as it might arise in other as-applied challenges based on particular special circumstances or the factual circumstances of individual cases." (Maj. opn., ante, 318 Cal. Rptr.3d at p. 538, 543 P.3d at p. 981; see id. at p. 537, 543 P.3d at p. 980 ["While we do not foreclose the possibility of other challenges to the distinctions drawn by the special circumstance statute based on a more robust record and a more focused as applied inquiry."].) It is not clear what the court means in dangling this possibility. Does it mean that Hardin or someone like Hardin could bring a challenge to the felony murder special circumstance by comparing such cases with simple murder cases in which the felony murder special circumstance was not charged or not found true? Or does it mean that a litigant in Hardin’s position would have to bring a challenge specifically to the robbery-murder special circumstance, since that is the one he was convicted of? Or does it mean that a litigant would have to focus the equal protection claim not simply on a "particular special circumstance] ]" but more narrowly on a comparison to "individual cases" (id. at p. 538, 543 P.3d at p. 981) that did not result in a special circumstance charge or finding but involved similar facts?

Although the court’s lack of elaboration may be understandable, its gesture of purportedly leaving the door open to future challenges should invite some skepticism. For it is not clear how a litigant can succeed on a "more focused" claim (maj. opn., ante, 318 Cal.Rptr.3d at p. 537, 543 P.3d at p. 980) without running into the argument that even if there is disparate treatment within a subset of similar cases (defined by a particular special circumstance or set of factual circumstances), the overall classification of persons convicted of special circumstance murder as more culpable than persons convicted of simple murder is a " ‘gross generalization[ ] and rough accommodation[ ] that the Legislature seems to have made’ " and that " ‘we must accept’ " (id. at p. 532, 543 P.3d at p. 976). Indeed, given a prison population as large as ours, many young offenders serving LWOP could cite one or more cases involving equally if not more culpable conduct that received less punitive treatment. Presumably it would be easy for the court to reject such claims by moving the analysis up one or more levels of generality. What the court envisions as the proper level of "focus" is a mystery. The opendoor language reads like an attempt to cushion the blow of today’s decision; time will tell whether it offers only hollow hope.

IV.

Amidst all the legal reasoning in this case, what is missing are the facts of Tony Hardin’s crime, committed in 1989 at age 25, and a real-life understanding of why the Legislature raised the cutoff for youth offender parole eligibility to age 26. According to the Court of Appeal decision affirming his conviction, the evidence showed that Hardin was friends with his elderly neighbor, Norma Barber. One night, Hardin killed Barber in her home by strangling her and stole her jewelry, VCR, Vantage cigarettes, car keys, and other items because he was "desperate to buy drugs and had no money to do so." Hardin also stole two cans of beer from her fridge. The next morning, Hardin tried to buy cocaine from a drug dealer using Barber’s necklace. The dealer refused but offered Hardin drugs in exchange for a ride to pick up a drug supply. Hardin complied and drove Barber’s car, which had a personalized license plate. After they picked up the drugs, Hardin drove to a pawn shop and parked the car illegally in the lot. Inside the shop, Hardin exchanged Barber’s jewelry for $15 and filled out the pawn slip with his name and provided a thumbprint. Meanwhile, the car was ticketed by a traffic officer.

Hardin used the $15 to buy cocaine from the dealer. That evening, Hardin asked for more cocaine and used the VCR as pay- ment. He also offered Barber’s microwave and brought the dealer to Barber’s apartment. There, Hardin held the door open while the dealer took the microwave and other items. Hardin also lent the dealer Barber’s car, which he parked nearby, for additional cocaine. Sometime after, police came to question Hardin at his home. Hardin spoke to the police and denied ever driving Barber’s car. During the interview, police saw Vantage cigarette butts on an ashtray as well as cans of beer that Barber’s son had reported missing. Hardin was ultimately arrested, charged, and convicted. It does not take much to recognize that Hardin’s crime, driven by a drug habit and devoid of any sophistication, exhibited many of the hallmark features of youth: "recklessness, impulsivity, and heedless risk-taking" (Miller, supra, 567 U.S. at p. 471, 132 S.Ct. 2455); "inability to assess consequences" (id. at p. 472, 132 S.Ct. 2455); "inability to deal with police officers" (id. at p. 477, 132 S.Ct. 2455); and vulnerability to substance abuse (id. at p. 478, 132 S.Ct. 2455).

It is now almost 35 years later, and Hardin is 60 years old. I do not know whether he is a changed man. But no one disputes that the basic propositions of "science," "social science," and "common sense" (Miller, supra, 567 U.S. at p. 471, 132 S.Ct. 2455) on which the Legislature relied in expanding youth offender parole eligibility apply equally to all young adults, whatever their crimes. The Attorney General, citing amicus briefs filed by neuroscience scholars and by Human Rights Watch, says "[t]he People have never disputed that ‘[f]undamental changes in brain development’ may occur through age 25, or whether certain young adult offenders initially sentenced to life terms have been capable of demonstrating ‘remarkable’ reform." Some young offenders may be unable to show sufficient rehabilitation to gain parole; eligibility does not mean release. But the Legislature recognized that many young offenders — through normal maturation as well as years of hard work in coming to terms with their past conduct, repaying debts to victims and society, and bettering themselves and others — are capable of change and deserve a meaningful chance at life beyond prison walls.

Upon " ‘a serious and genuine judicial inquiry’ " (Newland, supra, 19 Cal.3d at p. 711, 139 Cal.Rptr. 620, 566 P.2d 254), I see no rational basis for extending youth offender parole eligibility to persons convicted of simple murder regardless of the number or severity of their crimes, while denying it to the Tony Hardins who have been condemned to die in prison for committing similar crimes in their youth. I again join numerous judges throughout the state in urging the Legislature to revisit this issue. (See, e.g., People v. Jackson (2021) 61 CalA.pp.5th 189, 202a–202b, 279 Cal.Rptr.3d 396 (conc. stmt. of Liu, J.) review den. June 9, 2021, S267812; see also maj. opn., ante, 318 Cal.Rptr.3d at pp. 538–539, 543 P.3d at p. 981.)

While I agree with the court’s decision to dispense with the "similarly situated" step of our equal protection doctrine, I dissent from the merits of today’s equal protection holding. I would affirm the judgment of the Court of Appeal.

Dissenting Opinion by Justice Evans

Tony Hardin committed a murder in 1989 when he was 25 years old. Hardin, who is African American, was convicted of special circumstance murder and sentenced to life without the possibility of parole (LWOP). In 2021, Hardin moved for a hearing pursuant to People v. Franklin (2016) 63 Cal.4th 261, 202 Cal.Rptr.3d 496, 370 P.3d 1053 (Franklin) to preserve evidence related to youth mitigating factors for an eventual youth offender parole hear- ing. The superior court denied the motion on the grounds he was ineligible for a youth offender parole hearing based on his LWOP sentence. (Pen. Code, § 3051, subd. (h).)^1a

The question presented is whether section 3051’s exclusion of youthful offenders from the youth offender parole eligibility scheme based on their LWOP sentence violates equal protection. I would hold that it does. The LWOP exclusion offends the Legislature’s only express and articulated purpose of the youth offender parole eligibility scheme and lacks rationality. The exclusion bears the taint of racial prejudice and perpetuates extreme racial disparities plaguing our juvenile and criminal justice systems. Thus, I conclude it fails any mode of rational basis review. With respect, I dissent.

I.

Level of Scrutiny & Lens of Deference

Several amici urge the court to apply strict scrutiny. Under the federal equal protection clause, strict scrutiny only applies where the challenged regulation involves a fundamental right or a suspect classification. (Massachusetts Bd. of Retirement v. Mv.rgia (1976) 427 U.S. 307, 312, 96 S.Ct. 2562, 49 L.Ed.2d 520.) Hardin concedes it does not apply here.

Even assuming strict scrutiny does not apply to the exclusion at issue here, rational basis review still requires us to engage in a " ‘serious and genuine’ " inquiry between the classification and the legislative objective at issue in this case. (Newland v. Board of Governors (1977) 19 Cal.3d 705, 711, 139 Cal.Rptr. 620, 566 P.2d 254.) While I generally agree that this exclusion fails under any form of rational basis review (see dis. opn. of Liu, J., ante), I write separately to explain that the nature of the deprivation and whether the classification can be attributed to bias should inform this court’s mode of deference. A rational basis review that considers racial disparities in classifications is particularly appropriate in cases such as this "where the challenged classification appears to impose a substantially disproportionate burden on the very class of persons whose history inspired the principles of equal protection." (State v. Russell (Minn. 1991) 477 N.W.2d 886, 889.)^2a

The LWOP exclusion disproportionately impacts Black and Brown youth. It perpetuates racial disparities in LWOP sentences for youthful offenders. While perhaps unintentional, it nonetheless embodies racial bias that has plagued our criminal and juvenile justice systems since their inception. The stakes could not be higher. The differential treatment means the difference between whether young people who are otherwise worthy of release will die in prison or will return to society following a grant of parole. The impact of that deprivation on an individual and their family cannot be overstated. As the majority recognizes, the deprivation also extends to a societal level. Society benefits from the return of rehabilitated individuals sentenced to LWOP to their communities. (See Human Rights Watch, I Just Want to Give Back: The Reintegration of People Sentenced to Life Without Parole (June 2023) p. 4 https://www.hrw.org/report/2023/06/28/i-just-want-to-give-back/reintegration-of-people-sentenced-to-life-without-parole> [as of Mar. 4, 2024].)^3a

Notably, a sister court has recognized that the equal protection guarantee of its state constitution "hold[s] lawmakers to a higher standard of evidence when a statutory classification demonstrably and adversely affects one race differently than other races, even if the lawmakers’ purpose in enacting the law was not to affect any race differently." (Fletcher Props. v. City of Minneapolis (Minn. 2020) 947 N.W.2d. 1, 19 (Fletcher).) In such cases, the Minnesota equal protection clause demands "actual (and not just conceivable or theoretical) proof that a statutory classification serves the legislative purpose." (Ibid.) The parties here do not present any argument that the equal protection guarantee of California’s constitution likewise requires this heightened degree of rational basis analysis. In my view, the resolution of that question is not determinative in this case. Our mode of deference, however, must take into account whether the challenged classification results in demonstrable and adverse racial discrimination.

With these principles in mind, I turn to the question of whether section 3051’s LWOP exclusion fails rational basis review.

II.

Application of the Rational Basis Standard

I begin with the articulated purpose of section 3051. The parties agree the articulated objective of section 3051 is to provide youthful offenders with a meaningful opportunity to obtain release upon a showing of maturation and rehabilitation. In extending youth offender parole eligibility to those who committed a crime before they were 26 years of age, the Legislature relied on brain science establishing the attributes of youth are maintained until age 26.^4a The LWOP exclusion — which impacts youthful offenders who were 18 to 25 years old at the time of the offense — is entirely at odds with that statutory objective and the brain science motivating the enactment and extension of youth offender parole eligibility. Youthful offenders sentenced to LWOP, as a class, have the same capacity for maturation and rehabilitation as their parole-eligible counterparts. Their youthful age — not their offense or sentence — is what makes them less morally culpable and more likely to rehabilitate themselves such that they should be entitled to a youthful offender parole hearing.

The majority speculates the Legislature excluded youthful offenders sentenced to LWOP to account for their culpability based on their offense of special circumstance murder. This purported purpose not only conflicts with the statute’s actual purpose, but there is nothing in the statute or its history indicating the Legislature was motivated by any "culpability" rationale. (Maj. opn., ante, 318 Cal.Rptr.3d at pp. 530, 531-532, 543 P.3d at pp. 974, 975–976; see dis. opn. of Liu, J., ante, 318 Cal.Rptr.3d at pp. 552–559, 543 P.3d at pp. 992–999.) Contrary to the majority’s hypothesis, the Legislature’s decision to teth- er the youthful offender parole eligibility date to a youthful offender’s controlling offense does not reflect rational judgments about culpability. (See dis. opn. of Liu, J., ante, 318 Cal.Rptr.3d at pp. 554-557, 543 P.3d at pp. 995-997.) What’s more, the framework has little to no relevance to the Legislature’s choice to enact the exclusion at issue before us. It is one thing to designate varying parole eligibility dates based on a youthful offender’s controlling offense. It is quite another to exclude a class of youthful offenders from parole eligibility entirely based on their sentence, given the underlying rationale for youth offender parole.

Even assuming we can impute a legislative rationale that is contrary to a statute’s purpose, a "culpability" rationale for the LWOP exclusion here is irrational. The hallmarks of youth and the heightened potential for rehabilitation are not crimespecific. (Miller v. Alabama. (2012) 567 U.S. 460, 473, 132 S.Ct. 2455, 183 L.Ed.2d 407.) In retaining the attributes of youth until age 26, as the Legislature recognized, youthful offenders "cannot with reliability be classified among the worst offenders." (Roper v. Simmons (2005) 543 U.S. 551, 569, 125 S.Ct. 1183, 161 L.Ed.2d 1.) As noted, youthful offenders who have been sentenced to LWOP are just as capable of becoming rehabilitated as their peers. Given the neuroscience, excluding youthful offenders from parole eligibility based on their offense does not make rational sense.

The imputed "culpability" rationale is also belied by uncontroverted evidence presented to the Legislature and this court.^5a As the Committee on Revision of the Penal Code noted, recent research has shown that 95 percent (in one study, 98 percent) of first degree murders could be charged as special circumstance murder and that factors other than culpability — including race — impact whether youth are convicted of special circumstance murder or simple first degree murder. (Com. on Revision of the Pen. Code, Annual Report and Recommendations (2021) pp. 51–52 http://clrc.ca.gov/CRPC/Pub/Reports/CRPC_AR2021.pdf> [as of Mar. 4, 2024] (Annual Report).) According to the Committee, 96 percent of LWOP sentences are based on a special circumstance murder. (Id. at p. 51.) Hardin was convicted of a felony murder special circumstance. Felony murder special circumstance accounts for over 50 percent of the LWOP sentences for special circumstance murder. (Id. at p. 52.) Unlike other special circumstances, the felony murder special circumstance does not require a mens rea of intent to kill. (See People v. Anderson (1987) 43 Cal.3d 1104, 1138-1139, 240 Cal. Rptr. 585, 742 P.2d 1306.)

Black people are disproportionately convicted of the felony-murder special circumstance. (Annual Report, supra, at p. 52.) As the Annual Report noted, 42 percent of people convicted of felony murder special circumstance are Black "compared to only 34% of the overall first-degree murder population and 26% of the second-degree murder population." (Ibid.) "Rates of felony murder special circumstance convictions also vary by the intersection of race and age. … Black individuals sentenced to LWOP for felony murder are much more likely to be younger at the time of offense than their White counterparts. In fact, almost half of people who were sentenced to LWOP through felony murder for offenses that took place when they were under the age of 21 are Black." (Special Circumstances Conviction Project, Life Without Parole and Felony Murder Sentencing in California (2023) p. 9 https://csw.ucla.edu/wp-content/uploads/2023/08/SCCP-Report11.pdf> [as of Mar. 4, 2024].)

While racial disparities exist across age groups, racial disparities are most prevalent "among people who were 25 or younger at the time of the offense and received a life without parole sentence — 86% are people of color." (Annual Report, supra, at p. 53.) The total LWOP population in California is over 5,000, and 62 percent are youthful offenders. (Id. at pp. 50, 53.) Of the roughly 3,100 youthful offenders sentenced to LWOP, 38 percent are Black, 38 percent are Latinx, 14 percent are White, 2 percent are Asian or Pacific Islander, 2 percent are American Indian/Alaskan Native, and 7 percent are "other." (Id. at p. 51.) In contrast, for the overall LWOP population, 35 percent are Black, 35 percent are Latinx, and 21 percent are White, with the same percentages for the remaining demographic groups. (Ibid.) The seven percent point differential (86 percent of youthful offenders sentenced to LWOP are people of color, compared to 79 percent of the overall LWOP population are people of color) is due to an increased rate in sentencing Black and Latinx youth to LWOP, and a decreased rate in sentencing White youth to LWOP.^6a (Ibid.) "African American youth are sentenced to life without

parole at a rate that is 18.3 times the rate for whites. Hispanic youth in California are sentenced to life without parole at a rate that is five times that for white youth in the state." (Human Rights Watch, When I Die, They’ll Send Me Home: Youth Sentenced to Life Without Parole in California (Jan. 2008) Vol. 20, No. 1 (G) p. 4 http://www.hrw.org/reports/2008/us0108/us0108web.pdf> [as of Mar. 4, 2024].) This racially disparate impact makes it especially important that we evaluate the classification against the Legislature’s articulated purpose and reject the exclusion as irrational. (See Fletcher, supra, 947 N.W.2d at 19.)

The Legislature was well aware of the racial disparities in LWOP sentences for youthful offenders, as the legislative history of various bills relating to the youth offender parole eligibility scheme includes discussion of these disparities.^7a Although the Legislature has enacted various remedial measures to address racism in our justice and carcerai systems, the question remains: why did the Legislature ignore the brain science and disparate impact of the LWOP exclusion on young people of color? The legislative history does not provide a definitive answer to this question. The LWOP exclusion, however, perpetuates severe racial disparities and, given its historical context, bears the taint of prejudice against Black and Brown youth.

III.

The LWOP Exclusion Perpetuates Racial Bias Against Black and Brown Youth

The historical context of the LWOP exclusion illuminates its origin and the moti- vating force behind it. "To determine the validity of the enactment … it must be viewed in light of its historical context and the conditions existing prior to its enactment." (Mulkey v. Reitman (1966) 64 Cal.2d 529, 534, 50 Cal.Rptr. 881, 413 P.2d 825.) "[T]he judicial branch can rely on history and context on issues of race to the same extent that courts have always relied on history and context to analyze all other issues." (State v. Hawkins (2022) 200 Wash.2d 477, 519 P.3d 182, 196.) "The way to stop discrimination on the basis of race is to speak openly and candidly on the subject of race, and to apply the Constitution with eyes open to the unfortunate effects of centuries of racial discrimination." (Schuette v. Coalition to Defend Affirmative Action Integration and Immigrant Rights and Fight for Equality By Any Means Necessary (2014) 572 U.S. 291, 381, 134 S.Ct. 1623, 188 L.Ed.2d 613 (dis. opn. of Sotomayor, J.); see id. at pp. 380–381, 134 S.Ct. 1623.)

The historical context of the LWOP exclusion demonstrates it was motivated — consciously or not — by racial bias, including racial stereotypes and myth. The provision excluded a group of youthful offenders based on their LWOP sentence — a metric that is not reflective of their culpability or their potential for rehabilitation, and a sentence with a significant disparate impact on Black and Brown youth. The Legislature enacted the LWOP exclusion against the backdrop of the now-debunked "superpredator" myth. The myth distorted policy makers’ collective understanding of youth as a mitigating circumstance and instead treated it as an aggravating circumstance and specifically demonized young Black males. The LWOP exclusion tracks that myth.

In the mid-1990s, Princeton University Professor John J. DiIulio, Jr., warned of an approaching violent crime surge perpetrated by "tens of thousands of severely morally impoverished" and "super crimeprone young males … on the horizon." (DiIulio, The Coming of the Super-Predators (Nov. 27, 1995) Washington Examiner https://www.washingtonexaminer.com/magazine/1558817/the-coming-of-the-super-predators/> [as of Mar. 4, 2024].) According to DiIulio, "[A]s long as their youthful energies hold out, they will do what comes ‘naturally’: murder, rape, rob, assault, burglarize, deal deadly drugs, and get high." (Ibid.) Criminologist James Alan Fox likewise warned that " ‘[u]nless we act today, we’re going to have a bloodbath when these kids grow up.’ " (Mills et al., Juvenile Life Without Parole in Law and Practice: Chronicling the Rapid Change Underway (2016) 65 Am. U. L.Rev. 535, 582.)

The superpredator myth particularly focused on Black youth. DiIulio claimed "[t]he surge in violent youth crime has been most acute among black inner-city males" (DiIulio, supra, Washington Examiner) and predicted that "as many as half of [the] juvenile super-predators could be young black males." (DiIulio, My Black Crime Problem, and Ours (Spring 1996) 6 City J. 19 https://www.city-journal.org/html/my-black-crime-problem-and-ours-11773.html> [as of Mar. 4, 2024].) In reversing a trial court’s denial of a motion to correct an illegal sentence, the Connecticut Supreme Court explained, "[T]he superpredator myth employed a particular tool of dehumanization — portraying Black people as animals. [Citation.] … The superpredator metaphor invoked images of packs of teens prowling the streets. The news coverage in the mid-1990s, which depicted ‘young Black males, showing them [handcuffed] and shackled, held down by [the] police, or led into courtrooms wearing orange jumpsuits’ … [citation] … left little doubt that the ‘packs’ were Black teens." (State v. Belcher (2022) 342 Conn. 1, 268 A.3d 616, 626, 627 (Belcher).) "Under its influence, all too many Black and brown children were explicitly or tacitly classified as ‘juvenile superpredators’ and treated as irredeemable monsters." (State v. Anderson (2022) 200 Wash.2d 266, 516 P.3d 1213, 1227 (dis. opn. of González, C. J.).)

The superpredator myth "turn[ed] upside down the constitutional mandate of Roper and its progeny. By labeling a juvenile as a superpredator, the very characteristics of youth that should serve as mitigating factors in sentencing — impulsivity, submission to peer pressure, deficient judgment — are treated instead as aggravating factors justifying harsher punishment." (Belcher, supra, 268 A.3d at p. 629.) "The ‘superpredator’ was constructed as the ultimate other, as possessing all the characteristics that innocent young children do not. … And because the ‘superpredator’ was the antithesis of childhood, it was slyly constructed as young, Black, and male." (Nunn, The End of Adolescence: The Child as Other: Race and Differential Treatment in the Juvenile Justice System (2002) 51 DePaul L.Rev. 679, 713.)

Empirical evidence quickly demonstrated that the superpredator myth was baseless and false. Between 1994 and 2009, the juvenile crime rate dropped by half. (Southerland, Youth Matters: The Need to Treat Children Like Children (2015) 27 J. Civ. Rights & Economic Development 765, 777 (Southerland).) There was "a fifty-six percent decline in homicides committed by juveniles from 1993 to 1998, and a thirty percent decline in overall juvenile crime during the same period." (Barton, Reconciling the Burden: Parental Liability for the Tortious Acts of Minors (2002) 51 Emory L.J. 877, 879.) In California, "from 1980 to 2016, the arrest rate among those 17 or younger dropped by 84 percent." (Lofstrom et al., Public Policy Institute of Cal., New Insights into California Arrests: Trends, Disparities, and County Differences (Dec. 2018) p. 3 https://www.ppic.org/wp-content/uploads/new-insights-into-california-arrests-trends-disparities-and-county-differences.pdf> [as of Mar. 4, 2024].) "Moreover, the predictions that youth of color would be primarily responsible for increases in violent crime were proven false. The fluctuations in juvenile homicide rates during the last two decades have not been specific to any demographic groups, peaking in 1994 for both African-American and white teenagers before falling through the year 2000." (Southerland, supra, 27 J. Civ. Rights & Economic Development at p. 777.)

Despite the drop in juvenile crime and arrest rates, the overall size of the incarcerated juvenile population grew in response to the superpredator myth, disproportionately impacting youth of color. "[F]our out of five youth newly held in detention between 1983 and 1997 were juveniles of color. The transfer of juveniles of color to adult court was equally, if not more, disproportionate. … These numbers persist even today." (Moriearty & Carson, Cognitive Warfare on Young Black Males in America (2012) 15 J. Gender, Race & Justice 281, 300–301, fn. omitted.) "Ultimately, the sinister connections between race, crime, and youth led to punitive sanctions, like life without parole, for young offenders." (Southerland, supra, 27 J. Civ. Rights & Economic Development at p. 781.) As a result, Black and Brown young people were sentenced to LWOP at extreme and disparate rates in California.^8a While empirical evidence demonstrated that the superpredator myth was baseless and false, the myth "tapped into and amplified racial stereotypes that date back to the founding of our nation." (Belcher, supra, 268 A.3d at p. 626.) The superpredator myth "relied heavily on ‘racist imagery and stereotypes’ and harkened back to ‘historic representations of African Americans [and other people of color] as violence-prone, criminal and savage." (Southerland, supra, 27 J. Civ. Rights & Economic Development at p. 773.)

The superpredator myth is one of many incarnations of racism that have plagued our criminal and juvenile justice systems since their inception. For example, once juvenile courts became more accessible to Black youth in the mid-1900s, the justice system shifted away from a rehabilitative objective and became more punitive. (Lapp, Young Adults & Criminal Jurisdiction (2019) 56 Am. Crim. L.Rev. 357, 386, citing Ward, The Black Child-Savers: Racial Democracy and Juvenile Justice (2012) p. 4.) "[T]he increase in disproportionate minority contact with juvenile court overlaps with the decline of the rehabilitative ideal and the rise of a more punitive juvenile court." (Lapp, at p. 386.) At the same time, politicians began treating youth not as individuals in need of guidance, support, and perhaps treatment — but as looming forces threatening to destroy public safety. (See Henning, The Challenge of Race and Crime in a Free Society: The Racial Divide in Fifty Years of Juvenile Justice Reform, 86 Geo. Wash. L.Rev. 1604, 1618-1620 (Henning).) The superpredator myth was an apex of racial prejudice in criminal and juvenile justice policy and catalyzed "nearly every state in the country to step up the sentencing and punishment of juveniles." (Belcher, supra, 268 A.3d at p. 628; see Henning, supra, 86 Geo. Wash. L.Rev. at p. 1620.) It also coincided with a proliferation in the use of LWOP sentences — particularly for Black and Brown youthful offenders. (See pp. 7–8, ante; see also Mills et al., The Phillips Black Project, No Hope: Re-Examining Lifetime Sentences for Juvenile Offenders (2015) p. 10 https://staticl.squarespace.com/static/55bd511ce4b0830374d25948/t/5600cc20e4b0f36b5caabe8a/1442892832535/JLWOP+2.pdf> [as of Mar. 4, 2024].) "Starting in 1992, the height of the superpredator panic, a black juvenile arrested for homicide has been twice as likely to be sentenced to LWOP as his white counterpart." (Id. at p. 9.)

In California, the superpredator myth animated legislation underpinning LWOP sentences for youthful offenders. For example, as a direct result of the superpredator myth, voters passed Proposition 21 in 2000. (See de Vries, Guilt By Association: Proposition 21's Gang Conspiracy Law Will Increase Youth Violence in California (2002) 37 U.S.F. L.Rev. 191, 197; see also Assem. Com. on Public Safety, Analysis of Sen. Bill No. 1391 (2017–2018 Reg. Sess.) as amended May 25, 2018, pp. 4–5 [acknowledging the shift towards punitive treatment of youth "was fueled by media’s portrayal of youth as ‘superpredators,’ consistent with the era’s tough on crime attitude. … In 2000, Proposition 21 again dramatically shifted California’s criminal justice policies"].) Proposition 21, among other things, created the gang-murder special circumstance and allowed — and sometimes mandated — charging children as young as 14 years old directly in adult criminal court. In advocating for the initiative’s passage, proponents adopted the language of the superpredator myth, perpetuating racial prejudice and capitalizing on dire "predictions of a juvenile crime wave." (Voter Information Guide, Primary Elec. (Mar. 7, 2000), argument in favor of Prop. 21, p. 48.) Contrary to the ballot material claims, juvenile crime had, in fact, been declining since 1993. (See pp. 10–11, ante.)

Even after the superpredator myth was exposed as false and the system began to refocus on rehabilitation, Black youth continue to disproportionately "experience the devastating effects of legislative and policy shifts that undermined the core rehabilitative philosophy of American juvenile courts in the wake of the superpredator myth." (Henning, supra, 86 Geo. Wash. L.Rev. at p. 1622.) They continue to be viewed as older and more culpable than White youth (id. at p. 1627) and, as noted above, have experienced the disproportionate imposition of LWOP sentences. Passed in the wake of the superpredator myth, the LWOP exclusion at issue here is part of this legacy of dehumanization and harm against Black and Brown youth.

* * *

The LWOP exclusion perpetuates extreme racial disparities in our criminal and juvenile justice systems. The historical and invidious discrimination against Black and Brown youth in criminal and juvenile justice policy provides important context when analyzing whether the exclusion has

a rational basis. The LWOP exclusion is particularly striking since the Legislature otherwise recognizes that youthful offenders as a class have diminished moral culpability. Particularly given the context of the LWOP exclusion’s enactment and its discriminatory impact, the statutory classification must serve the Legislature’s expressed purpose — to provide youthful offenders with a meaningful opportunity to obtain release upon a showing of maturation and rehabilitation. In light of that purpose and lack of any difference in the brain development or capacity for rehabilitation between excluded and non-excluded young people, the LWOP exclusion is irrational.

This case calls on us to correct a legacy of casting Black and Brown youth as predatory, remorseless, and irredeemable, older than they are, and treated differently from White youth. The equal protection clause demands that lawmakers extend the mercy, dignity and grace embodied in the youthful offender parole eligibility scheme to all youth — regardless of the crimes of which they were convicted. As a class, they all are less morally culpable and are more likely to become rehabilitated based on accepted scientific evidence regarding adolescent brain development. The majority has avoided this heed with the hollow promise of another day. I urge the Legislature to correct itself by ridding section 3051 of the LWOP exclusion and extending youth offender parole eligibility to all individuals who were convicted in their youth.

I respectfully dissent.