ORDER
On November 4, 2008, the California electorate passed Proposition 9, the “Victims’ Bill of Rights Act of 2008: Marsy’s Law.” Proposition 9 changed numerous aspects of California’s parole system, including the availability and frequency of parole hearings for prisoners not initially found suitable for parole.
Plaintiffs, eight individuals seeking to represent a class of California state prisoners currently serving life sentences with possibility of parole, bring a claim under 42 U.S.C. section 1983 arguing that these changes violate both the Ex Post Facto and Due Process clauses of the United States Constitution. Plaintiffs have also brought eight other claims not at issue in this order.
Plaintiffs move for a preliminary injunction barring enforcement of these changes. *1109 Defendants conversely move to dismiss plaintiffs’ challenge to Proposition 9, and by the same motion seek dismissal of plaintiffs’ ninth claim, which challenges Article V, Section 8(b) of the California Constitution. The ninth claim and the challenge thereto are the subject of a separate concurrently filed order.
The court resolves these motions on the papers and after oral argument. For the reasons stated herein, the court grants plaintiffs’ motion for a preliminary injunction, and grants in part and denies in part defendants’ motion to dismiss.
I. BACKGROUND
Proposition 9 changed numerous aspects of the California parole system. Plaintiffs challenge only the changes to the scheduling of parole hearings for prisoners who are not found suitable for parole at their initial hearing. To describe those changes, the court first surveys the California parole system, then discusses the hearing deferral process prior to Proposition 9, and finally describes the new system.
A. California’s Process for Parole Determinations
California’s parole system is overseen largely by the Board of Parole Hearings. The Board of Parole Hearings was formerly the Board of Prison Terms, and is hereinafter referred to as the Board. The Board has “the power to allow prisoners imprisoned in the state prisons ... to go upon parole outside the prison walls and enclosures.” Cal. Pen.Code § 3040. The penal code provides statutory guidance for the exercise of this power, and the Board has enacted regulations further specifying the parole process, pursuant to its authority under California Penal Code § 5076.2.
All prisoners whose sentences include life but do not include “without possibility of parole,” become “eligible” for parole after serving a minimum number of years, marked by the arrival of the prisoner’s “minimum eligible parole date.” Cal.Code Regs. tit. 15 §§ 2000(b)(3), (b)(67). This date is determined by the prisoner’s sentence and any credit the prisoner has earned for good conduct. A prisoner with a straight “life” sentence and maximum good conduct credit becomes eligible for parole in seven years. Cal. Pen.Code § 3046(a)(1). Parole eligibility comes later for prisoners whose sentences specify a minimum term, e.g., “15 to life.” For prisoners convicted of first or second degree murder prior to 1998, the minimum term is the term of years provided by their sentence — in the example above, fifteen years — reduced by up to one third for good behavior.
Once a prisoner is eligible for parole, the Board convenes a panel to determine whether the prisoner is to be released on parole, and if so, when. 1 The first inquiry is whether the prisoner is “suitable” for release. Cal. Pen.Code § 3041. The panel “shall” find the prisoner suitable for parole, and move on to the next step of setting a release date,
unless it determines that the gravity of the current convicted offense or offenses, or the timing and gravity of current or past convicted offense or offenses, is such that consideration of the public safety requires a more lengthy period of incarceration for this individual, and that a parole date, therefore, cannot be fixed at this meeting.
Id.
§ 3041(b). In making this determination, “the fundamental consideration in parole decisions is public safety ... [which]
*1110
involves an assessment of an inmate’s
current
dangerousness.”
In re Lawrence,
Because suitability is a function of a prisoner’s current dangerousness, a finding that the prisoner is unsuitable for parole is not permanent. If the panel determines that the prisoner is presently unsuitable for parole, the panel “defers” the parole hearing until a later date, at which point a panel will determine whether intervening changes have rendered the prisoner suitable. Cal. Pen.Code § 3041.5. The deferral process is discussed in the following sections.
If the panel instead determines that the prisoner is suitable for parole, the next step is to set a date for release. The panel determines the “base term” for the prisoner, which is based on “the gravity of the base offense, taking into account all of the circumstances of the crime.” CaLCode. Regs. tit. 15 § 2282(a). The actual calculation is performed by reference to matrices enumerating base terms for the circumstances of various offenses.
Id.
§§ 2282-2289. The panel then considers whether a departure from the base term is warranted by any of a number of other factors specifically enumerated by the regulations, such as the prisoner’s good behavior.
Id.
§ 2404. If the time already served by the prisoner exceeds the term calculated by this method, the Board tentatively designates the prisoner for release. If not, the board will set a release date for a time after the completion of this term.
See In re Jackson,
This system plainly contemplates that some prisoners will be found suitable for parole prior to the expiration of the term calculated in the second step. However, plaintiffs assert that “Since 1990, ... it is rare for a life prisoner to be found suitable for and granted parole before he is well beyond any punishment term that can be imposed for his offense.” 3 Pis.’ Mem. *1111 Supp. Mot. Prelim. Inj., 16 n. 11. Defendants have not contested this assertion. Plaintiffs further assert that “All the named plaintiffs are beyond any terms that would be (or have been) set for them and all will be released as soon as parole is granted and the Governor lets the grant stand.” Id. The court notes, however, that the class definition is not so limited, and the class potentially includes unnamed class members who have not completed such terms.
After the panel makes a decision as to suitability or unsuitability, the Governor has an opportunity to review and reverse or modify that decision.
See
Cal. Const. Art. V § 8(b); Cal.Penal Code §§ 3041.1, 3041.2;
In re Johnson,
The effect of the Penal Code section 3041 and the implementing regulations is to vest all California prisoners “whose sentences provide for the possibility of parole with a constitutionally protected liberty interest in the receipt of a parole release date.”
Irons v. Carey,
B. The Deferral Process Prior to Proposition 9 4
Prior to the amendments provided by Proposition 9, when a panel determined that a prisoner was unsuitable for parole, the length of deferral was determined by California Penal Code section 3041.5(b)(2) (2008). This section provided that when the Board found a prisoner unsuitable for parole,
The board shall hear each case annually thereafter, except the board may schedule the next hearing no later than the following:
(A) Two years after any hearing ... if the board finds that it is not reasonable to expect that parole would be granted at a hearing during the following year ...
(B) Up to five years after any hearing at which parole is denied if the prisoner has been convicted of murder, and the board finds that it is not reasonable to expect that parole would be granted at a hearing during the following years .... If the board defers a hearing five years, the prisoner’s central file shall be reviewed by a deputy commissioner within three years at which time the deputy commissioner may direct that a hearing be held within one year.
*1112 In determining how long to defer a hearing, and in making suitability determinations at the subsequent hearings, the panel applies the same criteria used for the initial suitability determination. Cal. Pen. Code § 3041.5(b)(2)(A)-(B) (2008), Cal. Code Regs. tit. 15, §§ 2268(b), 2270(d). 5 Thus, the panel evaluates whether the factors informing its assessment of the prisoner’s potential threat to public safety are likely to change; if so, when; and whether these changes will be sufficient to render the prisoner suitable for parole.
Plaintiffs provide several statistics illustrating the manner in which this statute was applied. According to the Board, in 2007, 35 percent of prisoners who were denied parole following a hearing were given deferral periods of one year, 32 percent received deferrals of two years, and 33 percent received deferrals of three years or more. In 2008 (through December 10), these percentages were 40, 33, and 27 percent, respectively. 6 Pis.’ Mem. Supp. Mot. Prelim. Inj. 14, n. 9.
Once a deferred hearing date had been set, that hearing date could potentially be advanced. If the deferral was for five years, the Board was obliged to review the prisoner’s situation at three years to determine whether the hearing should be advanced. Cal. Pen.Code § 3041.5(b)(2)(B). Moreover, a prisoner could separately request an advanced hearing date, although the former statute provided no formal mechanism for such requests.
In re Jackson,
C. The Proposition 9 Amendments to The Deferral Process
Proposition 9 drastically altered the deferral process, replacing former subsection 3041.5(b)(2) with the following, now codified at subsection (b)(3):
The board shall schedule the next hearing, after considering the views and interests of the victim, as follows:
(A) Fifteen years after any hearing at which parole is denied, unless the board finds by clear and convincing evidence that the criteria relevant to the setting of parole release dates enumerated in subdivision (a) of Section 3041 are such that consideration of the public and victim’s safety does not require a more lengthy period of incarceration for the prisoner than 10 additional years.
(B) Ten years after any hearing at which parole is denied, unless the board finds by clear and convincing evidence that ... consideration of the public and victim’s safety does not require a more lengthy period of incarceration for the prisoner than seven additional years.
(C) Three years, five years, or seven years after any hearing at which parole is denied, because ... consideration of the public and victim’s safety requires a more lengthy period of incarceration for the prisoner, but does not require a more lengthy *1113 period of incarceration for the prisoner than seven additional years.
Soon after Proposition 9 was passed, the Board issued an Administrative Directive identifying various effects of the proposition. BPH Administrative Directivé No. 08/01, Regulatory Sections Impacted by Proposition 9, December 8, 2008. One such effect, which was repeatedly stated in the directive, was that the Board now had “no discretion to set a denial period for any term other than those enumerated,” i.e., for a period other than “15, 10, 7, 5, or 3 years.” Id.
The amended statute provides two ways in which a deferred hearing date may be changed once it has been set. First, the Board
may in its discretion, after considering the views and interests of the victim, advance a hearing set pursuant to paragraph (3) to an earlier date, when a change in circumstances or new information establishes a reasonable likelihood that consideration of the public and victim’s safety does not require the additional period of incarceration of the prisoner provided in paragraph (3).
Cal. Pen.Code § 3041.5(b)(4). Second, an inmate may request that the board advance the hearing. Id. § 3041.5(d). Such a request must “set forth the change in circumstances or new information” required by subsection (b)(4). Id. § 3041.5(d)(1). The statute limits when a prisoner may make such a request:
An inmate may make only one written request as provided in paragraph (1) during each three-year period. Following either a summary denial of a request made pursuant to paragraph (1), or the decision of the board after a hearing described in subdivision (a) [ 7 ] to not set a parole .date, the inmate shall not be entitled to submit another request for a hearing pursuant to subdivision (a) until a three-year period of time has elapsed from the summary denial or decision of the board.
Id. § 3041.5(d)(3). After the court held oral argument on the motion for a preliminary injunction, the Board promulgated a second administrative directive, which interpreted this language as imposing a three-year delay only once the prisoner has filed a request for an advanced hearing, such that a prisoner need not wait three years before filing an initial request for an advanced hearing. BPH Administrative Directive No. 09/01, “Penal Code Statutes Enacted by Proposition 9 That Allow An Advanced Hearing Date,” February 5, 2009. Plaintiffs do not challenge this interpretation in this motion. 8
This second directive also states that pursuant to section 3041.5(b)(4), when the Board advances a hearing date, there is not “a minimum time period that must be served from the hearing at which the denial length was determined.” Thus, the Board contends that when a hearing date is advanced, whether on the Board’s initiative under (b)(4) or on a prisoner’s request under (d), the Board is not limited to the time periods specified for deferral of the hearing.
*1114 D. Alleged Punitive Purpose of Proposition 9
As part of their argument that Proposition 9 violates the Ex Post Facto Clause, plaintiffs also allege that Proposition 9 has a demonstrably punitive purpose. Plaintiffs offer the following evidence in support of this argument. Section 2 of the Act, the Findings and Recommendations, stated that “Crime victims are entitled to ... above all, the right to an expeditious and just punishment of the criminal wrongdoer.” ¶ 1. These findings further state that the prior system created a “failure to impose actual and just punishment upon [victims’] wrongdoers.” Id. at ¶ 9.
Section 3 of the Act, the “statement of purposes and intent,” provides that
It is the purpose of the People of the State of California in enacting this initiative measure to:
1. Provide victims with rights to justice and due process.
2. Invoke the rights of families of homicide victims to be spared the ordeal of prolonged and unnecessary suffering, and to stop the waste of millions of taxpayer dollars, by eliminating parole hearings in which there is no likelihood a murderer will be paroled, and to provide that a convicted murderer can receive a parole hearing no more frequently than every three years, and can be denied a follow-up parole hearing for as long as 15 years.
E. Procedural History
This suit was initially filed by plaintiff Richard Gilman on April 7, 2005. At the time, Gilman was proceeding
pro se.
As his claim was about to proceed to trial in June of 2008, the Magistrate Judge assigned to this case appointed counsel, and granted leave to file an amended complaint naming additional plaintiffs. After several stipulated extensions, plaintiffs moved for leave to file a fourth amended and supplemented complaint. Plaintiffs concurrently filed a motion for class certification and the instant motion for a preliminary injunction. The court heard argument on these motions on January 30, 2009. The court granted the motion to amend and the motion for class certification. Order Granting Leave to Amend (Dkt. No. 183),
Defendants then filed a petition to appeal this court’s class certification order under Fed.R.Civ.P. 23(f), and concurrently moved to dismiss plaintiffs’ eighth claim (challenging Prop. 9) and plaintiffs’ ninth claim (challenging Article V, section 8(b) of the California Constitution). The court heard argument on the motion to dismiss on May 18, 2009. The Ninth Circuit granted permission to appeal class certification on May 28, 2009.
On August 11, 2009, the court ordered the parties to provide supplemental briefing addressing the appeal’s effect on the pending motions. Both parties agreed that the court could continue to entertain the motions, but plaintiffs conceded that if a preliminary injunction is granted, it should extend solely to the named plaintiffs. See Plaintiffs’ Brief Regarding Effect of Interlocutory Appeal, Sept. 1, 2009 (Dkt. No. 207). 9
*1115
Defendants then moved to stay proceedings pending resolution of the interlocutory appeal. The court denied this motion on October 15, 2009. (Dkt. No. 212),
II. STANDARDS
A. Standard for a Motion for a Preliminary Injunction
The purpose of the preliminary injunction as provided by Fed.R.Civ.P. 65 is to preserve the relative positions of the parties — the status quo — until a full trial on the merits can be conducted.
See University of Texas v. Camenisch,
“A plaintiff seeking a preliminary injunction must establish that he is [1] likely to succeed on the merits, [2] that he is likely to suffer irreparable harm in the absence of preliminary relief, [3] that the balance of equities tips in his favor, and [4] that an injunction is in the public interest.”
Am. Trucking Ass’ns v. City of Los Angeles,
B. Standard for a Motion to Dismiss
A Fed.R.Civ.P. 12(b)(6) motion challenges a complaint’s compliance with the pleading requirements provided by the Federal Rules. Under Fed.R.Civ.P. 8(a)(2), a pleading must contain a “short and plain statement of the claim showing that the pleader is entitled to relief.” The complaint must give defendant “fair notice of what the claim is and the grounds upon which it rests.”
Bell Atlantic v. Twombly,
“Plausibility,” as it is used in
Twombly
and
Iqbal,
does not refer to the likelihood that a pleader will succeed in proving the allegations. Instead, it refers to whether the non-conclusory factual allegations, when assumed to be true, “allow[ ] the court to draw the reasonable inference
*1116
that the defendant is 'liable for the miscom duct alleged.”
Iqbal,
III. ANALYSIS
Evaluation of both the motion to dismiss and the motion for a preliminary injunction begins with the merits of plaintiffs’ claim. As explained below, the court concludes that plaintiffs’ due process claim fails, but that plaintiffs have demonstrated a likelihood of success on their Ex Post Facto argument. The court then turns to the remaining preliminary injunction factors, and concludes that an injunction is warranted.
A. Merits
Plaintiffs advance two separate legal theories, arguing that the changes effectuated by Proposition 9 violate both the Ex Post Facto Clause and plaintiffs’ substantive due process rights. These theories are closely related: both argue that Proposition 9 will cause some prisoners who have ceased to pose a threat to public safety to remain in prison because the prisoners will be unable to receive a timely suitability hearing. Plaintiffs’ Ex Post Facto claim argues that such confinement is more likely after Proposition 9, such that the likely period of incarceration has increased. Plaintiffs’ substantive due process claim argues that this confinement is not reasonably related to California’s stated purpose for denial of parole, the protection of public safety.
1. Ex Post Facto
Article 1, section 10 of the United States Constitution provides that “No state shall ... pass any ... ex post facto Law.” This clause prohibits any law that “makes more burdensome the punishment for a crime after its commission.”
Beazell v. Ohio,
The Supreme Court has twice discussed what it means for a change to be detrimental in the context of Ex Post Facto claims broadly similar to the one here.
Garner v. Jones,
In both cases, the Court “declined to articulate a single ‘formula’ ” for determining sufficiency.
Morales,
a. Parole Changes in Morales
In
Morales
a California prisoner brought a habeas corpus petition challenging the 1981 amendments to section 3041.5 of the California Penal Code. Prior to this amendment, all life prisoners whose sentences included possibility of parole received annual parole hearings. The 1981 amendment authorized the Board to defer subsequent suitability hearings for up to three years, but only for prisoners convicted of “more than one offense which involves the taking of a life,” and only where the Board found “that it was not reasonable to expect that the parole would be granted at a hearing during the following years and state[d] the bases for the finding.”
Morales,
The Court provided three reasons for this conclusion. Most importantly, the prisoners whose deferral periods were increased would be unlikely to have been found suitable at an earlier date.
Id.
The Court noted that in general, prisoners convicted of multiple offenses involving the taking of a life were particularly unlikely to be found suitable for parole.
Id.
at 511,
Because the terms of the 1981 amendment increased deferral of subsequent suitability hearings only in cases that would be unlikely to delay any actual finding of suitability, because advanced hearings in light of changed circumstances were available, and because a delay of a finding of suitability might not delay actual release, the Court concluded that “the narrow class of prisoners covered by the amendment cannot reasonably expect that their prospects for early release on parole would be enhanced by the opportunity of annual hearings.”
Id.
at 512,
*1118 b. Parole Changes in Garner
The Court considered a similar decrease in the frequency of parole hearings in
Garner,
The Court held that two features of the changed regulation, both of which were also present in
Morales,
precluded a finding that the regulation’s terms presented an Ex Post Facto problem.
Id.
at 254,
c. Lessons from Morales and Garner
As illustrated above, Morales and Garner did not attempt to quantify the risk of prolonged incarceration. Instead, each looked to whether prisoners who could expect release (or a significant chance thereof) at one time under the former rule would be released only at a later time under the new rule. In both cases, the Court found that subsequent hearings would be delayed only when there was no appreciable likelihood of an earlier release.
Neither the Supreme Court nor the Ninth Circuit have clearly addressed whether a plaintiff must show that he personally faces a significant risk of prolonged punishment.
Gamer
stated that “[i]n the case before us, [the prisoner] must show that as applied to his own sentence the law created a significant risk of increasing his punishment.”
The Seventh Circuit has confronted this issue, holding that
The proper question to ask, then, is whether the new procedure creates a significant risk of increased punishment for [petitioner] Glascoe. This is not to be confused with the question of whether the new parole practice is harsher for a class of prisoners generally; we must focus on the consequence of the new practice on the sentence of the particular inmate bringing the challenge!.]
Glascoe v. Bezy,
The Seventh Circuit’s opinion in
Glascoe
did not cite the Supreme Court’s opinion in
Weaver,
and it appears that
Glascoe
cannot be reconciled with
Weaver’
s admonition not look to “special circumstances” that may blunt a law’s impact on specific individuals. As recognized by Justice Souter’s dissent in
Gamer,
because sentencing “is often a mix of rules and discretion,” it can be difficult to determine whether a change has affected the sentence of a particular prisoner.
Garner,
*1120 d. Ex Post Facto Analysis of Proposition 9
As demonstrated above, the Court’s decisions in
Gamer
and
Morales
were informed by the particular features of the laws under consideration.
See Morales,
One difference between Proposition 9 and the laws at issue in
Morales
and
Garner
is the number of distinct changes effectuated by Proposition 9. Both
Morales
and
Gamer
considered mere extensions of the maximum possible deferral date. Proposition 9(1) increased the minimum deferral, (2) reduced, in various ways, the Board’s ability to defer for less than the maximum possible term, and (3) increased the maximum deferral. Although
Gamer
explained that the operation of challenges laws must be analyzed “within the whole context of Georgia’s parole system,”
i. Increase to the Minimum Deferral Period
Garner
and
Morales
both emphasized that under the new laws, a longer deferral would be imposed only when the parole board had found it unreasonable to expect that parole would be granted in the interim. Ga
rner,
Moreover, in this case, the Board’s practice prior to Proposition 9 implied that the Board reasonably expected that many prisoners would become suitable, and potentially be released, in one or two years. 12 Defendants do not contest plaintiffs’ assertion that unlike in Morales, many, if not most, of the prisoners affected by the challenged provisions have been already served their “base terms” such that, if the prisoners were found suitable for parole, the Board would set a prompt release date. Thus, plaintiffs argue that if these prisoners are not able to demonstrate their suitability for three years, they are exposed to a significant risk of prolonged confinement.
Defendants effectively concede this risk, but argue that the risk is ameliorated by the Board’s ability to advance hearing dates in light of changed circumstances or *1121 new information and prisoners’ ability to request such advancement. By definition, a deferral is only set when the Board concludes that the prisoner is not presently suitable for parole. Thus, a subsequent hearing will result in the prisoner’s release only if some fact changes to render that prisoner suitable. 13 Under the former system, the Board would set a period of less than three years if it thought a change might occur in that time. 14 According to defendants, although the three year minimum prevents the Board from scheduling an earlier hearing based on this possibility, if a change actually occurs, that occurrence will entitle the prisoner to an advanced hearing. Thus, defendants argue that in all the circumstances when a prisoner would actually be released under the former system, the prisoner will also be released under the new system, albeit pursuant to a different procedure.
The possibility of advanced hearings was one of the several factors considered in
Garner
and
Morales.
Neither case suggests that advanced hearings are themselves sufficient.
Garner,
Plaintiffs argue that for expected changes, advanced hearings are an inadequate substitute for scheduled hearings, and plaintiffs’ argument suffices to survive a motion to dismiss and to demonstrate a likelihood of success on the merits. At the time these motions were argued, there was no mechanism in place for initiating or accepting petitions to advance a hearing. Once a petition is filed, the Board must recéive the view of a homicide victim’s next of kin before deciding whether to grant a request for an advanced hearing. Cal. Pen.Code § 3041.5(d)(2). If the Board decides to advance the hearing, it must calendar it, for a time which provides the victim’s family with 45 days notice. Id. § 3041. For all of these reasons, there will necessarily be a delay between any meritorious request for an advanced hearing and the grant of such a hearing, and plaintiffs contend, with some evidence, that this delay will likely exceed a year. See Pis.’ Reply to Defs.’ Supp. Evid., Dkt. 180, at 5.
When a change in circumstances renders a prisoner suitable for parole, if he is unable to secure an advanced hearing until *1122 after the time at which he would have received a scheduled suitability hearing under the former statute, the prisoner’s incarceration will have been prolonged. Because the Board’s past practices demonstrate that it expected a significant number of prisoners to become suitable for parole in less than three years, the risk of prolonged incarceration for prisoners who would have received shorter deferral appears significant.
Plaintiffs additionally argue that advanced hearings are inadequate because the amendments do not restrict the Board’s discretion to advance hearings, such that the Board may decline to advance a hearing even when a prisoner demonstrates a change in circumstance. Insofar as plaintiffs challenge Proposition 9 on its face, i.e., “by its own terms,” I must assume that the Board will exercise a neutral grant of discretion in a manner consistent with the Ex Post Facto Clause.
Garner,
In summary, Proposition 9, unlike the changes considered in
Morales
and
Garner,
increases the minimum deferral period. Plaintiffs are likely to succeed in showing that many prisoners who have currently been found unsuitable for parole, including the named plaintiffs here, have a significant chance of becoming suitable in less than three years. Plaintiffs have also made a sufficient showing that many prisoners, including the named plaintiffs, have served them base terms and would be granted immediate release if found suitable for parole.
C.f. Morales,
ii. Limits to the Board’s Discretion
Aside from raising the minimum deferral period, Proposition 9 limited the Board’s ability to impose the minimum period. Under the former rule, the default was the minimum period, and the Board had discretion to impose a longer deferral only when “it [was] not reasonable to expect that parole would be granted at a hearing during the following year[s].” Former Cal. Pen.Code § 3041.5(b)(2)(A)-(B). Further, the statute provided that the Board “may” impose a longer deferral in these circumstances.
Id.
The Board thus had discretion to impose less than the maximum even when it was not reasonable to expect that parole would be granted sooner. Finally, because the statute was silent as to the standard of proof to be used in forming these “reasonable expectations,” the Board had discretion to determine this standard.
Shaputis,
Proposition 9 restricts the Board’s discretion in regard to each of the above matters. The default deferral is now the maximum. Cal. Pen.Code § 3041.5(b)(3)(A). The Board may set a *1123 lesser deferral only where it “finds by clear and convincing evidence that .... consideration of the public and victim’s safety does not require a more lengthy period of incarceration.” Id. Because consideration of the public safety is the determinant of parole suitability, Proposition 9 allows a deferral for less than the maximum only when clear and convincing evidence indicates that parole will actually be granted at the next hearing. Regardless of whether this “clear and convincing evidence” standard of proof differs from the Board’s interpretation of the prior “reasonable expectation” standard, the Board no longer has discretion to schedule an earlier hearing even when it does not expect parole to be granted. 15
Removing discretion to set earlier hearings, rather than expanding discretion to set deferred hearings, sharply distinguishes this case from
Gamer
and
Morales.
Both
Gamer
and
Morales
considered changes that, like California’s prior system, granted the Board discretion to postpone hearings when the Board made specific findings that an earlier release was unlikely.
Garner,
Here, plaintiffs are likely to succeed in showing that there is a significant chance that some prisoners not presently suitable for parole will become suitable in the fewer than the next fifteen years, but that these prisoners are unable to currently demonstrate by clear and convincing evidence that this change will occur, or when. Moreover, for the reasons stated in the previous section, plaintiffs are likely to succeed in showing that there is a significant risk that forcing these prisoners to demonstrate their changed status through an advanced hearing will result in longer incarceration than would result under the former rule, which would have provided for an earlier scheduled hearing either because. of the Board’s exercise of discretion *1124 or because of the prior rule’s lower maximum.
iii. Increase in Maximum
As noted above, Proposition 9 increased the maximum deferral period from 5 years to 15 years.
Gamer
and
Morales
each demonstrated that an increase in the maximum deferral period, standing alone, does not violate the Ex Post Facto Clause. In both cases, however, the increase in the maximum deferral would only affect prisoners who were not likely to receive parole during the length of their deferral. Here, where the increase in the maximum is accompanied by changes that limit the Board’s ability to impose less than the maximum, the increase further contributes to the risk of prolonged incarceration. As noted above,
Gamer
held that changes must be reviewed “within the whole context of [the state’s] parole system.”
Garner,
iv. Conclusions Regarding Plaintiffs’ Ex Post Facto Challenge
Both Morales and Gamer indicated that changes to the timing of parole suitability hearings could violate the Ex Post Facto Clause. Although the Court found no violation in those cases, Proposition 9 is factually distinct in numerous ways. Unlike the laws reviewed in Morales and Gamer, Proposition 9 increases the minimum deferral period — here, from one to three years. Proposition 9 farther sharply limits the Board’s discretion to set deferrals for less than the maximum. Finally, Proposition 9 drastically increases the maximum deferral period. These changes create a risk of prolonged incarceration, and plaintiffs are likely to succeed in showing that this risk is significant despite the possible availability of advanced hearings.
Plaintiffs separately argue that Proposition 9 has an explicitly punitive purpose, independent of its punitive effect. Because the court concludes that plaintiffs have shown a likelihood of success as to punitive effect, and because defendants’ motion to dismiss does not address this argument, the court does not further discuss any possible punitive purpose here.
2. Substantive Due Process
Plaintiffs separately argue that Proposition 9’s changes to the parole deferral process violate the Substantive Due Process Clause.
See, e.g., Lawrence v. Texas,
In a substantive due process challenge, the initial inquiry is whether the asserted right is one that is “fundamental.”
Wash. v. Glucksberg,
Plaintiffs approach rational basis review by first identifying the government’s interest. They first argue that under California Penal Code section 3041, California has declared that the only interest furthered by a denial of parole (once the minimum term is served) is prevention of a prisoner’s threat to public safety. Plaintiffs then argue that whenever a prisoner’s circumstances change such that he is no longer a threat to public safety, any further incarceration of the prisoner does not further this interest. Pis.’ Mem. Supp. Mot. Prelim. Inj. 21. Plaintiffs then reiterate much of their Ex Post Facto argument, arguing that as a result of Proposition 9’s increased deferral periods, prisoners whose circumstances have changed will often remain in custody for prolonged periods while waiting for a suitability hearing at which the prisoner may demonstrate this change.
Although plaintiffs accurately characterize section 3041, other Penal Code sections are not so narrow. For example, the California Supreme Court has recognized that one purpose behind the 1982 amendments to section 3041.5 was to save the expense of additional hearings.
Jackson,
At its heart, plaintiffs’ argument is that Penal Code section 3041 ties the state’s hands, trumping any other statutes purporting to assert other interests. This argument fails, because federal courts lack “the authority to enjoin the administration of, or to declare invalid, state laws on the grounds that they are incompatible with other state laws, much less to rule that state laws are contrary to the state legislature’s intent.”
McQuillion v. Schwarzenegger,
The state’s interest in saving expenses is rationally related to any resulting deprivation of prisoners’ interests in parole. The state could rationally conclude that the savings realized from avoiding frequent parole hearings is great enough to offset any added expense incurred by continued incarceration of those few prisoners who would have been found suitable for parole.
B. Likelihood of Irreparable Injury
Even where plaintiffs are likely to succeed on the merits, a preliminary injunction may not issue unless plaintiffs have shown that they are “likely to suffer irreparable harm in the absence of preliminary relief.” Winter,-U.S. at-,
Winter
differs from this case-and from many other cases seeking injunctive relief — because in
Winter,
plaintiffs’ showing as to the merits had little connection to a likelihood of specific irreparable harm.
Winter
concerned a claim under the National Environmental Policy Act, 42 U.S.C. § 4321
et seq.
(“NEPA”). Plaintiffs challenged the Navy’s use of long-range sonar, arguing that the Navy had not adequately studied the sonar’s effects on marine mammals.
Winter,
— U.S. at-,
In this case, unlike in Winter, the merits of the claim and the likelihood of irreparable injury overlap in that both consist of potential confinement in violation of the Ex Post Facto Clause. Above, the court concluded that plaintiffs are likely to succeed in showing a “significant risk” of such confinement. Although Winter did not quantify “likely,” in the preliminary injunction context, “likely” would appear to require no more than “significant.”
The relevant likelihood, however, is not the likelihood of injury absent any adjudication of this case. Instead, the question is whether such injury is likely to occur before final resolution of this matter.
As is already apparent, this case raises complex issues. It would be optimistic to conclude that proceedings before this court will be resolved within another year, and it seems clear that both sides intend to appeal any judgment in the other side’s favor.
As to plaintiff Masoner, there is evidence demonstrating likelihood of injury during the pendency of this case. 17 He had routinely received one-year deferrals prior to the passage of Proposition 9, had once been found suitable by the Board only to have the grant of parole reversed by the Governor. At his 2008 hearing, the panel conceded that he was approaching suitability even as it denied parole. Under Proposition 9, he will not receive a scheduled suitability hearing until late 2011. Enjoining the enforcement of Proposition 9 would almost certainly enable Masoner to receive scheduled annual hearings, and he has shown that it is sufficiently likely that he would be released on parole at such a hearing prior to the completion of this suit.
While there is no evidence concerning the remaining plaintiffs, the operative complaint alleges that five of these seven plaintiffs have previously been found suitable for parole only to have this determination reversed by the governor. Each plaintiffs’ most recent deferral was for one or two years. Although neither party significantly discussed these allegations in connection with the motion for a preliminary injunction, it does not appear that these allegations are in dispute. I therefore conclude that each of the named plain *1127 tiffs has shown a likelihood of irreparable injury absent a preliminary injunction. 18
C. Balance of Hardships
The hardships faced by plaintiffs are longer prison terms and violations of their constitutional rights. The hardships faced by the defendant are the cost of administering additional parole hearings. Defendants further argue that a preliminary injunction would impair public safety, causing a hardship to the defendants, who are charged with protecting the public. However, the injunction sought would not cause any prisoner to be released absent a determination that the prisoner was not a threat to public safety and therefore suitable for parole. 19 Balancing the other hardships, the court concludes that notwithstanding the State of California’s present budget problems, the balance favors plaintiffs.
Defendants also implicitly rely on a type of psychological injury described by Proposition 9, the “prolonged and unnecessary suffering” parole hearings inflict upon collateral victims of the inmates’ offense (i.e., relatives of the decedent). See Proposition 9, § 3(2). Presumably, this language refers to the stress and anguish these persons experience resulting from attendance of parole hearings, or simply from the knowledge that such hearings are being held. The court is not aware of any authority recognizing injury of this type. Even if this type of injury is cognizable and properly considered in the preliminary injunction analysis (which the court does not decide), either under the balancing of hardships or as an aspect of the public interest, the injury is not sufficient to tip the scales against plaintiffs.
D. Public Interest
In this case, both parties invoke lofty public interests. Plaintiffs note the public’s interest in upholding the constitution. Defendants argue that the public has an interest in seeing its will enforced. Proposition 9 was enacted through the exercise of California’s initiative processes. As I have previously observed, however, that happenstance provides the statute with no special insulation from review for asserted constitutional infirmity. Rather, the Supreme Court has held that “‘it is irrelevant that the voters ... enacted [a statute], because the voters may no more violate the Constitution by enacting a ballot measure than a legislative body may do so by enacting legislation.’ ”
Service Employees Int’l Union v. Fair Political Practices Com.,
*1128
Defendants cite several cases concerning injunctions that would directly affect, respectively, the public’s ability to conduct an upcoming election or to receive electricity.
Cardona v. Oakland Unified School Dist., 785
F.Supp. 837, 842 (N.D.Cal.1992),
Sierra Club v. Georgia Power Co.,
IY. CONCLUSION
Plaintiffs have shown a likelihood of success on the merits as to their Ex Post Facto challenge to Proposition 9’s amendments to California’s parole process. Plaintiffs have shown that these amendments are likely to cause irreparable injury absent a preliminary injunction. These factors, as well as the balance of hardships and the public interest, demonstrate that plaintiffs are entitled to a preliminary injunction barring enforcement of Proposition 9 as to the named plaintiffs.
Plaintiffs have failed to state a claim that Proposition 9 violates the substantive due process clause.
For these reasons, the court ORDERS as follows:
1. Plaintiffs’ motion for a preliminary injunction, Dkt. No. 157, is GRANTED as to the named plaintiffs. Defendants and their agents are ENJOINED from enforcing the provisions of Proposition 9 that amend former California Penal Code section 3041.5(b)(2)(A) as to the named plaintiffs.
2. Defendants’ motion to strike, Dkt. No. 164, is DENIED.
3. Defendants’ motion to dismiss, Dkt. No. 187, is GRANTED IN PART. Plaintiffs’ substantive due process challenge to Proposition 9 is DISMISSED WITHOUT PREJUDICE. See Vess v. Ciba-Geigy Corp. USA,317 F.3d 1097 , 1108 (9th Cir.2003) (Initial dismissal under Fed.R.Civ.P. 12(b)(6) should ordinarily be without prejudice). Defendants’ motion to dismiss is DENIED as to plaintiffs’ Ex Post Facto challenge to Proposition 9. As explained in the concurrently filed order, defendant’s motion to dismiss is also DENIED as to plaintiffs’ ninth claim, challenging Cal. Const. Art. V, § 8(b).
IT IS SO ORDERED.
Notes
. California Penal Code section 3041(a) provides that this meeting shall occur one year prior to the minimum eligible parole date. Plaintiffs assert that in practice, this initial meeting occurs after the date has passed. Such a delay, if any, is not at issue in the present motion.
. Plaintiffs state, without citation to any authority, that the panel’s suitability determination is made on a preponderance of the evidence standard. Although it may be that this standard is implicit in the penal code, plaintiffs have not demonstrated that this is so, and no standard of proof appears in either section 3041 or in Lawrence, Shaputis, or Rosenkrantz.
. Plaintiffs assert that the function of the second step in the parole process is to ensure that the prisoner serves a sentence long enough to satisfy the punitive purpose of in *1111 carceration. Although this is certainly a plausible interpretation, no party has provided authority indicating that this is the intended function of the delayed release dates.
. Plaintiffs' motion for a preliminary injunction discusses thirty years of various amendments to the parole hearing process. The present motion seeks to enjoin the amendments caused by Proposition 9, and not any of the prior amendments. Therefore, the history of these amendments is not directly relevant to the present' motion. The court describes the law as it existed immediately prior to the passage of Proposition 9.
. Prior to Proposition 9, the Board had the power to set regulations specifying the factors to be considered in determining the length of deferral; the Board's regulations specified that these factors were the same as those used in determining suitability. Proposition 9 incorporates these regulations in to the statute itself. Thus, the penal code now specifies that public safety is the determinant of both suitability and the deferral period.
. These figures are apparently undisputed.
. Subdivision (a) refers to "all hearings for the purpose of reviewing a prisoner’s parole suitability.” Cal. Pen.Code § 3041.5(a).
. In their moving papers (submitted prior to the promulgation of this directive), plaintiffs adopted a contrary interpretation of the statute, interpreting it as obliging prisoners to wait three years after any unsuitability finding before submitting a request for an advanced hearing. Plaintiffs have since responded to the administrative directive by stating that while they suspect that the directive contradicts the clear language of the statute, they do not challenge it in the present motion. Accordingly, the court assumes for purposes of this motion only that the administrative directive is valid and presents a controlling interpretation of the statute.
. The court expresses some reservation about limiting the current motion to the named plaintiffs. As explained in the October 15, 2009 denial of defendants' motion for a stay, the court remains convinced that class certification is proper. Moreover, prisoners in various other proceedings are requesting to join in this action, or are being referred thereto.
See
Motion to Intervene by Michael L. Hughes (Dkt. No. 200);
see also Bair v. Warden, Deuel Vacational Inst.,
No. 2:09-cv-1730,
. Here, "class” refers to the class of persons affected by a change, rather than the particular class certified in this case.
. If, contrary to the analysis above, plaintiffs were required to show an individualized significant risk of prolonged incarceration for purposes of the merits of the Ex Post Facto *1120 chalienge, the court would find that such a showing had been made for the reasons underlying the court's irreparable injury analysis.
. Plaintiffs note that the Board granted one or two year deferrals to approximately two thirds of prisoners found unsuitable for parole. However, this statistic does not demonstrate that the Board had expected all these prisoners to become suitable within this time. The former statute permitted the Board to impose a longer deferral only prisoners convicted of murder. Even for those prisoners, the Board had discretion to impose a shorter deferral even absent an expectation that the prisoner would become suitable within two years.
. Plaintiffs candidly note that a prisoner may be found suitable for a parole at a subsequent hearing even if the circumstances have not changed, because the panel at the subsequent hearing may be constituted of individuals who view the same circumstances differently.
Although there is an undeniable realism to plaintiffs’ argument, the argument was foreclosed by
Morales. Morales
held, for example, that changes in the membership of the state parole board, although they may predictably effect the likelihood of release on parole, would not ordinarily constitute an Ex Post Facto violation.
. In the following section, part III(A)(l)(d)(ii), the court discusses the particular standards of proof used by the Board in making this determination.
. The parties have provided little discussion of the effect of the Board’s prior discretion to schedule earlier hearings even absent a reasonable expectation that parole would be granted. It appears likely to the court that in many cases, the evidence will not support a prediction one way or the other regarding future suitability for parole. In such a case, the prior system afforded the Board discretion to expect the unexpected and schedule an earlier hearing. Proposition 9 eliminates this discretion, putting the burden of proof on the prisoner to clearly and convincingly predict future suitábility.
. The court reiterates that it is likely to be hard to predict, from a decade in advance, when many prisoners will become suitable for parole. It may be that a significant-number of prisoners become eligible even though their eligibility could not have been clearly and convincingly predicted. This possibility demonstrates the importance of holding periodic suitability hearings regardless of individual predictions. Gamer and Morales allowed periodic hearings to be skipped, but both implied that the burden of proof was on the state to provide specific reasons for believing that parole would not be granted at these hearings. Proposition 9, on the other hand, puts the burden of proof on the prisoner. If, as the court expects, it is frequently impossible to make a prediction one way or the other, allocation of the burden of proof is crucial. Proposition 9's shifting of the burden of proof, together with its 15 year default deferral, effectively removes the prior presumption of periodic scheduled hearings.
. Defendants move to strike the discussion of these facts included in plaintiffs' memorandum in support of their motion. (Dkt. No. 164.) Because a Fed.R.Civ.P. 12(f) motion to strike may only be brought as to "pleadings,” the court construes this motion as an evidentiary objection. Plaintiffs have supplied evidence to support their claims regarding Masoner. For the reasons stated in plaintiffs’ opposition to defendants’ motion to strike, the motion to strike is denied.
. The defendants further argue that injunctive relief is unnecessary because the individuals can file habeas petitions. The court first notes that plaintiffs' claims may be properly brought under 42 U.S.C. section 1983 — indeed,
Gamer
concerned an analogous section 1983 Ex Post Facto claim.
. It is conceivable that the Board may occasionally err in its suitability determinations. If errors are possible, then an increase in the number of parole hearings (as would result from an injunction) may also increase the likelihood of an error that mistakenly releases a dangerous prisoner. However, this risk is too speculative and slight to weigh into the balance of hardships analysis.
. Indeed, it can be argued that the public interest favors an injunction, since “it is 'in the public interest to terminate the unconstitutional application’ of a statute.”
Levine v. Fair Political Practices Comm’n,