Gomez v. Singletary

733 So.2d 499 (1998) Alejandro GOMEZ, Petitioner, v. Harry K. SINGLETARY, Jr., Respondent. Alex David Goodwin, Petitioner, v. Harry K. Singletary, Jr., Respondent. Steven Kivett, Petitioner, v. Harry K. Singletary, Jr., Respondent. Jeffrey Lynn Hock, Petitioner, v. Harry K. Singletary, Jr., Respondent. Giovanni Galvis, Petitioner, v. Harry K. Singletary, Jr., Respondent. Nos. 90,642, 90,654, 90,655, 90,759 and 90,829. Supreme Court of Florida. December 24, 1998. Rehearing Denied May 20, 1999. John C. Schaible and Christopher M. Jones, Florida Institutional Legal Services, Inc., Gainesville, Florida, for Petitioners. Susan A. Maher, Deputy General Counsel, Department of Corrections, Tallahassee, Florida, for Respondent. *500 REVISED OPINION OVERTON, J. This case concerns the consolidated petitions for writ of habeas corpus filed by petitioners Alejandro Gomez, Alex David Goodwin, Steven Kivett, Jeffrey Lynn Hock, and Giovanni Galvis. We have jurisdiction pursuant to Article V, section 3(b)(9) of the Constitution of Florida. We find that United States Supreme Court decisions mandate that we grant the petitions. We must do so because the substitution of the Control Release Program by the State for the statutory overcrowding programs in effect at the time of the petitioners' offenses improperly curtailed the availability of future credits. Consequently, under Weaver v. Graham, 450 U.S. 24, 101 S.Ct. 960, 67 L.Ed.2d 17 (1981), as reaffirmed in Lynce v. Mathis, 519 U.S. 433, 117 S.Ct. 891, 137 L.Ed.2d 63 (1997), this action constituted a violation of the Ex Post Facto Clause of the United States Constitution. See U.S. Const. art I, § 10. We note that this case is clearly distinguishable from Meola v. Department of Corrections, 732 So.2d 1029 (Fla.1998), and Thomas v. Singletary, 729 So.2d 369 (Fla. 1998), in which we are denying relief.[1] I. The Various Overcrowding Programs Since the early 1970s, Florida has had a severe prison overcrowding problem which has resulted in the closing of Florida's prison system several times. See Costello v. Wainwright, 397 F.Supp. 20, 22 (M.D.Fla.1975), aff'd 525 F.2d 1239 (5th Cir.1976). In 1980, the Florida Department of Corrections (the Department) entered into a consent decree that provided for the capping of the prison population within certain limits. Id. As explained below, over the years, the legislature enacted a maze of overcrowding gain time statutes with differing names and requirements. Pursuant to the consent decree in Costello, and as a part of the Correctional Reform Act of 1983, the legislature enacted the Emergency Gain Time statute. See § 944.598, Fla. Stat. (1983). The Emergency Gain Time statute was the first overcrowding gain time statute. It became effective on June 16, 1983. However, under those provisions, the statute did not become operative until prison overcrowding surpassed 98% of lawful capacity (the triggering percentile threshold). Effective June 2, 1986, the triggering percentile threshold was increased to 99% capacity. See ch. 86-46, §§ 1, 3, at 166, 168, Laws of Fla.; § 944.598, Fla. Stat. (Supp.1986). *501 The Emergency Gain Time statute was superseded by the enactment of the Administrative Gain Time statute. See § 944.276, Fla. Stat. (1987). This statute went into effect on February 5, 1987. See ch. 87-2, §§ 1, 2, 3, at 3-4, Laws of Fla. While the Emergency Gain Time statute technically remained in effect, the Administrative Gain Time statute effectively superseded the Emergency Gain Time statute because the Administrative Gain Time had a triggering percentile threshold of 98%, which was lower than the 99% threshold of the Emergency Gain Time statute. The State never implemented the Emergency Gain Time statute, so no Emergency Gain Time was ever awarded. See Blankenship v. Dugger, 521 So.2d 1097, 1098 (Fla.1988). The Emergency Gain Time statute was repealed effective June 17, 1993. See ch. 93-406, §§ 32, 44, at 2966, 2974, Laws of Fla. (reflected in the Table of Repealed and Transferred Sections (only)—Index at D-97, Fla. Stat. (1993)). The Department of Corrections (hereafter the Department) made its first overcrowding awards under the Administrative Gain Time statute in February of 1987. It awarded a total of 720 days of Administrative Gain Time to individual prisoners between February of 1987 and June 30, 1988. Effective July 1, 1988, the Administrative Gain Time statute was repealed[2] and the Provisional Credits statute took its place, with a triggering percentile threshold of 97.5. See § 944.277, Fla. Stat. (Supp. 1988); ch. 88-122, § 5, at 535-37, 572, Laws of Fla. The Provisional Credits statute was in effect from July 1, 1988, through June 16, 1993. During this time the Department awarded a total of 1860 days of Provisional Credits to individual prisoners between July 1988 and January 18, 1991. It is very significant that the award of credits under both Administrative Gain Time and Provisional Credits programs was administered pursuant to a computer program without an individual review of inmate records. It proved to be unpopular both with the public and with state officials because several notorious early releasees committed new violent crimes. In an effort to provide a better, more tailored method of reducing prison overcrowding, the Control Release Program was enacted by the legislature, effective September 1, 1990, with a triggering percentage threshold of 97.5%. See § 947.146, Fla. Stat. (1989); ch. 89-526, §§ 1, 2, 52, at 2659-61, 2690, Laws of Fla. At the same time that the Control Release program was created, the threshold for the Provisional Credits program was increased to 98%. Id. Although Control Release was similar to the Provisional Credits statute because it provided for the award of overcrowding allotments which advanced the inmate's release date, it was different in several respects. First, the Control Release program was administered by the Parole Commission, sitting as the Control Release Authority, rather than the Department. See § 947.146(1), Fla. Stat. (1989). Second, Control Release allotments were not awarded across-the-board to all eligible inmates based a computer program, as was done with the prior programs. Third, even if inmates were statutorily eligible for Control Release consideration, some inmates' release dates were not advanced as allotments were awarded, because they had been assigned a "non-advanceable" control release date. See § 947.146(2), Fla. Stat. (1989); Fla. Admin. Code R. 23-22.006-22.008. The Commission assigned "non-advanceable" control release dates to inmates it considered to be bad release risks to the community, taking into consideration such things as the prior number and type of convictions and prior behavior on supervision, as well as the discretionary consideration of aggravating and/or mitigating factors, including victim input. See § 947.146(5), Fla. Stat. (1989); Fla. Admin. Code R. 23-22.006-22.008. In *502 addition, control release dates could be modified based on new information tending to indicate an inmate's negative or increased release risk prognosis. Moreover, inmate release dates could be extended or advanced, i.e., allotments could be given or taken away, as the prison overcrowding levels changed. See § 947. 146(6), Fla. Stat. (1989); Fla. Admin. Code R. 23-22.010. In sum, while Control Release was similar to the previous programs because it provided for reduced sentences, it was different because it incorporated several of the more discretionary aspects of a traditional parole-type program.[3] In many ways, the Control Release program was better than the previous programs because it helped to ensure that only the least dangerous inmates would be released early when prison overcrowding occurred. On January 18, 1991, the Department stopped awarding overcrowding credits under the 98% triggering threshold of the prior credits program because the Commission was awarding allotments when the 97.5% percentage threshold was reached. In 1993, the legislature repealed the Provisional Credits statute. See § 944.278, Fla. Stat. (1993)(note 2); ch. 93-406, §§ 32, 44, at 2966, 2974, Laws of Fla. Then, between 1993 and 1995, the legislature progressively increased the percentage threshold for Control Release. In addition, during several periods of time, beginning in 1993, prison overcrowding exceeded all of the prior thresholds. See § 947.146(2), Fla. Stat. (1993); § 947.146(2), Fla. Stat. (Supp.1994). At the time of their offenses, all five petitioners were eligible for prison overcrowding credits (either Emergency Gain Time, Administrative Gain Time, or Provisional Credits). However, by the time all but petitioner Hock had reached the Department's custody, the petitioners had already either been made effectively ineligible, or the entire program had been repealed. Accordingly, petitioners Galvis, Kivett, Gomez, and Goodwin were never awarded any overcrowding credits.[4] The following is a summary of the five slightly different scenarios of each of the petitions: Petitioner Galvis was convicted of a drug trafficking offense committed on September 24, 1992. At that time, several programs were in effect: the Emergency Gain Time statute, with a threshold of 99%; the Provisional Credits program, with a threshold of 98%; and the Control Release program, with a threshold of 97.5%. Galvis was eligible for Emergency Gain Time, but ineligible for any other credits until after he had finished serving the fifteen-year mandatory minimum.[5] Petitioner Kivett was convicted of manslaughter committed on October 18, 1987. At that time the following overcrowding programs were in effect: the Emergency Gain Time program, with a threshold of 99%; and the Administrative Gain Time program, with a threshold of 98%. Kivett was eligible for both of those programs. Kivett was offense-eligible for Control Release consideration but after consideration, the Commission found Kivett to be a release risk and assigned him a non-advanceable *503 date. This designation meant that when the Commission awarded allotments (days), they did not affect Kivett's actual release date. See § 947.146(2), Fla. Stat. (1989); Fla. Admin. Code R. 23-22.006-22.008. Petitioner Gomez was convicted of a second-degree murder committed on September 11, 1986. At that time, the only overcrowding program in effect was the Emergency Gain Time program, with a threshold of 99%. Gomez was eligible under that statute. However, the Department did not receive custody of Gomez until March 9, 1993. At that time, overcrowding allotments were being awarded under the Control Release program and Gomez was not eligible for Control Release. Petitioner Goodwin was convicted of a DUI manslaughter committed on May 6, 1989. At that time, the Emergency Gain Time statute was in effect with a threshold of 99%; the Provisional Credits statute was also in effect, with a threshold of 97.5%; and Goodwin was eligible for those programs. However, Goodwin was not received into the Department's custody until April 28, 1994, which was after the Provisional Credits program had been repealed. Goodwin, like Kivett, was offense-eligible for Control Release consideration, but was deemed to be a release-risk and assigned a non-advanceable date. Petitioner Hock was convicted of a second-degree murder committed on October 1, 1988. At that time the following programs were in effect: the Emergency Gain Time statute, with a threshold of 99%, and the Provisional Credits statute with a threshold of 97.5%. Hock was eligible for both those programs and received 360 days of Provisional Credits between May 1990 and January 18, 1991. After that date, overcrowding allotments were awarded pursuant to the Control Release program and Hock was ineligible for that program. Hock's 360 days of Provisional Credits were canceled in 1993, but restored in 1997, as explained in footnote 4. In this case, the petitioners argue that under Weaver v. Graham, 450 U.S. 24, 101 S.Ct. 960, 67 L.Ed.2d 17 (1981), as reaffirmed in Lynce, they should have been awarded overcrowding gain time credits because, at the time of their offenses, they were eligible for at least one of the various versions of overcrowding gain-time credits. Except for Hock, they were never awarded any overcrowding credits because such credits were no longer available to them by the time they entered the Department's custody and they were effectively ineligible for Control Release. They allege that they are entitled to relief because the statutes which were enacted after their offenses were more onerous than the ones in effect at time of their offenses and the application of those statutes resulted in their having to serve longer sentences. They assert that not only should they have been awarded either Emergency Gain Time, Administrative Gain Time or Provisional Credits, but the State should not have been able to make them effectively ineligible for any type of overcrowding credits by substituting the Control Release Program for the prior programs. They further assert that prison overcrowding continued for a number of years even after the substitution of the Control Release program. They conclude, therefore, that all the petitioners should have received and continued receiving either Emergency Gain Time, Administrative Gain Time or Provisional Credits (as the case may be) until late 1994 or even later. The State argues that the petitioners are not entitled to these credits and argues that the United States Supreme Court's decision in Lynce only requires the reinstatement of credits for those inmates who actually received credits which were subsequently cancelled. Since these inmates never received any overcrowding credits[6]*504 the State asserts that they are not entitled to any credits. The State further argues that even if the prior programs had not been repealed or replaced, Secretary Singletary would not have exercised his discretion to implement them because it was clear that the Control Release program was a much better way to ensure that only the least dangerous inmates were released early when prison overcrowding reached crisis proportions. II. Analysis under Lynce v. Mathis In Lynce, petitioner Lynce argued that, like regular gain time,[7] prison overcrowding credits could constitute part of an inmate's sentence because a "prisoner's eligibility for reduced imprisonment is a significant factor entering into both the defendant's decision to plea bargain and the judge's calculation of the sentence to be imposed." Lynce, 519 U.S. at 445-46, 117 S.Ct. 891 (quoting Weaver v. Graham, 450 U.S. 24, 32, 101 S.Ct. 960, 67 L.Ed.2d 17 (1981)). The State argued that with prison overcrowding credits, the above-mentioned reliance argument was not present because at the time of the plea bargain and sentencing, the petitioner could not have reasonably expected to receive any such credits given that the State could have alleviated the prison overcrowding problem by other means by the time the inmate reached prison. Accordingly, the State argued, unlike regular gain time, the existence or amount of prison overcrowding is always speculative. The United States Supreme Court found this argument unpersuasive in Lynce's case because "the actual course of events [made] it unnecessary to speculate about what might have happened." Lynce, 519 U.S. at 446, 117 S.Ct. 891. Stated another way, in Lynce, no one had to "speculate" about whether the contingency of the relevant prison overcrowding threshold might be reached during Lynce's incarceration so as to allow for such awards because it was reached during Lynce's incarceration, and he was awarded a specific number of Provisional Credits. The Court, therefore, made clear that, for ex post facto purposes, there is no difference between regular gain time and prison overcrowding gain time. In this case as well, there is no need to engage in such speculation because prison overcrowding did reach the relevant percentages in 1993 and but for the effective repeals of the earlier programs, the petitioners would have received credits and their sentences would have been reduced. There is a distinction, however, between this case and Lynce. In Lynce, the issue was whether the petitioners' credits should be restored. In this case, the issue is whether the petitioners should receive the overcrowding credits they would have received if the actual overcrowding mechanism which was in effect at the time of their offense had continued in effect. In other words, in this case, there was no cancellation of overcrowding credits already awarded. Accordingly, it is at least arguable that the circumstances present in Lynce are not present here. Nevertheless, the Lynce Court specifically relied on and cited extensively to Weaver v. Graham, 450 U.S. 24, 101 S.Ct. 960, 67 L.Ed.2d 17 (1981), which is quite similar to this case. In Weaver, the Department attempted to alter the manner of awarding basic and incentive gain time which resulted in a reduction in the availability of gain time. The Supreme Court found that because "the new provision constricts the inmate's opportunity to earn early release, [it] *505 thereby makes more onerous the punishment for crimes committed before its enactment. This result runs afoul of the prohibition against ex post facto laws." Weaver, 450 U.S. at 35-36, 101 S.Ct. 960 (emphasis added). This case is similar because the substituted overcrowding program (Control Release) significantly constricted the petitioners' opportunity to earn early release under the prior statutory provisions by making them effectively ineligible. As mentioned above, the Department asserts that the petitioners are not entitled to relief because Secretary Singletary would not have exercised his discretion under the Provisional Credits statute to continue to award Provisional Credits, because he felt that the Control Release program was a better program. While we agree that the Control Release program certainly seems to have been better designed to ensure public safety and that the Secretary might have had the discretion to refuse to award any type of overcrowding credits, we do not believe that the Ex Post Facto Clause gives the Secretary the discretion to choose to allow the implementation of the later, more onerous Control Release statute to these inmates in place of the statutes which were in effect at the time of their offenses. Accordingly, we are compelled by the United States Supreme Court's decision in Lynce to conclude that for these inmates, the substitution of the Control Release program for the earlier programs resulted in an ex post facto violation under Weaver. Our decision today is consistent with our recent decision in Britt v. Chiles, 704 So.2d 1046 (Fla.1997). In that case, this Court rejected the State's argument that a change in the gain time forfeiture statute which increased the amount of gain time forfeitable was not subject to the Ex Post Facto Clause because it was too contingent upon future events. This Court granted the petition because the statute "eliminate[d] an inmate's opportunity to earn incentive gain-time" for a certain amount of time after a disciplinary infraction was committed. Britt, 704 So.2d at 1047-48. This Court specifically likened the situation in Britt to that in Weaver, because by reducing the ability to earn gain time after a disciplinary infraction, the Department was essentially "curtailing the availability of future credits [and therefore] it effectively postpone[d] the date when [an inmate] would become eligible for early release." Britt, id. at 1046-47 (first brackets added)(quoting Lynce, 519 U.S. at 442, 117 S.Ct. 891). Accordingly, we grant the instant consolidated petition and order the Department to apply the version of the overcrowding statute in effect at the time of each petitioner's offense. III. The Remedy The Department has provided detailed charts showing the actual prison population over the years as compared to the relevant lawful capacity levels. The difference between these numbers is the amount of overcrowding present during the pertinent years. The Department also set forth charts presenting several possible ways of providing for the award of credits were we to grant the petition. The Department has made its calculation based both on the actual levels of prison overcrowding and on certain data projections for time periods for which the Department does not have actual numbers. The petitioners contest the Department's proposed relief because they argue that the use of the Control Release statute to reduce prison overcrowding by releasing the less risky inmates reduced the actual level of prison overcrowding, thereby reducing the amount of credit now due to the petitioners. We conclude that if the legislature had maintained the percentage threshold for Control Release at below 97.5%, by releasing the less dangerous inmates on Control Release at that lower level, that action would probably have prevented the overcrowding levels from reaching the higher threshold levels. If those thresholds *506 had not been reached, the required percentage contingency would not have come into fruition and there would probably have been no ex post facto violation.[8] Due to the later increases in the Control Release thresholds, however, prison overcrowding periodically did exceed the relevant threshold levels in 1993 and onward for a number of years. Nevertheless, to the extent that the State did succeed in reducing the prison population by releasing the less dangerous inmates on Control Release, we find no impropriety. Therefore, for any time-frames in which the prison population did not exceed the relevant percentage thresholds, the Department need provide no relief.[9] While we recognize that it may be impossible to calculate the exact amount of prison overcrowding and the exact number of credits that would have been issued to the petitioners nearly a decade ago had the State utilized the overcrowding statutes in effect at the time of each inmate's offense, we believe this opinion sets forth sufficient guidelines so that the Department may endeavor to calculate a reasonable estimate of the relief now due to each inmate. The petitioners also argue that for the years in which Control Release allotments were awarded they should receive the much higher number of Control Release allotments that were given to the Control Release eligible inmates. We conclude that under the Ex Post Facto Clause, the petitioners are only entitled to the number of credits that should have been awarded under the prior statutes, not what other inmates were actually awarded under the Control Release program. To do otherwise would result in an undue windfall for the petitioners.[10] The petitioners also argue that those inmates who were only eligible for Emergency Gain Time at the time of their offenses should be awarded continuing amounts of early release credits totaling much more than the amount determinable by the Emergency Gain Time statute itself. See § 944.598, Fla. Stat. (1983-Supp. 1992). We disagree. According to subsections (1) and (2) of section 944.598, when the threshold level was surpassed, the Secretary was given authority to award up to thirty days Emergency Gain Time, in five-day increments, to the entire eligible inmate population. Subsection (2), however, makes clear that these "across-the-board" awards were limited to a certain period of time. If after fifteen days overcrowding continued despite the awards made under subsection (2), subsection (3) became activated.[11] Under subsection (3), prison overcrowding was further alleviated by awarding credits to a different, more limited group of inmates. These inmates would then be *507 placed on "compulsory conditional release." If an inmate was eligible for Emergency Gain Time under subsection (2), but was not eligible for continuing awards under subsection (3), then that inmate would be limited to the amount of credits awardable under subsection (2). We believe that the United States Supreme Court's somewhat different interpretation of that statute in Lynce v. Mathis, 519 U.S. 433, 448 n. 18, 117 S.Ct. 891, 137 L.Ed.2d 63 (1997), was not intended to control the manner in which the appropriate remedy should be calculated (in fact, the Supreme Court did not mention subsection (3) at all). On the contrary, the Supreme Court's discussion of that statute was merely a reaffirmation of the "core Ex Post Facto" argument that all the later overcrowding statutes were really the same, at least for purposes of deciding whether a later statute was merely a "revamping" of the prior statute. In other words, we believe that the Supreme Court's discussion of that statute was only meant to reject the Department of Corrections' assertions in that case that all the separate overcrowding statutes had nothing to do with each other and that an inmate's entitlement to overcrowding credits under one statute terminated upon the enactment of a new statute. There is one final issue that must be addressed before the State can calculate the exact number of credits due to each inmate. That issue concerns the three different definitions of prison overcrowding. Prison overcrowding occurs when the prison population exceeds the "lawful capacity" of the prison system.[12] Under all the overcrowding statutes, the award of credits was activated when the prison population exceeded the relevant percentage (either 97.5%, 98%, or 99%) of "lawful capacity." "Lawful capacity," however, was defined differently during various time periods. When the Emergency Gain Time statute was first enacted in 1983, the statute defined "lawful capacity" as "the total capacity of all institutions and facilities in the prison system as determined either by the Legislature or by the courts." See § 944.598(7)(b), Fla. Stat. (1983). In 1983, however, the legislature had not actually determined an exact definition, nor had any court done so. As discussed above, in the early 1980s, the Department was embroiled in the Costello prison overcrowding lawsuit. According to the Costello settlement agreement, effective July 1, 1985, the definition of "lawful capacity" was determined to be 133% of total design capacity. See § 944.598(6)(b) (note 1), Fla. Stat. (Supp.1992). Effective June 10, 1995, the definition was changed to 150%. See §§ 947.146(2); 944.096(4)(b); 944.023(1)(b), Fla. Stat. (1995); ch. 95-251, §§ 1-6, at 2235-38, Laws of Fla. The Department argues that it can change and apply retrospectively the definition of prison overcrowding by changing the percentage over which it is still "lawful" to exceed the total design capacity. The Department makes this assertion because it believes that it has always been understood that the definition of overcrowding could be changed. We cannot agree. We conclude that the definition of overcrowding, as it was defined at the time of the inmate's offense, is part and parcel of the relevant version of the overcrowding statute that should be applied to each inmate. We agree that the legislature may change the definition of prison overcrowding; however, since that change as proven by the Department's charts can drastically reduce or even eliminate the credits an inmate might have received under the prior definition, the Department cannot apply the definition retrospectively to inmates whose offenses were committed prior to the effective date of the change. Accordingly, the first definition of "lawful capacity" applies to inmates whose offenses occurred on or after June 16, 1983, *508 but before July 1, 1985. That definition does not apply to any of the petitioners in this case and thus, we decline to address it. The second definition is 133% of the total design capacity. That definition must be applied to inmates whose offenses were committed on or after July 1, 1985, but before June 10, 1995. See § 944.598(6)(b) (note 1), Fla. Stat. (Supp.1992). Accordingly, it applies to these petitioners. The third definition is 150% of total design capacity.[13] This definition, however, cannot be applied to these petitioners as concerns Emergency Gain Time, Administrative Gain Time, and Provisional Credits. See §§ 947.146(2); 944.096(4)(b); 944.023(1)(b), Fla. Stat. (1995); ch. 95-251, §§ 1-6, at 2235-38, Laws of Fla. Therefore, for each version of the pertinent overcrowding statute there is both a relevant percentage threshold (97.5%, 98%, 99%, 100%, etc.) over "lawful capacity" when the overcrowding statutes would activate, and there is a relevant definition of "lawful capacity" (either actual prison population, 133%, or 150%). IV. Conclusion Long before Lynce, the United States Supreme Court made clear in Weaver that, as concerns gain time, if a "new provision constricts the inmate's opportunity to earn early release, [it] thereby makes more onerous the punishment for crimes committed before its enactment [and] runs afoul of the prohibition against ex post facto laws." Weaver, 450 U.S. at 35-36, 101 S.Ct. 960. Lynce merely clarified that overcrowding credits must be treated the same as regular gain time. By retrospectively subjecting the petitioners to the more restrictive eligibility requirements of the subsequently enacted Control Release program, the State effectively repealed the prior programs in effect at the time of the petitioners' offenses and severely curtailed the availability of such credits to the petitioners, in violation of both Weaver and Lynce. The State must now rectify its error. We are cognizant of the tremendous effort the recalculation required under this opinion will entail on the part of the Department. Nevertheless, we believe this result is compelled by the decisions in Lynce and Weaver. We are bound to follow the law as expressed by the United States Supreme Court. This Court has repeatedly accepted the State's view concerning retroactive legislation restricting gain time. Each of those times the United States Supreme Court vacated our decisions. See Weaver v. Graham, 376 So.2d 855 (Fla.1979), rev'd, 450 U.S. 24, 101 S.Ct. 960, 67 L.Ed.2d 17 (1981); Calamia v. Singletary, 686 So.2d 1337 (Fla.1996), vacated, 520 U.S. 1141, 117 S.Ct. 1309, 137 L.Ed.2d 473 (1997); State v. Lancaster, 687 So.2d 1299 (Fla. 1997), vacated, ___ U.S. ___, 118 S.Ct. 37, 139 L.Ed.2d 5 (1997). We are compelled, therefore, to grant the petitioners' petitions for writ of habeas corpus.[14] Because we trust that the State will fully comply with the dictates of this opinion, we withhold issuance of the writ. It is so ordered. HARDING, C.J., and SHAW and ANSTEAD, JJ., concur. WELLS, J., dissents with an opinion. PARIENTE, J., recused. WELLS, J., dissenting. I dissent in this case because I believe that the majority has unnecessarily and unreasonably extended the United States Supreme Court's decision in Lynce v. *509 Mathis, 519 U.S. 433, 117 S.Ct. 891, 137 L.Ed.2d 63 (1997), far beyond that necessary to comply with that Court's mandate. The majority erroneously ignores the question which the Supreme Court expressly states that it decided in Lynce: The question presented by this case is whether a 1992 statute canceling such credits for certain classes of offenders after they had been awarded-indeed, after they had resulted in the prisoners' release from custody-violates the Ex Post Facto Clause of the Federal Constitution. Id. at 435, 117 S.Ct. 891 (emphasis added). Fundamental to a correct analysis is the recognition that Lynce and this case are patently factually distinguishable. Completely unlike the question presented in Lynce, here we have a situation in which the receipt of the subject credits was not only speculative at the time of the petitioners' offenses but they were simply never awarded. Since they were never awarded, they could not have been retroactively canceled. Furthermore, contrary to Lynce's situation, these prisoners were never released. Therefore, neither of the two essential factual components of Lynce (actual award and then actual release) are present. The reason these two basic factual components were absent is that the petitioners were not in the custody of the Department at the time that such credits were being awarded.[15] To give the petitioners prison overcrowding reduction credits when they were not even in prison during the time frame in which such credits were being awarded is to engage in a purely fictionalized scenario of what the petitioners might have received if they had been where they were not. I do not believe the United States Supreme Court intended for us to engage in such fantasy. I am quite aware that the Supreme Court in Weaver v. Graham, 450 U.S. 24, 101 S.Ct. 960, 67 L.Ed.2d 17 (1981), held in respect to the earned gain-time statutes that a law need not impair a "vested right" in order to violate the ex post facto prohibition. This clearly comports with reason in respect to the availability of earned gain time. As the United States Supreme Court wrote, a prisoner is "disadvantaged by the reduced opportunity to shorten his time in prison simply through good conduct." Id. at 33-34, 101 S.Ct. 960. However, contrary to the majority's reflexive reaction, I am not persuaded that the United States Supreme Court's decision in Lynce forces an illogical equating of earned gain-time credits and unpredictable prison overcrowding credits for every ex post facto analysis purpose. Rather, the United States Supreme Court merely said that the motivation of the legislature in passing the various gain-time statutes was immaterial for purposes of ex post facto analysis. 519 U.S. at 442, 117 S.Ct. 891. This should not be translated to mean that every person who ultimately ends up in the prison system is entitled to the benefit of such credits regardless of that person's actual factual circumstances. While the Supreme Court's decision in Lynce and its reversal in Calamia v. Singletary, 520 U.S. 1141, 117 S.Ct. 1309, 137 L.Ed.2d 473 (Fla.1997), of this Court's decision in Calamia v. Singletary, 686 So.2d 1337 (Fla.1996), may have eliminated the distinction between such credits after they have been awarded, this simply does not require the conclusion that there is no longer any distinction between such credits even before they have been awarded. Obviously, the credits in Lynce's situation were no longer speculative since the credits had been specifically in his records, were awarded to him, and had been acted upon to release Lynce. Lynce was not really a decision about a prisoner's "opportunities" for early release but rather a decision about the retroactive cancellation *510 of something which was no longer speculative. As the Supreme Court noted, the question concerned the "fact that [Lynce] was actually awarded 1,860 days of provisional credits and the fact that those credits were retroactively canceled." Lynce, 519 U.S. at 446, 117 S.Ct. 891. In other words, I believe Lynce should be read to say that while a prisoner may not have a right to expect something before he has it because it is speculative, after he receives it and uses it to attain his liberty, it is no longer speculative. Again, I would point to the two essential parts of the Supreme Court's question: whether an ex post facto violation occurs when there is (1) an actual award of credits; and (2) actual release from custody based on those credits. We simply do not have either of those two necessary elements in this case. Accordingly, there can be no violation. Even after Lynce, several courts have indicated that a proper ex post facto analysis still includes a determination of whether the sought-after early release opportunity is or was "speculative." In Hallmark v. Johnson, 118 F.3d 1073 (5th Cir.), cert. denied, ___ U.S. ___, 118 S.Ct. 576, 139 L.Ed.2d 415 (1997), for example, the court concluded that although the new gain-time provision at issue there had the potential to constrict the petitioners' opportunities for early release to some degree, that contingency would only occur if and when a number of additional speculative variables came into fruition. Accordingly, the court found no ex post facto violation. In Wottlin v. Fleming, 136 F.3d 1032 (5th Cir. 1998), the court found that since Wootlin's eligibility for early release had always been subject to the discretion of prison officials, his opportunity for release had been speculative from the beginning. The prison officials' determination that they would not exercise their discretion for violent offenders like Wootlin did little to Wootlin's hope for release since he might never have been chosen for release under the previous policy. Id. at 1037. Accordingly, as in Hallmark, the court found no ex post facto violation. The Supreme Court seems to have recognized in Lynce itself that the "speculative" component expounded upon in California Department of Corrections v. Morales, 514 U.S. 499, 506-513, 115 S.Ct. 1597, 131 L.Ed.2d 588 (1995), remains intact when it reasoned that "[i]n this case, unlike in Morales, the actual course of events makes it unnecessary to speculate about what might have happened." Lynce, 519 U.S. at 446, 117 S.Ct. 891. In Morales, the petitioner's chances of receiving parole had always been subject to the discretion of the parole board. Since Morales was serving sentences for several murder convictions, his chances for early release had never been very good. A later change in the parole rules extending the time period between parole considerations for persons convicted of more than one murder did little to change Morales' chances for early release. The Supreme Court found the rule changes had only a speculative effect on Morales and concluded that such changes do not violate the Ex Post Facto Clause. Morales, 514 U.S. at 507-509, 518, 115 S.Ct. 1597. I believe that the same conclusion should apply to this case. In this case, all of the petitioners committed their offenses while the emergency gain time statute was in existence. However, that statute could not go into effect until the Secretary of the Department of Corrections both certified certain information concerning the prison population and prison capacity and advised the Governor of the existence of a state of emergency. See § 944.598(1), Fla. Stat. (1983-Supp.1992). At that point, the Governor would certify the above information by letter. Id. Only then could the Secretary of the Department of Corrections actually declare a state of emergency. Id. However, even at that point, the award of credits was discretionary on the part of the Secretary. Id. The Department of Corrections has provided information indicating that no such *511 state of emergency was ever declared. Accordingly, the requisite preconditions never occurred, and the statute never went into effect. The declaration of a state of emergency was a necessary condition or contingency. This contingency was speculative at the time of the petitioners' offenses, and it never occurred. Furthermore, these inmates never even received the "substitute" credit that Lynce received.[16] Accordingly, all necessary contingencies failed to come into fruition. Thus, contrary to the situation in Lynce, the sought-after early release opportunity never materialized for these inmates. The fact that other credits actually materialized for other inmates should not be allowed to substitute for the requisite concreteness necessary to find an ex post facto violation.[17] Some of the petitioners claim credits under several of the later-enacted overcrowding statutes administered by the Department, as discussed in detail in the majority opinion. While these statutes did not require a declaration of a state of emergency, they did require a certification from the Secretary that the pertinent prison capacity level had been reached. At that point, the Secretary was given discretion to award credits. See § 944.276(1), Fla. Stat. (1987) (administrative gain time); § 944.277(1), Fla. Stat. (Supp.1988-Supp. 1992) (provisional credits). In hindsight, we know that the Secretary did actually award credits to a number of inmates. However, he never awarded credits to these inmates. Again, I do not think Lynce requires that we excuse the fact that credits never materialized for these inmates by pointing to other inmates who actually received credits. In sum, while I understand that the majority concludes that it is compelled to grant these petitions, having been reversed in other gain-time cases by the Supreme Court, I believe that this reflexive reaction is inappropriate here because what is presented is clearly distinguishable from Lynce. *512 Gomez Appendix A: Overcrowding relief programs — Dates — Thresholds ---------------------------------------------------------------------------------------------------------------------------------------------------------------- Arranged by Programs Result Offense Date ---------------------------------------------------------------------------------------------------------------------------------------------------------------- Petitioner Offense Emergency Emergency Admin. Prov. Prov. Cont. R. Cont. R. Cont. R. Cont. R. Gain Time Gain Time Gain Time Credits Credits below at or below between between 98% 99% 98% 97.5% 98% 97.5% 99% 99-100% 99-100% (TC) 6/16/83- 6/2/86- 2/5/87- 7/1/88- 9/1/90- 9/1/90- 6/17/93- 4/25/94- 6/10/956/1/86 6/16/93 6/30/88 8/31/90 6/16/93 6/16/93 4/24/94 6/9/95 present-TC Gomez 2* Murder X Entitled only to 9/11/86 limited credits under Emerg. Gain Time Kivet Manslaughter X X Entitled to both 10/18/87 Emerg. GT & Admin. GT** Hock 2* Murder X X Entitled to both 10/1/88 Emerg. GT & Prov. Credits** Goodwin DUI X X Entitled to both 5/6/89 Manslaughter Emerg. GT & Prov. Credits** Galvis Drag X Not eligible Not eligible Entitled only to 9/24/92 Trafficking until until limited credits mandatory mandatory under Emerg. Gain completed completed Time ** but no double credits for same period of time *513 Gomez Appendix B — Overcrowding Awards by Offender Group ---------------------------------------------------------------------- Effective Date Group 31 Group 42 Group 53 ---------------------------------------------------------------------- 06/30/1991 76 days 07/31/1991 109 days 10/31/1991 180 days 11/30/1991 172 days 04/30/1992 184 days 05/31/1993 220 days 06/30/1993 40 days 40 days 07/15/1993 07/31/1993 16 days 16 days 08/31/1993 51 days 51 days 09/30/1993 32 days 32 days 10/31/1993 36 days 36 days 11/30/1993 51 days 51 days 12/31/1993 56 days 56 days 01/31/1994 36 days 36 days 1 Group 3 includes offenders whose offenses were committed on or after June 2, 1986 but before June 17, 1993 (6/2/86-6/16/93) and who are eligible for Emergency Gain Time (at the 99% threshold), but who are not eligible for Administrative Gain time, Provisional Credits, or Control Release. 2 Group 4 includes offenders whose offenses were committed on or after February 5, 1987 but before July 1, 1988 (2/5/87-6/30/88) and who are eligible for Administrative Gain Time (at the 98% threshold), but who are not eligible for Provisional Credits or Control Release. 3 Group 5 includes offenders whose offenses were committed on or after July 1, 1988 but before September 1, 1990 (7/1/88-8/31/90) and who are eligible for Provisional Credits at the 97.5% threshold), but who are not eligible for Control Release. *514 02/28/1994 46 days 46 days 03/31/1994 32 days 32 days 04/28/1994 30 days 05/13/1994 *30 days 04/30/1994 34 days 34 days 05/31/1994 26 days 26 days 06/30/1994 33 days 33 days 07/31/1994 28 days 28 days 08/31/1994 48 days 48 days 09/30/1994 30 days 30 days 10/31/1994 37 days 37 days 11/30/1994 42 days 42 days 12/31/1994 37 days 37 days 01/31/1995 37 days 37 days 02/28/1995 37 days 37 days 03/31/1995 37 days 37 days 06/30/1995 1 day 08/31/1995 20 days 09/30/1995 25 days 10/31/1995 20 days 11/30/1995 1 day ---------------------------------------------------------------------- Total 60 days 822 days 1830 days ---------------------------------------------------------------------- * Assuming Offender meets criteria under section 944.598(3), Florida Statutes (Supp.1986). NOTES [1] In Meola v. Department of Corrections, we address gain time in the context of prisoners who had their Administrative Gain Time and/or Provisional Credits canceled. We hold that the petitioners are not entitled under ex post facto principles to the reinstatement of the overcrowding credits that they had actually been awarded, but which were subsequently canceled. We also hold that the petitioners were accorded due process in the cancellation of the credits and that there was no equal protection violation. In Thomas v. Singletary, [729 So.2d 369] (Fla.1998), we address the extension and then cancellation of prisoner's release dates under the Control Release program. We hold that since inmates were always on notice that their control release dates could be changed to a later date, the legislative amendments to the program and, ultimately, the cancellation of their release dates did not result in an ex post facto violation. In State v. Lancaster, No. 86,312 [687 So.2d 1299] (Fla. Dec. 24, 1998), in which we are granting relief, we address gain time in the context of prisoners who had their Administrative Gain Time or Provisional Credits forfeited upon revocation of probation. We hold that the State has statutory authority to forfeit their credits, but only for criminal offenses committed after a certain date and that the 1993 statute called the Safe Streets Initiative cannot be used to cancel their gain time as concerns inmates who were released on supervision before its enactment. While each of these opinions are similar because they discuss the effect that the United States Supreme Court decision in Lynce v. Mathis has had on gain time caselaw, they are clearly distinguishable, as discussed above. [2] See ch. 88-122, § 6, at 537, Laws of Fla. [3] See, e.g., §§ 947.165, 947.173, 947.174, Fla. Stat. (discussing Parole Commission's duties to periodically review and revise inmate parole dates). [4] Hock was awarded 360 days Provisional Credits but those credits were subsequently cancelled. In Calamia v. Singletary, 686 So.2d 1337 (Fla.1996), we affirmed that cancellation. That decision, however, was vacated by the United States Supreme Court in Calamia v. Singletary, 520 U.S. 1141, 117 S.Ct. 1309, 137 L.Ed.2d 473 (1997). On remand, we granted Hock's petition in light of the United States Supreme Court decision in Lynce v. Mathis, 519 U.S. 433, 117 S.Ct. 891, 137 L.Ed.2d 63 (1997). See Calamia v. Singletary, 694 So.2d 733 (Fla.1997). The Department has now restored Hock's 360 days of Provisional Credits, but has not released him. In this case, Hock is seeking the award of credits he was never awarded. [5] We have been informed by the Department that Galvis' prison sentence is made up entirely of a mandatory sentence. [6] As noted previously, none of the petitioners, with the exception of Hock, received any credits. Hock received 360 days Provisional Credits and has already had those credits restored. In this case, Hock is seeking additional credits that he never received. [7] Regular gain time awards (incentive and basic gain time) are determined in part by the date of offense, the type of offense and the inmate's behavior. The amount of regular gain time awardable is not statutorily tied to the amount of prison overcrowding. See generally § 944.275, Fla. Stat. (1997). [8] The Supreme Court seems to have indicated in Lynce that had the State been able to release other less dangerous inmates and therefore avoid the occurrence of the requisite levels of prison overcrowding, it might have avoided the necessity of releasing the more dangerous inmates pursuant to the prior statutes. It stated: The State, after all, could have alleviated the overcrowding problem in various ways: It could have built more prisons; it could have paroled a large category of nonviolent offenders; or it might have discontinued prosecution of some classes of victimless crimes. Lynce, 519 U.S. at 446, 117 S.Ct. 891 (emphasis added). [9] The Department must ensure, however, that its calculations have been made based on the relevant definition of prison overcrowding, as discussed below. [10] In addition, even if an inmate is eligible for more than one type of credit, that inmate should not be awarded duplicate (or double) credits for the same time period. [11] The time-restricted authority under section 944.598(2) contrasts with the later overcrowding statutes which specifically provided the State with continuing authority to award credits "across-the board" until the prison population fell below a certain percentage threshold. See § 944.276(2), Fla. Stat. (1987)(Administrative Gain Time); § 944.277(2), Fla. Stat. (Supp.1988)(Provisional Credits); § 947.146(2), Fla. Stat. (1989)(Control Release). [12] The term "lawful capacity" was later changed to "total capacity" in 1995. See §§ 947.146(2), 944.023(b), Fla. Stat. (1995); ch. 95-251, §§ 1-6, at 2235-38, Laws of Fla. [13] The 150% definition contains certain exclusions. For example, medical and mental facilities must remain at or below design capacity, i.e., they may not exceed 100% of design capacity. See § 944.023(1)(b)(1-6), Fla. Stat. (1995). [14] The general relief to be accorded is reflected in Appendices A and B. Inmates not parties to this action are generally bound by these appendixes but retain their right to challenge the applicability of the charts to their particular situations. [15] As noted in the majority opinion, petitioner Hock was in custody during a portion of the relevant time frames and received some credits. However, he has already had those credits restored, and this case does not concern those credits. [16] While Lynce was eligible at the time of his offense for only emergency gain time and never received that particular type of overcrowding credit, he eventually did receive another type of overcrowding credit (provisional credits). The Supreme Court concluded that for ex post facto purposes the statutes were essentially the same. 519 U.S. at 449, 117 S.Ct. 891 (noting that "[t]he changes in the series of statutes authorizing the award of overcrowding gain-time do not affect petitioner's core ex post facto claim.") [17] Petitioner Hock did receive some "substitute" credits, however, as noted above, he has already had those credits restored.