Coleman v. Brown

STEPHEN REINHARDT, Circuit Judge, LAWRENCE K. KARLTON, THELTON E. HENDERSON, Senior District Judges.

On June 20, 2013, this Court issued an Opinion and Order once again directing defendants to comply with our August 2009 Population Reduction Order by reducing the prison population to 137.5% design capacity by December 31, 2013. June 20, 2013 Op. & Order, 952 F.Supp.2d 901, 2013 WL 3326872 (E.D.Cal.2013) (ECF No. 2659/4662).1 The Population Reduction Order, although almost four years old, has still not been complied with by defendants. On June 28, 2013, defendants requested a stay of the June 20 Order pending appeal to the United States Supreme Court. Defs.’ Mot. to Stay (ECF No. 2665/4673). For the reasons set forth below, we DENY defendants’ motion for a stay.

It is worth stating at the outset that by its underlying appeal defendants (sometimes referred to as “the State”) seek to relitigate a thoroughly reasoned decision of the Supreme Court, Brown v. Plata, issued two years ago. That decision holds that within two years the State must re*1060duce its prison population to 137.5% of design capacity because, when a higher number of prisoners is confined in the prisons, the prison conditions result in medical and mental health care that violates the Eighth Amendment. - U.S. •-■, 131 S.Ct. 1910, 179 L.Ed.2d 969 (2011). Because of the State’s resistance to complying with that decision, and in order to avoid the necessity of contempt proceedings against the Governor and other state officials, this Three-Judge Court has repeatedly declined to initiate such proceedings and has even sua sponte extended the time for defendants to comply with the Population Reduction Order issued in conformity with Brown v. Plata. This Court has repeatedly directed defendants to adopt specific plans that will serve to reduce the prison population to the designated figure by the specified date. Until now, the State has insisted that it is unable (read unwilling) to comply with the Population Reduction Order. In the present motion, however, it has finally acknowledged that it will comply if the Supreme Court denies the stay it will request from that Court. Defs.’ Mot. to Stay at 2 (ECF No. 2665/4673). Accordingly, with anticipation that the Supreme Court’s denial of the stay will finally bring defendants into compliance with the Population Reduction Order and the Eighth Amendment (subject to the durability of its compliance), we further explain our reasons for denying defendants’ motion.

I. PROCEDURAL HISTORY

The history of this litigation is of defendants’ repeated failure to take the necessary steps to remedy the constitutional violations in its prison system, violations that have still not been remedied after 23 years. The litigation began with two separate class actions. The first, Coleman v. Brown, began in 1990 and concerns California’s failure to provide constitutionally adequate mental health care to its prison population. The second, Plata v. Brown, began in 2001 and concerns California’s failure to provide constitutionally adequate medical care to its prison population. The district courts in both cases found constitutional violations and ordered injunctive relief. In 1995, in Coleman, the district court found that defendants were violating the Eighth Amendment rights of mentally ill prisoners. Coleman v. Wilson, 912 F.Supp. 1282 (E.D.Cal.1995). The court appointed a Special Master to supervise defendants’ efforts to remedy the constitutional violations. Id. at 1323-24. In 2005, in Plata, after a stipulated injunction failed to remedy the Eighth Amendment violations, the district court placed defendants’ prison medical care system in a receivership. Oct. 3, 2005 FF & CL, 2005 WL 2932253, at *31. Now, 23 years later in one case and 12 years later in the other, despite the extensive efforts we have made to bring about compliance with our Population Reduction Order, which has been approved by the Supreme Court, defendants remain delinquent.

“After years of litigation, it became apparent that a. remedy for the constitutional violations would not be effective absent a reduction in the prison system population.” Plata, 131 S.Ct. at 1922. In 2006, the Coleman and Plata plaintiffs independently filed motions to convene a three-judge court capable of issuing a population reduction order under the PLRA. Both motions were granted, and on July 26, 2007, the cases were assigned to the same Three-Judge Court, made up of the district judges overseeing Plata and Coleman and one circuit judge appointed in conformance with Court Rules by the Chief Judge of the Circuit. After a fourteen-day trial, this Three-Judge Court issued a 184-page opinion ordering defendants to reduce the institutional prison population to 137.5% design capacity within two years. Aug. 4, 2009 Op. & Order, 922 *1061F.Supp.2d 882 (E.D.Cal.2009) (ECF No. 2197/3641) (“Population Reduction Order”).

In issuing the Population Reduction Order, this Court found that “no relief other than a prisoner release order is capable of remedying the constitutional deficiencies at the heart of these two cases,” id. at 962, and that “there was overwhelming agreement among experts for plaintiffs, defendants, and defendant-intervenors that it is ‘absolutely’ possible to reduce the prison population in California safely and effectively,” id. at 974. We did not instruct defendants how to reduce the prison population. We left this question to defendants but ordered them to submit a plan for compliance within 45 days of our Population Reduction Order. Id. at 1003-04. Defendants did not comply; they submitted a plan for reducing the population to 137.5% within five years, not two. Defs.’ Population Reduction Plan (ECF No. 2237/3678). This Court ordered defendants to comply by providing a two-year plan. Oct. 21, 2009 Order Rejecting Defs.’ Proposed Population Plan (ECF No. 2269/3711). Defendants responded with a plan for compliance by which they would reduce the prison population to 167%, 155%, 147%, and 137.5% at six-month benchmarks. Defs.’ Response to Three-Judge Court’s Oct. 21, 2009 Order (ECF No. 2274/3726). On January 12, 2010, this Court issued an order accepting defendants’ two-year timeline, but stayed the date of the order while defendants appealed to the Supreme Court. Jan. 12, 2010 Order to Reduce Prison Population, 2010 WL 99000 (ECF No. 2287/3767).

In June 2011, the Supreme Court affirmed this Court’s Population Reduction Order, holding that “the court-mandated population limit is necessary to remedy the violation of prisoners’ constitutional rights.” Plata, 131 S.Ct. at 1923. Although the Population Reduction Order, the Supreme Court stated, was “of unprecedented sweep and extent,” and the release of prisoners a matter of “undoubted, grave concern,” so too “is the continuing injury and harm resulting from these serious constitutional violations.” Id. The Supreme Court rejected defendants’ argument that a population reduction order was not required because the overcrowding could be eliminated through construction and other efforts. The Supreme Court called such options “chimerical,” id. at 1938-39, and noted that defendants’ troubled history in this litigation belied placing trust in them. The Supreme Court said:

Attempts to remedy the violations in Plata have been ongoing for 9 years. In Coleman, remedial efforts have been ongoing for 16. At one time, it may have been possible to hope that these violations would be cured without a reduction in overcrowding. A long history of failed remedial orders, together with substantial evidence of overcrowding’s deleterious effects on the provision of care, compels a different conclusion today.

Id. at 1939. The Supreme Court also rejected defendants’ argument that population reduction would adversely affect public safety, citing this Court’s extensive factual findings to the contrary. Id. at 1942-43. The Supreme Court specifically endorsed expanding good time credits, stating that “[ejxpansion of good time credits would allow the State to give early release to only those prisoners who pose the least risk of reoffending,” id. at 1943, and cited positive evidence from other jurisdictions that had successfully implemented good time credits, id. at 1942-43. The Supreme Court concluded that “[t]he relief ordered by the three-judge court is required by the Constitution and was authorized by Congress in the PLRA,” and ordered defendants to “implement the order without further delay.” Id. at 1947.

*1062Following the Supreme Court’s decision, this Court mandated a two-year schedule for defendants to reduce the prison population to 137.5% design capacity: 167% design capacity by December 27, 2011; 155% design capacity by June 27, 2012; 147% design capacity by December 27, 2012; and 137.5% design capacity by June 27, 2013. June 30, 2011 Order Requiring Interim Reports at 1-2 (ECF No. 2375/4032). Defendants responded by informing this Court that they would reach these benchmarks primarily through “Realignment,” a- measure authorized by Assembly Bill 109 that shifted criminals who had committed “non-serious, non-violent, and non-registerable sex crimes” from state prisons to county jails. Defs.’ Resp. to Jan. 12, 2010 Court Order (ECF No. 2365/4016). Realignment went into effect in October 2011 and enabled defendants to comply with the first benchmark shortly after the December 27, 2011 deadline. Defs.’ Jan. 6, 2012 Status Report (ECF No. 2411/4141).

It soon became apparent, however, that Realignment alone would not be sufficient to meet the 137.5% design capacity benchmark by June 2013. In February 2012, plaintiffs filed a motion asking defendants to show cause as to how they would reach this benchmark. They insisted that based on the California Department of Corrections and Rehabilitation’s (“CDCR’s”) own Fall 2011 population projections, defendants would not meet the benchmark. Pis.’ Mot. for an Order Requiring Defs. to Demonstrate How They Will Achieve the Required Population Reduction by June 2013 at 2-3 (ECF No. 2420/4152). Defendants responded that the CDCR’s Fall 2011 population projections were not reliable and that the forthcoming Spring 2012 population projections would be more accurate. Defs.’ Opp’n to Pis.’ Mot. for Increased Reporting in Excess of the Court’s June 30, 2011 Order at 2-4 (ECF No. 2423/4162). This Court accepted defendants’ argument and denied plaintiffs’ motion without prejudice. Mar. 22, 2012 Order Denying Pis.’ Feb. 7, 2012 Mot. (ECF No. 2428/4162). Two months later, plaintiffs renewed their motion, correctly observing that the Spring 2012 population projections were not significantly different from the Fall’s. Pis.’ Renewed Mot. for an Order Requiring Defs. to Demonstrate How They Will Achieve the Required Population Reduction by June 2013 (ECF No. 2435/4180). Plaintiffs also informed this Court of a new public report, “The Future of California Corrections” (“The Blueprint”), in which defendants stated that they would not meet the 137.5% June 2013 benchmark and would seek modification of this Court’s Population Reduction Order. See CDCR, The Future of California Corrections: A Blueprint to Save Billions of Dollars, End Federal Oversight, and Improve the Prison System, Apr. 2012, available at http://www.cdcr.ca.gov/2012plan/ does/plan/complete.pdf Plaintiffs asked that defendants be held in contempt. Defendants responded, informing us that they intended to seek modification of our Population Reduction Order to increase the final benchmark from 137.5% to 145% design capacity. Defs.’ Opp’n to Pis.’ Renewed Mot. for an Order Requiring Defs. to Demonstrate How They Will Achieve the Required Population Reduction by June 2013 at 2 (ECF No. 2442/4192).

This Court ordered two rounds of supplemental briefing regarding defendants’ anticipated motion to modify the Population Reduction Order. June 7, 2012 Order Requiring Further Briefing (ECF No. 2445/4193); Aug. 3, 2012 2d Order Requiring Further Briefing (ECF No. 2460/4220). In response, defendants retreated, stating that they believed it would be premature to begin modification proceedings before the prison population reached 145% design capacity, which they predicted would happen in February or March of 2013. Defs.’ Resp. to Aug. 3, 2012 2d Order Requiring *1063Further Briefing at 9-10 (ECF No. 2463/4226). In September 2012, we again denied without prejudice plaintiffs’ request that defendants be held in contempt. We also asked defendants to answer questions they had failed to respond to in their supplemental briefing, namely how long it would take them to develop a system for identifying low-risk offenders for early release (“Low-Risk List,” a list recommended by the Supreme Court in Plata, 131 S.Ct. at 1947), and whether they could comply with our Population Reduction Order by June 2013, and if not, when the earliest time they could comply by would be. Sept. 7, 2012 Order Granting in Part & Denying in Part Pis.’ May 9 and Aug. 22, 2012 Mots. (ECF No. 2473/4235). Defendants responded that they needed six months to develop the Low-Risk List and that they could comply with our Population Reduction Order with a six-month extension, largely by maintaining the out-of-state program. Defs.’ Resp. to Sept. 7, 2012 Order at 5-6 (ECF No. 2479/4243). Believing that resolution was close, this Court ordered both parties to meet and develop plans to reduce the prison population to 137.5% design capacity by (a) June 27, 2013, and (b) December 27, 2013. Oct. 11, 2012 Order to Develop Plans to Achieve Required Prison Population Reduction at 1 (ECF No. 2485/4251).

On January 7, 2013, both parties filed plans to meet the 137.5% design capacity benchmark. Defendants stated that they could comply by December 2013 without the release of prisoners. Defs.’ Resp. to Oct. 11, 2012 Order (ECF No. 2511/4284). But, despite this promising report, not long after this filing defendants refused to take further action to comply with our Population Reduction Order. First, in their January, February, and March status reports, defendants stated that they would take no further action to comply with the Order. See Defs.’ Jan. 2013 Status Report at 1 (ECF No. 2518/4292); Defs.’ Feb. 2013 Status Report at 1 (ECF No. 2538/4342); Defs.’ March 2013 Status Report at 1 (ECF No. 2569/4402). Second, the Governor terminated his emergency powers, declaring that the crisis in the prisons was resolved. Gov. Edmund G. Brown Jr., A Proclamation by the Governor of the State of California, Jan. 8, 2013, available at http://gov.ca.gov/news.phpTid=17885. As a result, defendants were no longer able to contract to house approximately 9,500 prisoners in out-of-state prisons, forcing a scheduled partial return of these prisoners during 2013, and a consequent increase in the prison population. Third, defendants filed a motion in the Coleman court to terminate all injunctive relief in that case. Mot. to Terminate & to Vacate J. & Orders (Coleman ECF No. 4275). Fourth, defendants filed a motion to vacate or modify our Population Reduction Order. Defs.’ Mot. to Vacate or Modify Population Reduction Order (ECF No. 2506/4280) (“Three-Judge Motion”). This motion did not await the prison population’s reaching a design capacity of 145%, as defendants had said they would, see supra p. 1062. In fact, the prison population has still not reached that figure.

This Court stayed consideration of defendants’ Three-Judge Motion on January 29, 2013. At the same time, we granted defendants a six-month extension to comply with our Population Reduction Order, extending the final 137.5% design capacity benchmark to December 31, 2013. Jan. 29, 2013 Order at 2-3 (ECF No. 2572/4317). On April 11, 2013, this Court denied defendants’ Three-Judge Motion, as modified,2 and ordered that they take *1064all steps to comply with the Population Reduction Order. Apr. 11, 2013 Op. & Order, 922 F.Supp.2d at 1008-09 (ECF No. 2590/4541). We gave three reasons for denying defendants’ Three-Judge Motion. First, it was barred by res judicata as an improper attempt to relitigate the 137.5% figure that we had determined in 2009 and that the Supreme Court had explicitly affirmed. Id. at 1031-32. Second, defendants did not meet their burden under Federal Rule of Civil Procedure 60(b)(5) to prove a “significant and unanticipated change in factual conditions warranting modification.” Id. at 1034 (citing United States v. Asarco Inc., 430 F.3d 972, 979 (9th Cir.2005) (summarizing Rufo v. Inmates of Suffolk Cnty. Jail, 502 U.S. 367, 384-86, 112 S.Ct. 748, 116 L.Ed.2d 867 (1992))). Third, defendants failed to demonstrate a “durable” solution that would justify this Court’s vacating a prior order. Id. at 1043-44.

To ensure compliance with our Population Reduction Order, this Court asked defendants to submit a list of all population reduction measures discussed as possible remedies during the course of the Three-Judge Court proceedings (“List”) and, from that List, suggest a plan for compliance with the Population Reduction Order (“Plan”), without regard to whether defendants had the authority to implement the measures designated. We further ordered defendants to use their best efforts to implement the Plan, and to inform us of their progress in their monthly reports. Finally, we ordered defendants to develop a “Low-Risk List” that they might use, if necessary, to comply with the Population Reduction Order by releasing low-risk offenders. Apr. 11, 2013 Order (ECF No. 2491/4542).

On May 2, 2013, defendants submitted their List and Plan. Defs.’ Resp. to April 11, 2013 Order (ECF No. 2609/4572). Defendants’ response did not comply with our April 11 Order, as they did not provide a Plan that would reach the 137.5% population benchmark by December 31, 2013. Defendants conceded as much, id. at 5 n. 3, 37 (acknowledging that their latest Plan will not achieve the 137.5% figure by December 31, 2013), although they underestimated the scope of their noncompliance. They estimated that their Plan would result in a prison population of 140.7% design capacity by December 31, 2013; in fact, their Plan might at best achieve a prison population of 142.6% design capacity by December 31, 2013-4,170 prisoners short of the 137.5% benchmark. See June 20, 2013 Op. & Order, 952 F.Supp.2d at 920-23, at *17-19(ECF No. 2659/4662) (explaining this discrepancy). Thus, well into the third decade of litigation, it was clear that defendants remained unwilling to implement a plan that would comply with the *1065Population Reduction Order and the Supreme Court’s 2011 decision.

II. JUNE 20, 2013 OPINION & ORDER

On June 20, 2013, in response to defendants’ proposed Plan that would not in any event achieve compliance, and facing a “long and unhappy history of litigation,” this Court entered a “comprehensive order to insure against the risk of inadequate compliance.” June 20, 2013 Op. & Order, 952 F.Supp.2d at 926, at *21 (ECF No. 2659/4662) (quoting Hutto v. Finney, 437 U.S. 678, 687, 98 S.Ct. 2565, 57 L.Ed.2d 522 (1978)). We ordered defendants to implement an “Amended Plan” consisting of the measures in their proposed Plan plus an additional measure consisting of the expansion of good time credits, prospective and retroactive, to all prisoners, as set forth in defendants’ List, but not in their Plan. This additional measure would provide the 4,170 prisoners needed to bring defendants’ Plan into compliance, assuming that the compliance would be durable. We carefully explained our reason for choosing this particular measure. First, extensive testimony at the 2009 trial revealed that good time credits were the most promising measure for reducing overcrowding. Id. at 927, at *22. The measure would in many cases reduce the prison population by allowing prisoners to shorten their lengths of stay in prison by as little as a few months. At trial, plaintiffs’ experts — Doctors Austin and Kris-berg, and Secretaries Woodford, Lehman, and Beard — were unanimous in their agreement that “such moderate reductions in prison sentences do not adversely affect either recidivism rates or the deterrence value of imprisonment.” Aug. 4, 2009 Op. & Order, 922 F.Supp.2d at 976 (ECF No. 2197/3641). Defendants’ one expert in opposition, Dr. Marquart, did not in fact oppose good time credits. Id. at 975-76. His only criticism — that good time credits expansion might reduce the opportunity for prisoners to complete rehabilitative programming — was, in our final determination, “a note about the factors that should be considered in designing an effective expanded good time credits system. It is entitled to little, if any, weight as an observation about the possible negative effect on public safety of such a system.” Id. at 977. Based on this and other testimony, we concluded following trial that early release through good time credits does not increase the crime rate but rather “affects only the timing and circumstances of the crime, if any, committed by a released inmate.” Id. at 977. We further “credited] the opinions of the numerous correctional experts that the expansion of good time credits would not adversely affect but rather would benefit the public safety and the operation of the criminal justice system.” Id. at 979.

Second, we rejected defendants’ arguments against expanding the good time credits measure by applying it retroactively and to all offenders. June 20, 2013 Op. & Order, 952 F.Supp.2d at 928-30, at *23-24 (ECF No. 2659/4662). Defendants first argued that, although prospective application of good time credits for prisoners convicted of non-violent offenses is safe, retroactive application of these credits to these same prisoners is not safe. Defendants provided no support for this proposition. Moreover, both the Plata Receiver and the State’s own CDCR Expert Panel had recommended making the good time credits changes retroactive. See Receiver’s 23rd Report at 33 (ECF No. 2636/4628); CDCR Expert Panel, A Road-map for Effective Offender Programming in California: A Report to the California Legislature, June 2007, at 95.3 Defen*1066dants further argued that good time credits should not be afforded to prisoners convicted of violent offenses. Yet not a single expert at trial distinguished between inmates convicted of violent and non-violent crimes for the purposes of good time credits, and the CDCR Expert Panel specifically recommended expanding good time credits for all prisoners, “including all sentenced felons regardless of their offense or strike levels.” Id. at 92. Based on these observations, we concluded that defendants’ arguments against expanding the good time credit measure were without merit.

Third, we noted the success other jurisdictions experienced in safely expanding their good time credits programs. June 20, 2013 Op. & Order, 952 F.Supp.2d at 908-09, 930 & n. 26, at *5-6, *24 & n. 26 (ECF No. 2659/4662). California has instituted good time credit programs in 21 counties between 1996 and 2006, resulting in approximately 1.7 million inmates having been released by court order without an increase in the crime rate. Id. at 908, at *5 (citing testimony by Dr. Krisberg, Aug. 4, 2009 Op. & Order, 922 F.Supp.2d at 978-79 (ECF No. 2197/3641)). Washington expanded its good time credits program and Secretary Lehman, the former head of corrections for Washington, testified at trial that “these measures did not have any ‘deleterious effect on crime’ or public safety.” Id. (citing Aug. 4, 2009 Op. & Order, 922 F.Supp.2d at 980-81 (ECF No. 2197/3641)). Illinois, Nevada, Maryland, Indiana, and New York all successfully implemented good time credits expansion without adversely affecting public safety. Id. In New York, the prison population decreased due in part to the expansion of programs awarding good time credits, and the crime rate declined substantially. Id.

Finally, we pointed out that the Supreme Court had expressly endorsed the good time credits measure: “Expansion of good-time credits would allow the State to give early release to only those prisoners who pose the least risk of reoffending.” Plata, 131 S.Ct. at 1943. The Supreme Court also approvingly discussed the empirical and statistical evidence from other jurisdictions that had successfully implemented good time credits. Id. at 1942-43 (listing the experience in certain California counties, Washington, etc.). In endorsing the good time credits measure, the Supreme Court stated that this Court’s factual findings on public safety were to be credited over the contrary views of defendants. Id. at 1942. The Supreme Court was in clear agreement with this Court that defendants could reduce the prison population to 137.5% design capacity without adversely affecting public safety, specifically through the expansion of good time credits. For these reasons, we ordered defendants to implement an “Amended Plan” consisting of their Plan and the expanded good time credits measure.

Although this Court ordered defendants to implement the Amended Plan, including good time credits, we emphasized that we desired to “continue to afford a reasonable measure of flexibility to defendants, notwithstanding their failure to cooperate with this Court or to comply with our orders during the course of these proceedings.” June 20, 2013 Op. & Order, 952 F.Supp.2d at 918, at *15 (ECF No. 2659/4662). To this end, this Court offered defendants three methods of making substitutions to the measures in the Amended Plan. First, defendants may, if they prefer, revise the good time credit measure currently proposed such that it does not result in the release of violent offenders, so *1067long as the revision results in the release of at least the same number of prisoners as would the current good time credit measure.4 This may be accomplished in part by adjusting the credits to levels awarded by other states or counties. Second, defendants may substitute for any group of prisoners who are eligible for release under the Amended Plan a different group of prisoners consisting of no less than the same number of prisoners selected pursuant to the Low-Risk List, with the substitution being in the order in which the prisoners are listed, individually or by category on that list. Third, defendants may substitute any group of prisoners from the List of all population reduction measures identified in this litigation, submitted by defendants on May 2, 2013, for any groups contained in a measure listed in the Amended Plan, should defendants conclude by objective standards that they are no greater risk than the prisoners for whom they are to be substituted. Id. We provided examples of such substitutions: “Lifers,” who, due to age or infirmity, are adjudged to be “low risk” by CDCR’s risk instrument; prisoners who have nine months or less to serve of their sentence who could serve the duration of their sentences in county jails rather than in state prisons; or prisoners who could be reassigned from state prisons to leased jail space. Id. at 936, at *30.5 By “Lifers,” we refer to the category of prisoners who are serving sentences of a fixed number of years to life and are eligible for parole. As of 2011, there were 32,000 Lifers in California state prisons. Lifers made up 20% of the California prison population, an increase from 8% in 1990. A 2011 study by the Stanford Criminal Justice Center reported that “the incidence of commission of serious crimes by the recently released lifers has been minuscule.” Robert Weisberg, Debbie A. Mukamal & Jordan D. Segall, Life in Limbo: An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California at 3^1 (Stanford Criminal Justice Center, Sept. 2011).

Our June 20, 2013 Order was not the first time we have given defendants a broad choice in determining how to comply with our Population Reduction Order. Over the past four years, this Court has done everything possible to ensure that defendants have flexibility in adopting measures that will attain compliance. We have never ordered defendants to select any particular measures; rather, we have consistently offered defendants the choice as to how they will reach the 137.5% design capacity benchmark. Our Population Reduction Order merely asked for a plan for compliance. Aug. 4, 2009 Op. & Order, 922 F.Supp.2d at 1003-04 (ECF No. 2197/3641). Our January 2010 Order accepted defendants’ two-year timeline for compliance without ordering them to im*1068plement any specific measures. Jan. 12, 2010 Order to Reduce Prison Population at 4 (ECF No. 2287/3767). Our April 11, 2013 Order deferred to defendants for a Plan for reaching the 137.5% design capacity benchmark that they found most acceptable. Finally, and as described in detail above, although our most recent order, issued on June 20, 2013, makes suggestions as to how defendants could reduce the prison population to 137.5% by December 31, 2013, it leaves defendants significant flexibility in deciding how to reach this cap. See June 20, 2013 Op. & Order, 952 F.Supp.2d at 924, at *20 (ECF No. 2659/4662) (“We are willing to defer to [defendants’] choice for how to comply with our Order, not whether to comply with it.”).

Further, this Court has twice extended deadlines for compliance for defendants, even without their formally requesting that we do so. In January 2010, when this Court ordered defendants to reduce the prison population to certain benchmarks every six months, we sua sponte stayed this order pending appeal to the Supreme Court. Jan. 12, 2010 Order to Reduce Prison Population at 6 (ECF No. 2287/3767). Because the Supreme Court’s decision was not issued until June 2011, defendants gained an additional two years with which to comply with this Court’s Population Reduction Order-an additional two years that the Supreme Court recognized in endorsing our two-year timeline. Plata, 131 S.Ct. at 1946 (noting that “defendants will have already had over two years to begin complying with the order of the three-judge court”). Then, on January 29, 2013, again without any formal request by defendants, this Court once more extended the deadline, giving defendants six additional months to comply with our Population Reduction Order. Jan. 29, 2013 Order at 2-3 (ECF No. 2527/4317). As a result, defendants will have had well over four years to comply with our Population Reduction Order — more than twice the amount of time contemplated in that Order.

Despite our repeated efforts to assist defendants to comply with our Population Reduction Order, they have consistently engaged in conduct designed to frustrate those efforts. They have continually sought to delay implementation of the Order. At the time of the Population Reduction Order, defendants asked this Court to wait for “chimerical” possibilities. Plata, 131 S.Ct. at 1938. As the Order was appealed to the Supreme Court, defendants insisted that this Court had been convened prematurely and that alternative remedies to a prisoner release order existed. The Supreme Court rejected these arguments, ordering defendants to “implement the order without further delay.” Id. at 1947. Defendants have doné no such thing. They have refused to follow the Supreme Court’s order. They took one action, Realignment, and when it became apparent that this action would be insufficient to comply with the Population Reduction Order, defendants refused to take any further steps to reduce the prison population to 137.5% design capacity. Instead, they moved to terminate all prospective relief granted by the Coleman court under the PLRA’s termination provision, moved to vacate the Population Reduction Order issued by this Court under Federal Rule of Civil Procedure 60(b)(5), voluntarily terminated their own emergency powers to house prisoners out of state, and reported in their monthly status reports that they would no longer take actions to comply with the Population Reduction Order. Governor Brown declared, notwithstanding the orders of this Court, that the crisis in the prisons was resolved. See Gov. Edmund G. Brown Jr., A Proclamation by the Governor of the State of California, Jan. 8, 2013, http://gov.ca.gov/news.php? id=17885. Finally, when asked to submit a *1069Plan for compliance, defendants submitted, instead, a Plan for noncompliance — a Plan that fell far short of the required figures.

In defense of their actions, defendants equivocate regarding the facts and the law. For example, defendants have repeatedly asserted that they have reduced the prison population by “more than 42,000 inmates since 2006.” Defs.’ Resp. to Pis.’ Resp. & Req. for Order to Show Cause Regarding Defs.’ Resp. to Apr. 11, 2013 Order at 3 (ECF No. 2640/4365); see also Defs.’ Resp. to Apr. 11, 2013 Order at 39 (ECF No. 2609/4572) (same). This statistic is misleading, as it includes reductions made between 2006 and 2009, before we issued our Population Reduction Order. Similarly, in defense of their May 2013 Plan for noncompliance, defendants stated that they have “taken all actions in [their] power” to reach the December 2013 population cap, arguing that they are either without authority to take further measures or that such measures would threaten public safety. Defs.’ Resp. to Pis.’ Resp. & Req. for Order to Show Cause Regarding Defs.’ Resp. to Apr. 11, 2013 Order at 1 (ECF No. 2640/4365). In making this statement, defendants failed to recognize that they could have met the 137.5% cap by increasing capacity, a measure that would have reduced overcrowding without releasing any prisoners; asked the legislature to modify the restrictions to which they adverted in submitting an insufficient Plan; requested changes to sentencing policies that would have reduced the prison population substantially; or retained, instead of surrendering, emergency authority regarding housing prisoners out of state. Given defendants’ history of noncompliance, it comes as no surprise that they have requested a last-minute stay of our June 20, 2013 Order, rather than making any effort to comply with the 2011 mandate of the Supreme Court. Of crucial importance, however, defendants now state that absent a stay by this Court and the Supreme Court, they will comply with the Population Reduction Order. Defs.’ Mot. to Stay at 2 (ECF No. 2665/4673). Such compliance, if durable, will bring the California prison population into conformity with the Eighth Amendment.

As to the timeliness of defendants’ request for a stay, as plaintiffs point out in their thorough and thoroughly reasoned response, see Pis.’ Am. Resp. to Defs.’ Mot. to Stay (ECF No. 2669/4677), defendants’ sense of urgency appears to be as newly developed as their sense of urgency regarding the appeal of the Population Reduction Order, which is now over four years old. This Court’s April 11, 2013 Order, the order immediately prior to the one at issue here, and one of a number of orders directing defendants to comply with the Population Reduction Order, was appealed by defendants to the Supreme Court on May 13, 2013, yet no stay was requested following that appeal. Defs.’ Notice of Appeal to the Supreme Court at 3 (ECF No. 2621/4605). Moreover, despite the familiarity defendants’ counsel undoubtedly have with this case after at least four years of uninterrupted litigation, they felt compelled to obtain a 45-day extension of time in which to file a jurisdictional statement before the Supreme Court, thus belying the need for urgency in resolving the appeal. See Brown v. Plata, Sup.Ct. Docket No. 13A5, available at http://www.supremecourt.gov/Search. aspx? FileN ame =/docketfiles/13a5.htm (noting that, on July 1, 2013, Justice Kennedy granted defendants’ June 25, 2013 application to extend the time to file a jurisdictional statement on appeal from July 12, 2013, to August 26, 2013).

III. DISCUSSION

In considering an application for a stay, this Court considers: (1) whether the stay applicant has made a strong show*1070ing that he is likely to succeed on the merits; (2) whether the applicant will be irreparably injured absent a stay; (3) whether issuance of the stay will substantially injure the other parties interested in the proceeding; and (4) where the public interest lies. Nken v. Holder, 556 U.S. 418, 434, 129 S.Ct. 1749, 173 L.Ed.2d 550 (2009); Humane Soc. of U.S. v. Gutierrez, 558 F.3d 896, 896 (9th Cir.2009). Applying these factors to this case, this Court has no difficulty in denying defendants’ request for a stay.

First, defendants have not made a strong showing that they are likely to succeed on the merits. Defendants appear to make two arguments regarding this factor in their application for a stay: (1) there are no longer any underlying constitutional violations; and (2) even if constitutional violations remain, additional population reductions are not necessary to remedy these violations. Defs.’ Mot. to Stay at 8-9 (ECF No. 2665/4673). The first argument is not raised before this Three-Judge Court. As explained supra p. 1063 & n. 2, although defendants initially posed this question in their January 7, 2013 Three-Judge Motion to Vacate the Population Reduction Order, they later modified the motion by removing any constitutional question from the purview of this Court. Defendants have also never made this argument before the Plata court. That is, they have not asked the Plata court or this Court to determine that defendants are no longer failing to provide constitutionally adequate medical health care to its prison population or to vacate injunctive relief on that ground. They have, in fact, made this argument only once. They did so before the Coleman court, on January 7, 2013. They asked that court to terminate all injunctive relief in Coleman on the ground that California’s mental health care system for prisoners no longer violates the Eighth Amendment. See Mot. to Terminate & Vacate J. & Orders at 28 (Coleman ECF No. 4275). The Coleman court denied defendants’ motion to terminate on the ground that “ongoing constitutional violations remain” “in the delivery of adequate mental health care.” Apr. 5, 2013 Order Denying Defs.’ Mot. to Terminate at 67 (Coleman ECF No. 4539). Moreover, we have made clear that our Population Reduction Order relied on each of the two cases individually and collectively, and that if the constitutional violations exist in either case, they exist for the purposes of this Three-Judge Court.6 Thus, even if plaintiffs were to file a motion to dismiss in the Plata case on the ground that the medical health care system in California *1071prisons no longer violates the Eighth Amendment, and even if they were to succeed on that motion, which appears to be highly unlikely, our Population Reduction Order would still be necessary to remedy the constitutional violations that remain in Coleman.7

Defendants’ second argument, that even if constitutional violations remain, additional population reductions are not necessary to remedy these violations, has been raised properly, see Three-Judge Mot. (ECF No. 2506/4280), but is without merit. In 2011, the Supreme Court affirmed in full this Court’s finding that the only way to remedy the ongoing constitutional violations in California prisons is to reduce the prison population to 137.5% of design capacity. Plata, 131 S.Ct. at 1945 (“There are also no scientific tools available to determine the precise population reduction necessary to remedy a constitutional violation of this sort. The three-judge court made the most precise determination it could in light of the record before it.”). In fact, describing the evidence before the Three-Judge Court, the Supreme Court said that the evidence supported “an even more drastic remedy,” i.e., a population cap lower than 137.5% design capacity. Id. at 1945. Defendants have not met the 137.5% design capacity benchmark. The current California prison population is at 149.2% design capacity. CDCR, Weekly Rpt. of Population, July 1, 2013, available at http://www. cdcr.ea.gov/Reports_Research/Offender_ Information_Services_Branch/WeeklyWed/ TPOPlA/TPOPlAdl30626.pdf.

When defendants first made this argument before this Court in January 2013, we rejected it on the ground that they had not provided evidence of any significant and unanticipated change in circumstances to rebut the Supreme Court’s determination that only a population reduction to 137.5% design capacity would remedy the underlying constitutional violations. Apr. 11, 2013 Op. & Order Denying Defs.’ Mot. to Vacate or Modify Population Reduction Order at 55-56 (ECF No. 2590/4541). Defendants provide no further support for such a contention, and therefore are unlikely to succeed on the merits. Defs.’ Mot. to Stay at 9 (ECF No. 2665/4673) (stating only that “additional population reductions are unnecessary to prevent death or needless suffering or to ensure that the quality of medical and mental health care does not pose a substantial risk of serious harm to the two certified classes of inmates”).

Second, defendants will not be irreparably injured absent a stay. The Amended Plan that we have ordered defendants to implement consists largely of measures in their proposed Plan. See Defs.’ Resp. to Apr. 11, 2013 Order at 28-33 (ECF No. 2609/4572). Further, we have already determined that the one additional measure we have suggested they implement, the full expansion of good time credits, will not cause irreparable injury. As explained in detail supra pp. 1065-66, this Court carefully considered the question of whether the expansion of good time credits was consistent with public safety in our August 2009 Opinion & Order. We heard extensive testimony from the leading experts in the country, all of whom— *1072including the now Secretary of CDCR Dr. Beard — testified that the expansion of good time credits could be implemented safely. The Supreme Court affirmed this conclusion, crediting our factual findings, Plata, 131 S.Ct. at 1942, and endorsing our determination that expansion of good time credits would reduce overcrowding “with little or no impact on public safety” by allowing the State “to give early release to only those prisoners who pose the least risk of reoffending,” id. at 1943.

Defendants’ “new evidence” in their request for a stay is not to the contrary. Defendants cite an article by two Stanford Law School professors for the proposition that “even inmates that CDCR has considered ‘low risk’ recidivate such that 41% are returned to California prisons within three years, and that 11% of such ‘low risk’ offenders have been ‘rearrested for a violent felony within 3 years of release.’ ” Defs.’ Mot. to Stay at 6-7 (ECF No. 2665/4673) (citing Joan Petersilia & Jessica Greenlick Snyder, Looking Past the Hype: 10 Questions Everyone Should Ask About California’s Prison Realignment, 5(2) Cal. J. Politics Policy 266, 295 (2013)). This sole law journal article, not subject to cross-examination, of course, is not sufficient to rebut the extensive testimony this Court considered after fourteen days of trial in 2009. This aside, the professors’ statistics, even if correct, are irrelevant to the question of whether releasing prisoners early will have a different effect on their behavior than releasing them later. The statistics that defendants cite indicate the percentage of prisoners who are likely to recidivate, but they do not suggest that there is a difference in the percentage of low-risk prisoners who recidivate when they are released early compared to when they are released at the time originally scheduled. At trial, after considering extensive testimony on the question of whether early release through good time credits increases the crime rate, the evidence showed overwhelmingly that it does not, and that it “affects only the timing and circumstances of the crime, if any, committed by a released inmate.” Aug. 4, 2009 Op. & Order, 922 F.Supp.2d at 977 (ECF No. 2197/3641). In fact, an argument can be made that the early release of prisoners may even decrease the crime rate. The State could well use the funds it saves by caring for fewer prisoners to fund reentry programs such as drug rehabilitation, job training, housing assistance, education, and other programs that reduce recidivism. The absence of such reentry assistance is far more likely to increase recidivism than release on a date earlier than initially scheduled.

Moreover, although this Court believes the expanded good time credits measure is the simplest and best way for defendants to comply with our Population Reduction Order, we have not required defendants to implement this measure. Rather, we have afforded them flexibility, allowing them to modify the good time credits measure by, for example, increasing the amount of such credits that can be awarded to particular sets of individuals and limiting the number of prisoners who will be eligible to receive them. We have also allowed defendants to substitute for measures on the Amended Plan (including the good time credits measure) other measures from their List, or to substitute prisoners from the Low-Risk List. For example, defendants might reassign prisoners to leased jail space— one of the measures included on their List. June 20, 2013 Op. & Order, 952 F.Supp.2d at 935, at *30 (ECF No. 2659/4662). We also suggested that defendants consider substituting, for prisoners who fall within the Amended Plan, “Lifers” who, due to age or infirmity, are adjudged to be “low risk” by CDCR’s risk assessment and a number of whom may be physically and mentally unable to commit future crimes. *1073Id. Our only requirement is that the substituted measures result in defendants’ reaching the 137.5% design capacity benchmark by December 31, 2013.

Third, issuance of a stay of our June 20, 2013 Order will substantially injure plaintiffs. The Plata and Coleman courts have both determined that mental and medical health care conditions in the California state prisons violate plaintiffs’ constitutional rights, and this Court and the Supreme Court have held that the only way to remedy these constitutional violations is to reduce prison overcrowding to 137.5% design capacity. Recent reports by the Receiver in Plata, Clark Kelso, confirm this finding. Kelso recently reported that “we do not have appropriate and adequate healthcare space at the current population levels. We need population levels to reduce to 137.5% of design capacity as ordered by the Three Judge Panel.” Receiver’s 23rd Report at 31 (ECF No. 2636/4628). Granting a stay would result in continuing injury to plaintiffs by maintaining the prison population at the current level of 149.2%, far above the constitutional level determined by this Court and affirmed by the Supreme Court in 2011 to be necessary to the safety and welfare of those in the custody of the State.8

Fourth, the public interest lies in denying defendants’ request for a stay. “[I]t is always in the public interest to prevent the violation of a party’s constitutional rights.” Melendres v. Arpaio, 695 F.3d 990, 1002 (9th Cir.2012) (internal quotation marks and citation omitted). Here, the public interest lies in obviating the ongoing constitutional violations in the mental and medical health care systems in California’s prisons — violations that this Court and the Supreme Court have determined will be eliminated only when defendants reduce the prison population from its current state of 149.2% design capacity to 137.5% design capacity. Finally, the public interest lies in denying the stay because defendants have informed this Court that, absent a stay, they will comply with the Population Reduction Order. Defs.’ Mot. to Stay at 2 (ECF No. 2665/4673). Conformity with the Order, if durable, will satisfy the requirements of the Eighth Amendment.9

IV. CONCLUSION

Granting defendants a stay of our June 20, 2013 Order would serve to resolve this litigation in defendants’ favor. The stay, which would last through the Supreme Court’s determination whether its previous 2011 decision was warranted, would last well past December 31, 2013, the date by which defendants have been ordered to reduce the prison population to 137.5% design capacity. Put differently, granting the stay would mean that at the end of the period by which defendants have been or*1074dered to comply, defendants will have been excused from meeting the requirements of this Court’s Population Reduction Order. Only denial of the stay by this Court and the Supreme Court will, defendants concede, cause them to comply with the Population Reduction Order issued in August 2009 and approved by the Supreme Court in June 2011. Specifically, only denial of the stay will cause defendants to implement the Plan it has selected along with an additional measure, whether the additional measure be the expansion of good time credits, a measure recommended by numerous experts at trial, which other states have had success in safely implementing, and which the Supreme Court endorsed in Brown v. Plata; use of the Low-Risk List; or any of a number of other measures of defendants’ choice.

Coleman was initiated 23 years ago, and Plata 12 years ago. The district court in Coleman has issued over 100 substantive orders in an attempt to bring defendants into compliance with the Eighth Amendment of the Constitution. Apr. 5, 2013 Order Denying Defs.’ Mot. to Terminate at 31 (Coleman ECF No. 4539). The district court in Plata has issued over 50 such orders, see Docket Sheet, Plata v. Brown, No. C01-1351 TEH (N.D.Cal.), and undoubtedly would have issued many more had a Receiver not been appointed in 2006. After this long history of defendants’ noncompliance, this Court cannot in good conscience grant a stay that would allow defendants to both not satisfy the Population Reduction Order and relitigate the Supreme Court’s emphatic decision in the very case before us. A denial of the stay by this Court and the Supreme Court will, however, at least result in the State’s obeying the orders of the federal judiciary and bringing the prison system into compliance with the Eighth Amendment, should the measures it selects prove durable.

For the above reasons, defendants’ motion to stay this Court’s June 20, 2013 Order is DENIED.

IT IS SO ORDERED.

. All filings in this Three-Judge Court are included in the individual docket sheets of both Plata v. Brown, No. C01-1351 TEH (N.D.Cal.), and Coleman v. Brown, No. 90-cv-520-LKK (E.D.Cal.). In this Order, when we cite to these filings, we list the docket number in Plata first, then Coleman. When we cite to filings in the individual cases, we include the docket number and specify whether the filing is from Plata or Coleman.

. Contrary to defendants’ representation in their motion to stay this Court’s June 20, 2013 Order, defendants’ Three-Judge Motion was not based on "evidence showing that underlying Eighth Amendment deficiencies in medical and mental health care had been remed*1064ied.” Defs.' Mot. to Stay at 3 (ECF No. 2665/4673). Although defendants initially asked this Court to decide this constitutional question, they later modified their motion by withdrawing this request. Defs’. Resp. to Jan. 29, 2013 Order at 4 (ECF No. 2529/4332) (“The issue to be decided by this Court is not constitutional compliance.’’); Defs.’ Reply Br. in Supp. of Three-Judge Mot. at 11 (ECF No. 2543/4345) (“Defendants’ motion did not seek a determination of constitutionality.”). With this request withdrawn, the only argument that defendants made for vacatur was that crowding was no longer the primary cause for any underlying constitutional violations. Id.

Defendants made a similar misrepresentation in their notice of appeal to the Supreme Court of our April 11, 2013 Order. In that notice, they stated that they would appeal in part because we "did not fully or fairly consider the evidence showing that the State's prisoner health care now exceeds constitutional standards.” Defs.’ Notice of Appeal to the Supreme Court at 3 (ECF No. 2621/4605). We did not consider this evidence because, as stated above, defendants explicitly modified their motion so as to withdraw any constitutional questions from this Court’s consideration.

. The members of the CDCR Expert Panel included various leading experts in crime and incarceration, such as Doctors Petersilia, Krisberg, and Austin; current CDCR Secretary Jeffrey Beard; and many other senior *1066officials of correctional programs throughout the country.

. These modifications to the proposed good time credit program would not affect the inclusion of retroactivity. They would only affect aspects such as the amount of good time credit to be received by various categories of offenders, all non-violent, and the amount of credit to be received for the various activities for which good time credit is rewarded. For example, defendants could extend 2-for-l-credit earning to prisoners other than those held in fire camps and minimum custody facilities (their current proposal), increase the credit earning limit for milestone completion credits, or increase the credit earning capacity of non violent offenders above 34 percent. Other states have taken similar measures to expand their good time credit programs for non-violent offenders without a subsequent increase in recidivism. June 20, 2013 Op. & Order, 952 F.Supp.2d at 930 & n. 26, at *24 & n. 26 (ECF No. 2659/4662).

. The prisoners now housed out of state who were due to be returned this year are already accounted for in the Plan. No other prisoners housed out of state will be considered as part of any substitute measure.

. See Aug. 4, 2009 Op. & Order, 922 F.Supp.2d at 922-24 (ECF No. 2197/3641) (discussing how crowding causes "general problems in the delivery of medical and mental health care”); id. at 924-26 (discussing how overcrowded reception centers result in insufficient medical care); id. at 926-27 (discussing the especially grave consequences of overcrowded reception centers for individuals with mental illness); id. at 926-29 (discussing the effect of insufficient treatment space and the inability to properly classify inmates on both medical and mental health care); id. at 928-30 (discussing lack of space for mental health beds); id. at 930-32 (discussing how conditions of confinement result in the spread of diseases); id. at 931-32 (discussing how conditions of confinement exacerbate mental illness); id. at 932-34 (discussing shortages in medical health care staff); id. at 934-35 (discussing shortages in mental health care staff); id. at 936-37 (discussing medication management issues in both Plata and Coleman); id. at 938 (discussing the effect of lockdowns on the provision of medical health care); id. at 938-39 (discussing the effect of lockdowns on the provision of mental health care); id. at 938-40 (discussing the need for medical records in medical and mental health care); id. at 940-41 (discussing the increasing acuity of mental illness); id. at 941-42 (discussing suicides); id. at 941-42 (discussing preventable deaths).

. One three-judge court was convened, instead of two, for practical reasons only. The individual district courts recommended consolidation "[flor purposes of judicial economy and avoiding the risk of inconsistent judgments.” July 23, 2007 Order in Plata, 2007 WL 2122657, at *6; July 23, 2007 Order in Coleman, 2007 WL 2122636, at *8. The Supreme Court agreed, stating that there was a “certain utility in avoiding conflicting decrees and aiding judicial consideration and enforcement.” Plata, 131 S.Ct. at 1922. It was a "limited consolidation” only and, most important, “[t]he order of the three-judge District Court is applicable to both cases.” Id.

. In their motion for a stay, defendants state that "population reduction is just one of many existing remedies directed at the alleged Eighth Amendment violations at issue; the other remedies will remain in place irrespective of any stay here." Defs.’ Mot. to Stay at 7 (ECF No. 2665/4673). This does not change our finding, affirmed by the Supreme Court, that the only way to completely alleviate the ongoing constitutional violations is to reduce the prison population to 137.5% design capacity.

. It is not enough simply to meet a specific target number of prisoners on a specific date. Durability is necessary to ensure compliance with both the Order and the Constitution, and can be determined only after a period of time in which this Court can examine whether the ratio of prisoners to design capacity is stable. Changes in penological policies and procedures, as well as other matters, may have a significant effect on the prisoner to design capacity ratio. We maintain jurisdiction over the question for a reasonable period of time in order to resolve that issue.