Plata et al v. Schwarzenegger et al, No. 3:2001cv01351 - Document 2197 (N.D. Cal. 2009)

Court Description: OPINION AND ORDER by three-judge court re: prison overcrowding. Signed by Judges Stephen Reinhardt, Lawrence K. Karlton, and Thelton E. Henderson on 08/04/09. (tehlc3, COURT STAFF) (Filed on 8/4/2009)

Download PDF

1 IN THE UNITED STATES DISTRICT COURTS 2 FOR THE EASTERN DISTRICT OF CALIFORNIA 3 AND THE NORTHERN DISTRICT OF CALIFORNIA 4 UNITED STATES DISTRICT COURT COMPOSED OF THREE JUDGES 5 PURSUANT TO SECTION 2284, TITLE 28 UNITED STATES CODE 6 7 8 RALPH COLEMAN, et al., 9 Plaintiffs, 10 v. 11 NO. CIV S-90-0520 LKK JFM P ARNOLD SCHWARZENEGGER, et al., 12 Defendants. 13 14 MARCIANO PLATA, et al., NO. C01-1351 TEH 15 16 17 THREE-JUDGE COURT Plaintiffs, THREE-JUDGE COURT v. 18 ARNOLD SCHWARZENEGGER, et al., 19 Defendants. 20 21 22 23 24 25 26 27 28 OPINION AND ORDER 1 TABLE OF CONTENTS 2 3 I. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 4 II. FACTUAL AND PROCEDURAL BACKGROUND . . . . . . . . . . . . . . . . . . . . . 10 5 A. Plata (Medical Care) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 6 1. Complaint, Stipulation, and Order for Injunctive Relief . . . . . . . . . . 11 7 2. Appointment of Court Experts and Their Findings . . . . . . . . . . . . . . 13 8 3. Periodic Status Conferences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 9 4. Proceedings To Determine Whether a Receiver Should Be 10 Appointed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 11 5. Findings of Fact and Conclusions of Law Concerning Continuing 12 Failure To Meet Constitutional Standards and Necessity of a 13 Receivership . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 14 6. Interim Remedies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 15 7. Appointment of the Plata Receiver . . . . . . . . . . . . . . . . . . . . . . . . . . 21 16 B. Coleman (Mental Health Care) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 17 1. Findings of Eighth Amendment Violations . . . . . . . . . . . . . . . . . . . . 23 18 2. Remedial Orders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 19 a. Mental Health Care Beds and Treatment Space . . . . . . . . . . . 28 20 b. Transfers to Appropriate Level of Care . . . . . . . . . . . . . . . . . . 32 21 c. Staffing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 22 23 3. C. Special Master s 2006 Monitoring Reports . . . . . . . . . . . . . . . . . . . . 34 Crowding in California s Prison System . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 24 1. The Increasing California Prison Population . . . . . . . . . . . . . . . . . . . 37 25 2. Studies Commissioned by the State of California To Examine 26 Prison Crowding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 27 3. Defining the Capacity of California Prisons . . . . . . . . . . . . . . . . . . . . 39 28 4. Crowding in Relation to Capacity . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 2 1 D. Governor Schwarzenegger s Emergency Proclamation . . . . . . . . . . . . . . . . 43 2 E. Motions To Convene Three-Judge Court and Subsequent Prison Studies 3 by the State of California . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 4 1. Motions To Convene and Initial Proceedings . . . . . . . . . . . . . . . . . . 44 5 2. Intervening Reports on Prison Crowding . . . . . . . . . . . . . . . . . . . . . . 44 6 3. Final Hearing and Rulings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 7 F. 8 III. LEGAL FRAMEWORK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 9 A. 10 11 Proceedings Before this Three-Judge Court . . . . . . . . . . . . . . . . . . . . . . . . . 48 The PLRA Standard for Prisoner Release Orders: Primary Cause and No Other Relief . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 B. 12 The PLRA Standard for All Prospective Relief: Need-NarrownessIntrusiveness and Consideration of Public Safety . . . . . . . . . . . . . . . . . . . . . 52 13 C. 14 IV. CROWDING AS PRIMARY CAUSE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 15 A. 16 17 The Remedial Nature of the Three-Judge Court Proceeding . . . . . . . . . . . . . 54 General Problems in the Delivery of Medical and Mental Health Care Caused by Crowding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 B. Space Issues Affecting the Delivery of Care . . . . . . . . . . . . . . . . . . . . . . . . . 60 18 1. Reception Centers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 19 2. Treatment Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 20 3. Inability To House Inmates by Classification . . . . . . . . . . . . . . . . . . 67 21 4. Beds for Mentally Ill Inmates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 22 C. Conditions of Confinement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 23 D. Other Access to Care Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 24 1. Staffing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 25 a. Medical Staff . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 26 b. Mental Health Staff . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 27 c. Custodial Staff . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 28 2. Medication Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 3 1 3. Specialty Medical Care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 2 4. Lockdowns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 3 E. Medical Records . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 4 F. Increasing Acuity of Mental Illness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 5 G. Extreme Departures from the Standard of Care and Preventable or Possibly 6 Preventable Deaths, Including Suicides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 7 H. Expert Opinions Regarding Causation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 8 I. Findings and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 9 V. NO OTHER RELIEF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 10 A. 11 Alternatives to a Prisoner Release Order . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 1. Inadequacy of Construction as a Remedy . . . . . . . . . . . . . . . . . . . . 102 12 a. Prison Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 13 b. Construction of Re-entry Facilities . . . . . . . . . . . . . . . . . . . . 103 14 c. Medical Facilities and Prison Expansion . . . . . . . . . . . . . . . 105 15 d. Construction as a Means of Compliance . . . . . . . . . . . . . . . . 108 16 2. Inadequacy of Additional Hiring . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 17 3. Insufficiency of the Plata Receivership and Coleman Special 18 Mastership . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 19 4. Other Proposals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 20 B. Expert Testimony . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 21 C. Findings and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 22 VI. NARROWLY DRAWN, LEAST INTRUSIVE REMEDY THAT EXTENDS 23 NO FURTHER THAN NECESSARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 24 A. Scope of Relief . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 25 B. Form of Relief . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 26 C. The Required Population Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124 27 28 4 1 VII. POTENTIAL POPULATION REDUCTION MEASURES AND THEIR 2 IMPACT ON PUBLIC SAFETY AND THE OPERATION OF THE 3 CRIMINAL JUSTICE SYSTEM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131 4 A. Criminogenic Nature of Overcrowded Prisons . . . . . . . . . . . . . . . . . . . . . . 133 5 B. Potential Population Reduction Measures . . . . . . . . . . . . . . . . . . . . . . . . . . 137 6 1. Early Release Through Expansion of Good Time Credits . . . . . . . . 139 7 2. Diversion of Technical Parole Violators . . . . . . . . . . . . . . . . . . . . . 145 8 3. Diversion of Low-Risk Offenders with Short Sentences . . . . . . . . . 149 9 4. Expansion of Evidence-Based Rehabilitative Programming in 10 Prisons or Communities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152 11 5. 12 13 Sentencing Reform and Other Potential Population Reduction Measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154 C. Impact of Proposed Measures on Communities . . . . . . . . . . . . . . . . . . . . . 157 14 1. Investigation and Prosecution of Crime . . . . . . . . . . . . . . . . . . . . . . 157 15 2. Effect on Jail Population . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160 16 3. Effect on Parole Supervision Resources . . . . . . . . . . . . . . . . . . . . . . 162 17 4. Impact on Community Corrections, Rehabilitative Services, and 18 Re-entry Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164 19 5. Impact on Integrity of Criminal Justice System . . . . . . . . . . . . . . . . 166 20 6. Weight To Be Given Public Safety . . . . . . . . . . . . . . . . . . . . . . . . . 167 21 D. Feasibility Notwithstanding the Present Fiscal Crisis . . . . . . . . . . . . . . . . . 167 22 E. Inclusion of Mentally Ill Inmates in Any Population Reduction Order . . . 168 23 F. Empirical Evidence on Incarceration and Crime Rates . . . . . . . . . . . . . . . . 172 24 G. Findings and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177 25 VIII. CONCLUSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181 26 ORDER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183 27 28 5 1 I. INTRODUCTION 2 California s correctional system is in a tailspin, the state s independent oversight 3 agency has reported. Ex. P3 at i (Jan. 2007 Little Hoover Commission Report, Solving 4 California s Corrections Crisis: Time Is Running Out ).1 Tough-on-crime politics have 5 increased the population of California s prisons dramatically while making necessary reforms 6 impossible. Id. at ii, 2-5, 9, 20. As a result, the state s prisons have become places of 7 extreme peril to the safety of persons they house, Ex. P1 at 7-8 (Governor 8 Schwarzenegger s Oct. 4, 2006 Prison Overcrowding State of Emergency Declaration), while 9 contributing little to the safety of California s residents, Ex. P3 at ii. California spends 10 more on corrections than most countries in the world, but the state reaps fewer public 11 safety benefits. Id. at 14. Although California s existing prison system serves neither the 12 public nor the inmates well, the state has for years been unable or unwilling to implement the 13 reforms necessary to reverse its continuing deterioration. 14 In this proceeding, we address two particular problems that every day threaten the 15 lives and health of California prisoners. First, the medical and mental health care available to 16 inmates in the California prison system is woefully and constitutionally inadequate, and has 17 been for more than a decade. The United States Constitution does not require that the state 18 provide its inmates with state-of-the-art medical and mental health care, nor does it require 19 that prison conditions be comfortable. California must simply provide care consistent with 20 the minimal civilized measure of life s necessities, Rhodes v. Chapman, 452 U.S. 337, 347 21 (1981) care sufficient to prevent the unnecessary and wanton infliction of pain or death, 22 Estelle v. Gamble, 429 U.S. 97, 103-04 (1976). Tragically, California s inmates have long 23 been denied even that minimal level of medical and mental health care, with consequences 24 that have been serious, and often fatal. Inmates are forced to wait months or years for 25 medically necessary appointments and examinations, and many receive inadequate medical 26 care in substandard facilities that lack the medical equipment required to conduct routine 27 1 The Little Hoover Commission is a state agency charged with preparing reports and recommendations regarding the structure and operation of state government in order to 28 improve its economy, efficiency, and service. Cal. Gov t. Code §§ 8501, 8521-8522. 6 1 examinations or afford essential medical treatment. Seriously mentally ill inmates languish 2 in horrific conditions without access to necessary mental health care, raising the acuity of 3 mental illness throughout the system and increasing the risk of inmate suicide. A significant 4 number of inmates have died as a result of the state s failure to provide constitutionally 5 adequate medical care. As of mid-2005, a California inmate was dying needlessly every six 6 or seven days. 7 California s inmates face a second everyday threat to their health and safety: the 8 unprecedented overcrowding of California s prisons. Since reaching an all-time population 9 record of more than 160,000 in October 2006, the state s adult prison institutions have 10 operated at almost double their intended capacity. As Governor Schwarzenegger observed in 11 declaring a prison state of emergency that continues to this day, this creates conditions of 12 extreme peril that threaten the health and safety of the men and women who work inside 13 [severely overcrowded] prisons and the inmates housed in them . . . . Ex. P1 at 1, 8. 14 Thousands of prisoners are assigned to bad beds, such as triple-bunked beds placed in 15 gymnasiums or day rooms, and some institutions have populations approaching 300% of 16 their intended capacity. In these overcrowded conditions, inmate-on-inmate violence is 17 almost impossible to prevent, infectious diseases spread more easily, and lockdowns are 18 sometimes the only means by which to maintain control. In short, California s prisons are 19 bursting at the seams and are impossible to manage. 20 It is the relationship between these two critical problems that lies at the heart of the 21 cases before us. We must answer the question whether overcrowding is the primary cause of 22 the unconstitutional medical and mental health care to which California prison inmates are 23 currently subjected. Two federal lawsuits have brought the crisis in California s prisons to 24 this three-judge court. Plaintiffs in the two lawsuits contend that a reduction in the prison 25 population is necessary to bring the California prison system s medical and mental health 26 care into constitutional compliance. In both Plata v. Schwarzenegger and Coleman 27 28 7 1 v. Schwarzenegger,2 the federal courts initially issued narrow orders requiring California to 2 develop and implement remedial plans to meet this objective. However, as the state time and 3 again failed to meet its own remedial targets let alone to achieve constitutional compliance 4 both courts were forced to adopt increasingly drastic remedies, culminating in the Plata 5 court s 2005 appointment of a receiver to manage the prison medical system. Ultimately, by 6 late 2006 it became apparent that the overcrowding in California s prisons rendered the 7 efforts of the courts, the Coleman Special Master, and the Plata Receiver utterly insufficient. 8 At the request of the Plata and Coleman courts, the Chief Judge of the United States Court of 9 Appeals for the Ninth Circuit convened this three-judge court to consider the plaintiffs 10 request for a court-ordered reduction in the California prison population. 11 During the pendency of this proceeding, the outlook for California s prisons has only 12 grown dimmer. The state is now in the throes of a fiscal crisis that renders it unable or 13 unwilling to commit the necessary resources to fix the problems in its prisons. As Matthew 14 Cate, Secretary of the California Department of Corrections and Rehabilitation3 and a 15 defendant here, recently put it, California cannot at this time become further indebted for 16 correctional healthcare. 4 Ex. 1 to Defs. July 1, 2009 Response to Court s June 18, 2009 17 Order, filed in Coleman, at 1. 18 Federal law makes any prisoner release order, including the population reduction 19 order requested by plaintiffs, a remedy of last resort, H.R. Rep. No. 104-21, at 25 (1995) 20 2 Plata involves the prison system s constitutionally inadequate medical care, while 21 Coleman involves the constitutional deficiencies in mental health care provided to California inmates. 22 3 Until 2005, California s adult prisons were run by the California Department of 23 Corrections, which was a department within the state s Youth and Corrections Agency. On July 1, 2005, the agency was reorganized and renamed the California Department of 24 Corrections and Rehabilitation ( the CDCR ). Ex. P5 at ix. In this opinion and order, we refer to the agency as the CDCR except when quoting orders issued prior to the 25 reorganization. 4 26 California has reduced spending on education, health care, the social safety net, and services for the needy, the blind, and children to the breaking point. Under these 27 circumstances, we would be reluctant to direct the state to allocate additional funds to its prisons or to rehabilitative services at the expense of others to whom it has a legal and moral 28 obligation. 8 1 (report of the House Committee on the Judiciary on the Violent Criminal Incarceration Act of 2 1995), and imposes various conditions upon the issuance of such an order. See 18 U.S.C. 3 § 3626(a)(3). As we explain below, those conditions have been met here: (1) crowding is the 4 primary cause of the state s failure to provide its inmates with constitutionally sufficient 5 medical and mental health care; (2) no relief besides a prisoner release order can bring the 6 California prison system into constitutional compliance; (3) an order requiring the state to 7 reduce the population of its adult institutions to a lower percentage of their combined design 8 capacity than presently exists a population cap is narrowly tailored to the constitutional 9 violations identified by the Plata and Coleman courts, extends no further than necessary to 10 remedy those violations, and is the least intrusive possible remedy; and (4) the state can 11 comply with such an order with little or no impact on public safety and the operation of the 12 criminal justice system. There are numerous means by which the state can reduce the prison 13 population, from parole reform and the diversion of technical parole violators and low-risk 14 offenders to sentencing reform and the expansion of good time credits and rehabilitative 15 programming. There is no need for the state to release presently incarcerated inmates 16 indiscriminately in order to comply with our order. Much of the relief can be achieved 17 instead by reducing prison intake in a manner recommended by the state s own experts. 18 We recognize the gravity of the population reduction order we issue herein, and we do 19 not intervene in matters of prison population lightly. Nonetheless, when federal court 20 intervention becomes the only means by which to enforce rights guaranteed by the 21 Constitution, federal courts are obligated to act. Without this, all the reservations of 22 particular rights or privileges would amount to nothing. The Federalist No. 78 (Alexander 23 Hamilton). California s prisoners have long been denied constitutionally adequate medical 24 and mental health care, often with tragic consequences, and the overcrowding in California s 25 prisons, which have become criminogenic, must be reduced if the prison system is to achieve 26 constitutional compliance. California s prisoners, present and future, (and the state s 27 population as a whole) can wait no longer. 28 9 1 II. FACTUAL AND PROCEDURAL BACKGROUND 2 Because the courts prior remedial efforts are of profound relevance in understanding 3 the effect of prison overcrowding and the inadequacy of forms of relief that do not address 4 that problem, we begin with a detailed history of the individual Plata and Coleman cases. 5 We then describe the crowded conditions in California s prison system and the history of the 6 three-judge court proceeding before turning to the legal questions before us. 7 A. 8 The history of Plata involves extensive remedial efforts over the last seven years that Plata (Medical Care) 9 have faltered because of the severe overcrowding in California s prisons. 10 The Plata class action was filed on April 5, 2001, and plaintiffs filed an amended 11 complaint on August 20, 2001, alleging constitutional violations in the delivery of medical 12 care to inmates confined in California state prisons, as well as violations of the Americans 13 with Disabilities Act and § 504 of the Rehabilitation Act. Ex. D1059. Plaintiffs asserted that 14 the unconstitutional conditions caused by defendants failure to properly care for and treat 15 the prisoners in [their] custody . . . caused widespread harm, including severe and 16 unnecessary pain, injury and death. Id. ¶ 1. The Plata plaintiffs and defendants negotiated 17 a stipulation for injunctive relief, which the Plata court approved by court order.5 18 However, defendants proved incapable of or unwilling to provide the stipulated relief. 19 Three years after approving the stipulation as an order of the court, the Plata court conducted 20 an evidentiary hearing that revealed the continued existence of appalling conditions arising 21 from defendants failure to provide adequate medical care to California inmates. The Court 22 found that defendants had been given every reasonable opportunity to bring [the] prison 23 medical system up to constitutional standards, and it [was] beyond reasonable dispute that the 24 State ha[d] failed. Oct. 3, 2005 Findings of Fact & Conclusions of Law Re: Appointment of 25 Receiver ( FF&CL ), 2005 WL 2932253, at *1 (Ex. D1063).6 Following that hearing, the 26 27 5 The stipulation resolved all of plaintiffs claims, including their Rehabilitation Act and ADA claims. 6 All references to court orders in this section of our opinion and order are to orders 28 filed in the district court in Plata. 10 1 Plata court concluded that it had no choice but to place the CDCR s medical health care 2 delivery system in receivership. The Plata Receivership continues to this date, but, as we 3 explain below, severe crowding throughout California s prison system renders the Receiver 4 unable to resolve the constitutional violations at issue in Plata. 5 6 1. Complaint, Stipulation, and Order for Injunctive Relief In their amended complaint, the Plata plaintiffs alleged that a number of specific 7 deficiencies in the CDCR s prison medical care system rendered the system as a whole 8 unconstitutional. The alleged deficiencies included inadequate medical screening of 9 incoming prisoners; delays in or failure to provide access to medical care, including 10 specialist care; untimely responses to medical emergencies; the interference of custodial staff 11 with the provision of medical care; the failure to recruit and retain sufficient numbers of 12 competent medical staff; disorganized and incomplete medical records; a lack of quality 13 control procedures, including lack of physician peer review, quality assurance and death 14 reviews ; a lack of protocols to deal with chronic illnesses, including diabetes, heart disease, 15 hepatitis, and HIV; and the failure of the administrative grievance system to provide timely 16 or adequate responses to complaints concerning medical care. Ex. D1059 ¶ 192.7 17 Prior to filing suit, the Plata plaintiffs had been in informal negotiations with 18 defendants since July 1999. Ex. D1060 ¶ 3 (June 13, 2002 Stip. & Order). After Plata was 19 filed, the parties ultimately agreed to a stipulation for injunctive relief, which the Plata court 20 entered as an order on June 13, 2002. Defendants agreed to and were ordered to implement 21 certain policies and procedures on a staggered basis, with seven prisons to complete 22 implementation in 2003. Id. ¶¶ 4-5. In each subsequent year, defendants were to complete 23 24 25 26 7 As we explain below, infra Section II.A.2, it became apparent early in the Plata 27 litigation that, given the size of their populations, California s prisons lacked the space and facilities required to deliver constitutionally adequate medical care or to remedy the above 28 deficiencies. 11 1 implementation at five additional prisons, such that statewide implementation would be 2 achieved by the end of 2008. Id. ¶ 5.8 3 The stipulated policies and procedures, which defendants filed with the Plata court on 4 February 15, 2002, and supplemented on May 30, 2002, are approximately 800 pages long 5 and contained in 11 volumes. Mar. 10, 2003 Order at 2. Although the stipulated policies 6 and procedures were designed to meet or exceed the minimum level of care necessary to 7 fulfill the defendants obligation to plaintiffs under the Eighth Amendment of the United 8 States Constitution, the stipulation require[s] defendants to provide only the minimum 9 level of medical care required under the Eighth Amendment. Ex. D1060 ¶ 4. 10 The stipulation for injunctive relief provided inter alia for regular audits of 11 defendants compliance. Id. ¶¶ 19-23. These audits were to include a review of no less than 12 180 inmate health records at each prison. Id. ¶ 21(a). Medical assessments or treatment 13 plans contained in those records would be deemed substantially in compliance with the 14 settlement agreement if they were consistent with the policies and procedures or with the 15 community standard of care imposed under the laws of the State of California upon health 16 care providers licensed to practice in California. Id. ¶ 22(b) & at 11 n.3. Compliance with 17 the agreement would also require conducting minimally adequate death reviews and quality 18 management proceedings, having tracking, scheduling and medication administration 19 systems adequately in place, and the absence of any pattern or practice that is likely to 20 result in serious problems [where] those problems are not being adequately addressed. Id. 21 ¶¶ 22(c)-(e). 22 Had the stipulated policies and procedures been implemented, they would have 23 resulted in comprehensive improvements to nearly all aspects of the medical delivery system 24 in California s prisons, including quality management; health records management; 25 8 This roll-out plan did not include Pelican Bay State Prison, which was under the 26 Plata court s jurisdiction in a separate action, Madrid v. Cate, Case No. C90-3094 TEH (N.D. Cal.), and was not included in the Plata case. However, on June 6, 2008, the parties 27 stipulated, and the court ordered, that prisoners housed at Pelican Bay State Prison be included in the Plata class. Thus, the Plata case now includes all thirty-three adult 28 institutions within the CDCR. 12 1 infectious disease control; staffing; inter-institution transfers; and the timing and manner in 2 which inmates are provided with physician and nurse care, as well as with necessary 3 medications. Unfortunately, defendants utterly failed to comply with the implementation 4 schedule to which they had stipulated. As of May 10, 2005, when implementation should 5 have been completed at twelve prisons, not a single prison ha[d] successfully completed 6 implementation. May 10, 2005 Order to Show Cause ( OSC ), 2005 WL 2932243, at *2 7 (Ex. D1062). The same remains true now, more than seven years after the court approved 8 the parties settlement agreement. As we explain below, this is due in large part to the severe 9 overcrowding in California s prisons. 10 11 2. Appointment of Court Experts and Their Findings In addition to stipulating to an implementation schedule, the parties agreed to the 12 appointment of medical and nursing experts to advise the Court on the adequacy and 13 implementation of defendants Policies and Procedures and any other matter that 14 appropriately may be the subject of the experts testimony. Ex. D1060 ¶¶ 16-17. The 15 experts routinely reviewed defendants progress towards implementing the stipulated 16 injunctive relief and periodically communicated their findings and recommendations to the 17 Plata court. 18 In their July 16, 2004 report to the court, the experts identified a pattern of serious 19 deficiencies relating to physician quality at California prisons, and defendants agreed to 20 address those deficiencies in a stipulation entered as an order of the court on September 17, 21 2004 ( Patient Care Order ), Ex. D1061. 22 23 24 25 26 The Patient Care Order required defendants to engage an independent entity to (a) evaluate the competency of physicians employed by the CDCR and (b) provide training to those physicians found to be deficient. It also required defendants to undertake certain measures with respect to the treatment of highrisk patients, to develop proposals regarding physician and nursing classifications and supervision, and to fund and fill Quality Management Assistance Teams ( QMAT ) and other support positions. 27 Oct. 3, 2005 FF&CL, 2005 WL 2932253, at *2. However, [d]efendants failed to come 28 close to meeting the terms of the Patient Care Order, even with generous extensions of time 13 1 from the Court. Id. The experts noted one example of defendants failure to comply after 2 visits to the Substance Abuse and Treatment Facility in February and March 2005: 3 4 5 6 [N]ot only has little progress been made in the implementation of Plata, but the initial morale and enthusiasm in utilizing QMAT has evaporated in large part because of the inability of the California Department of Corrections to provide the necessary staff and support to this process. This has delivered an unspoken message that no change will occur. 7 May 16, 2005 Experts Report on Substance Abuse Treatment Center, at 3 (filed in Plata on 8 May 19, 2005). 9 The experts reports following visits to San Quentin State Prison were no better. 10 Following a February 2005 visit to that facility, the court s nursing experts observed that 11 clinics in housing areas were sometimes nothing more than an office used by correctional 12 officers and lacked basic medical equipment and supplies. Apr. 9, 2005 Nursing Experts 13 Report on San Quentin, at 2 (filed in Plata on May 10, 2005). The [m]ost disturbing 14 conditions were in one unit where 15 16 17 18 19 20 21 22 [t]he area used for nursing triage [was] a small room at the end of the tier that the nurse accesses by walking through a gate and into the men s showers. . . . Because of a clogged shower drain, standing water was present outside the clinic door. Inside, the room was filthy. The furniture was old and in disrepair. There was no examination table, medical equipment or supplies, or handwashing facilities. According to staff, equipment (otoscope [an instrument used to examine the ear]) requested for this area had been denied. As well, there was no telephone or computer access. Prior to this room being used, a broom closet on the fourth tier was used for nurse triage. These conditions are deplorable and have no resemblance to a medical setting whatsoever. 23 Id. at 2-3. Following their visits to San Quentin in January and February 2005, the medical 24 experts noted that [m]edical record reviews demonstrate[d] multiple instances of 25 incompetence, indifference, cruelty, and neglect. Ten deaths were reviewed. All showed 26 serious problems; most deaths were preventable. . . . Routine medical care [was] replete with 27 numerous errors resulting from both system failures as well as physician mistakes. Apr. 8, 28 2005 Medical Experts Report on San Quentin, at 13 (filed in Plata on May 10, 2005). 14 1 Perhaps most damning was the medical experts conclusion that overall compliance 2 with the Stipulated Order and subsequent Court Orders was non-existent [at San Quentin]. In 3 fact, it was clear that for most areas we reviewed there has been indifference to beginning the 4 process required in the Stipulated Order, id. at 2 despite the fact that the prison was to 5 have completed that process by the time of the experts site visit. The experts ultimately 6 concluded that San Quentin was so old, antiquated, dirty, poorly staffed, poorly maintained, 7 with inadequate medical space and equipment and over-crowded that it is our opinion that it 8 is dangerous to house people there with certain medical conditions and is also dangerous to 9 use this facility as an intake facility. Id. According to the experts, the overcrowding and 10 facility life-safety and hygiene conditions create a public health and life-safety risk to 11 inmates who are housed there. Id. 12 13 3. Periodic Status Conferences Beyond receiving periodic reports from the experts, the Plata court also conducted 14 regular status conferences with the parties to help monitor and facilitate implementation of 15 the stipulated injunctive relief, as well as to assess defendants ability and willingness to 16 comply with the court order approving such relief. Based on the experts dismal reports of 17 defendants progress, the court increased the frequency of these conferences and, in February 18 2005, started meeting with the parties on a monthly basis. To facilitate these meetings, 19 which typically involved large numbers of CDCR staff housed in Sacramento, the court 20 rotated the location of these meetings between San Francisco and Sacramento. 21 22 23 24 25 26 The Court invited the parties during [the] monthly status conferences to contribute ideas as to possible remedies, and the Court especially encouraged defendants to consider ways in which they could take the actions necessary to solve the medical care problems through measures within their own control, including use of the extraordinary powers of the Governor. The Court went to the length of requesting that defendants present it with a series of proposed orders so that the Court could help empower them to overcome some of their bureaucratic hurdles on their own. Defendants did not submit a single proposed order. 27 Oct. 3, 2005 FF&CL, 2005 WL 2932253, at *26 (citation omitted). 28 15 1 2 4. Proceedings To Determine Whether a Receiver Should Be Appointed Ultimately, the Plata court found itself with no alternative but to issue an order to 3 show cause ( OSC ) why defendants should not be found in civil contempt and why a 4 receiver should not be appointed to manage medical care delivery for the CDCR. As the 5 court noted when it issued the OSC on May 10, 2005: 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 In the four years since this case was filed, which includes the year and a half that this Court has been meeting with the parties on a regular basis, two things have become ever increasingly clear: (1) the Governor has appointed, and the State has hired, a number of dedicated individuals to tackle the difficult task of addressing the crisis in the delivery of health care in the California Department of Corrections ( CDC ), and, (2) despite the best efforts of these individuals, little real progress is being made. The problem of a highly dysfunctional, largely decrepit, overly bureaucratic, and politically driven prison system, which these defendants have inherited from past administrations, is too far gone to be corrected by conventional methods. The prison medical delivery system is in such a blatant state of crisis that in recent days defendants have publicly conceded their inability to find and implement on their own solutions that will meet constitutional standards. The State s failure has created a vacuum of leadership, and utter disarray in the management, supervision, and delivery of care in the Department of Corrections medical system. Defendants have devised a long-term strategy to contract out health care management and much of the delivery of care. However, full implementation of that plan is, by defendants own estimates, years away. In the meantime, roughly 162,000 prisoners are being subjected to an unconstitutional system fraught with medical neglect and malfeasance. Defendants themselves have conceded that a significant number of prisoners have died as a direct result of this lack of care, and it is clear to the Court that more are sure to suffer and die if the system is not immediately overhauled. .... Since the entry of the Stipulated Injunction in June 2002, the most notable characteristic of this case has been defendants failure to achieve any substantial progress in bringing the medical care system even close to minimal constitutional standards. 26 May 10, 2005 OSC, 2005 WL 2932243, at *1-2. Even following issuance of the OSC on 27 the brink of possible contempt and the imposition of a Receivership defendants were able 28 16 1 to enact only very limited and piece-meal measures, with no prospect for system-wide reform 2 or restructuring. Oct. 3, 2005 FF&CL, 2005 WL 2932253, at *26. 3 Beginning on May 31, 2005, and concluding on June 9, 2005, the Plata court 4 conducted a six-day evidentiary hearing concerning the OSC. Id. at *2. The court 5 considered eighty-two exhibits, id., and heard testimony from the court experts; relevant state 6 officials, including Undersecretary of Corrections Kevin Carruth and Dr. Renee Kanan, the 7 Acting Director of Health Care Services for the CDCR; and defendants medical expert 8 Dr. Ronald Shansky.9 Following the hearing, the parties submitted legal briefs addressing 9 both contempt and the appointment of a receiver, and several unions representing state prison 10 medical personnel filed an amicus brief. Id. Defendants did not dispute that the Plata court 11 had the power to appoint a receiver; instead, they argued only that a receivership was an 12 extraordinary remedy to be used only if less intrusive remedies had failed or were likely to 13 fail. Defs. June 20, 2005 Response to OSC at 2, 25. 14 On June 30, 2005, the Plata court heard argument on the OSC. Oct. 3, 2005 FF&CL, 15 2005 WL 2932253, at *2. Based on the arguments of counsel, the evidence presented, the 16 full record in this case, and the Court s own observations on prison tours [of two facilities, 17 accompanied by counsel for the parties], the Court delivered an oral ruling at the conclusion 18 of the hearing that it would take control of the medical delivery system of the CDCR and 19 place it under the auspices of a Receivership. Id. 5. 20 Findings of Fact and Conclusions of Law Concerning Continuing 21 Failure To Meet Constitutional Standards and Necessity of a 22 Receivership 23 On October 3, 2005, the court issued findings of fact and conclusions of law setting 24 forth the detailed reasoning behind its oral ruling. As the court noted in its written decision: By all accounts, the California prison medical care system is broken beyond repair. The harm already done in this case to California s prison inmate population could not be more grave, 25 26 27 9 As noted in our discussion below, although Dr. Shansky testified as defendants expert witness in the proceedings before the Plata court, he testified as plaintiffs expert 28 witness in the proceedings before this three-judge court. 17 1 and the threat of future injury and death is virtually guaranteed in the absence of drastic action. The Court has given defendants every reasonable opportunity to bring its prison medical system up to constitutional standards, and it is beyond reasonable dispute that the State has failed. Indeed, it is an uncontested fact that, on average, an inmate in one of California s prisons needlessly dies every six to seven days due to constitutional deficiencies in the CDCR s medical delivery system. This statistic, awful as it is, barely provides a window into the waste of human life occurring behind California s prison walls due to the gross failures of the medical delivery system. 2 3 4 5 6 7 It is clear to the Court that this unconscionable degree of suffering and death is sure to continue if the system is not dramatically overhauled. Decades of neglecting medical care while vastly expanding the size of the prison system has led to a state of institutional paralysis. The prison system is unable to function effectively and suffers a lack of will with respect to prisoner medical care. 8 9 10 11 Id. at *1. 12 In its order, the court identified a number of serious problems in the care provided to 13 inmates. The court found that the CDCR failed to follow its own policies regarding access to 14 medical care, and inmates routinely lacked timely access to care, both in terms of screening 15 requests and in receiving care once it was determined that an appointment with a physician 16 was warranted. Id. at *13. Inmates needing specialty services to treat serious medical 17 problems were forced to wait inordinate and inexcusable amounts of time for appointments; 18 at one prison, inmates with consultation referrals from early 2004 had yet to be seen in May 19 2005. Id. at *16. In addition, the CDCR had failed to develop or implement a system to 20 track and treat inmates with chronic care needs, id. at *14, and the court s nursing expert 21 found that CDCR nurses often fail[ed] to perform basic functions, such as taking vital 22 signs, conducting examinations, and identifying urgent medical issues requiring immediate 23 referral to a physician. Id. at *9. 24 Not unexpectedly, death reviews revealed repeated gross departures from even 25 minimal standards of care. Id. at *7. The lack of adequate care also resulted in an 26 inordinately high level of morbidity, defined as any significant injury, harm or medical 27 complication that falls short of death, among CDCR inmates. Id. at *8-9. For example: 28 18 1 2 3 4 5 6 [I]n 2004 a San Quentin prisoner with hypertension, diabetes and renal failure was prescribed two different medications that actually served to exacerbate his renal failure. An optometrist noted the patient s retinal bleeding due to very high blood pressure and referred him for immediate evaluation, but this evaluation never took place. It was not until a year later that the patient s renal failure was recognized, at which point he was referred to a nephrologist on an urgent basis; he should have been seen by the specialist within 14 days but the consultation never happened and the patient died three months later. 7 Id. (citations omitted). This incident was simply a representative example[] of the grossly 8 inadequate care that could be found throughout the prison system. Id. at *6. Many prisoners 9 were the victims of similar treatment, or worse. 10 Beyond these quality of care problems, the court noted a number of additional 11 deficiencies in the prison medical system. Prison medical facilities lack[ed] the necessary 12 medical equipment to conduct routine examinations and to respond to emergencies, id. at 13 *15, and were also completely inadequate for the provision of medical care : 14 15 16 17 18 Many clinics [did] not meet basic sanitation standards. Exam tables and counter tops, where prisoners with infections such as Methicillin-Resistant Staph Aureus (MRSA) and other communicable diseases are treated, [were] not routinely disinfected or sanitized. Many medical facilities require[d] fundamental repairs, installation of adequate lighting and such basic sanitary facilities as sinks for hand-washing. In fact, lack of adequate hygiene ha[d] forced the closure of some operating rooms. 19 Id. at *14 (citations omitted). Likewise, the management of prison pharmacy operations was 20 unbelievably poor. Id. at *16. No statewide coordination between pharmacies existed, and 21 there were serious, long-standing problems with dispensing medication, renewing 22 prescriptions, and tracking expired prescriptions. Id. Medical records in most CDCR 23 prisons were either in a shambles or non-existent. . . . mak[ing] even mediocre medical care 24 impossible, id. at *14 (citation omitted), and the resulting lack of access to inmates medical 25 histories result[ed] in dangerous mistakes, delay in patient care, and severe harm. Id. 26 Furthermore, the reception center intake process, which was designed to allow medical staff 27 to identify inmates medical issues, including communicable diseases posing a risk of 28 transmission to other inmates and staff, was woefully inadequate. Id. at *12-13. 19 1 The court also noted a number of serious personnel problems. Qualified medical staff 2 were sorely lacking at every level. According to one court expert, 20-50% of physicians at 3 the prisons provide[d] poor quality of care. Id. at *5. However, the CDCR was incapable 4 of recruiting qualified personnel to fill the significant vacancies that existed throughout the 5 system, id. at *11, and the CDCR s lack of a medical credentialing policy resulted in many 6 CDCR clinicians practicing outside of their areas of medical expertise. Id. at *21. The 7 CDCR also lacked medical leadership, both at the central office and at individual prisons, 8 and the resulting lack of supervision foster[ed] a culture of non-accountability and non9 professionalism whereby the acceptance of degrading and humiliating conditions bec[ame] 10 routine and permissible. Id. at *10 (internal quotations, citation, and alteration omitted). 11 Bases on these findings, the Plata court concluded that the establishment of a 12 Receivership, along with those actions necessary to effectuate its establishment, are narrowly 13 drawn to remedy the constitutional violations at issue, extend no further than necessary to 14 correct a current and ongoing violation of a federal right, and are the least intrusive means to 15 correct these violations. Id. at *33. The court recognized that: 16 17 18 19 20 the imposition of a Receivership is a drastic measure. But it is not a measure that the Court has sought, nor is it one the Court relishes. Rather, the Court is simply at the end of the road with nowhere else to turn. Indeed, it would be fair to say that the Receivership is being imposed on the Court, rather than on the State, for it is the State s abdication of responsibility that has led to the current crisis. Since the Court has jurisdiction over this matter, it has no choice but to step in and fill the void. 21 Id. at *31 (citation omitted). The court held the contempt remedy in abeyance after 22 concluding that a finding of contempt was not a prerequisite to the appointment of a receiver. 23 Id. at *33. Nevertheless, it sought to employ all feasible means other than a prisoner release 24 to remedy the constitutional violations. 25 26 6. Interim Remedies On plaintiffs motion, the court considered appointing a temporary receiver but 27 ultimately opted instead to appoint a correctional expert pending the search for and 28 appointment of a receiver. Id. at *34-35. On November 14, 2005, the Correctional Expert 20 1 filed a report and recommendations on interim remedies concerning clinical staffing and 2 death reviews. [T]he Correctional Expert s report powerfully underscore[d] the depth of the 3 crisis in the delivery of health care services in the CDCR . . . . Ex. D1065 at 1 (Dec. 1, 2005 4 Order). Over defendants objections, the court ordered a series of discrete, urgently needed, 5 remedial measures that could be undertaken immediately to improve recruitment and 6 retention of clinical staff. Id. at 1, 6-15. 7 8 7. Appointment of the Plata Receiver With the parties participation, the Plata court engaged in a national search for a 9 receiver. On February 14, 2006, the court appointed Robert Sillen as Receiver, with an 10 effective date of April 17, 2006. In its order of appointment, the court conferred broad 11 authority on the Receiver to provide leadership and executive management of the California 12 prison medical health care delivery system with the goals of restructuring day-to-day 13 operations and developing, implementing, and validating a new, sustainable system that 14 provides constitutionally adequate medical care to all class members as soon as practicable. 15 Ex. P313 at 2 (Feb. 14, 2006 Order). The Receiver was assigned the duty to control, 16 oversee, supervise, and direct all administrative, personnel, financial, accounting, 17 contractual, legal, and other operational functions of the medical delivery component of the 18 CDCR, id., and was granted all powers vested by law in the Secretary of the CDCR as they 19 relate to the administration, control, management, operation, and financing of the California 20 prison medical health care system. Id. at 4. On January 23, 2008, the Court appointed 21 J. Clark Kelso to replace Sillen as Receiver, and he has served in that capacity to date. 22 The Receivers have implemented substantial changes in the CDCR s prison medical 23 care system and have issued regular reports documenting their progress. For example, the 24 Receiver has increased recruitment and retention of clinical staff, implemented a new 25 pharmacy system, and instituted pilot programs to improve medical screening at reception 26 centers and management of chronic care. Nonetheless, as we describe below, fundamental 27 unconstitutional deficiencies, caused primarily by overcrowding, continue to exist and 28 prevent the delivery of constitutionally adequate medical care to California s inmates. 21 1 B. 2 While the Plata court has struggled to bring the CDCR s medical system into Coleman (Mental Health Care) 3 constitutional compliance for more than seven years, the Coleman action has lasted even 4 longer almost two decades. The first five years of litigation culminated in a finding that the 5 CDCR was violating the Eighth Amendment by failing to provide constitutionally adequate 6 mental health care to inmates with serious mental disorders. The past fourteen years have 7 involved continual efforts to remedy the constitutional violations. 8 At the time of the Coleman trial, the Eighth Amendment violations stemmed in large 9 part from the state s complete failure to identify with any accuracy the number of mentally ill 10 inmates in the prison population, despite several expert reports addressing the issue. Early in 11 Coleman s remedial phase, the state developed a screening mechanism to identify mentally 12 ill inmates and plans for a system that could deliver mental health care to the thousands of 13 inmates suffering from serious mental disorders. There are currently over 34,000 inmates 14 identified as seriously mentally ill in the state s prisons. Ex. P243 at 900124 (collection of 15 monthly CDCR mental health population placement reports, dated between December 2006 16 and August 2008). However, California remains unable to deliver constitutionally adequate 17 mental health care for these inmates. 18 After fourteen years of remedial efforts under the supervision of a special master and 19 well over seventy orders by the Coleman court, the California prison system still cannot 20 provide thousands of mentally ill inmates with constitutionally adequate mental health care, 21 and critically mentally ill inmates [are] languishing in horrific conditions without access to 22 immediate necessary mental health care. May 2, 2006 Order at 2.10 The relentless growth 23 of the inmate population has prevented the state from meeting its obligations under the 24 Eighth Amendment and has led, inexorably, to the proceeding before this court. 25 26 27 10 All references to court orders in this section of our opinion and order are to orders filed in the district court in Coleman. Coleman docket numbers are listed when multiple, 28 untitled orders were issued on the same day. 22 1 2 1. Findings of Eighth Amendment Violations The Coleman action was filed on April 23, 1990. On July 25, 1991, plaintiffs filed an 3 amended complaint, Ex. D1036, raising claims under the Eighth and Fourteenth 4 Amendments to the United States Constitution and the Rehabilitation Act, 29 U.S.C. § 794. 5 These claims were based on serious inadequacies in the delivery of mental health care to 6 inmates in the California adult prison system. Id. The Coleman court subsequently certified 7 a class consisting of inmates with serious mental disorders.11 Nov. 14, 1991 Order at 4-5. 8 The matter proceeded to trial before a United States Magistrate Judge, and in June 1994 the 9 magistrate judge found that defendants delivery of mental health care to class members 10 violated the Eighth Amendment. On September 13, 1995, the district court adopted the 11 magistrate judge s decision, with modifications. Coleman v. Wilson, 912 F. Supp. 1282 12 (E.D. Cal. 1995).12 13 In adopting the magistrate s findings, the Coleman court identified several significant 14 deficiencies in the delivery of mental health care to California s inmates. First, the court 15 found delays in access to necessary mental health care at each level of the mental health 16 care delivery system as it exist[ed] in the CDC, which result[ed] in exacerbation of illness 17 and patient suffering. Id. at 1308, 1309. Evidence specifically noted by the Coleman court 18 19 20 21 22 23 24 25 26 27 11 The class certified in 1991 consisted of all inmates with serious mental disorders who are now or who will in the future be confined within the California Department of Corrections (except the San Quentin State Prison, the Northern Reception Center at Vacaville and the California Medical Facility-Main at Vacaville). Nov. 14, 1991 Order at 4-5. On July 12, 1995, the Coleman class was decertified as to the Rehabilitation Act claim, which was dismissed with prejudice. July 12, 1995 Order at 2. The class definition was subsequently amended to include all inmates with serious mental disorders who are now, or who will in the future, be confined within the California Department of Corrections. July 23, 1999 Order & Stip. & Order Amending Plaintiff Class & Application of Remedy appended thereto at 2. Inmates suffering from serious mental disorders include those with Organic BrainSyndrome-Severe, Schizophrenia, Major Depression [or] the Bipolar Disorders, those who currently or within the last three years . . . [have] had a significant disorder of thought or mood which substantially impairs or substantially impaired reality testing, judgment or behavior, and those who currently do[] not have the ability to meet the functional requirements of prison life without psychiatric intervention, including psychotropic medication. Coleman, 912 F. Supp. at 1300 nn.15-16 (internal quotations and citations omitted). 12 The district court s order was issued following de novo review by that court of the 28 magistrate judge s findings and recommendations. Coleman, 912 F. Supp. at 1293, 1297. 23 1 included backlogs of 300-400 inmates awaiting transfer to enhanced outpatient programs at 2 California Men s Colony [( CMC )] or California Medical Facility [( CMF )] and a defense 3 exhibit describing the problem of the backlog of male inmates awaiting transfer to CMF and 4 CMC for mental health services as approaching the crisis level. Id. at 1309 (internal 5 quotations omitted). 6 In addition, defendants did not have a systematic program for screening and 7 evaluating inmates for mental illness. Id. at 1305. Instead, they relied on mechanisms that 8 were either used haphazardly, or depend[ed] for efficacy on incomplete or non-existent 9 medical records, self-reporting, or the observations of custodial staff inadequately trained in 10 the signs and symptoms of mental illness. Id. at 1305-06. As a result, thousands of 11 inmates suffering from mental illness [were] either undetected, untreated, or both. Id. at 12 1306. 13 Furthermore, the Coleman court found that defendants supervision of the use of 14 medication [was] completely inadequate; prescriptions [were] not timely refilled, there [was] 15 no adequate system to prevent hoarding of medication, . . . inmates on psychotropic 16 medication [were] not adequately monitored, and it appear[ed] that some very useful 17 medications [were] not available because there [was] not enough staff to do necessary post18 medication monitoring. Id. (internal quotations and citation omitted); see also id. at 1310. 19 The court also found violations of a constitutional magnitude in the involuntary medication 20 of inmates. Id. at 1313. In addition, the court found significant deficiencies in medical 21 record keeping, including disorganized, untimely and incomplete filing of medical records, 22 insufficient charting, and incomplete or nonexistent treatment plans at most prisons. Id. at 23 1314 (internal quotations and citation omitted); see also id. at 1315. The court found that 24 inmates [were] typically transferred between prisons without even such medical records as 25 might exist. Id. at 1314 (internal quotations and citation omitted); see also id. at 1315. 26 The Coleman court also found that the California Department of Corrections [was] 27 significantly and chronically understaffed in the area of mental health care services. Id. at 28 1307. Relying on the testimony of a defense expert, the Coleman court further found that 24 1 defendants [could not] provide adequate mental health care without some form of quality 2 assurance program to ensure the competence of their mental health care staff, but that the 3 CDCR lacked any such program. Id. at 1308. 4 These findings led the Coleman court to conclude that defendants lacked all of the 5 basic, essentially common sense, components of a minimally adequate prison mental health 6 care delivery system, id. at 1298 (citing Balla v. Idaho State Bd. of Corr., 595 F. Supp. 7 1558, 1577 (D. Idaho 1984) (citing Ruiz v. Estelle, 503 F. Supp. 1265, 1339 (S.D. Tex. 8 1980)), including proper screening; timely access to appropriate levels of care; an adequate 9 medical record system; proper administration of psychotropic medication; competent staff in 10 sufficient numbers; and a basic suicide prevention program. Id. at 1298 n.10. The Coleman 11 court found that the CDCR was seriously deficient in each of the first five components and 12 that the CDCR s suicide prevention program was adequate in design but inadequately 13 implemented due to severe and chronic understaffing throughout the CDCR. Id. at 14 1305-15.13 15 On the basis of its findings, the Coleman court entered an order for injunctive relief 16 requiring defendants to develop plans to remedy the constitutional violations under the 17 supervision of a special master. Id. at 1323-24; see also Fact #5, Nov. 17, 2008 Joint 18 Statement of Undisputed Facts. 19 20 21 13 In addition, the Coleman court found inappropriate use of disciplinary and 22 behavioral control measures directed towards the members of plaintiff class. Id. at 1319-20. Seriously mentally ill inmates were being treated with punitive measures by the custody 23 staff to control the inmates behavior without regard to the cause of the behavior because custody staff was inadequately trained in the signs and symptoms of serious mental illness. 24 Id. at 1320. Defendants placement of Coleman class members in administrative segregation and segregated housing units ( SHUs ) was found to violate the Eighth Amendment because 25 mentally ill inmates were placed in such units without any evaluation of their mental status, because such placement [caused] further decompensation, and because inmates [were] denied 26 access to necessary mental health care while they [were] housed in administrative segregation and/or segregated housing. Id. at 1320 (internal quotations and citation 27 omitted). The court also found unconstitutional defendants policy permitting the use of tasers and 37mm guns on Coleman class members without consideration of the impact of 28 such measures on mental illness. Id. at 1321-23. 25 1 2 2. Remedial Orders On December 11, 1995, the Coleman court appointed a special master to oversee the 3 remedial phase of the action. Dec. 11, 1995 Order Appointing a Special Master at 2. The 4 specific duties of the Special Master included working with defendants to develop a remedial 5 plan to address the constitutional violations identified by the court, monitoring defendants 6 implementation of and compliance with the remedial plan, and submitting interim reports on 7 the progress of the remedial plan and defendants compliance. Dec. 11, 1995 Order of 8 Reference at 3-4. 9 Eighteen months later, the Special Master submitted a report to the court accompanied 10 by remedial plans, policies, procedures, and forms collectively identified as the Mental 11 Health Services Delivery System Program Guides (hereafter Program Guides ). June 6, 12 1997 Special Master s Report on Plans, at 1-2.14 The court accepted the Special Master s 13 report, ordered two specific modifications recommended by the Special Master, gave 14 provisional approval to the Program Guides, and directed the Special Master to forthwith 15 commence monitoring defendants implementation of and compliance with the delivery of 16 mental health care services as set forth in the Program Guides. June 27, 1997 Order at 2-3. 17 Following the court s provisional approval of the Program Guides, defendants 18 continued to work with the Special Master to implement and revise the guides.15 In early 19 2006, the Special Master submitted a report and recommendations regarding a Revised 20 Program Guide that defendants concurrently submitted for final approval. See Jan. 2006 21 Revised Program Guide (Coleman docket # 1753). On March 3, 2006, the Coleman court 22 23 14 A reformatted copy of the Program Guides was filed in January 1998. Coleman docket # 913. 15 As the Coleman Special Master explained when defendants Revised Program Guide 24 was submitted for final approval, at the start of the remedial phase the basic program guides were a work in progress, hence their provisional adoption. Many of the programmatic 25 components of the defendant s mental health system were still embryonic and needed much nurturing. . . . All agreed that their implementation needed close scrutiny and analysis over 26 the next several years. During the subsequent implementation process, many aspects of the provisionally approved plans, policies, and protocols were revisited and amended by the 27 court, while some other provisions were modified and upgraded by the defendants on their own initiative. Feb. 3, 2006 Special Master s Report & Recommendations on Defs. 28 Revised Program Guide at 2. 26 1 gave final approval to all undisputed provisions of the Revised Program Guide and ordered 2 their immediate implementation. Mar. 3, 2006 Order at 1-2.16 3 Operating under the framework established by the Program Guides, the Coleman 4 court has engaged in extensive efforts to address the identified constitutional violations 5 through means other than a prisoner release order. Since June 1997, the Coleman Special 6 Master has filed twenty monitoring reports and fifty-six other reports. During the same 7 period, the Coleman court has issued well over seventy orders concerning the matters at the 8 core of the remedial process. As discussed in detail below, the vast majority of the orders by 9 the Coleman court have been directed at accurately projecting short-, medium-, and long10 range bed needs; creating a sufficient number of beds at the higher levels of the mental health 11 care delivery system; reducing delays in transfers to necessary levels of care; and ensuring 12 adequate staffing.17 In addition, the court has issued several orders addressing deficiencies at 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 16 The Revised Program Guide approved by the Coleman court in March 2006 contains specific provisions for an annual revision process. See Jan. 2006 Revised Program Guide (Coleman docket # 1753-2) at 12-1-14; see also Dezember Trial Aff. ¶ 24 ( The Program Guide is now subject only to an annual revision process. ). The Coleman court has specifically approved at least one additional modification to the Revised Program Guide. See Sept. 11, 2006 Stip. & Order at 3. The parties have offered three separate versions of the Revised Program Guide into evidence. Defendants have offered as Exhibit D1147 a document they represent to be the Revised Program Guide approved by the Coleman court in March 2006. See Dezember Trial Aff. ¶ 16. Plaintiffs have offered as Exhibit P9 a document identified as the September 2006 Revised Program Guide. Defendants have also offered as Exhibit D1148 a version of the 2008 Revised Program Guide to which is appended a redline document showing edits from a Draft August 2008 revision. Defendants represent that at the time of filing the 2008 Revised Program Guide had been through the annual revision process to enable [its] publication and that distribution of the final 2008 Revised Program Guide to the field [was] in the offing. Dezember Trial Aff. ¶ 24. Unless otherwise noted, all citations in this opinion and order are to the 2008 Revised Program Guide, Ex. D1148. 17 At the earliest stages of the remedial phase, the Special Master reported that defendants plan for screening inmates at reception centers represented a vast improvement over the screening procedures that existed at the time of trial, and that defendants had chosen an effective screening instrument. Mar. 12, 1996 First Report of the Special Master on the Remedial Plan at 6-7. Implementation of screening practices was slow at the start of the remedial phase, but by mid-1997 defendants screening process had improved. Evidence offered at the Coleman trial showed that, in July 1987, approximately 2,966 inmates had been identified with a psychiatric classification and/or placement in psychiatric facilities used by the CDCR, while, conservatively, over 4,000 inmates with serious mental disorders were undetected. See Coleman, 912 F. Supp. at 1306 n.29. By July 1997, 14,293 inmates with serious mental disorders had been identified. See Feb. 3, 2006 Special Master s Report & Recommendations on Defs. Revised Program Guide at 2. The Special Master s second monitoring report, filed in October 1998, reflected increasing institutionalization of, and 27 1 specific institutions.18 Finally, the court has issued several orders concerning suicide 2 prevention efforts, including, in the last five years, orders addressing a rising number of 3 inmate suicides, particularly in administrative segregation units.19 a. 4 5 Mental Health Care Beds and Treatment Space As the remedial phase of Coleman began and thousands of inmates with serious 6 mental disorders were identified, the need for additional treatment space at every level of the 7 mental health care delivery system became manifest. See Ex. D1292 (Special Master s 8 Response to Court s May 17, 2007 Request for Information) at 5 (noting emergence in mid9 and late-1990s of a need for much expanded mental health care and the space needed to 10 provide it ).20 11 12 compliance with, the mental health screening system, see Ex. D1108 (compilation of summaries and recommendations from the Coleman Special Master s twenty monitoring 13 reports) at DEFS059840-DEFS059849. By August 2008, there were 34,319 inmates with serious mental disorders identified in California s prison system. Ex. P243 at 900124. 14 18 E.g., Nov. 19, 1998 Order at 1-2 (regarding California Rehabilitation Center 15 ( CRC ), Mule Creek State Prison ( Mule Creek ), Salinas Valley State Prison ( SVSP ), Wasco State Prison ( Wasco ), Deuel Vocational Institution ( DVI ), California Institution 16 for Men ( CIM ), California Institution for Women ( CIW ), and California State PrisonSolano ( CSP-Solano )); Oct. 26, 2001 Order at 1-2 (regarding California Substance Abuse 17 Training Facility ( SATF ) and California State Prison-Los Angeles County ( CSP-LAC )); Apr. 25, 2002 Order at 2-3 (regarding CSP-LAC); June 13, 2002 Order (Coleman docket # 18 1384) at 1-2 (regarding CIM, SATF, California State Prison-Corcoran ( CSP-Corcoran ), CSP-LAC, CSP-Solano, San Quentin State Prison ( San Quentin ), and SVSP); Mar. 8, 2005 19 Order at 3-4 (regarding CSP-Corcoran, San Quentin, and Richard J. Donovan Correctional Facility ( R.J. Donovan )). 20 19 E.g., Dec. 22, 2000 Order at 4 (requiring Special Master to report on whether 21 defendants have adequate mechanisms for disciplining staff whose conduct contributes to inmate suicide); Oct. 1, 2001 Order at 2 (directing implementation of Suicide Reporting and 22 Review Policy); Jan. 12, 2004 Order at 2-3 (requiring several training and planning measures for suicide prevention); June 10, 2005 Order at 1-2 (Coleman docket # 1668) (requiring 23 implementation of several suicide prevention measures); June 8, 2006 Order at 2-3 (requiring defendants to develop a plan to deal within rising percentage of suicides in administrative 24 segregation and a budget and implementation schedule); Aug. 8, 2006 Stip. & Order at 1-2 (regarding use of video-monitoring for suicide watch observation); Sept. 11, 2006 Stip. & 25 Order at 3 (extending time to submit final plan regarding suicides in administrative segregation). 26 20 Plaintiffs also offered this document into evidence as Exhibit P35. Because we 27 discuss the reports of the Coleman Special Master and the Plata Receiver throughout this opinion and order, we note that, at trial, both plaintiffs and defendants introduced various 28 reports from the Receiver and the Special Master without objection. 28 1 At the time of the Coleman trial, mental health care delivery to inmates in California s 2 prison system was limited to a few institutions and involved some 3,200 designated mental 3 health care beds, Defs. Proposed Finding of Fact # 45 (citing Ex. D1273 at 43-44; 4 Dezember Trial Aff. ¶ 70), including beds for inpatient hospital care provided by the 5 Department of Mental Health ( DMH ) at CMF and Atascadero State Hospital. Dezember 6 Trial Aff. ¶ 70. After the Coleman trial, defendants undertook to implement plans 7 for the delivery of a continuum of mental health services, including long-term inpatient care (provided through the department s contract with the California Department of Mental Health), short-term inpatient care (the department s Mental Health Crisis Bed program), intensive outpatient care (the Enhanced Outpatient Program) and routine outpatient care (the Correctional Clinical Case Management program). 8 9 10 11 Mar. 12, 1996 First Report of the Special Master on the Remedial Plan at 2-3. Defendants 12 planned regional mental health care service areas, with [i]nitial entry to the service 13 continuum . . . provided primarily through a uniform screening process at each of the 14 CDCR s reception centers.21 Id. at 3. 15 Defendants remedial plans were built around the Mental Health Services Delivery 16 System ( MHSDS ) set forth in the original Program Guides and the Revised Program 17 Guide. The MHSDS is designed to provide mental health care to all inmates with current 18 symptoms of any of the Axis I serious mental disorders identified in the current Diagnostic 19 and Statistical Manual,22 inmates who need mental health treatment to protect life and/or 20 treat significant disability/dysfunction resulting from a diagnosed or suspected mental 21 disorder, and inmates with a diagnosis or recent episode of exhibitionism. Ex. D1148 at 22 23 24 21 We describe reception centers in more detail below when we discuss whether crowding is the primary cause of the constitutional violations at issue. See infra 25 Section IV.B.1. 22 26 As listed in the Revised Program Guide, these are: Schizophrenia (all subtypes); Delusional Disorder; Schizophreniform Disorder; Schizoaffective Disorder; Brief Psychotic 27 Disorder; Substance-Induced Psychotic Disorder (excluding intoxication and withdrawal); Psychotic Disorder Due to a General Medical Condition; Psychotic Disorder Not Otherwise 28 Specified; Major Depressive Disorders; and Bipolar Disorders I and II. Ex. D1148 at 12-1-6. 29 1 12-1-6. The MHSDS has the same basic structure as the embryonic 23 system first reported 2 by the Coleman Special Master in March 1996. The system is designed around four levels of 3 care: the Correctional Clinical Case Management Services program ( CCCMS or 3CMS ), 4 the Enhanced Outpatient Program ( EOP ), Mental Health Crisis Bed ( MHCB ) Placement, 5 and DMH Inpatient Hospital Care. Ex. D1148 at 12-1-7 to 12-1-9.24 6 A significant amount of remedial effort in Coleman has been spent on the as yet 7 unsuccessful endeavor to develop a sufficient number of mental health care beds at the EOP, 8 MHCB, and inpatient levels of care,25 as well as to provide adequate treatment space for all 9 inmates with serious mental health disorders.26 The Coleman court has issued numerous 10 orders addressing the need for mental health care beds and treatment space, including orders 11 23 Feb. 3, 2006 Special Master s Report & Recommendations on Defs. Revised 12 Program Guide at 2. 24 13 The CCCMS level of care is for inmates whose symptoms are under control or in partial remission and can function in the general prison population, administrative 14 segregation, or segregated housing units. Ex. D1148 at 12-1-7. The EOP level of care is for inmates who suffer Acute Onset or Significant Decompensation of a serious mental disorder 15 characterized by increased delusional thinking, hallucinatory experiences, marked changes in affect, and vegetative signs with definitive impairment of reality testing and/or judgment, 16 and who are unable to function in the general prison population but do not require twentyfour hour nursing care or inpatient hospitalization. Id. at 12-1-7 to 12-1-8. MHCBs are for 17 inmates who are markedly impaired and/or dangerous to others as a result of mental illness, or who are suicidal, and who require 24-hour nursing care. Id. at 12-1-8 to 12-1-9. The 18 MHCB level of care is also for inmates awaiting transfer to a hospital program and for inmates being stabilized on medication prior to transfer to a lower level of care. Id. 19 Finally, DMH inpatient care is for inmates who cannot be successfully treated at a lower level of care; both intermediate and acute levels of inpatient care are to be provided. Id. at 20 12-1-9. 25 21 CCCMS inmates are housed in the general prison population. 22 23 24 25 26 27 28 26 At a relatively early stage in the remedial process, defendants recognized the need to develop an adequate method of forecasting the need for such beds. However, according to Robin Dezember, Chief Deputy Secretary of CDCR s Correctional Healthcare Services Division at the time of trial, there was a period of several years prior to 2006 where there seemed to be a lack of continuous attention to this program. Rep. Tr. at 862:12-14. In 2002, a health care consulting firm designed a mental health bed demand forecast methodology for the CDCR. . . . This method projects future bed needs based on several variables that drive bed usage, including total overall prison population, length of stay and discharge rates of patients in inpatient status, and growth in outpatient demand proportional to the historical prevalence of outpatients in the total prison population. Defs. Statewide Mental Health Bed Plan, April 2006, filed April 17, 2006, at 3. In 2006, defendants acknowledged that the forecasting methodology developed in 2002 needed to be updated. May 2, 2006 Order at 2 n.1. 30 1 directing defendants to assess the need for beds and treatment space throughout the mental 2 health care delivery system and to plan for and develop the necessary number of beds as well 3 as sufficient space at each level of care.27 4 When the state s growing prison population reached a record of more than 160,000 in 5 2006, the shortage of beds and space reached a crisis level. In March 2006, defendants were 6 ordered to submit a plan to meet both the immediate and long-term need for mental health 7 care beds. Mar. 3, 2006 Order at 3-4. During a subsequent hearing on the adequacy of 8 defendants proposed plan, the CDCR s then-Director of Health Care Services reported a 9 shortage of 75 MHCBs and 125 intermediate inpatient beds and repeatedly referred to the 10 shortage as a crisis. May 2, 2006 Order at 2. The Coleman court found that defendants 11 plan entirely failed to address the CDCR s immediate bed needs: The special master reports, the record reflects, and defendants admit, that the plan presented to the court in no way adequately responds to the severe shortage of intermediate care facility beds and mental health crisis beds that currently exists in the CDCR. It is undisputed that the shortage is leaving critically mentally ill inmates languishing in horrific conditions without access to immediately necessary mental health care. 12 13 14 15 16 Id. The court further found that defendants long-range plan for the provision of acute and 17 intermediate care beds and mental health crisis beds appeared sound in principle, but 18 required revision because it was based on population figures that were already out of date. 19 Id. Defendants plan for EOP beds was not approved because it describe[d] a shortfall of 20 over 1000 such beds in the year 2011. Id. at 4. Following the hearing, the court ordered 21 defendants to file an amended long-term plan and to include with that plan a list of any 22 projects that could be accelerated; to file a plan for the interim provision of intermediate 23 inpatient beds and mental health crisis beds; and to maintain, open, or create intermediate 24 inpatient and mental health crisis beds at specific prison locations. Id. at 4-6. The Coleman 25 court has subsequently issued several orders concerning the provision of EOP, MHCB, and 26 27 E.g., May 21, 1998 Stip. & Order at 4; Sept. 14, 2000 Order at 2; Apr. 4, 2001 Order at 4; June 27, 2001 Order at 2; Dec. 20, 2001 Order at 1-2; Mar. 4, 2002 Order at 1; May 7, 27 2002 Order at 1-2; Oct. 8, 2002 Order at 2; Jan. 12, 2004 Order at 2; Apr. 5, 2004 Order at 3; July 9, 2004 Order at 3-4; Oct. 5, 2004 Order at 2; Jan. 27, 2005 Order at 2; Mar. 3, 2006 28 Order (Coleman docket # 1772) at 3-4. 31 1 inpatient beds, all of which are in critically short supply, including an extensive order 2 concerning defendants long-range and interim plans for the provision of these beds.28 3 However, providing the beds is obviously infeasible without the necessary space in which to 4 locate them, especially in light of the constantly increasing need for such beds as a result of 5 the substantial, if unanticipated, growth in the prison population. b. 6 7 Transfers to Appropriate Level of Care Throughout Coleman s remedial phase, the state s delivery of mental health care to its 8 inmates has been plagued by delays in the transfer of inmates to higher levels of care. Both 9 the original Program Guides and the Revised Program Guide include timelines for post10 referral transfers to EOP programs, mental health care crisis beds, and DMH inpatient beds. 11 See Coleman docket # 913 at 1-4, 4-13, 5-13, 6-4; Ex. D1148 at 12-1-16.29 Unfortunately, 12 the state remains unable to transfer inmates to required care in a timely fashion, and the 13 14 28 Oct. 20, 2006 Order; see also July 20, 2006 Order (Coleman docket # 1904) at 1; Aug. 23 2006 Order. 29 15 The timelines in the Revised Program Guide are as follows: 17 Reception Centers: EOP transfers should occur within 60 days, or 30 days if clinically indicated. CCCMS transfers should occur within 90 days, or 60 days if clinically indicated. 18 MHCB: MHCB transfers should occur within 24 hours of referral. 19 22 DMH: Transfers to DMH acute placements should occur within 10 days of referral, if accepted to DMH. Referral must be completed within 2 working days of identification. Transfers to DMH intermediate care placements should occur within 30 days of referral, if accepted to DMH. Referral must be completed within 5-10 working days. 23 EOP: Transfers to general population ( GP ) EOP programs should occur within 60 days, or 30 days if clinically indicated. 24 26 EOP Administrative Segregation Unit ( ASU ) Hub: EOP inmates housed in the regular ASU should transfer to an EOP ASU Hub within 30 days of placement in the regular ASU or within 30 days of referral to EOP level of care. 27 PSU: EOP inmates housed in the ASU who are endorsed for the PSU must be transferred within 60 days of endorsement. 16 20 21 25 28 Stewart Expert Report ¶ 153; see Ex. D1148 at 12-1-16. 32 1 Coleman court has issued numerous orders directed at expediting transfers and reducing 2 delays.30 c. 3 4 Staffing A final focus of the remedial effort in Coleman over the last decade has been the 5 development and retention of sufficient numbers of competent mental health care clinicians. 6 In June 1998, the Coleman court issued the first of numerous orders aimed at remedying the 7 substantial understaffing of the CDCR s mental health care system, directing defendants to 8 show improvement in the quality and quantity of contracted psychiatric services and/or the 9 implementation of a recruitment program sufficient[] to fill vacancies in presently 10 authorized positions. June 16, 1998 Order at 1. In the same order, the court directed the 11 Coleman Special Master to recommend the staffing ratios necessary to a constitutionally 12 adequate mental health care delivery system. Id. at 2.31 Since then, the court has repeatedly 13 ordered defendants to create the necessary positions and to hire staff to fill those positions.32 14 In addition, the court has issued orders designed to assure the competence of staff, primarily 15 by requiring the state to develop and implement a quality assurance and peer review 16 process.33 17 After two years of compliance monitoring, it became apparent that orders setting 18 staffing ratios and requiring defendants to fill clinical positions would not be sufficient to 19 30 E.g., July 26, 1999 Order at 5-6; Jan. 13, 2000 Order (Coleman docket # 1111) at 4; 20 Apr. 27, 2000 Order at 5; July 3, 2000 Order at 6; Sept. 14, 2000 Order at 2; Apr. 4, 2001 Order at 3-4; Jan. 12, 2004 Order at 2; Mar. 25, 2004 Order at 2-3; Mar. 8, 2005 Order at 2; 21 Oct. 20, 2006 Order at 3. 22 31 In July 1999, the court approved several mental health staffing ratios and required defendants to adopt and implement specific mental health care staffing ratios for 23 administrative segregation units. July 26, 1999 Order at 4-5. 32 24 E.g., Aug. 25, 1998 Order at 1; Jan. 19, 1999 Order at 2; July 26, 1999 Order at 4; Jan. 13, 2000 Order (Coleman docket # 1111) at 4; Apr. 27, 2000 Order at 5; July 3, 2000 25 Order; Aug. 28, 2000 Order (Coleman docket # 1198) at 3; Apr. 4, 2001 Order at 4; Oct. 26, 2001 Order at 1; June 13, 2002 Order (Coleman docket # 1383) at 4; June 13, 2002 Order 26 (Coleman docket # 1384) at 2; Mar. 3, 2006 Order (Coleman docket # 1772) at 3; Mar. 9, 2006 Order (Coleman docket # 1774) at 1-2. 27 33 E.g., June 16, 1998 Order at 2; Aug. 12, 1998 Order at 1-2; June 13, 2002 Order 28 (Coleman docket # 1384) at 2. 33 1 remedy the constitutional violations. Accordingly, the Coleman court ordered defendants to 2 develop a plan to retain CDCR psychiatrists. July 26, 1999 Order at 4. Over the next eight 3 years, as part of its ongoing effort to ensure that California hires and retains sufficient 4 clinical staff, the court issued several orders concerning recruitment and retention bonuses, as 5 well as salary increases for mental health clinicians.34 3. 6 7 Special Master s 2006 Monitoring Reports By the end of the first decade of remedial work in Coleman, the state had made some 8 progress but still had not met its constitutional obligation to provide Coleman class members 9 with adequate mental health care. July 23, 2007 Order, 2007 WL 2122636, at *3. Worse, 10 two monitoring reports filed by the Coleman Special Master in 2006 reflected a troubling 11 reversal in the progress of the remedial efforts of the preceding decade and demonstrated the 12 profound impact of population growth on the state s ability to meet its constitutional 13 obligations to seriously mentally ill inmates. 14 On January 23, 2006, the Coleman Special Master filed his Fifteenth Monitoring 15 Report, which included findings made at monitoring visits to all CDCR institutions between 16 early August 2004 and late May 2005. Jan. 23, 2006 Fifteenth Monitoring Report at 2-3. 17 The report was grim. The Special Master reported rising vacancy rates in staffing, as well as 18 a growing crisis in accessibility to a MHCB level of care and the continuing inadequacy of 19 access to DMH programs highlighted by the unmet needs assessment that was conducted and 20 concluded during the period. Ex. D1108 (compilation of summaries and recommendations 21 from the Coleman Special Master s twenty monitoring reports) at DEFS06022122 DEFS060222. The Special Master also reported that suicides in CDCR escalated 23 significantly during the monitoring period for reasons that are just beginning to be subjected 24 to analysis. Id. at DEFS060222. 25 26 34 E.g., Jan. 13, 2000 Order (Coleman docket # 1111) at 4-5; July 25, 2003 Order at 6; Mar. 8, 2005 Order at 1-2; June 10, 2005 Order (Coleman docket # 1667) at 1-2; Mar. 9, 27 2006 Order (Coleman docket # 1774) at 1-2; Dec. 15, 2006 Order at 1-2; Feb. 7, 2007 Order at 2; May 23, 2007 Order (Coleman docket # 2236) at 5; June 28, 2007 Order (Coleman 28 docket # 2301) at 3. 34 1 The Special Master further reported that transfers to more intensive levels of mental 2 health programming and treatment had deteriorated sharply and widely. Id. at 3 DEFS060252. The availability of MHCBs, the department s sole internal resource for 4 providing short-term crisis care for unstable and suicidal inmates, had declined to the point 5 that it became by mid-2005 a critical issue with severe impact on CDCR s most seriously 6 mentally disordered inmates. Id. In addition, the waiting list for the admission to 7 Psychiatric Service Units (PSUs) for EOPs with a SHU [Segregated Housing Unit] term, 8 imposed on inmates who are viewed as a danger to themselves or others, expanded steadily, 9 and mental health caseload inmates continued to spend long periods in reception awaiting 10 transfer to EOP and 3CMS general population programs. Id. at DEFS06025211 DEFS060253. 12 Taken together, the expanding wait lists, critical shortage of beds, and identification of 13 hundreds of inmates in need of clinical referrals meant that a growing number of the most 14 seriously mentally ill inmates in the CDCR were not receiving in a timely fashion the levels 15 of care they needed. Id. at DEFS060253. To explain this backward slide in the progress 16 made under the Coleman court s supervision, the Special Master pointed to the prison 17 system s expanding population. For example, none of the [CDCR s] planning documents . . 18 . addressed the department s need to expand its capacity to provide acute inpatient DMH care 19 to meet the expanding need being pushed, among other causes, by an inexorably rising 20 MHSDS population commensurate with CDCR s growing overall population. Id. at 21 DEFS060258. Likewise, progress in the timely transfer of mentally ill inmates from 22 reception centers into general population programs had been largely cancelled by the 23 recently escalating growth in the overall CDCR population and the concomitantly increasing 24 number of MHSDS inmates in reception. Id. at DEFS060272-DEFS060273. 25 Defendants did not object to the Special Master s Fifteenth Monitoring Report or the 26 recommendations contained therein, including the Special Master s finding as to the role 27 played by the rapidly growing prison population and the resulting lack of space necessary to 28 35 1 provide the requisite care to mentally ill inmates. Mar. 3, 2006 Order (Coleman docket 2 # 1772) at 1. 3 As compliance work continued in 2006, the population pressures identified by the 4 Special Master in his Fifteenth Monitoring Report were evident: Compliance became more 5 difficult and the gains made by defendants in the first decade receded. On December 14, 6 2006, the Coleman Special Master filed his Sixteenth Monitoring Report. That report, which 7 covered a monitoring period from the summer of 2005 until March 2006, Ex. D1108 at 8 DEFS060302, revealed that serious shortages in staffing and bed space, as well as substantial 9 delays in transfers to necessary levels of care, continued unabated. 10 Among other findings, the Special Master reported that the inexorably expanding 11 demand for services resulting from the bulging population had caused a continuing 12 deterioration of mental health staffing. Id. at DEFS060303. According to the Special 13 Master, [t]welve years after the determination that mental health treatment in CDCR was 14 unconstitutional, the defendants still lacked clinical resources to meet the needs of some 25 15 to 30 percent of inmates identified as seriously mentally disordered. Id. at DEFS060304. 16 17 18 19 20 Furthermore, the Special Master reported that [t]he general breakdown in transfers was another transcendent issue in the 16th round of review. As the overall caseload population continued to increase, so too did the percentage of the caseload in need of program beds with intensive care and high security, including specifically DMH inpatient beds, MHCBs, PSU beds and EOP administrative segregation placements. 21 id. at DEFS060306. [A]ccess to appropriate levels of care for seriously mentally ill inmates 22 remained a problem in almost every CDCR institution. Id. at DEFS060307. 23 Although Defendants filed a response to two recommendations contained in the 24 Sixteenth Monitoring Report, they did not object to any of the above findings, once more 25 including the Special Master s determination that the escalating growth in the overall 26 CDCR population was a major cause of the CDCR s reversal of progress. Id. at 27 DEFS060273; see Defs. Dec. 7, 2006 Response to Special Master s Sixteenth Report. 28 36 1 C. 1. 2 3 Crowding in California s Prison System The Increasing California Prison Population Since the mid-1970s, California s prison population has increased by over 750 4 percent, rising from approximately 20,000 inmates to an all-time high in October 2006 of 5 over 170,000 inmates, with more than 160,000 housed in the state s adult prison institutions. 6 Ex. P1 at 1 (Governor Schwarzenegger s Oct. 4, 2006 Prison Overcrowding State of 7 Emergency Declaration); Ex. P5 at 62 (May 2006 California Policy Research Center Report, 8 Understanding California Corrections ); Fact # 9, Nov. 17, 2008 Joint Statement of 9 Undisputed Facts; Ex. D1259-1. Much of this population expansion occurred during the time 10 in which the Plata and Coleman courts have monitored the medical and mental health care in 11 California s prisons. In 1991, when the Coleman plaintiffs filed their amended complaint, 12 the state s prison system housed approximately 100,000 inmates. Ex. P410 at 2 (CDCR 13 Offender Information Services Branch Data Analysis Unit, Institution and Camp Design Bed 14 Capacity and Population, June 30, 1987 - June 30, 2007). As of August 27, 2008, 156,352 15 inmates were housed in in-state prison institutions. Fact # 10, Nov. 17, 2008 Joint Statement 16 of Undisputed Facts.35 17 The expansive growth of the prison population in California is due, in part, to the 18 state s adoption of determinate sentencing in the 1970s, Ex. P5 at 61-62, and the countless 19 increases in criminal sentences enacted by the legislature or in initiative measures in 20 succeeding years, Ex. P3 at 68 (Jan. 2007 Little Hoover Commission Report, Solving 21 California s Corrections Crisis: Time Is Running Out ) (detailing increases in California 22 sentencing since the Determinate Sentencing Act became effective in 1977). In addition, 23 California s prison population has increased because of its post-sentencing practices. The 24 state has [] been widely criticized for not doing a better job of preparing inmates to return to 25 35 In this opinion and order, we will hereafter consider only figures and percentages relating to the CDCR s thirty-three in-state adult prison institutions. We do not consider 26 camps, community correction centers, or Department of Mental Health state hospitals, all of which also house CDCR inmates. It is the thirty-three in-state adult prison institutions that 27 are the subject of the Governor s Prison Overcrowding State of Emergency Proclamation and were the focus of the evidence at trial before this court. All references to system and 28 systemwide encompass only those thirty-three adult institutions. 37 1 society. Ex. P4 at 121 (June 2004 Corrections Independent Review Panel Report, 2 Reforming Corrections ). 3 Approximately 90 percent of state prison inmates are eventually released on parole, and at present, more than half return to prison. A 2003 study by the Little Hoover Commission concluded that inmates are not prepared for their release from prison. Department of Corrections reports show that 43 percent of inmates released from prison in 1999 were sent back to prison within a year and that 56 percent returned within two years. Many of those returned to prison are parolees who are sent back for violating the conditions of parole, rather than for committing new crimes, and many of those go back for relatively short periods of time an average of 5½ months. 4 5 6 7 8 9 Id. The consequences of the state s failure to prepare inmates for re-entry are significant: 10 The vast numbers of parolees returning to prison help drive both the size of the prison 11 population and the cost of the system. In 2001 more than 74,000 (47 percent) of the average 12 daily prison inmate population of 157,000 was made up of parole violators. Id. Finally, 13 also significant are the actions of the parole board and the Governor in declining to release 14 prisoners serving terms of 15 or 25 years to life who have served their minimum sentence or 15 more with unblemished records and are determined by prison officials not to constitute a risk 16 to society. 17 18 19 2. Studies Commissioned by the State of California To Examine Prison Crowding The California legislature has recognized prison crowding as a serious problem since 20 at least 1987, when it convened a Blue Ribbon Commission on Inmate Population 21 Management. See Ex. P2 at 78. The commission issued its final report in 1990, with thirty22 eight recommendations, including alternative sanctions, and more programming [and] 23 reentry programs. Id. Between 1990 and 2006, more than a dozen commissions and other 24 groups issued reports with proposals to solve the overcrowding problem in California s 25 prison system. Id. at 3, 10, 78-79. As Joan Petersilia, co-chair of the expert panel convened 26 by the CDCR in 2007, noted, all of the reports recommended essentially the same ten 27 things, including diverting non-violent, non-serious offenders and technical parole violators 28 from prison; using a risk and needs assessment tool to match inmates with resources and 38 1 programming; expanding rehabilitative programs; reforming California s determinate 2 sentencing system; transferring low-risk prisoners in the later part of their sentences to 3 community-based reintegration facilities; establishing a sentencing commission; reforming 4 parole; creating partnerships between state and local corrections agencies; requiring that all 5 programs be based on solid research evidence; and promoting public awareness regarding 6 California s prison system. Id. at 77. 7 One of the most exhaustive reports completed during this period was the June 2004 8 report of the Corrections Independent Review Panel, which was appointed by Governor 9 Schwarzenegger; chaired by former California Governor and Attorney General George 10 Deukmejian, who had a reputation as tough on crime; and composed of forty independent 11 correctional consultants and representatives from state agencies. Ex. P4 at i. The Panel 12 noted that California s correctional system has grown to become the largest in the nation, 13 rivaling in size and numbers even those of most other countries, and that [n]ot surprisingly, 14 this massive system shows the strains of both its age and its decades-long growth. Id. at 15 199. The Panel found that [a]dult prisons are severely overcrowded, imperiling the safety 16 of both correctional employees and inmates. Id. Consequently, a number of the Panel s 237 17 recommendations, including the enhancement of earned credits, the expansion of 18 rehabilitative programming, the identification of older inmates for early release, and the 19 diversion of certain parole violators, were aimed at inmate population reduction. See id. at 20 122-61. 21 22 3. Defining the Capacity of California Prisons In its report, the Corrections Independent Review Panel discussed three distinct 23 measures of prison capacity: design capacity, operable capacity, and maximum safe and 24 reasonable capacity. Ex. P4 at 123-124. First: 25 26 27 Design capacity is the term used for the past 50 years to designate the number of inmates a prison is designed to accommodate according to standards developed by the Commission on Accreditation and the American Correctional Association. [Footnote omitted.] The number can be based on any combination of single-occupancy cells, double-occupancy 28 39 1 2 3 cells, single- or double-bunked multiple occupancy rooms, or dormitories. The standards take into account the need for humane conditions, as well as the need to prevent violence and move inmates to and from programs, such as mental health care, education classes, and drug abuse treatment. 4 Id. at 123. In California, design capacity is based on one inmate per cell, single bunks in 5 dormitories, and no beds in space not designed for housing. Id. 6 California has never limited its prison population to 100% design capacity, id. at 123 7 n.1, and has in some respects planned for inmate population levels that exceed 100% design 8 capacity. The staffing packages for California s prison facilities have two parts: the 9 initial staffing package, which is based on population at 100% design capacity, or one 10 inmate per cell, and the overcrowding package which, depending on the level of the facility 11 being built, could be 150 percent, 175 percent, 190 percent or 200 percent. Rep. Tr. at 12 540:24-541:4 (Raymond). The overcrowding package is a staff enhancement of the 13 design bed package. Id. at 548:4-7. The combined staffing package shows the size of the 14 staff necessary for a facility at 100% design capacity and the additional staff required as the 15 facility becomes more crowded. Id. at 545:10-13. 16 Similarly, prisons built between 1985 and 1998, when the design capacity of the 17 CDCR s adult institutions and camps increased from 29,042 to near its present level of 18 approximately 80,000 inmates, Ex. P212 at Table 10, were designed and built to 19 accommodate population growth with respect to some infrastructure components 20 specifically the water, wastewater, electrical and mechanical components, needed to meet 21 anticipated overcrowding of as much as 190 percent in cells and 140 percent in 22 dormitories. Dezember Trial Aff. ¶ 72 (quoting Ex. D1292, Coleman Special Master s 23 May 31, 2007 Response to Court s May 17, 2007 Request for Information, at 5). However, 24 these same prisons were not designed and made no provision for any expansion of medical 25 care space beyond the initial 100% of [design] capacity. Id. (quoting Ex. D1292 at 4-5). 26 Even worse, none of the 19 CDCR institutions planned and built in the boom of the 80s 27 and 90s gave any thought to the space that might be needed for mental health purposes. Id. 28 40 1 (quoting Ex. D1292 at 5).36 A similar failure in design vision occurred with the Department 2 of Mental Health, the sole provider of inpatient mental health care for CDCR inmates, 3 which discovered in 1998 that it had no facilities of its own in which to provide the level of 4 inpatient care needed by CDCR for high custody inmates with a history of violence or 5 escape. Id. (quoting Ex. D1292 at 8). Thus, even though the infrastructure of California s 6 newer prisons was built to accommodate inmate populations greater than 100% design 7 capacity, no similar accommodation was made for the provision of medical and mental health 8 care in California s prisons. 9 The second measure of prison capacity, operable capacity, refers to the maximum 10 capacity of the prisons to house inmates safely and securely while providing effective 11 education, training, and treatment. Ex. P4 at 122. Operable capacity . . . takes into account 12 space needed for effective programming in addition to safety and security. Id. at 124. 13 Based on input from a group of experienced California prison wardens, the Corrections 14 Independent Review Panel determined that the operable capacity of California s prison 15 system is 145% design capacity. Id. Notably, however, operable capacity does not take into 16 account the space required to provide medical and mental health care. See id. at 161 n.3; 17 Nov. 9, 2007 Scott Report ¶ 46. 18 The third measure, maximum safe and reasonable capacity, refers to the 19 maximum number of inmates who can safely and reasonably be housed in the prison 20 system. Ex. P4 at 124. This definition takes into account only the safe and reasonable 21 capacity of individual housing units according to inmate custody levels, staffing levels, and 22 the physical structure of the units. Id. Units for inmates at higher custody levels have a 23 24 25 26 27 28 36 But see Sept. 3, 2008 Tilton Dep. at 60:10-61:17 (testifying that in the 1980s and 1990s, the CDCR would make sure [it] provided programs based on the population, and that this testimony referred to prisons at somewhere between 100% and 140% design capacity). We do not credit Tilton s testimony on this point because he also testified that the CDCR operated fully-programmed facilities at that time. Id. at 61:16-17. As is clear from our discussion of the history of the Plata and Coleman cases, the CDCR was not operating fully-programmed facilities with regard to medical and mental health care. Moreover, even if Tilton s testimony were to be credited, he acknowledged that certain facilities lost the ability, in terms of space, to deliver adequate programs to the inmates when populations exceeded 140% design capacity. Id. at 62:14-19. As we note below, the California prison population well exceeds 140% design capacity, and indeed is approaching 200%. 41 1 lower maximum safe and reasonable capacity than units for inmates who present a lower 2 security risk. Id. at 124. 3 The Department of Corrections has determined the maximum safe and reasonable capacity of the general population and reception center housing to be 190 percent of design capacity, while other housing can be filled only to between 100 and 160 percent of design capacity. Overall, the Department has determined that the maximum safe and reasonable capacity of the state s male prisons is . . . 179 percent of design capacity. 4 5 6 7 Id. Maximum safe and reasonable capacity does not take into account the need for 8 humane conditions incorporated into design capacity, or the need for programming space 9 incorporated into both design and operable capacity. See id. at 123-124. More important for 10 present purposes, that classification does not take into account the space or facilities required 11 to provide medical or mental health care. 12 13 4. Crowding in Relation to Capacity California s inmate population has far exceeded the design capacity of the state s 14 prison system for over twenty-five years. See, e.g., Ex. P268 at 2 (Institution and Camp 15 Design Bed Capacity and Population, June 30, 1983 - June 30, 2003); Ex. P410 at 2; 16 Ex. D1259-1. By October 2006, the state s adult prisons, excluding camps, were operating at 17 200.2% design capacity with 162,792 inmates.37 Ex. D1149 at 1 (CDCR weekly population 18 report as of October 25, 2006). As of August 27, 2008, the population of these institutions 19 was reduced to 195.9% design capacity with 156,352 inmates, largely as a result of shipping 20 several thousand prisoners to Mississippi and other contract states. Ex. P135 at 1 (CDCR 21 weekly population report as of August 27, 2008). The current level of crowding far exceeds 22 even the maximum safe and reasonable capacity of the California prison system, which, by 23 CDCR s own determination, is 179% design capacity for prisons holding male prisoners. 24 Ex. P4 at 124. 25 26 27 37 The state also operates several prison camps, housing just over 4000 inmates. These camps are less crowded than the adult institutions and operate at between 100% and 110% 28 design capacity. Ex. P20 at 1; Ex. P21 at 1. 42 1 D. 2 In response to the severity of the prison crowding problem, Governor Arnold Governor Schwarzenegger s Emergency Proclamation 3 Schwarzenegger, a primary defendant in both Plata and Coleman, declared a state of 4 emergency on October 4, 2006. Ex. P1. In his Prison Overcrowding State of Emergency 5 Proclamation, the Governor declared that all 33 of CDCR s prisons are now at or above 6 maximum operational capacity, and 29 of the prisons are so overcrowded that the CDCR is 7 required to house more than 15,000 inmates in conditions that pose substantial safety risks ; 8 that the severe overcrowding in 29 CDCR prisons has caused substantial risk to the health 9 and safety of the men and women who work inside these prisons and the inmates housed in 10 them ; that the overcrowding crisis gets worse with each passing day, creating an 11 emergency in the California prison system ; and that immediate action is necessary to 12 prevent death and harm caused by California s severe prison overcrowding. Id. at 1, 6, 8. 13 The risks enumerated by the Governor in his Proclamation include increased, 14 substantial risk for transmission of infectious illness ; security risks caused by line-of-sight 15 problems for correctional officers, particularly in areas where inmates are triple-bunked and 16 in tight quarters ; and thousands of gallons of sewage spills and environmental 17 contamination from overloading the prisons sewage and wastewater systems. Id. at 2. 18 Governor Schwarzenegger also declared that the suicide rate in the 29 severely overcrowded 19 prisons [was] approaching an average of one per week. Id. at 6. 20 In addition, the Proclamation described three separate proposals by the Governor to 21 address the overcrowding crisis, including a proposal for two new prisons and space for 22 83,000 prisoners to address California s current and future incarceration needs. Id. at 7. 23 The California Legislature rejected all of these proposals. Id. As a result, the Governor 24 invoked his powers under the California Emergency Services Act to call for immediate 25 efforts to transfer inmates to out-of-state correctional facilities, as well as the suspension of 26 state contracting laws so that the CDCR could contract for all goods and services needed to 27 immediately mitigate the severe overcrowding and the resulting impacts within California. 28 Id. at 8-9. 43 1 The California Correctional Peace Officers Association ( CCPOA ), a plaintiff- 2 intervenor in this case, challenged the validity of the Proclamation in state court. On June 4, 3 2008, the California Court of Appeal upheld the Proclamation, finding that the Governor 4 acted within his authority, in part because the declaration of emergency was based on 5 conditions that presented extreme peril to the safety of persons and property. CCPOA v. 6 Schwarzenegger, 163 Cal. App. 4th 802 (2008). The Proclamation declaring a state of 7 emergency remains in effect. Fact # 12, Nov. 17, 2008 Joint Statement of Undisputed Facts. 8 9 10 11 E. Motions To Convene Three-Judge Court and Subsequent Prison Studies by the State of California 1. Motions To Convene and Initial Proceedings Following the Governor s issuance of the State of Emergency Proclamation, the 12 plaintiffs in Plata and Coleman filed motions to convene a three-judge court to limit the 13 prison population.38 The Plata court continued the hearing on its motion to provide 14 defendants with an opportunity to outline specific measures they were taking or planned to 15 take to alleviate crowding, as well as to allow the Plata Receiver to analyze the effects of 16 crowding on his remedial efforts. Feb. 15, 2007 Order in Plata at 4-5. Similarly, the 17 Coleman court, after oral argument, continued the hearing for six months to permit 18 defendants to demonstrate sufficient progress in their remedial efforts and in relieving prison 19 overcrowding such that convening a three-judge court would not be necessary. Dec. 11, 20 2006 Rep. Tr. in Coleman, passim; Dec. 12, 2006 Order in Coleman at 1. 21 22 2. Intervening Reports on Prison Crowding During the period in which the motions to convene a three-judge court were pending, 23 two more reports concerning prison overcrowding were presented to the California 24 Legislature. First, in January 2007, the Little Hoover Commission, a bipartisan and 25 independent state body charged with conducting research and preparing recommendations to 26 38 The Prison Litigation Reform Act of 1996 ( PLRA ) provides that a prisoner release 27 order may be issued only by a three-judge court. 18 U.S.C. § 3626(a)(3)(B). We discuss in more detail below, infra Section III, the meaning of the term prisoner release order and 28 other relevant provisions of the PLRA. 44 1 improve the economy, efficiency, and service of California state government, Cal. Gov t. 2 Code §§ 8501, 8521-8522, echoed the concerns in the Governor s State of Emergency 3 Proclamation, stating that California s prisons are out of space and running out of time. 4 Ex. P3 at 1. In its report, entitled Solving California s Corrections Crisis: Time Is Running 5 Out, the Commission, which had previously issued a series of reports on California s 6 prisons, id. at 13, again offered comprehensive recommendations to reduce the prison 7 population, improve public safety, and manage public dollars, id. at 1. Second, in June 2007, 8 the Expert Panel on Adult Offender Recidivism Reduction Programming a panel convened 9 by the CDCR and consisting of the CDCR s Chief Deputy Secretary for Adult Programs and 10 a number of academic experts, consultants, and former and current secretaries of corrections 11 in Pennsylvania, Arizona, Washington, Ohio, and Maine,39 Ex. P2 at ii issued a report 12 recommending a course of action to reduce the prison population while at the same time 13 reducing recidivism and generating savings. Ex. P2. 14 The first recommendation of both the Little Hoover Commission and the CDCR 15 Expert Panel was to reduce prison overcrowding. Ex. P3 at iv; Ex. P2 at 10. Both panels 16 noted that the state had received numerous reports over the past two decades containing 17 recommendations for reducing the state s prison population. Ex. P3 at iv; Ex. P2 at 10 & 18 App. A. Although the Expert Panel was convened to make recommendations for reducing 19 California s high recidivism rate and improving the programming in California s prison and 20 parole system, Ex. P2 at vii, and not for solving the overcrowding problem, id. at 10, the 21 panel nonetheless found that California s prisons were dangerously overcrowded and that 22 reducing overcrowding was a pre-condition to [the] success of its mission, id. at viii. 3. 23 24 Final Hearing and Rulings On June 27, 2007, the Plata and Coleman courts jointly heard oral argument on 25 plaintiffs motions to convene a three-judge court. Persuaded that the state had not 26 39 Several members of the Expert Panel appointed by defendants, including James Austin, Ph.D., Jeffrey Beard, Ph.D., Joseph Lehman, and Barry Krisberg, Ph.D., testified for 27 plaintiffs at the trial of this matter. Another member of the Expert Panel, James Gomez, was the Director of the California Department of Corrections from 1991 to 1996, during the 28 merits phase of the Coleman action. 45 1 adequately addressed its prison overcrowding crisis so as to make possible the remedying of 2 the constitutional violations, and that consideration of a population reduction order was 3 necessary in order to achieve that objective in both cases, both courts granted plaintiffs 4 motions. July 23, 2007 Order in Plata, 2007 WL 2122657; July 23, 2007 Order in Coleman, 5 2007 WL 2122636. 6 The Plata court found that although the Receiver has made much progress since his 7 appointment, the establishment of the Plata Receivership did not require the court to wait 8 more time, potentially years, to see whether the Receiver s plans will succeed or fail. 9 July 23, 2007 Order in Plata, 2007 WL 2122657, at *3. It found that the unconstitutional 10 conditions that led to the Receiver s appointment continued to exist. The Plata court 11 explained: 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Had the Receiver reported to the Court that he did not view overcrowding to be a substantial impediment to implementing the reforms required in this case, the Court may well have reached a different conclusion regarding the appropriateness of convening a three-judge court to consider a prisoner release order. However, quite to the contrary, the Receiver s reports indicate that overcrowding is a serious problem that impacts, for example, his ability to develop adequate reception centers and health facilities because of the high numbers of inmate transfers and the inadequate amount of available health care beds and other physical space. Receiver s Report Re: Overcrowding [Ex. D1092] at 26-28. Overcrowding also negatively impacts the Receiver s ability to hire and retain competent medical and managerial staff. Id. at 24-26. Beyond that, the Receiver reports that: Every element of the Plan of Action faces crowding related obstacles. Furthermore, overcrowding does not only adversely impact the Receiver s substantive plans, it also adversely impacts on the very process of implementing remedies because overcrowding, and the resulting day to day operational chaos of the CDCR, creates regular crisis situations which call for action on the part of the Receivership and take time, energy, and person power away from important remedial programs. 26 Id. at 28-29. . . . 27 Tellingly, the Receiver s concerns about the impacts of overcrowding on his ability to reform the medical health care 28 46 1 2 3 4 5 6 7 8 9 10 11 12 13 delivery system became even stronger in the weeks following his initial report. In his supplemental report, filed just four weeks after his initial report, the Receiver concluded that: Mission changes, yard flips, and prison-to-prison transfers, aggravated by the limited alternatives imposed by overcrowding, are now assuming a size, scope and frequency that will clearly extend the timeframes and costs of the receivership and may render adequate medical care impossible, especially for patients who require longer term chronic care. Receiver s Suppl. Report Re: Overcrowding [Ex. D1094] at 10 (emphases added). While the Court appreciates Defendants statements that greater coordination between the State and the Receiver will alleviate some of the Receiver s concerns, such sentiments only underscore the Receiver s expressed concerns that overcrowding presents serious problems not only because of the substantive ways in which it interferes with delivery of medical care, but also because of the amount of time and attention the Receiver must devote to dealing with crowding-related issues. It is clear to the Court that the crowded conditions of California s prisons, which are now packed well beyond their intended capacity, are having and in the absence of any intervening remedial action, will continue to have a serious impact on the Receiver s ability to complete the job for which he was appointed: namely, to eliminate the unconstitutional conditions surrounding delivery of inmate medical health care. 14 Id. at *4 (last emphasis added). 15 The Coleman court found that between 1997 and 2005, defendants had made slow 16 but evident progress toward constitutional compliance, but that, [i]n spite of the 17 commendable progress . . ., defendants mental health care delivery system has not come into 18 compliance with the Eighth Amendment at any point since this action began. July 23, 2007 19 Order in Coleman, 2007 WL 2122636, at *3. The Coleman court further found that: 20 21 22 23 Several prisons remain notable exceptions to the progress made at others, and delays in access to care at the highest level of need mental health crisis beds, acute inpatient care, and intermediate inpatient care have plagued the CDCR throughout the course of this litigation. Moreover, defendants efforts at long-range planning for the delivery of mental health care continues to be hampered by inadequacies in the capture and collection of data and the use of outdated methodologies to interpret that data. 24 25 26 27 . . . [O]n May 31, 2007, the Special Master reported that programming space, beds for mentally ill inmates, and staffing levels have all been impacted seriously by overcrowding. Special Master s Response to Court s May 17, 2007 Request for Information, filed May 31, 2007, at 4-14 ( Special Master s May 31, 2007 Response ). The staffing shortages alone mean that the CDCR only has enough staff to provide full mental 28 47 1 health services to roughly two-thirds of its mental health caseload, or two-thirds of required services to its full caseload, or, probably more realistically, some combination of reduced services to some segments of the caseload that can be covered with a third less clinicians than required. Id. at 11-12. While acknowledging the difficulties in quantifying precisely the scope of the unmet mental health needs, the Special Master reports that, 2 3 4 5 defendants cannot meet at least a substantial portion, amounting in some loose amalgam to about 33 percent, of acknowledged mental health needs with current staffing resources. Insufficient intensive mental health treatment beds and a chronic lack of programming space for mental health treatment contribute further to defendants inability to meet required mental health services. All three deficiencies are unquestionably exacerbated by overcrowding. 6 7 8 9 10 Id. at 14. With a mental health caseload of almost 33,000 inmates, id. at 2, this level of unmet needs is unconscionable. 11 12 Id. at *3-*4 (footnote omitted). 13 In their orders granting plaintiffs motions, the Plata and Coleman courts 14 recommended that the cases be assigned to the same three-judge court [f]or purposes of 15 judicial economy and avoiding the risk of inconsistent judgments. July 23, 2007 Order in 16 Plata, 2007 WL 2122657, at *6; see also July 23, 2007 Order in Coleman, 2007 WL 17 2122636, at *8. The Chief Judge of the United States Court of Appeals for the Ninth Circuit 18 agreed and, on July 26, 2007, convened the instant three-judge district court pursuant to 28 19 U.S.C. § 2284. 20 F. 21 In August and September 2007, this court granted motions to intervene on behalf of Proceedings Before this Three-Judge Court 22 defendants filed by groups of district attorneys; sheriffs, police chiefs, and probation officers 23 (collectively law enforcement intervenors ); counties; and Republican state Senators and 24 Republican Assembly Members. We note that the Republican state Senators and Republican 25 Assembly Members constitute just over a third of the membership of each respective body. 26 We also granted the CCPOA s motion to intervene on behalf of plaintiffs. 27 On November 1, 2007, we appointed a settlement referee, former state Court of 28 Appeal Justice Elwood Lui, and a settlement consultant, current state Court of Appeal Justice 48 1 and former Legal Affairs Secretary to Governor Schwarzenegger, Peter Siggins, to aid the 2 parties and intervenors in settlement discussions. Nov. 1, 2007 Order at 1-2. However, the 3 settlement efforts ultimately proved unsuccessful, as Justices Lui and Siggins reported to the 4 three-judge court on June 25, 2008. 5 On September 15, 2008, defendants filed a motion for summary judgment, which we 6 denied by written order on November 3, 2008. Trial commenced on November 18, 2008, 7 and concluded on December 19, 2008, after fourteen court days in which we heard testimony 8 from nearly fifty witnesses, received written testimony from several additional witnesses, and 9 received hundreds of exhibits into evidence. Following the close of evidence, we received 10 proposed findings of fact and conclusions of law from the parties and intervenors and heard 11 final argument on February 3 and 4, 2009. 12 To assist the parties in planning their further actions, we issued a tentative ruling on 13 February 9, 2009, explaining that plaintiffs had met their burden of proof and that a 14 population reduction order was necessary to remedy the constitutional violations concerning 15 the provision of medical and mental health care in California s prisons. We even gave the 16 state an indication of the range within which the population cap would fall. In our tentative 17 ruling, we once again asked whether a court-appointed settlement referee would be of 18 assistance. Plaintiffs and intervenors expressed a willingness to engage in further settlement 19 discussions, but the state defendants responded that they did not believe such efforts would 20 be fruitful. 21 After carefully reviewing all of the evidence and oral and written arguments presented 22 in this proceeding, we make the following findings of fact and conclusions of law and issue 23 the following order. This opinion and order supersedes the tentative ruling in its entirety. 24 25 III. LEGAL FRAMEWORK 26 Federal courts have long recognized that population reduction orders may sometimes 27 be necessary to ensure constitutional prison conditions. For example, in Duran v. Elrod, 713 28 F.2d 292 (7th Cir. 1983), the Seventh Circuit upheld a district court s order requiring a 49 1 reduction in the population of the Cook County Department of Corrections, finding that the 2 order was sensitive to[] . . . the principles of federalism, id. at 297, and that the district 3 court acted fairly and reasonably to ease a critical problem of overcrowding in the face of 4 substantial noncompliance by Cook County, id. at 298. Likewise, in Newman v. Alabama, 5 683 F.2d 1312 (11th Cir. 1982), the Eleventh Circuit found that, where Alabama s county 6 jails were unconstitutionally overcrowded, a cap on the state inmate population in the county 7 jails represent[ed] the proper balance between the duty of the district court to remedy 8 constitutional violations and the right of the State to administer its prison and parole 9 systems, id. at 1321. There are other examples as well, including a continuing cap on Los 10 Angeles County s jail population stipulated to by the parties in Rutherford v. Pitchess, 11 No. CV 75-4111 (C.D. Cal.). 12 Until 1996, federal courts relied upon general principles of equitable relief and 13 federalism in deciding whether to enter a population reduction order to remedy constitutional 14 violations. However, in 1996 Congress enacted the Prison Litigation Reform Act ( PLRA ), 15 Pub. L. No. 104-134, 110 Stat. 1321 (codified in relevant part at 18 U.S.C. § 3626). The 16 PLRA established a comprehensive set of [statutory] standards to govern prospective relief 17 in prison conditions cases. Gilmore v. California, 220 F.3d 987, 998 (9th Cir. 2000). 18 Because there is no dispute that both the Plata and Coleman lawsuits are civil action[s] with 19 respect to prison conditions, the matter before us is governed by the statutory requirements 20 of the PLRA. 18 U.S.C. § 3626(a)(1). 21 The PLRA contains two sets of requirements that are relevant here: one applicable to 22 all forms of prospective relief in federal prison conditions lawsuits, see id., and another 23 applicable only to prisoner release orders, see id. § 3626(a)(3). The PLRA defines a 24 prisoner release order as any order . . . that has the purpose or effect of reducing or 25 limiting the prison population, or that directs the release from or nonadmission of prisoners 26 to a prison. Id. § 3626(g)(4). Under this definition, a prisoner release order includes not 27 only an order requiring the release of presently incarcerated inmates, but also an order 28 requiring the diversion of convicted persons from prison, changing the treatment of parole 50 1 violators in order to prevent their return to overcrowded prisons, or imposing a cap on the 2 prison population or any part of it. See, e.g., Tyler v. Murphy, 135 F.3d 594, 595-96 (8th Cir. 3 1998) (finding a cap on the number of technical probation violators who could be admitted to 4 a particular facility to be a prisoner release order ). There is no dispute that the population 5 reduction order requested by the plaintiffs falls within the PLRA s definition of prisoner 6 release order because the order would have the purpose of limiting the prison 7 population. 18 U.S.C. § 3626(g)(4). Accordingly, this court can grant the plaintiffs 8 request for a population reduction order only if the proposed order meets both the PLRA s 9 specific standard for prisoner release orders and its general standard for prospective relief in 10 prison conditions cases. 11 12 13 A. The PLRA Standard for Prisoner Release Orders: Primary Cause and No Other Relief The PLRA does not prohibit courts from entering an order requiring a reduction in the 14 population of a prison or prison system. To the contrary, in enacting the PLRA, Congress 15 was clear to state that a court still retains the power to order [a population reduction order] 16 when such an order is truly necessary to prevent an actual violation of a prisoner s federal 17 rights. H.R. Rep. No. 104-21, at 25 (1995); cf. 141 Cong. Rec. S14419 (daily ed. Sept. 27, 18 1995) (statement of Sen. Abraham) (noting that the PLRA permits narrowly tailored 19 order[s] to correct constitutional violations and that the PLRA allows the courts to step in 20 where they are needed ).40 Rather than barring prisoner release orders altogether, the 21 PLRA simply makes such orders, including population caps and other population reduction 22 orders, the remedy of last resort. H.R. Rep. No. 104-21, at 25. 23 It does so by imposing a number of restrictions on the entry of prisoner release orders. 24 First, a court considering such an order must find that a court has previously entered an 25 40 In fact, a number of courts have entered consent decrees for prisoner release since 26 the enactment of the PLRA decrees that must meet the same set of requirements as any order entered by a court. See 18 U.S.C. §3626(c)(1); Roberts v. Mahoning County, 495 27 F. Supp. 2d 719 (N.D. Ohio 2007); John Boston, The Prison Litigation Reform Act, 67 Brook. L. Rev. 429, 446 n.67 (2001) (collecting orders for overcrowding relief entered by 28 consent decree after the enactment of the PLRA). 51 1 order for less intrusive relief that has failed to remedy the deprivation of the Federal right 2 sought to be remedied through the prisoner release order, and that the defendant has had a 3 reasonable amount of time to comply with the previous court orders. 18 U.S.C. 4 § 3626(a)(3)(A). If both of these requirements are met, the court must request that a three5 judge district court be convened to consider the propriety of the proposed order. Id. 6 § 3626(a)(3)(B). Finally, the three-judge court must find by clear and convincing evidence 7 (1) that crowding is the primary cause of the violation of the Federal right, and (2) that no 8 other relief will remedy the violation of the Federal right. Id. § 3626(a)(3)(E). 9 Before convening the present three-judge court, the Plata and Coleman courts found 10 that their prior orders for less intrusive relief had failed to remedy the unconstitutional denial 11 of adequate medical and mental health care to prisoners in California s prisons, and that the 12 defendants have had a more than reasonable amount of time to comply with those prior 13 orders. See July 23, 2007 Order in Plata, 2007 WL 2122657, at *3; July 23, 2007 Order in 14 Coleman, 2007 WL 2122636, at *2. Accordingly, the findings required by § 3626(a)(3)(A) 15 have been made. The procedural history described above clearly establishes that the Plata 16 and Coleman courts have previously entered orders for less intrusive relief that have failed to 17 remedy the constitutional deprivations at issue in each case despite the reasonable time given 18 to defendants to comply with those orders. In this opinion and order, we primarily consider 19 the requirements of § 3626(a)(3)(E) whether crowding is the primary cause of the 20 unconstitutional denial of adequate medical and mental health care to California s prisoners, 21 see infra Section IV, and whether any other form of relief could remedy those constitutional 22 violations, see infra Section V. 23 24 25 B. The PLRA Standard for All Prospective Relief: Need-NarrownessIntrusiveness and Consideration of Public Safety In addition to these specific limitations on the entrance of prisoner release orders, the 26 PLRA establishes a standard applicable to all forms of prospective relief in prison conditions 27 lawsuits. First, the PLRA requires that such relief [be] narrowly drawn, extend[] no further 28 than necessary to correct the violation of the Federal right, and [be] the least intrusive means 52 1 necessary to correct the violation of the Federal right. 18 U.S.C. § 3626(a)(1)(A). Rather 2 than imposing any new limitations on federal authority, this provision codifies the common3 law standard for injunctive relief, generally referred to as the need-narrowness4 intrusiveness standard. See H.R. Rep. 104-21, at 24 n.2 (1995) (explaining that the dictates 5 of [18 U.S.C. §3626(a)(1)] are not a departure from current jurisprudence concerning 6 injunctive relief ); see also Armstrong v. Davis, 275 F.3d 849, 872 (9th Cir. 2001); Smith v. 7 Ark. Dep t of Corr., 103 F.3d 637, 647 (8th Cir. 1996); Williams v. Edwards, 87 F.3d 126, 8 133 n.21 (5th Cir. 1996).41 Likewise, the PLRA requires that any prospective relief extend 9 no further than necessary to correct the violation of the Federal right of a particular plaintiff 10 or plaintiffs. 18 U.S.C. § 3626(a)(1)(A); cf. Lewis v. Casey, 518 U.S. 343, 357-360 (1996) 11 (holding that the remedy in a prison conditions case must remedy actual injuries that have 12 been identified by the court and suffered by the plaintiffs). In class action lawsuits such as 13 Plata and Coleman, the PLRA requires that the remedy be tailored to the actual injuries 14 suffered by class members. See Armstrong, 275 F.3d at 870-73. 15 Second, the PLRA requires that any court considering the entry of prospective relief 16 give substantial weight to any adverse impact the order might have on public safety or the 17 operation of the criminal justice system. 18 U.S.C. § 3626(a)(1)(A); see also H.R. Rep. No. 18 104-21, at 24 (1995) (stating that courts must give appropriate consideration to any 19 potential impact on public safety or the criminal justice system ). This requirement codifies 20 the longstanding common law requirement that federal courts pay particular regard for [] 21 public consequences in employing the extraordinary remedy of injunction. Weinberger v. 22 Romero-Barcelo, 456 U.S. 305, 312 (1982); see also Yakus v. United States, 321 U.S. 414, 23 440 (1944). 24 25 41 In lieu of changing the general standard for prospective relief in prison conditions 26 cases, the PLRA limits federal court authority in matters relating to prison conditions primarily by applying the need-narrowness-intrusiveness standard to consent decrees as 27 well as court orders, 18 U.S.C. § 3626(c)(1); by making it easier to terminate existing court orders or consent decrees, 18 U.S.C. § 3626(b); and by imposing distinct limitations on 28 prisoner release orders, 18 U.S.C. § 3626(a)(3). 53 1 We address the need-narrowness-intrusiveness standard in Section VI and consider 2 the impact of the order we adopt on public safety and the operation of the criminal justice 3 system in Section VII. 4 C. 5 The question before this three-judge court is whether the remedy requested by the The Remedial Nature of the Three-Judge Court Proceeding 6 plaintiffs is proper as a matter of federal law. The Plata and Coleman courts years ago 7 identified the constitutional deficiencies underlying this proceeding. Since that time, both 8 cases have been in their remedial phase. After prior remedial efforts failed, the Plata and 9 Coleman courts both faced the question whether an order requiring a reduction in the 10 population of California s prisons was necessary to remedy the previously identified 11 constitutional violations, and both concluded that such an order should be considered by a 12 three-judge court. 13 We need not yet again evaluate the state s continuing constitutional violations. In 14 requesting that this three-judge court be convened, the Plata and Coleman courts both found, 15 without objection from defendants, that the constitutional violations were ongoing. See 16 July 23, 2007 Order in Plata, 2007 WL 2122657, at *3; July 23, 2007 Order in Coleman, 17 2007 WL 2122636, at *4. That is sufficient under the PLRA. In addition, defendants have 18 never filed a motion to terminate under § 3626(b), the proper means for any challenge to the 19 existence of current and ongoing constitutional violations relating to the provision of 20 medical and mental health care in the California prisons. Moreover, even if we were 21 required to find independently that the requirements of § 3626(a)(3)(A) including its 22 requirement that prior orders have failed to remedy the deprivation of the Federal right 23 have been met, we did so in denying defendants motion for summary judgment, Nov. 3, 24 2008 Order at 6-7. Accordingly, the question we must answer in this opinion and order is 25 entirely remedial, i.e., whether the plaintiffs proposed remedy meets the imposing standards 26 established by the PLRA.42 27 42 Because this proceeding deals only with the plaintiffs requested remedy, we did not permit the introduction of evidence relevant only to determining whether the constitutional 28 violations found by the Plata and Coleman courts were current and ongoing. 54 1 IV. CROWDING AS PRIMARY CAUSE 2 The extent of overcrowding in the California prison system, approximately 190% of 3 systemwide design capacity, is extraordinary and almost unheard of. Rep. Tr. at 4 297:1-17, 298:19-20 (Haney). The problem is widespread and not restricted to just a few 5 institutions. It s occurred throughout the system. Id. at 297:23-25. There would seem to be 6 no dispute about the egregious nature of the overcrowding in this case. Under the PLRA, 7 however, the question is whether clear and convincing evidence establishes that the 8 overcrowding is the primary cause of the unconstitutional denial of adequate medical and 9 mental health care to California s prisoners. 18 U.S.C. § 3626(a)(3)(E)(i). Only if it is may 10 the court a three-judge court enter a population reduction order. Defendants do not 11 contest that prison crowding impedes the delivery of constitutionally adequate medical and 12 mental health care in the California prison system. They claim only that crowding is not the 13 primary cause of the violations of plaintiffs constitutional rights. E.g., Rep. Tr. at 2953:6-11 14 (closing argument by defendants counsel). 15 We accept defendants proposed definition of primary cause as the cause that is 16 first or highest in rank or importance; chief; principal. Random House Webster s 17 Unabridged Dictionary 1537 (2d ed. 1998) (defining primary ).43 We note, however, that 18 the PLRA does not require that crowding be the only cause of the constitutional violations at 19 issue. Probably it cannot be said of any event that it has a single causal antecedent; usually 20 there are many. 4 Harper, James and Gray on Torts § 20.2 (3d ed. 2007). The PLRA s 21 primary cause standard incorporates this basic aspect of causation. By requiring only that 22 crowding be the primary cause of the constitutional violations at issue, the PLRA s language 23 explicitly contemplates that secondary causes may exist. Had Congress intended to require 24 that crowding be the only cause, it would have used language to that effect for example, 25 exclusive or only instead of primary. 26 27 43 Unfortunately, the legislative history of the PLRA is sparse and provides no 28 meaningful insight into the meaning of primary cause or crowding. 55 1 As all of the parties to this proceeding have recognized, in the context of prison 2 conditions litigation crowding refers to the presence in a facility or prison system of a 3 prisoner population exceeding that facility or system s capacity. See, e.g., Doty v. County of 4 Lassen, 37 F.3d 540, 543 (9th Cir. 1994) (finding overcrowding where a jail s actual 5 population exceeded its design capacity by an average of approximately fifty percent); 6 Hoptowit v. Ray, 682 F.2d 1237, 1248-49 (9th Cir. 1982) (finding a penitentiary 7 overcrowded where its population exceeded its design capacity); see also Lareau v. Manson, 8 651 F.2d 96, 99-100 (2d Cir. 1981); cf. Random House Webster s Unabridged Dictionary 9 482 (2d ed. 1998) (defining crowded as filled to excess ). In other contexts, the term 10 overcrowding would ordinarily be used. Here, the words crowding and overcrowding have 11 the same meaning, and we use them interchangeably. 12 A prison system s capacity is not defined by square footage alone; it is also 13 determined by the system s resources and its ability to provide inmates with essential 14 services such as food, air, and temperature and noise control. Following the parties lead, we 15 will discuss the capacity of the California prison system primarily in terms of design 16 capacity. As the Corrections Independent Review Panel explained, design capacity 17 designate[s] the number of inmates a prison is designed to accommodate according to 18 standards developed by the Commission on Accreditation and the American Correctional 19 Association. Ex. P4 at 123. These standards take into account the need for humane 20 conditions, as well as the need to prevent violence and move inmates to and from programs, 21 such as mental health care, education classes, and drug abuse treatment. Id. 22 Taking into account the meaning of primary cause and the criteria governing 23 crowding, we must determine whether the presence in California s prison system of a 24 prison population almost double the system s design capacity is the principal cause of the 25 failure to provide constitutionally adequate medical and mental health care to the members of 26 the Plata and Coleman classes. 27 As we discuss below, the evidence presented at trial, including testimony from 28 defendants experts, admissions by defendants and their agents, and data maintained by 56 1 defendants, overwhelmingly establishes not only that crowding adversely affects every 2 aspect of prison administration, forcing a constant state of crisis management, but also that 3 crowding creates numerous barriers to the provision of medical and mental health care that 4 result in the constitutional violations we consider here. These barriers include severe space 5 and other shortages that prevent inmates from receiving the care they require. Crowding also 6 renders the state incapable of maintaining an adequate staff and an adequate medical records 7 system. In addition, crowding causes prisons to rely on lockdowns, which further restrict 8 inmates access to care, and it forces prisons to house inmates in non-traditional settings, 9 such as triple-bunks in gyms and dayrooms not designed for housing, that contribute to the 10 lack of care and the spread of infectious disease and that increase the incidence and severity 11 of mental illness among prisoners. 12 Multiple experts testified that crowding is the primary cause of the constitutional 13 violations at issue in Plata and Coleman. Most impressive, four current or former prison 14 administrators so testified. These four correctional experts had, collectively, administered 15 the correctional systems of five different states, including California.44 Three had never 16 before testified on behalf of a prisoner, and two were not paid for their time as experts. A 17 number of medical and mental health experts also testified that crowding is the primary cause 18 of the constitutional violations, and even defendants own mental health expert testified that 19 crowding is the primary cause of defendants inability to provide adequate care to the 20 Coleman class at reception centers. Dec. 10, 2007 Packer Report at 20. As the Secretary of 21 the Pennsylvania Department of Corrections testified, the biggest inhibiting factor right now 22 in California being able to deliver appropriate mental health and medical care is the severe 23 overcrowding of [the] system. Rep. Tr. at 219:7-10 (Beard). We agree. For the reasons we 24 discuss below, we conclude that clear and convincing evidence establishes that crowding is 25 26 44 The experts included Jeanne Woodford, former warden at San Quentin and former acting Secretary of the CDCR; Doyle Wayne Scott, former Executive Director of the Texas 27 Department of Criminal Justice; Joseph Lehman, former head of corrections in Pennsylvania, Washington, and Maine; and Jeffrey Beard, current Secretary of the Pennsylvania 28 Department of Corrections. 57 1 the primary cause of the unconstitutional denial of medical and mental health care to 2 California s prisoners. 3 4 5 A. General Problems in the Delivery of Medical and Mental Health Care Caused by Crowding Correctional experts agree that crowding affects virtually every aspect of a prison s 6 operation. Aug. 15, 2008 Lehman Report ¶ 10 (expert report from former head of 7 corrections in Pennsylvania, Washington, and Maine). Jeanne Woodford, the former head of 8 corrections in California, testified that, under crowded conditions, there are simply too 9 many issues that arise from such a large number of prisoners and staff. One result of this is 10 that management spends virtually all of its time fighting fires instead of engaging in 11 thoughtful decision-making and planning. This results in short-sighted decisions that create 12 even more crises. Nov. 9, 2007 Woodford Report ¶ 12. Doyle Wayne Scott, a thirty-year 13 employee of the Texas Department of Criminal Justice who served as its Executive Director 14 for five years, explained: 15 16 17 18 19 20 21 22 23 24 25 26 Overcrowding has burdened CDCR s inadequate management systems that underlie health care delivery. The excessive population leads to management failures in two ways. First, overcrowding engenders a state of perpetual crisis that causes management failures. Administrators spend their time doing damage control, rather than making sure the prison is operating properly and prisoners are getting the services that they need. . . . A population of 7,000 or more, as is found in some California prisons, is not manageable at all. The sheer size and complexities of managing a prison that size would be overwhelming for one manager especially with the limited resources in the areas of staffing and inadequate space for services to the offenders that I observed at all of the prisons I toured in California. One warden simply cannot know what he/she needs to know on a daily basis to make good informed management decisions. Second, overcrowding overwhelms management infrastructure. As I have read in numerous reports of the Receiver, the CDCR lacks the management information systems needed to adequately organize and track prisoner transfers for specialized medical and mental health care and public health related needs (for example, people with compromised immune systems not going to Valley Fever risk areas) in the severely overcrowded conditions. 27 Nov. 9, 2007 Scott Report ¶¶ 1, 76-77. Secretary Woodford concluded that crowding makes 28 it virtually impossible for the organization to develop, much less implement, a plan to 58 1 provide prisoners with adequate care ; [i]n [her] opinion, it is all but impossible to safely 2 and humanely incarcerate this many prisoners within the existing facilities. Nov. 9, 2007 3 Woodford Report ¶¶ 10, 12. 4 As put in the most simple terms by Secretary Woodford, who recently administered 5 the California prison system and who shortly before that was the warden at San Quentin, 6 [o]vercrowding in the CDCR is extreme, its effects are pervasive and it is preventing the 7 Department from providing adequate mental and medical health care to prisoners. Aug. 15, 8 2008 Woodford Supp. Report ¶ 31. While defendants dispute that crowding is the primary 9 cause of the ongoing constitutional violations in Plata and Coleman, they do not dispute that 10 crowding makes the delivery of adequate medical and mental health care in the California 11 prison system extremely difficult. Matthew Cate, the current head of the CDCR and a 12 defendant in this proceeding, stated that overpopulation makes everything we do more 13 difficult, Rep. Tr. at 1683:19-20, and further agreed that crowding continues to severely 14 hamper[] the Department s ability to provide inmates with adequate medical care in a 15 fiscally sound manner, id. at 1683:3-19 (testimony that statements in the Office of the 16 Inspector General s 2006 audit of the CDCR, issued when Cate was the Inspector General, 17 continue to be true today); Ex. P46 at ES-1 (April 2006 Office of the Inspector General 18 Accountability Audit, Review of Audits of the California Department of Corrections and 19 Rehabilitation Adult Operations and Adult Programs, 2000-2004). James Tilton, Cate s 20 predecessor as Secretary of the CDCR, likewise explained that it was clear to him that 21 crowding, and the resulting lack of space, adversely affected the delivery of medical and 22 mental health care. Sept. 3, 2008 Tilton Dep. at 80:5-25. Similarly, John Dovey, a former 23 CDCR official, testified before a state Senate committee in August 2006 that the risk of 24 catastrophic failure in a system strained from severe overcrowding is a constant threat. As 25 the Director of the Division of Adult Institutions [for the CDCR], it is my professional 26 opinion this level of overcrowding is unsafe and we are operating on borrowed time. 27 Ex. P72 at 15 (Aug. 15, 2006 CDCR Presentation to Senate Select Committee on Prison 28 Population Management and Capacity). Before this court, Robin Dezember, then the Chief 59 1 Deputy Secretary of the Correctional Healthcare Services Division of the CDCR, stated his 2 belief that we are terribly overcrowded in our prison system, and that crowding adversely 3 affects the delivery of mental health care services. Rep. Tr. at 853:13-15, 21-24. Even 4 defendants expert Dr. Ira Packer opined that the overcrowding in CDCR significantly 5 contributes to the difficulties in providing adequate mental health services. Dec. 10, 2007 6 Packer Report at 8. 7 B. 8 The evidence before us demonstrates that crowding causes a number of specific Space Issues Affecting the Delivery of Care 9 problems central to the ongoing violation of California inmates constitutional right to 10 adequate medical and mental health care. One of the clearest effects of crowding is that the 11 current prison system lacks the physical space necessary to deliver minimally adequate care 12 to inmates. This manifests itself in a variety of areas, each of which we discuss below. 13 14 1. Reception Centers The medical and mental-health related problems caused by crowding are immediately 15 apparent at the state s reception centers. Each year, California admits approximately 140,000 16 inmates into the state prison system. Rep. Tr. at 224:17-18 (Beard); see also Aug. 15, 2008 17 Austin Report ¶¶ 45-46 & Table 3; Ex. P18 at 3; Ex. P19 at 2; Ex. P75 at 3.45 The CDCR s 18 reception centers are the locus of the intake and classification functions for all of these 19 inmates. See Nov. 9, 2007 Austin Report ¶¶ 27-28; Aug. 15, 2008 Austin Report ¶ 97. The 20 CDCR has reception centers at twelve prisons, nine at male institutions and one at each of the 21 state s three female prison institutions. Ex. P135 at 3-4. As of August 2008, all but one of 22 these reception centers were near or over 200% design capacity, and two were over 300% 23 design capacity. Id. This severe crowding at the reception centers makes it impossible to 24 provide adequate medical and mental health services to inmates entering the California 25 45 As Dr. Austin explained, [t]here are two major types of prison admissions for the 26 CDCR new court commitments and parole violators. Aug. 15, 2008 Austin Report ¶ 45. At the cited pages, Exhibits P18, P19, and P75 set forth data on new admissions, felon parole 27 violators returned with a new term, and felon parole violators returned to custody during periods specified in each exhibit. The aggregation of these three numbers in each of the 28 exhibits ranges from 141,288 (Ex. P18) to 139,399 (Ex. P19) to 140,560 (Ex. P75). 60 1 prison system. In addition, severe crowding throughout the system forces prisons to house 2 inmates in these reception centers without adequate care for extended periods of time. 3 As the Plata Receiver explained, reception centers must be staffed and have the 4 appropriate clinical space to provide a level of medical care and clinical evaluations above 5 that of the general population institutions. Ex. D1092 at 19 (Plata Receiver s May 15, 2007 6 Report Re: Overcrowding).46 Each time an individual is admitted to the CDCR, whether for 7 the first time, by re-offending, or on a parole violation: 8 he or she returns to the CDCR through a reception center where a medical/mental health/dental health care appraisal must be performed. Once that appraisal is completed the newly received prisoner is transferred to an open bed at a prison which has been designated for his or her classification. However, none of the CDCR s designated reception centers were designed or constructed with adequate clinical space [to perform these functions]. 9 10 11 12 13 Id.; see also, e.g., Aug. 15, 2008 Haney Report ¶¶ 246-247 (describing space shortages at the 14 California Correctional Institution reception center, including a holding room containing 15 three holding cells that were originally intended to hold disciplinary cases but now have 16 been converted to mental health use ). To make matters worse, as the original prisons 17 designated for reception became overwhelmed by the influx of parole violators, the CDCR 18 was forced to convert general population prisons into reception centers. These 19 conversions, however, were not accompanied by adequate additions to clinical staff or 20 clinical space. Ex. D1092 at 19; see also, e.g., Nov. 9, 2007 Stewart Report ¶ 48 (because 21 reception center at DVI was not designed as a reception center, it has been difficult to find 22 space for various reception center functions ). 23 Without sufficient space, reception centers are unable to screen or treat inmates 24 adequately. For instance, as plaintiffs medical expert Dr. Ronald Shansky explained, the 25 number of prisoners who must be processed at the reception center at CIM exceeds the 26 number of patients that can be adequately treated, thereby forcing the prison to squeeze[] 27 too many prisoners and too many providers into the available treatment area. Nov. 9, 2007 28 46 Plaintiffs also offered this document into evidence as Exhibit P26. 61 1 Shansky Report ¶ 12. Exams are conducted in areas separated only by a thin white fabric 2 folding screen that is approximately five to six feet tall and conversations between 3 physicians and inmates can be overheard on the other side of the screen. Id. Similarly, at 4 North Kern State Prison, inmate health interviews are conducted in a small office, with 5 prisoners sitting back to back, separated only by a shoulder-high divider. Sept. 10, 2008 6 Shansky 2d Supp. Report ¶ 23. Such conditions do not allow for appropriate confidentiality, 7 causing prisoners to be less likely to provide accurate information about sensitive medical 8 and psychiatric conditions. Id. ¶ 24; see also Aug. 15, 2008 Haney Report ¶ 280 9 (psychiatrist and psychologist who work in the North Kern reception center must share a 10 converted cell that serves as their office ); Nov. 9, 2007 Stewart Report ¶ 48 (describing 11 observations of small classroom at DVI reception center where at any time six psychologists 12 simultaneously conduct reception center mental health assessments for new arrivals ); id. 13 ¶ 82 (noting that the [l]ack of adequate and appropriate space for reception center 14 psychological screening was also apparent at DVI ). 15 Moreover, at North Kern, follow-up physical examinations are conducted in rooms 16 that are so small that it would be very difficult if not impossible to perform an actual 17 physical examination in them, so that the exams that take place are in fact simply medical 18 interviews, primarily for the purpose of determining what type of housing is appropriate for 19 the prisoner. Sept. 10, 2008 Shansky 2d Supp. Report ¶ 25. This violates the basic 20 principle that incoming prisoners must undergo a comprehensive exam upon arrival so that 21 an adequate treatment plan may be developed and implemented. A physical exam, as 22 opposed to a medical interview, is necessary because some conditions can be identified and 23 confirmed only through physical examination of the patient. Id. ¶ 26; see also Rep. Tr. at 24 224:10-225:15 (Beard) (testifying that the number of people coming in through reception 25 centers may cause prisons to miss people who have certain needs and certain care needs that 26 aren t being dealt with ). The medical facilities at North Kern are also so inadequate that 27 defendants cannot comply with the Plata policies and procedures they agreed to implement, 28 62 1 which provide for a complete history and physical examination of inmates within fourteen 2 days of arrival at a reception center. Sept. 10, 2008 Shansky 2d Supp. Report ¶ 22. 3 The consequences of the state s inability to screen inmates properly at the reception 4 centers are obvious: If an inmate s health needs are not identified, they cannot be treated. In 5 addition, inmates whose needs are not identified may be placed in a setting that will 6 exacerbate existing but unidentified health problems. Likewise, if the lack of confidentiality 7 in the screening centers prevents inmates from reporting infectious diseases, the failure to 8 diagnose them at the reception center may result in their being spread throughout the prison 9 population. 10 In addition to preventing the reception centers from properly screening newly 11 admitted inmates, crowding at the reception centers prevents the provision of adequate care 12 to the inmates housed there. As numerous experts, including defendants own mental health 13 expert, testified, the number and types of inmates in the centers overwhelms their capacity to 14 provide adequate medical or mental health care services. Rep. Tr. at 1121:16-19 (Packer); 15 Rep. Tr. at 368:12-369:4 (Woodford) (mentally ill inmates did not receive really any 16 treatment at San Quentin reception center beyond identification as CCCMS or EOP and 17 certainly not anything to prevent further deterioration of people s mental illness ); Aug. 15, 18 2008 Stewart Supp. Report ¶ 136 ( CDCR s Reception Centers are dangerously overcrowded 19 and do not and cannot provide appropriate mental health care for anyone ); Sept. 10, 2008 20 Shansky 2d Supp. Report ¶ 79 ( acute staffing shortage, coupled with the lack of clinical 21 exam space prevents North Kern State Prison from providing incoming inmates with 22 comprehensive physical examination or follow-up appointments with primary care 23 providers); Rep Tr. at 368:12-22 (Woodford) (at San Quentin reception center, due to 24 vacancies and just the sheer numbers and lack of space, medical staff were unable to keep 25 up with physicals or providing any kind of chronic care follow-up ). 26 If California s inmates spent only a brief time at the reception centers before being 27 placed in other facilities, the centers inability to provide adequate medical and mental health 28 care to the inmates housed there would constitute a less substantial aspect of the 63 1 constitutional violations at issue in this proceeding. However, inmates in California are 2 tend[ing] to spend significantly longer periods of time in reception centers. Nov. 9, 2007 3 Stewart Report ¶ 24; see also Nov. 9, 2007 Austin Report ¶ 27 (reporting that the CDCR 4 routinely fails to meet its mandate to transfer inmates from reception centers to mainline 5 institutions within sixty days). As one of plaintiffs experts Dr. James Gilligan explained, 6 The dramatic levels of prison overcrowding through the state mean that individuals coming 7 into prison are housed in Reception Centers for extended periods of time, far longer than 8 intended. Aug. 15, 2008 Gilligan Report ¶ 26 (footnotes omitted). 9 The consequences of the increased lengths of stay at the reception centers along with 10 the lack of space in those centers are particularly grave for Coleman class members. 11 Dr. Packer, defendants mental health expert, reported that mentally ill individuals often 12 enter[] the prison system with a more acute mental health presentation, not having received 13 adequate treatment in the community and/or having abused substances there. Dec. 10, 2007 14 Packer Report at 20. These inmates are disproportionately represented among the parole 15 violators returning to custody for short sentences, id., and are thus likely to spend their entire 16 sentence at the reception center. Id. Because their sentences are so short, they are frequently 17 discharged before receiving treatment and fall into a vicious cycle, as they decompensate in 18 the community and quickly return . . . . Id.; see also Aug. 15, 2008 Haney Report 19 ¶¶ 358-59. ( [M]entally ill parolees often do not receive meaningful mental health treatment 20 when they are on parole. . . . Frequently as a result of their decompensation, many are 21 returned to prison, often for technical or minor violations. Thus, many of the parole 22 violations that return them to prison are directly related to their unmet mental health needs. 23 When they return to prison, these vulnerable prisoners are then packed into overcrowded 24 reception centers. ). 25 The absence of adequate mental health care at reception centers also has significant 26 adverse consequences for mentally ill inmates admitted to serve longer sentences in state 27 prison. The shortage of mental health care beds throughout the prison system which we 28 discuss in more detail below means that these inmates often spend months in a reception 64 1 center with little or no access to necessary mental health care while waiting for a bed to open 2 up. See, e.g., Nov. 9, 2007 Stewart Report ¶ 166 (discussing impact of delays in transfer of 3 mentally ill inmates from reception centers to necessary level of care); Aug. 15, 2008 Haney 4 Report ¶ 105 (discussing delays in transfer of EOP inmates out of CIM reception center); id. 5 ¶ 129 (discussing prolonged reception center stays and minimal treatment provided for EOP 6 and CCCMS inmates at CIM reported by Coleman Special Master). For example, the 7 number of inmates in reception centers needing an EOP level of care47 continues to grow, see 8 Ex. P243 at 900004-06, 900121-23, but the EOP program provided to these inmates falls far 9 below the care mandated by the Program Guide for EOP patients. Aug. 15, 2008 Haney 10 Report ¶ 29 (citing September 2006 Program Guide, Ex. P9 at 12-4-1); see also Nov. 9, 2007 11 Stewart Report ¶ 167. This is not surprising, given that the conditions in these reception 12 centers have been described as toxic, noxious, psychologically and medically unhealthy, 13 Rep. Tr. at 953:13-14 (Haney), and that a lack of treatment space severely impedes efforts to 14 provide even the most rudimentary forms of mental health care in reception centers. E.g., 15 Nov. 9, 2007 Stewart Report ¶ 80; Aug. 15, 2008 Haney Report ¶¶ 246, 247. 2. 16 17 Treatment Space The severe shortage of treatment space evident at CDCR reception centers affects the 18 provision of medical and mental health care throughout the state prison system. Dr. Stewart 19 reported that the problem of adequate office and treatment space is endemic in the CDCR, 20 Nov. 9, 2007 Stewart Report ¶ 190, and the Plata Receiver noted in his Turnaround Plan of 21 Action that investments in health care facilities have significantly lagged behind growing 22 inmate populations, so much so that available clinical space is less than half of what is 23 necessary for daily operations. Ex. D1133 at 25. In part, this is due to the CDCR s policy 24 and practice of anticipating that prisons will be filled beyond their design capacity, but not 25 including sufficient health care space to serve the anticipated population. Ex. D1092 at 20 26 (Plata Receiver s May 15, 2007 Report Re: Overcrowding). Compounding problems caused 27 by the lack of space, the space that does exist to provide health care services is often 28 47 The various levels of mental health care are defined supra note 24. 65 1 woefully inadequate. Through years of neglect, the facilities have long since passed the 2 time when modest investments could remedy the problem. We are dealing not with deferred 3 maintenance, but with some facilities that are literally falling apart. Ex. D1133 at 25. 4 The court received evidence of inadequate treatment space at a variety of prisons 5 statewide. At Avenal State Prison, staff must attempt to provide care for 7,525 inmates in 6 space designed for less than one-third of that number. Ex. D1233 at 25 (Plata Receiver s 7 Nov. 3, 2008 Analysis of Year 2007 Death Reviews).48 At Mule Creek State Prison, a Plata 8 Receivership team found that [a]ll of the Facility Clinics are undersized for the quantity of 9 inmate/patients seen on a daily basis and lack[] appropriate holding/waiting space for 10 inmate/patients ducated [scheduled] to be seen by health care providers. Ex. P101 at 7 11 (Plata Receiver s Custody/Security Assessment for Health Care Access at Mule Creek State 12 Prison). 13 One expert who testified at trial explained that crowding has so over-taxed the 14 clinical facilities at California Institution for Men that, as with the reception center at the 15 same prison, fundamental medical confidentiality rights are routinely ignored in the space 16 used to provide care to inmates housed at the prison: 17 In the West facility clinic at CIM, two PCPs [primary care physicians] share one room and simultaneously see patients for sick call and other encounters. A thin fabric folding screen separates the area in which the doctors see patients from a single exam table which the PCPs must share, as the room is not large enough to accommodate a second table. In the same clinic, the registered nurse conducts face-to-face triage appointments with patients in a large room that is shared by another nurse (who may be seeing patients) and an office technician. These arrangements cannot provide for minimally adequate patient-provider privacy. Moreover, the medical treatment area is so small that there is no medically appropriate waiting area, so sick patients must wait for appointments on a small bleacher outside the clinic, exposed to the elements. 18 19 20 21 22 23 24 25 Nov. 9, 2007 Shansky Report ¶ 24. Dr. Shansky also testified about the shortage of clinical 26 space he observed at several other prisons. Id. ¶¶ 16-23 (Valley State Prison for Women, 27 Avenal State Prison, and San Quentin); Sept. 10, 2008 Shansky 2d Supp. Report ¶¶ 31-46 28 48 Plaintiffs also offered this document into evidence as Exhibit G to Exhibit P413. 66 1 (North Kern State Prison, Substance Abuse and Treatment Facility, Pleasant Valley State 2 Prison, and California State Prison-Solano); see also Rep. Tr. at 663:22-665:10 (Rowlett) 3 (testifying about clinical space shortage at California State Prison-Solano). Similarly, 4 Secretary Woodford testified that space represents a serious obstacle to the delivery of 5 health care. Aug. 15, 2008 Woodford Supp. Report ¶ 27; see also id. ¶¶ 27-29 (discussing 6 space issues encountered on tours of the Correctional Training Facility ( CTF ) and 7 California State Prison-Los Angeles County (Lancaster)). 8 As the Plata Receiver concluded in his supplemental report on overcrowding, [t]here 9 is a dire need for additional clinical space . . . in the prisons because the existing capacity has 10 been swamped by the number of inmates in the system. Ex. D1094 at 2 (Plata Receiver s 11 June 11, 2007 Supp. Report Re: Overcrowding). On the basis of all of the evidence received 12 at trial, there is no doubt that crowding renders the existing clinical space in California s 13 prisons grossly inadequate. 14 15 3. Inability To House Inmates by Classification Crowding also negatively impacts the state s ability to house inmates according to 16 their proper classification, which in turn creates inadequacies in the medical and mental 17 health care that the system is capable of providing to inmates. 18 A prison classification system is an objective tool that allows correctional staff to 19 consider individual factors, including age, criminal history, educational levels or 20 deficiencies, mental health issues, [and] medical issues so that inmates can be placed in the 21 appropriate custody housing level. Rep. Tr. at 149:18-24 (Scott). Prisoners in California 22 are assigned to one of four levels of classification based on the length of their sentence, their 23 disciplinary history and other objective factors. Nov. 9, 2007 Woodford Report ¶ 13. 24 In an overcrowded system, this classification system breaks down. A well- 25 functioning system of classification should have no more than 10% of prisoners housed 26 outside their classification level. Nov. 9, 2007 Scott Report ¶ 69. In California, population 27 pressures have forced the CDCR to house an estimated 25% of inmates outside their 28 classification levels. Id. This failure to house inmates within such levels mak[es] it harder 67 1 to provide for their medical and mental health care needs, id. ¶ 68, and deprives the 2 overcrowded system of the flexibility needed to respond to inevitable crises. Id. As 3 Director Scott explained, this widespread rejection of CDCR s classification system has a 4 significant impact on medical and mental health care : 5 [I]t is harder to get health care appointments within the prison as well as out-of-prison specialty appointments, and more limited staff contact means that staff are less responsive to emergencies due to distrust, lack of understanding and compassion, and simple logistics: it is harder to get staff attention in a high-pressure, high-security unit. In addition, prisoners are subject to increased degrees of danger and potential for violence because they are placed with more dangerous and violent prisoners than their classification scores would warrant. 6 7 8 9 10 Id. ¶ 71. Similarly, Secretary Woodford testified that crowding makes it impossible to 11 move inmates where they need[] to be to address inmates medical and mental health needs. 12 Rep. Tr. at 375:4-6; see also id. at 227:4-13 (Beard) ( [H]uge overcrowding creates a 13 problem, because you have an individual who needs to go to Institution A, and Institution A 14 is full. So where do you put him. . . . [Y]ou end up having to put them somewhere that 15 maybe is not the most appropriate for that individual. ). Dr. Shansky likewise agreed that 16 the CDCR is currently unable to accommodate the housing needs of medical patients 17 requiring specialized placement. Nov. 9, 2007 Shansky Report ¶ 126. 18 19 4. Beds for Mentally Ill Inmates49 Crowding has also created severe bed shortages at every level of the CDCR s mental 20 health care system, causing inmates in need of higher levels of care to languish in clinically 21 inappropriate settings. It is not simply the beds themselves that the state does not possess, 22 but the space in which to place them. The need for such space is rapidly growing. From 23 December 2006 to August 2008, for example, the shortage of EOP beds more than tripled to 24 almost 1,000. Ex. P243 at 900007, 900124. Likewise, between June and September 2008, 25 49 Throughout this section, we cite evidence offered at trial on the number of mental health beds at each level of the mental health care delivery system above CCCMS. The 26 Coleman court recently approved several projects planned by defendants to increase, in the near term, the number of such beds. June 17, 2009 Order in Coleman. Nonetheless, it 27 remains to be seen whether and when defendants will comply with the Coleman court s June 17, 2009 order, and, in any event, compliance with that order alone will not suffice to 28 meet defendants constitutional obligations to the Coleman class. Id. at 5. 68 1 the CDCR s severe shortage of mental health crisis beds prevented more than two-thirds of 2 the inmates referred to such beds from actually being transferred.50 See Exs. P555, P586, 3 P587, P585 (mental health crisis bed referral data from June 2008 through September 2008 4 showing 391 transfers out of a total of 1,424 referrals). At the level of care reserved for the 5 most mentally ill, inmates sometimes wait as much as a year before being transferred to 6 inpatient beds. Aug. 15, 2008 Stewart Supp. Report ¶ 20. 7 The shortage of mental health beds throughout the system means that large numbers 8 of inmates in need of care cannot be transferred and do not receive the treatment their mental 9 illness requires. See, e.g., Aug. 15, 2008 Haney Report ¶ 216. Inmates requiring an EOP 10 placement often remain in general population yards receiving only limited mental health 11 treatment. Id. Many of them decompensate and require one or more admissions to a mental 12 health crisis bed for stabilization. Id. Because of the severe shortage of available mental 13 health crisis beds, however, inmates in need of such care are frequently placed in a variety 14 of temporary housing alternatives ranging from infirmaries to telephone-booth-sized 15 interview stalls typically placed in corridors. Ex. D1292 at 3. Most of these alternative 16 placements lack suitable staffing and/or the physical configuration needed for the continuous 17 monitoring or intensive treatment provided in a MHCB unit. Id. at 3-4. Suicidal inmates 18 referred to mental health crisis beds have spent from Thursday evening to . . . Monday 19 morning being transferred between so-called dry cells, which are tiny, freestanding 20 upright cages with mesh wiring surrounding them (and no toilet), during the day and 21 so-called wet cells, which are holding cells that have toilets, at night. Aug. 15, 2008 22 Haney Report ¶ 156. In several instances, inmates referred to mental health crisis beds have 23 committed suicide while awaiting transfer. E.g., Nov. 9, 2007 Stewart Report ¶¶ 173-75 24 (inmate referred to crisis bed hanged himself after several days on a suicide precaution 25 protocol in a mental health outpatient housing unit); Aug. 15, 2008 Stewart Supp. Report 26 ¶ 100 (after determination that neither restraint room nor crisis beds were available, an 27 50 All inmates referred to mental health crisis beds are suffering from severe decompensation or are a danger to themselves or others. A substantial proportion of these 28 inmate/patients are at a high risk for suicide. Ex. D1292 at 3; see also Ex. D1148 at 12-1-8. 69 1 inmate referred to mental health crisis bed and for possible involuntary medication returned 2 to administrative segregation cell where he hanged himself); id. ¶ 109 (inmate identified as 3 high suicide risk and referred to crisis bed hanged himself two days later in unlicensed 4 infirmary at CTF). 5 These shortages at every level, which are caused by the lack of space resulting from 6 overcrowding, have created a destructive feedback loop that is now endemic to the CDCR s 7 mental health care delivery system. Inmates denied necessary mental health placements are 8 decompensating and are ending up in mental health conditions far more acute than necessary 9 . . . . creat[ing] a cycle of sicker people being admitted, with greater resources necessary to 10 treat them, which then creates even further backlog in an already overwhelmed system. 11 Aug. 15, 2008 Stewart Supp. Report ¶ 92; see also Nov. 9, 2007 Stewart Report ¶¶ 31, 32; 12 Ex. D1292 at 9-10; Dec. 10, 2007 Packer Report at 11. Because overcrowding has led to a 13 significant, unaddressed demand for mental health services that only becomes more acute 14 over time, new mental health beds cannot be added quickly enough to address the system s 15 problems. [D]ue to the effects of overcrowding on the delivery of mental health care, any 16 reduction in the waitlists for higher levels of care will be temporary due to the pentup 17 demands in the system. Aug. 15, 2008 Haney Report ¶ 33. 18 C. 19 The severe overcrowding in California s prisons has also affected the conditions under Conditions of Confinement 20 which members of the Plata and Coleman classes are confined. One consequence of the 21 growing gap between the size of the CDCR population and the capacity of its prisons has 22 been a significant increase in the use of non-traditional or so-called ugly or bad beds. 23 Ex. P4 at 200; Rep. Tr. at 1912:8-14 (Kernan) (CDCR Undersecretary of Operations). These 24 include triple bunks, housing two inmates in cells designed for one inmate, and beds for 25 both low- and medium-risk inmates . . . crammed into gyms and dayrooms that were never 26 meant to be used for housing. Ex. P4 at 200; see also Rep. Tr. at 1912:15-17 (Kernan). The 27 court heard testimony and saw photographic and videographic evidence of these beds. See, 28 e.g., Rep. Tr. at 148:9-149:1 (Scott); id. at 269:11-25 (Lehman); Exs. P336, P339, P348, 70 1 P363. Director Scott, the former head of corrections in Texas, described some of the ugly 2 beds he saw on his tours of California prisons as truly appalling and reported that, [i]n 3 more than 35 years of prison work experience, I have never seen anything like it. Nov. 9, 4 2007 Scott Report ¶ 11. In the October 2006 Prison Overcrowding State of Emergency 5 Proclamation, Governor Schwarzenegger found that the CDCR was housing more than 6 15,000 inmates in these beds. Ex. P1 at 1. By August 2007, the number of inmates housed 7 in non-traditional beds had increased to approximately 19,600. Rep. Tr. at 1893:11-19 8 (Kernan); see also Ex. D1252-2 at 2. At the time of trial, the department was using 9 approximately 14,000 such beds. Rep. Tr. at 1911:9-14 (Kernan). 10 The use of non-traditional housing raises serious safety concerns, contributes to the 11 spread of infectious disease, and exacerbates mental illness. First, as Secretary Woodford, 12 former head of the CDCR, testified, the number of custodial staff is often grossly 13 inadequate to meet basic needs with often only two officers to supervise 200 prisoners in 14 a gym or a dorm. This is extremely dangerous for both the prisoners and the staff because 15 line of sight supervision is impossible under these circumstances and it does not permit the 16 staff the time to recognize that prisoners are in trouble from any number of causes, including 17 medical or mental illnesses. 51 Nov. 9, 2007 Woodford Report ¶ 17. One of her successors, 18 Secretary Tilton, similarly reported that because of overcrowded conditions, including the 19 high use of non-traditional beds, the risk of catastrophic failure in a system strained from 20 severe overcrowding is a constant threat. Ex. P104 at 15. 21 Second, crowding generates unsanitary conditions, overwhelms the infrastructure of 22 existing prisons, and increases the risk that infectious diseases will spread. See, e.g., Nov. 9, 23 2007 Scott Report ¶¶ 17-24. The Governor recognized such dangers when he issued his 24 emergency proclamation on crowding, declaring that current severe overcrowding in 29 25 CDCR prisons has caused substantial risk to the health and safety of CDCR staff, inmates, 26 and the public. Ex. P1 at 1-2. Similarly, Scott Kernan, then the Chief Deputy Secretary of 27 51 We discuss additional problems related to inadequate staffing below, infra 28 Section IV.D.1. 71 1 the Division of Adult Institutions for the CDCR, declared that overcrowding has led to 2 increased numbers of infectious disease outbreaks and riots and disturbances system-wide. 3 Ex. P11 ¶ 3 (May 16, 2007 Decl. of Scott Kernan filed in Plata) (noting eleven different 4 outbreaks, possible outbreaks, or exposure to tuberculosis at seven prisons). As plaintiffs 5 medical expert testified, the overcrowded housing conditions, and in particular, the 6 conditions in the non-traditional beds, including the converted gyms, create potential 7 breeding grounds for disease. Sept. 10, 2008 Shansky 2d Supp. Report ¶ 118; see also Rep. 8 Tr. at 270:7-12 (Lehman) (crowding contributes to the difficulties of healthcare delivery by 9 virtue of the fact that it increases the incidence of illnesses, [and] infectious disease ); id. at 10 257:15-22 (Beard) (while prisons may not always be incubators for disease, they could be if 11 your population densities get so intense, like if you have a gymnasium that you triple bunk 12 and put hundreds and hundreds of people in a closed dense area ); id. at 88:25-89:3 (Stewart) 13 (interviewed two Coleman class members who were suffering from staph infections that 14 they got while living in these unhealthy conditions ); Ex. P4 at 200 (non-traditional beds 15 create difficult, unsanitary living conditions where ventilation is poor, toilet access is 16 limited, and as many as 200 people might share six showers ). Until CDCR reduces its 17 population, it will remain highly vulnerable to outbreaks of communicable diseases, 18 including staph infections, tuberculosis and influenza. Nov. 9, 2007 Shansky Report ¶ 135. 19 Third, plaintiffs mental health experts also reported on the toxicity of non-traditional 20 housing for members of the Coleman class and other inmates. As Dr. Stewart explained, 21 [p]lacing inmates in overcrowded gym and dorm settings is often inappropriate for people 22 with mental health issues and can either exacerbate existing symptoms or, in some cases, 23 trigger symptoms in people who would not otherwise display them. Aug. 15, 2008 Stewart 24 Supp. Report ¶ 66; see also id. ¶ 52 (crowded dorm settings also may exacerbate mental 25 health conditions such as paranoia and create stressful environments for people who are 26 otherwise vulnerable due to mental health issues, including cognitive impairment ); Aug. 15, 27 2008 Haney Report ¶ 291 (reporting on extensive use of non-traditional or bad beds at 28 North Kern State Prison, which was operating at 200% design capacity, and describing 72 1 housing conditions as especially inappropriate for the confinement of mentally ill 2 prisoners ). 3 Finally, non-traditional beds are frequently created by converting activity space into 4 inmate housing areas, which adversely impacts all inmates by reducing the amount of space 5 available for programs. Ex. P4 at 124. As the Coleman Special Master has explained: 6 The inevitable result of severe overcrowding is that everyone also spends more and more time in their cells. General yards are more crowded, less well supervised and increasingly dangerous. There are not nearly enough walk-alone yards to provide statutorily required amounts of exercise for those who by choice or need require them. Gyms are no longer an option for time out of one s cell. Dayrooms share many of the same problems. Work or vocational opportunities shrink in the expanding population. Disturbances occur more frequently, with resulting increases in the number and duration of lockdowns. All inmates must spend increasingly larger chunks of their days in their cells, or much more dangerously, in one of those triple-bunked non-traditional spaces. None of this is conducive to the health and well-being of any inmate, much less a seriously mentally disordered inmate/patient . . . . 7 8 9 10 11 12 13 14 Ex. D1292 at 7-8. Instead, these conditions inevitably escalate[] the incidence of mental 15 illness and exacerbate[] the condition of those already mentally fragile and vulnerable. Id. 16 at 8. 17 D. 18 Beyond the issues arising from critical space shortages, crowding has other severe Other Access to Care Issues 19 impacts on access to medical and mental health care. 20 21 1. Staffing The level of crowding has rendered current staffing levels insufficient to handle the 22 health care needs of the overpopulated system: 23 24 25 26 27 Many CDCR prisons are unable to sustain the basic delivery of medical, mental health, and dental services because of limited staffing (clinical and custody) and an overwhelming number of prisoner/patients who require care. Every day, many California prison wardens and health care managers make the difficult decision as to which of the class actions, Coleman, Perez, Armstrong or Plata they will fail to comply with because of staff shortages and patient loads. 28 73 1 Ex. D1092 at 30 (Plata Receiver s May 15, 2007 Report Re: Overcrowding).52 Crowding 2 also makes it impossible for the CDCR to hire the additional staff necessary to provide 3 constitutionally adequate medical and mental health care to the current population. 4 5 a. Medical Staff Defendants own data demonstrates significant vacancy rates for medical staff. As of 6 August 2008, 20 percent of chief physician and surgeon positions, 25 percent of physician 7 positions, 19 percent of physician assistant positions, 39 percent of nurse practitioner 8 positions, 10 percent of registered nurse positions, and 18 percent of licensed vocational 9 nurse positions remained vacant. Ex. D1235-2 (charts summarizing staffing trends and 10 indicating number of positions and number of positions filled by full-time employees). The 11 statewide vacancy rate for primary care provider positions, which include physicians and 12 surgeons, nurse practitioners, and physician assistants, was 27 percent. Id.; Nov. 9, 2007 13 Shansky Report ¶ 37 (explaining positions that are considered primary care providers). 14 Some prisons have attempted to fill these vacancies with registry (contract) physicians, but 15 this practice is insufficient as a long-term solution. Because registry physicians tend to turn 16 over quickly, the prisons end up spending time doing extensive on-the-job training 17 repeatedly, which is time-consuming and detracts from patient care delivery. Sept. 10, 2008 18 Shansky 2d Supp. Report ¶ 66; see also Nov. 9, 2007 Shansky Report ¶ 45. 19 In any system, inadequate medical staffing, whether due to unfillable vacancies or 20 insufficient allocation of positions, will result in delayed care. In a dramatically 21 overcrowded system like the CDCR s the treatment delays become more acute. Nov. 9, 22 2007 Shansky Report ¶ 46. In overcrowded systems, prisoners experience significant 23 appointment delays, id., both in terms of seeing a primary care physician and even in being 24 triaged by a nurse to determine whether an appointment with a physician is necessary. E.g., 25 id. ¶¶ 46-49; Sept. 10, 2008 Shansky 2d Supp. Report ¶¶ 67-77. In addition, [w]ith too few 26 52 Perez v. Cate, Case No. C05-5241 JSW (N.D. Cal.), is a statewide class action 27 concerning dental care in California prisons. Armstrong v. Schwarzenegger, Case No. C94-2307 CW (N.D. Cal.), is a statewide class action concerning California prisoners and 28 parolees with hearing, vision, mobility, kidney, and learning impairments. 74 1 primary care providers to meet the most immediate needs of the current population, some 2 prisons are unable to develop required medical programs, including the chronic and 3 preventive care programs required by the Plata policies and procedures to which defendants 4 have committed themselves. Nov. 9, 2007 Shansky Report ¶¶ 51-53. Even the remedies the 5 state uses to alleviate crowding cause problems for an already overburdened staff. For 6 example, at CTF the medical department was swamped with work because they had been 7 ordered to review 1,500 medical files to determine which prisoners were eligible for transfer 8 to out-of-state prisons. Aug. 15, 2008 Woodford Supp. Report ¶ 6. 9 In addition to rendering current medical staffing levels seriously inadequate, crowding 10 makes it impossible for the CDCR to increase the number of clinical positions to the level 11 needed to provide adequate care to inmates. Staffing and space issues are inextricably 12 intertwined such that, given the overcrowding, hiring staff alone could not solve the problem. 13 As Secretary Lehman asked, [W]here are the providers going to work and how ? Rep. Tr. 14 at 272:1-13. A number of chief medical officers expressed the opinion that they would not 15 have sufficient space for clinical staff if all of the clinical positions currently budgeted were 16 filled, id. at 501:3-7 (Shansky), let alone if new positions were created and filled. 17 Moreover, crowding negatively impacts the recruitment and retention of clinical staff. 18 See, e.g., Nov. 9, 2007 Shansky Report ¶¶ 16, 20. Dr. Shansky explained that [t]he clinical 19 space allocated at San Quentin is so substandard and creates such a stressful environment 20 that . . . the prison s capacity to retain physicians is seriously jeopardized by both the 21 physicians perception of personal safety issues and the unprofessional conditions. Nov. 9, 22 2007 Shansky Report ¶ 23. More broadly, Dr. Shansky testified: 23 24 25 26 27 I believe that the hiring gains for clinicians made in the past year will be lost if these systemic issues [concerning overcrowding] are not addressed, because many newly-hired clinicians will be unwilling to risk their professional credentials and reputations by practicing in an environment where their patients are at risk of harm because among other things adequate clinical space is scarce, appointments are not scheduled, complete medical records are unavailable, and medications are not delivered. 28 75 1 Id. ¶ 136. The Plata Receiver likewise reported that crowding interferes with the ability to 2 recruit, hire and retain competent medical personnel. The overwhelming number of prisoners 3 needing care at the prisons, existing staffing shortages and inadequate clinical space are just a 4 few of the consequences that make developing a competent medical staff a daunting 5 challenge. Ex. D1094 at 2 (Plata Receiver s June 11, 2007 Supp. Report Re: 6 Overcrowding). 7 8 b. Mental Health Staff There are also staffing shortages at all clinical levels of the CDCR s mental health 9 care delivery system, and overcrowding in California s prisons has a profound impact on 10 mental health staffing levels. Rep. Tr. at 309:3-22 (Haney). Between March 2008 and 11 August 2008, for example, the total vacancy rate among existing mental health care positions 12 ranged from 22 percent to 36.1 percent, while the vacancy rate in psychiatrist positions was 13 particularly high, ranging from 30.6 percent to 54.1 percent. Ex. P245 at 1. Moreover, 14 CDCR significantly underestimated the staffing needed to implement critical portions of the 15 Coleman Program Guide requirements in its 2008-09 staffing requests. Aug. 15, 2008 16 Haney Report ¶ 336 (citing Ex. P485, July 12, 2008 letter from Coleman Special Master to 17 Robin Dezember and Coleman defense counsel reporting review of CDCR workload study). 18 Accordingly, these high vacancy rates understate the actual level of mental health 19 understaffing. Dr. Haney reported on the significant staff shortages at the eight facilities 20 he visited, all of which he described as [s]everely [o]vercrowded. Aug. 15, 2008 Haney 21 Report at 56 & ¶ 335. 22 As Dr. Haney explained, these shortages have a serious adverse effect upon the mental 23 health care provided to inmates. 24 25 26 27 [S]erious staffing shortages all translate into inadequacies in the mental health delivery system and, in some instances, an outright denial of needed and mandated mental health services. In many of the units this means that professional staff are doubling up on duties, performing more tasks than they should be called upon to handle, and managing far larger caseloads than is appropriate or effective. One psychologist at CIM told me I can t keep up with 28 76 1 everything. I ve been doing too much. We hired new staff, but that hasn t helped. He also told me in my opinion, we are doing about 50% of what we should be doing. 2 3 Id. ¶ 335. 4 Although defendants need additional clinical staffing to implement necessary mental 5 health programs, the CDCR ha[s] been unable to recruit and retain staff even to meet the 6 budgeted levels. Id. ¶ 336. This inability is directly related to the overcrowding in 7 California s prisons. [T]he serious deficiencies in office and treatment spaces I observed 8 throughout the system are themselves an obstacle to ever achieving appropriate clinical 9 staffing. The working conditions are terrible and there is no space, in any event, for more 10 clinicians. Id. Dr. Stewart explained that it is extremely difficult to recruit and retain 11 good clinical staff in a correctional environment in the best of times, but that, [i]n 12 overcrowded systems, with the attended violence, high acuity, [and] shortage of office space, 13 these ordinary recruitment problems are compounded and become significantly more difficult 14 to overcome. Nov. 9, 2007 Stewart Report ¶ 41. 15 16 c. Custodial Staff Crowding has also caused significant custodial staffing shortages in the CDCR that 17 have a direct impact on defendants ability to deliver constitutionally adequate medical and 18 mental health care to prison inmates. Custodial staff are essential to providing health care 19 to prisoners because they supervise prisoner movement to and from medical appointments, 20 they escort prisoners to services within an institution and they provide supervision when 21 prisoners are taken out of the prison to medical appointments, hospitals or they are 22 transferred to another institution. Nov. 9, 2007 Woodford Report ¶ 15. In addition, 23 custodial staff are responsible for alerting health care staff when prisoners complain of an 24 immediate serious problem and also are supposed to observe prisoners periodically to 25 identify actual or potential problems. Aug. 15, 2008 Woodford Supp. Report ¶ 25. 26 The California prison system lacks sufficient custodial staff to keep prisoners safe 27 from harm, id., or to provide prisoners with timely access to care and still perform other 28 essential functions, Nov. 9, 2007 Woodford Report ¶ 15. The paucity of correctional 77 1 officers in California, due to the low staffing rate and high number of vacancies, is 2 dangerous. Nov. 9, 2007 Scott Report ¶ 26 (footnote omitted). In fact, Director Scott 3 testified that every institution I toured had inadequate custodial staff on the ground to 4 address the needs of the prisoner population, including ensuring that health care services are 5 provided. Id. at 14 n.3. This is particularly dangerous for prisoners in need of medical 6 care . . . not just because staff are not available to escort prisoners or clinicians to 7 appointments, but because short-staffing can lead to forced overtime and burnout, such that 8 staff make poor decisions, particularly in health care emergencies. Id. ¶ 26; see also Nov. 9, 9 2007 Woodford Report ¶¶ 16-23 (discussing dangers of understaffing correctional officer 10 positions). As Director Scott testified: 11 12 13 14 15 16 17 18 [O]verworked staff without adequate back-up are less able to respond to emergencies and more likely to downplay prisoners concerns. In a housing unit such as San Quentin s H Unit Dorm 2 (one officer for 200 prisoners) or CIM s West Facility Cleveland Hall (two officers for 198 prisoners) or East Facility gym (two officers for 202 prisoners), staff in an emergency can only sound the alarm, make frantic telephone or radio calls, and hope for backup. An officer alone with several hundred inmates is unlikely, for example, to perform emergency first aid or CPR it is simply unsafe to do so with no backup, when prisoners could easily simulate an emergency as a diversion. The inability to perform basic lifesaving functions could have potentially devastating consequences on the life and health of a prisoner undergoing a medical or mental health emergency. This situation presents an unacceptable risk of harm to prisoners. 19 Nov. 9, 2007 Scott Report ¶ 59. 20 In addition, the Plata Receiver has explained that, [s]ystem-wide, CDCR lacks the 21 custody staff and organizational structure and processes to ensure that patient-inmates are 22 reliably escorted and/or transported to medical appointments. Ex. D1133 at 5 (Plata 23 Receiver s June 6, 2008 Turnaround Plan of Action). This results in denial of timely access 24 to health care services and substantially increas[es] the risk that patient-inmates health 25 will further deteriorate. Id. Dr. Shansky also concluded that lack of adequate custodial staff 26 causes significant delays in treatment . . . because there are not enough custody officers to 27 move the prisoners in and out of the clinics on a timely basis. Sept. 10, 2008 Shansky 2d 28 Supp. Report ¶ 107. Similarly, defendants own mental health expert testified that the 78 1 shortage of correctional officers statewide impedes the delivery of mental health care to 2 members of the Coleman class, particularly CCCMS inmates. Dec. 10, 2007 Packer Report 3 at 15-16. A reduction in the crowding of California s prisons would help ease the burden on 4 the custodial staff and permit staff members to better monitor inmates for medical or mental 5 health problems and to deliver inmates for necessary care. 6 7 2. Medication Management Next, crowding prevents defendants from achieving an adequate medication delivery 8 system that is marked by the timely delivery of the correct medication to the correct patient, 9 with accurate documentation of what has been administered. Nov. 9, 2007 Shansky Report 10 ¶ 79. Defendants medication delivery systems are inadequate for the size of the population 11 they serve, and are plagued by short-staffing at a number of prisons. . . . [Consequently,] 12 prisoners receive their medications late or not at all, and suffer as a result. Id. ¶ 80. The 13 shortcomings in the medication delivery system are rooted in overcrowding quite simply, 14 there are more patients requiring medications than the prison has the resources or staffing to 15 address. Id. ¶ 81. 16 Overcrowding affects the administration of both traditional medications to Plata class 17 members and psychotropic medications to Coleman class members. See, e.g., Rep. Tr. at 18 77:21-79:24 (Stewart); Aug. 15, 2008 Stewart Supp. Report ¶ 96. As Dr. Stewart testified, 19 [t]here are just too many people that are prescribed too many medications for the system to 20 handle. Rep. Tr. at 77:19-20. Following tours of Salinas Valley State Prison, California 21 Medical Facility, and Mule Creek State Prison, Dr. Stewart reported that: 22 23 24 25 26 27 28 First, due to the lack of adequate staff to distribute medications and the overwhelming number of inmates prescribed medications, staff members do not have sufficient time to adequately monitor whether inmates are taking medications properly. . . . Second, the clinical staff members who distribute medications are too understaffed to evaluate the efficacy and potential side effects of the prescribed medications. Every patient I talked to about the medication distribution system described the same drive-by process they received their medications in pill lines or at their cell doors from staff members who spent only a few seconds with them. The staff members never ask the patients about the efficacy of the medications or whether they are causing side effects. Third, psychiatrists are also overburdened and may 79 1 2 consist largely of contract employees that are unable to maintain consistent relationships with their patients due to constant movements between units or even prisons. 3 Aug. 15, 2008 Stewart Supp. Report ¶ 96; see also id. ¶ 95; Rep. Tr. at 670:9-673:10 4 (Rowlett) (correctional officer discussing similar problems based on her experience at 5 California State Prison-Solano). 6 The failure of the CDCR s medication delivery systems results in not only traditional 7 medical problems, but also high medication non-compliance rates among patients with 8 serious mental illness. Blood samples taken of seriously mentally ill CDCR inmates 9 admitted to DMH inpatient care units over more than two years show that the vast majority 10 of such inmates have little or no psychotropic medication in their systems. See Brewer Dep. 11 at 135:5-137:25; see also Aug. 15, 2008 Stewart Supp. Report ¶¶ 98-99 (reporting 12 information provided by Drs. Neill and Gandhi, DMH Program Directors for Salinas Valley 13 and CMF, respectively). The Executive Director of the DMH inpatient psychiatric programs 14 at Salinas Valley and CMF testified that this is a serious problem, Sept. 4, 2008 Brewer 15 Dep. at 127:17-18, the consequence of which is that acuity levels in mentally ill inmates 16 admitted to DMH units are rising, inmates admitted to inpatient care are taking longer to 17 stabilize on medications and often require orders for involuntary medication, and, upon 18 discharge, the inmates are then returned to the same system that fails to adequately monitor 19 medication compliance, thereby starting the cycle all over again. Aug. 15, 2008 Stewart 20 Supp. Report ¶ 100. 21 22 3. Specialty Medical Care The CDCR is also unable to provide access to specialty [medical] services, including 23 in urgent (high priority) cases, in accord with [its own] policy requirements. Nov. 9, 2007 24 Shansky Report ¶ 56. As a result of overcrowding, the number of prisoners who need such 25 services exceeds the capacity of the providers available to CDCR, and/or is so great that 26 CDCR cannot adequately track and schedule such cases. Id. For example, at Avenal State 27 Prison, Dr. Shansky reviewed two reports printed the day of his visit in the fall of 2007. Id. 28 ¶ 61. Those reports showed 1,293 pending specialty referrals, 316 urgent and 977 routine. 80 1 Id. Of the 316 pending urgent referrals, only approximately 105 had an appointment date, 2 with only 2 of the 316 urgent referrals a dismal 0.6 percent scheduled to take place within 3 the fourteen-day period required by CDCR policy for such appointments. Id. Of the 977 4 pending routine referrals, only approximately 285 had a scheduled appointment date, and 5 only approximately 135 of the 977 routine referrals approximately fourteen percent were 6 scheduled to occur within the three-month period required by CDCR policy for such 7 appointments. Id. ¶ 62. While Avenal provides the starkest numbers, the problem exists at 8 other prisons as well. E.g., id. ¶ 65 (more than 50 percent of urgent referrals on pending list 9 at High Desert State Prison were pending for longer than fourteen-day period required by 10 CDCR policy); Sept. 10, 2008 Shansky 2d Supp. Report ¶¶ 88-94 (discussing specialty care 11 problems at four prisons and concluding that [t]he demand for care, particularly for the high 12 priority cases, continues to overwhelm the resources available to the defendants ). 13 Prison staff at Avenal indicated that they were not confident that the reports [of 14 pending specialty care referrals] were entirely accurate, and that some referrals that 15 appeared as pending may have actually taken place but were not yet closed out in the system 16 as having been completed. Nov. 9, 2007 Shansky Report ¶ 63. However, 17 18 19 20 to the extent that the aging report data is incorrect, then it reflects that the prison has more patient data than it is capable of processing, leaving [Avenal State Prison] unable to determine who actually needs the services, with the distinct possibility of prisoners being double-scheduled (and thus delaying specialty services for other prisoners still actually in need of an appointment). 21 Id. ¶ 64. Put simply, even if the specialty care numbers are not as dire as the reports indicate 22 and it would be difficult to do worse than having only 0.6 percent of pending urgent 23 referrals scheduled within the fourteen-day period mandated by CDCR policy [Avenal s] 24 population exceeds its capacity for scheduling and tracking. Id. 25 26 4. Lockdowns Delays in access to care are even more acute during periods when prisons are in 27 lockdowns. See, e.g., Sept. 10, 2008 Shansky 2d Supp. Report ¶¶ 108-11 (discussing impact 28 of lockdowns at Pleasant Valley State Prison and High Desert State Prison). Because of 81 1 crowding, the California prison officials who administer the state s thirty-three adult prison 2 institutions 3 4 5 6 rely largely on lockdowns to control their system. . . . [I]n 2006, they had 449 lockdowns, which averaged 12 days a lockdown. And they had 20 or so of those lockdowns that were over 60 days. Those things impact upon your ability to properly deliver any service within an institution, including mental health and medical services. 7 Rep. Tr. at 218:18-25 (Beard); see also Nov. 9, 2007 Scott Report ¶ 63 ( Overcrowding 8 engenders a state of perpetual crisis that shuts down non-emergency prison functions. ). As 9 Dr. Haney explained: 10 11 12 13 14 15 16 Lockdowns are used in the California Department of Corrections, I believe, in large part because of the profound level of overcrowding at a level that is unheard of in corrections departments across the United States with which I m familiar. Lockdowns mean that prisoners, including EOP prisoners, if they are in a unit that is locked down, are essentially without programs during the periods of time that the lockdown is in place. There are housing units in the California Department of Corrections that are locked down more often than they are unlocked. 17 Rep. Tr. at 316:23-317:9; see also id. at 70:4-6 (Stewart) ( The fact that there s too many 18 inmates at the Salinas Valley State Prison in the general population yard . . . resulted in an 19 almost continuous lockdown. ). 20 Lockdowns require a radically different form of medical delivery than the services 21 provided under normal general population conditions. Ex. D1092 at 29 (Plata Receiver s 22 May 15, 2007 Report Re: Overcrowding). When a prison is in lockdown, inmates housed in 23 the general population are unable to leave their housing units to go to yard clinics to access 24 medical care; instead, clinical staff must go from cell to cell to see the prisoner/patient, or 25 small groups or individual prisoners must be escorted by correctional officers to and from 26 clinic areas. Id. at 29-30. California prisons are not staffed for this type of situation, 27 where staff must escort prisoners to every service or bring the service to them. Nov. 9, 28 2007 Woodford Report ¶ 25. 82 1 Likewise, lockdowns affect the delivery of mental health care in several ways. 2 Inmates frequently cannot leave their cells to attend necessary treatment programs. Rep. Tr. 3 at 881:4-10 (Dezember); Nov. 9, 2007 Stewart Report ¶ 138; Aug. 15, 2008 Stewart Supp. 4 Report ¶¶ 34, 38. Additionally, lockdowns prevent staff from supervising the intake of 5 psychotropic medications. Aug. 15, 2008 Stewart Supp. Report ¶ 96 (noting that lockdowns 6 result[] in distributions of medications through food ports or otherwise at cell doors, where 7 it is difficult to monitor compliance with medication regimens ). Finally, some mentally ill 8 prisoners cannot handle the severe stress of locked-down confinement and may 9 decompensate or become suicidal as a result. Aug. 15, 2008 Haney Report ¶ 166. 10 E. 11 Another deficiency in the delivery of medical and mental health care to California Medical Records 12 inmates concerns medical records. For example, Director Scott testified that the CDCR 13 cannot track and transfer essential health care records, because the record system lacks the 14 capacity to deliver records regarding this many prisoners. Nov. 9, 2007 Scott Report ¶ 78. 15 As he explained, 16 17 18 19 given the extraordinary number of prisoners in these facilities, it is simply impossible to manually file so many records on a timely basis. In my experience, such extraordinary pressure on staff also leads to serious filing errors, which means that even records that have been filed might not be available to clinicians, and might be impossible ever to locate. 20 Id. 21 Dr. Shansky s observations similarly revealed medical records that were dangerously 22 incomplete. Nov. 9, 2007 Shansky Report ¶ 101; see also id. ¶ 106 (at Avenal State Prison, 23 the amount of documents generated had simply overwhelmed the staff s capacity to timely 24 and properly place documents in prisoners [unit health records] ); id. ¶ 107 (at High Desert 25 State Prison, 107 inches of loose filing remained, and even those documents that are filed are 26 only placed rather than fastened into inmates health files, which greatly increases the 27 chance of documents being lost or misplaced ); cf. Nov. 9, 2007 Scott Report ¶ 78 (noting 28 observation of four feet of loose filing waiting to be placed in prisoners health records at 83 1 Avenal, a prison at over 200% of design capacity). Dr. Shansky summarized his 2 observations as follows: 3 4 5 6 7 At each of the prisons I inspected, I found that the medical records were unwieldy, rarely organized chronologically and, in general, poorly maintained. Retrieving useful information from the files invariably requires considerable time sifting through extraneous reports, misfiled documents and outdated materials. At the same time, certain documents that would be extremely useful, such as an updated Problem List for each file, which is required by the court-ordered Plata Policies and Procedures [to which defendants stipulated], are typically missing. 8 Sept. 10, 2008 Shansky 2d Supp. Report ¶ 99. Dr. Shansky further testified that CDCR s 9 tracking and information systems cannot keep up with the overwhelming data requirements 10 in the system s overcrowded prisons. Id. ¶ 102; see also Nov. 9, 2007 Shansky Report 11 ¶ 110 (noting that adequate care requires timely appointments, which in turn requires an 12 effective scheduling and tracking system, something that CDCR has proven itself incapable 13 of developing . . . due in large part to the sheer numbers of patients and their vast and 14 growing need for coordinated appointments ). 15 The deficiencies in the management of medical records extend to mental health care 16 as well. Defendants mental health expert Dr. Packer described several such deficiencies: 17 18 19 20 21 In several institutions there were difficulties in clinical staff obtaining charts in a timely manner (that is, the charts were not available when needed for a clinical assessment) as well as difficulties in updating the charts (that is, delays in notes being placed in the records). Staff referred to a category of Flimsy charts, meaning that they sometimes had limited information available when doing an assessment. The documentation I reviewed also described institutions in which there were significant numbers of charts unfiled. 22 Dec. 10, 2007 Packer Report at 19. Dr. Packer opined that the medical records problem is a 23 direct effect of overcrowding, and that [t]his problem does impact directly on the ability to 24 provide timely and appropriate care. Id. at 19-20; see also Rep. Tr. at 1119:2-5 (expressing 25 his opinion that the sheer number of inmates in the system is the most direct cause resulting 26 in the difficulty of CDCR to manage their medical records appropriately ). 27 Plaintiffs expert Dr. Stewart reviewed approximately sixty medical records during his 28 tours and reported that he consistently found the records to be unwieldy, disorganized and 84 1 bulky, with loose papers floating around in the files, and that [i]t was exceedingly difficult 2 to follow the clinical course of treatment of the patients because of the size and 3 disorganization of the files. Aug. 15, 2008 Stewart Supp. Report ¶ 102. He also reported a 4 few instances where [he] found other patients records in the files [he] reviewed. Id. He 5 described these problems as typical in an overwhelmed and overcrowded system. Id. 6 As Dr. Stewart explained, 7 [a]ccurate well organized medical records are a critical element of medical and mental health care. They are even more essential in a complex and overcrowded system such as the CDCR which is characterized by frequent transfers of patients, high turnover of clinical staff and overuse of contract clinicians who lack familiarity with the patients and the system itself. 8 9 10 11 Id. According to Dr. Shansky, [u]nless medical records and scheduling information are 12 managed, organized, and maintained effectively, appropriate health care services cannot be 13 provided. Overcrowding makes it impossible for CDCR to perform these essential 14 functions. Nov. 9, 2007 Shansky Report ¶ 97 (emphasis added). 15 F. Increasing Acuity of Mental Illness 16 Finally, and alarmingly, the evidence shows that crowded conditions, and the bed and 17 staffing shortages and delays in access to necessary care that result from crowding, intensify 18 the acuity of mental illness among inmates throughout the California prison system. As 19 Dr. Stewart explained: 20 21 22 23 24 [I]nsufficient access to higher levels of care has created a system which is overwhelmed by the acuity of its patients at every level of care. EOP units house many patients in need of inpatient care, MHCB s house patients in need of inpatient hospitalization, intermediate care facility units house many patients in need of acute hospital care and so on. When and if these patients finally reach the level of care they require, their mental health conditions may be far more serious, resulting in longer stays and more resources in order to stabilize and get well. 25 Aug. 15, 2008 Stewart Supp. Report ¶ 43 (footnote omitted); see also id. ¶ 88 ( It was clear 26 that the severe shortage of mental health beds has created a system that houses a significant 27 portion of Coleman class members at lower levels of care than the patients clinically 28 require. ). Dr. Stewart reported that he was struck by the very high acuity of the patients 85 1 [he] encountered during [his] tours because they were much sicker, as a whole, than the 2 Coleman class members [he] encountered between 1990 and 2000, when he served as a 3 court-appointed monitor at CMF. Id. ¶ 88. Dr. Stewart attributed this directly to 4 overcrowding: 5 The reality of the current MHSDS system, as demonstrated by my interviews with these class members and the admitted shortage of EOP, MHCB and inpatient beds, is that too many people are housed in places that simply cannot provide them with the level of mental health care they require. This is a direct result of overcrowding there are too many people in the system and too few resources to treat them. This in turn means that the acuity level at every level of care is higher than it would be in a system that has sufficient inpatient beds. 6 7 8 9 10 Id. ¶ 91. 11 Dr. Haney similarly reported that there is evidence that the worsening prison 12 overcrowding crisis has had a corresponding effect on the quality of mental health care. 13 Aug. 15, 2008 Haney Report ¶ 373 (emphasis in original). At trial, he testified that mentally 14 ill inmates in need of higher levels of mental health care are getting sicker as a result of 15 their inability to get the appropriate level of care, Rep. Tr. at 304:16-19, and he agreed that 16 overcrowding in California s prisons is resulting in more significant mental illness than one 17 would find at a properly run prison with proper population and an adequately functioning 18 mental health care delivery system. Id. at 305:24-306:5. 19 G. Possibly Preventable Deaths, Including Suicides 20 21 Extreme Departures from the Standard of Care and Preventable or All of the above problems, caused by crowded conditions, ultimately contribute to 22 unacceptably high numbers of both preventable or possibly preventable deaths, including 23 suicides, and extreme departures from the standard of care. 24 In 2006, California had a prisoner suicide rate of 25.1 suicides per 100,000 inmates, 25 compared to the national average of 14 per 100,000. Ex. P58 at 9 (Coleman Special Master s 26 Report on Suicides Completed in the CDCR in Calendar Year 2006).53 In reviewing these 27 53 Because the record does not contain evidence of the number of suicides in 2007 or 2008, it is unclear from the record whether California s inmate suicide rate has risen or 28 declined since 2006. It appears, however, that the suicide rate is not appreciably lower, as 86 1 suicides, the Special Master found that 72.1 percent of completed suicides in 2006 involved 2 some measure of inadequate treatment or intervention and were, therefore, most probably 3 foreseeable and/or preventable. Id. at 8. Since 2003, the percentage of suicides found to be 4 foreseeable or preventable each year has remained relatively constant at around 75 percent, 5 representing marked increases over the 45 percent rate of inadequate treatment that was 6 found for suicides that occurred in 2002. Id. at 8. While we do not suggest that crowded 7 conditions are the sole cause of the increase in the suicide rate among California inmates, the 8 evidence demonstrates that crowding throughout the prison system has a significant effect on 9 many of the risk factors that contribute to inmate suicides. Major contributing factors to 10 foreseeable and preventable suicides include inadequate clinical assessments, inappropriate 11 interventions, incomplete referrals, missed appointments and appointments that were not 12 rescheduled, unsupported diagnoses, failure to review records, assignments to inappropriate 13 levels of mental health care, failure to provide protective housing, and the provision of 14 inadequate or untimely resuscitation efforts. Ex. D1281 at 680. As our discussion above 15 makes clear, crowding is a major cause of nearly all of these factors. 16 The Plata Receiver also reviews inmate deaths to determine whether any deaths were 17 preventable or possibly preventable. In 2007, of the 110 deaths considered to be unexpected 18 and not the result of homicide or self-inflicted injuries, 44 deaths (40 percent) were found to 19 be preventable or possibly preventable, Ex. D1233 at 8 (Plata Receiver s Nov. 3, 2008 20 Analysis of Year 2007 Death Reviews), meaning that better medical management or a better 21 system of care would likely have or may have prevented the patient s death, id. at 5. 22 Dr. Shansky testified that this rate was extremely high. Rep. Tr. at 428:23-429:7. The 23 Receiver also examined extreme departures from the standard of care, defined as lapse[s] 24 in care that a reasonable and competent clinician would not render under the same or similar 25 circumstances. Ex. D1233 at 5, 15. He found extreme lapses in nearly 60 percent of the 26 the CDCR reported 31 apparent suicides to the Coleman Special Master during the first ten 27 months of 2008. See Ex. P171-R; Ex. P506. If annualized, this equates to approximately twenty-four suicides per 100,000 inmates for calendar year 2008, based on an in-state inmate 28 population of 156,352. 87 1 inmate deaths he reviewed. See id. at 9-13. Dr. Shansky testified that this is an 2 extraordinarily high rate, and that in the Illinois prison system he would see extreme 3 departures from the standard of care in only five to ten percent of inmate deaths. Rep. Tr. at 4 428:9-17. According to Dr. Shansky, there was no question that a number of the lapses were 5 related to crowding. Id. at 427:17-428:4, 430:21-431:3. 6 Defendants presented evidence that California had the fourteenth lowest average 7 annual illness mortality [rate] per 100,000 state prisoners from 2001 to 2004 in the United 8 States. Rep. Tr. at 1272:12-21 (Mumola). However, these statistics failed to control for 9 demographics of each state s inmate population; the statistics are therefore of limited value in 10 comparing states. Aug. 27, 2008 Reingold Report ¶¶ 10-11, 15, 24. Furthermore, California 11 has the fourth lowest death rate among all fifty states, and the fifth lowest after controlling 12 for age. Id. ¶ 12. Thus, while California has a very low death rate for its general 13 population, its death rate for state prisoners is relatively higher. Id. ¶ 14. In any event, 14 serious deficiencies continue to exist in the California prison system such that California 15 inmates are not receiving adequate care. This is true regardless of where California might 16 rank in a valid comparison of inmate death rates among the states. 17 H. 18 Based on their observations of crowded conditions in California s prisons, including Expert Opinions Regarding Causation 19 those discussed above, as well as on their extensive experience in working in or managing 20 crowded prisons,54 seven experts testified that crowding is the primary cause of California s 21 inability to provide constitutionally adequate medical and mental health care to its inmates.55 22 Four of the experts are current or former state prison system administrators who have led 23 correctional agencies in five states, including California. Jeanne Woodford who worked 24 for the CDCR for twenty-seven years in various capacities, including as warden at San 25 26 54 E.g., Nov. 9, 2007 Woodford Report ¶ 7; Nov. 9, 2007 Scott Report ¶¶ 4-5; Aug. 15, 2008 Lehman Report ¶ 7; Rep. Tr. at 263:24-267:12 (Lehman); id. at 209:9-14 (Beard). 55 As we have previously explained, while the primary cause issue is ultimately a 27 question of law for the three-judge court to decide, the Federal Rules of Evidence allow experts to express opinions that embrace the ultimate issue in a case, including the issue of 28 causation. Nov. 3, 2008 Order at 10-11 (citing Fed. R. Evid. 704(a) and other authority). 88 1 Quentin and as acting Secretary of the CDCR, Nov. 9, 2007 Woodford Report ¶ 1 testified 2 that she absolutely believe[s] the primary cause is overcrowding. Rep. Tr. at 376:3-9; see 3 also id. at 383:4-10; Nov. 9, 2007 Woodford Report ¶ 6. Doyle Wayne Scott who worked 4 for thirty years for the Texas Department of Criminal Justice, the second largest prison 5 system in the United States after California, including five years as its Executive Director, 6 and has served as an expert consultant to the National Institute of Corrections and seven 7 prison systems throughout the United States and Puerto Rico, Nov. 9, 2007 Scott Report 8 ¶¶ 1, 4 similarly testified that: 9 10 11 12 13 14 Everything revolves around overcrowding. The deficiencies in the classification plan, the deficiencies in the unavailability of staff because they are doing other tasks associated with overcrowding problems to do onsite medical appointments or offsite medical appointments, the wear and tear on the infrastructure. I know there have been electrical outages because of the overload that the large number of offenders is causing at institutions. There s also been water problems at a number of the institutions, and I think the Governor s proclamation clearly described a lot of those issues. 15 Rep. Tr. at 152:6-15. Director Scott therefore opined that overcrowding is the primary 16 cause of the medical and mental health care violations in California prisons. Id. at 152:1-6; 17 see also Nov. 9, 2007 Scott Report ¶ 80. Joseph Lehman who has over thirty-five years of 18 experience in corrections, including fifteen combined years as head of corrections in 19 Pennsylvania, Washington, and Maine, Aug. 15, 2008 Lehman Report ¶ 1 also rendered his 20 expert opinion that crowding is the primary cause of the inability to provide [medical and 21 mental health] services. It s overwhelming the system both in terms of sheer numbers, in 22 terms of the space available, in terms of providing healthcare. Rep. Tr. at 270:25-271:6. 23 Likewise, Jeffrey Beard a licensed psychologist who has worked for the Pennsylvania 24 Department of Corrections for over thirty-six years, including serving as its Secretary since 25 2001, id. at 200:15-201:7 testified that, in his opinion, the biggest inhibiting factor right 26 now in California being able to deliver appropriate mental health and medical care is the 27 severe overcrowding of [the] system. Id. at 219:7-10. 28 89 1 At least three of these four experts who had headed state prison systems had never 2 before testified on behalf of a prisoner, and at least two of them were not paid for their time 3 as experts in this case. Rep. Tr. at 153:12-14 (Scott) (never testified on behalf of a prisoner 4 or class of prisoners); id. at 230:2-10 (Beard) (never testified for plaintiffs in thirty-six years 5 as a corrections professional, and not paid for testimony in this case); id. at 273:6-10 6 (Lehman) (never testified on behalf of prisoners in thirty-five years of experience); id. at 7 385:12-14 (Woodford) (not paid for testimony in this case). They decided to testify on 8 plaintiffs behalf in this case because the situation in California is so egregious, id. at 9 273:11-12 (Lehman); and because the prisons aren t safe, nobody seems to be willing to 10 step up to the plate and fix the problem, and if there s anything I can do to help see that 11 California moves in [the right] direction . . . that s why I m here today, id. at 231:13-20 12 (Beard). Secretary Woodford, the former warden at San Quentin and acting Secretary of the 13 CDCR, explained that she testified: 14 15 16 because I truly believe that we can do better than we are in California. I think it s unbelievable that in this state that we have the kind of overcrowded conditions that we have; that we do little or nothing to prepare people for the return to society in spite of the fact that we parole 10,000 people a month from our prison system. 17 18 19 20 21 22 And I absolutely believe that we make people worse, and that we are not meeting public safety by the way we treat people. And that I believe overcrowding is prohibiting us from providing quality medical care and mental healthcare to inmates in our system. And for California to be in the shape that it s in is just unbelievable. 23 Id. at 385:17-386:5 (Woodford). 24 In addition to these present or former heads of state prison systems, three other experts 25 testified on plaintiffs behalf that crowding is the primary cause of the constitutional 26 violations at issue in Plata and Coleman. Dr. Ronald Shansky a physician who has worked 27 primarily in correctional health care for over thirty-six years, including twelve years as 28 Medical Director of the Illinois Department of Corrections and five years as a medical 90 1 consultant to the CDCR, and who has been involved with over two dozen other correctional 2 systems as either a court-appointed expert/monitor/special master or as a consultant retained 3 by the correctional system, including five years as a court-appointed receiver of the District 4 of Columbia Jail Medical and Mental Health Program, Nov. 9, 2007 Shansky Report ¶¶ 2-3 5 explained that: 6 7 8 9 the CDCR s medical care delivery system cannot provide a constitutional level of care because the prison system incarcerates far more prisoners than can be adequately treated with the resources, staffing and facilities available in the CDCR. In short, it is my opinion that overcrowding is the primary cause of the constitutional violations in the CDCR for Plata class members. 10 Sept. 10, 2008 Shansky 2d Supp. Report ¶ 7; see also Nov. 9, 2007 Shansky Report 11 ¶¶ 136-38. Dr. Shansky is confident in his conclusion. Rep. Tr. at 423:8-14. 12 Dr. Craig Haney a professor of psychology at the University of California, Santa 13 Cruz, who has studied the psychological effects of living and working in institutional 14 environments for thirty-five years and has toured, inspected, and analyzed conditions of 15 confinement in prisons in twenty states, three maximum security federal prisons, and prisons 16 in five other countries, Aug. 15, 2008 Haney Report ¶¶ 1-3 similarly testified that: 17 18 19 20 Because of the tremendous importance of overcrowding and its impact on virtually every aspect of prison life, it is my opinion that it is the primary cause of the continuing constitutional violations that plague the California prison system, including the CDCR s inability to provide medical and mental health care for state prisoners that meets the relevant constitutional minimum standards. 21 Id. ¶ 17; see also id. ¶ 364 (overcrowding is a crisis that now consumes the CDCR and 22 prevents it from discharging its constitutional responsibilities ). Dr. Haney further explained 23 that: 24 25 26 27 28 I don t believe in a system this overcrowded at this magnitude of overcrowding with overcrowding as widespread as it has been in California for as long a period that it has been that there s any other plausible or credible explanation for the failure of the system to provide constitutionally-adequate mental healthcare. The court s been monitoring this issue for many, many years. There have been many, many court orders, and there have been 91 1 2 3 4 5 many activities that have been engaged in in trying to bring this system s mental health care delivery into constitutional compliance. In the face of all of those efforts there has been this overwhelming overcrowding problem of such a degree, magnitude and duration that it has incapacitated the system s ability to deliver constitutionally-adequate care. 6 Rep. Tr. at 317:18-318:7. 7 Dr. Pablo Stewart a licensed psychiatrist and clinical professor of psychiatry at the 8 University of California, San Francisco, with over twenty years of experience in correctional 9 psychiatry, including service as a court-appointed expert in several federal class action 10 lawsuits concerning the delivery of mental health care in prisons and jails, Nov. 9, 2007 11 Stewart Report ¶¶ 1-15 testified that the conclusion that overcrowding is the primary 12 cause of the constitutional violations in Coleman is inescapable. See id. ¶ 196; Aug. 15, 13 2008 Stewart Supp. Report ¶ 111. Dr. Stewart s opinion is predicated on the persistent 14 nature of the constitutional violations in Coleman: 15 16 17 18 19 20 21 22 23 [T]aken together, the range of Constitutional violations . . . including inadequate suicide monitoring and prevention, inability to timely access appropriate levels of care, inability to timely access mental health clinicians due to staffing shortage, and inadequate medication management practices are unusual in a system that has been under Court supervision for more than ten years. These serious, dangerous violations this late in the remedial process are typical indicators of a system plagued by severe overcrowding. In a non-crowded system, the Constitutional violations are more readily addressed by such interventions as increased staff and increased programming. However, in a system overwhelmed by crowding, these traditional remedies are woefully inadequate. This appears to be the case in the CDCR where remedial efforts have resulted in significant expansions of staffing and programming activities, yet the constitutional violations persist or even worsen. 24 Id. ¶ 112. Dr. Stewart s opinion is also based on the fact that the percentage of persons with 25 serious mental illness in the CDCR is increasing faster than the overall CDCR population, a 26 phenomenon that is typical of overcrowded systems because . . . overcrowding creates new 27 mental health needs and exacerbates existing mental health needs. Id. ¶¶ 114, 116. Finally, 28 Dr. Stewart found that: 92 1 The causal link between overcrowding and unconstitutional mental health care is clear and direct in the many CDCR housing units where space shortages from overcrowding directly result in long-term living arrangements that are harmful to the mental health of Coleman class members. . . . These same harsh conditions, as discussed earlier, also increase the demand for mental health services in the general population who, in a properly operating, not overcrowded system, would not need mental health services. Isolation, seclusion, idleness, violence, fear and stress plague the prisoners in the CDCR as a direct result of overcrowding. These conditions exacerbate mental illness and are serious barriers to the provision of minimally adequate mental health and medical care. 2 3 4 5 6 7 8 Id. ¶ 117. 9 Defendants expert Dr. David Thomas an ophthalmologist for almost forty years 10 who served in various capacities at the Florida Department of Corrections for nine years, 11 most recently as Assistant Secretary for Health Services and Director of Health Services, and 12 who now serves as a professor of surgery and correctional medicine at Nova Southeastern 13 University in Fort Lauderdale, Florida, Nov. 9, 2007 Thomas Report ¶¶ 1-2 was the only 14 expert who testified that crowding was not the primary cause of constitutional deficiencies in 15 the delivery of medical care in California s prisons. See, e.g., Rep. Tr. at 1217:11-13. 16 Instead, Dr. Thomas testified that the single most important item in achieving a sound 17 Constitutional level of care is a culture that fosters providing care at that level. Nov. 9, 18 2007 Thomas Report ¶ 11 (emphasis omitted). In his opinion, the empowerment of [health 19 care] staff unlike in the past, when security services dominated the prison system and 20 program services existed only at the whim of security services is the crux of having a 21 constitutional level of health care. Addendum to Thomas Report ¶ 1. He further explained 22 that: 23 24 25 26 The culture was such prior to appointing of the Receiver that this was a security-driven system without regard for any other programs or any other constitutional requirements. Since the Receiver has been appointed, . . . there is clear indication that the culture is shifting in the department to understand the need for a correctional healthcare system that works on a constitutional level of healthcare. 27 Rep. Tr. at 1215:21-1216:3. According to Dr. Thomas, a constitutional system of delivering 28 medical care cannot be developed without the change away from a custody-oriented culture 93 1 that is now underway, and a reduction in crowding might make it easier to develop such a 2 system, but it is not necessary and will not, without more, lead to a constitutionally adequate 3 system of care. See, e.g., Addendum to Thomas Report ¶ 4. 4 We find the testimony of Dr. Thomas to be unpersuasive for several reasons. First, 5 Dr. Thomas s testimony that reducing crowding will not, without more, remedy the 6 constitutional violations at issue in Plata does not mean that crowding is not the primary 7 cause of those violations. Indeed, we find that reducing crowding is a necessary but not 8 sufficient condition for eliminating the constitutional deficiencies in the provision of medical 9 care to California s inmate population. Other steps will be necessary to fully remedy the 10 deficiencies in the CDCR s medical and mental health care services. Nonetheless, a problem 11 that has multiple causes will ordinarily still have a primary cause. As Dr. Shansky explained, 12 Reducing overcrowding is not a panacea, but crowding is the primary cause of the ongoing 13 inadequate medical care in the CDCR system. Overcrowding is the one factor that 14 negatively impacts almost every other matter that must be addressed to create a minimally 15 adequate medical care delivery system for California s prisons. Sept. 10, 2008 Shansky 2d 16 Supp. Report ¶ 9; see also supra Section IV.A. 17 Second, as Dr. Beard testified, a culture that allows custodial interference with the 18 delivery of care is problematic, but you have to realize that the culture grew out of the 19 overcrowding. Rep. Tr. at 221:17-222:9. Crowded conditions force prison administrators 20 to take a strong custodial approach. . . . They have to rely on the lockdowns. They have to 21 rely on guns, gas, those kinds of things, to control the prisons so they re safe for the staff and 22 for their inmates. Id. at 222:14-21. Thus, although we agree with Dr. Thomas that a 23 custody-dominated culture is a barrier to delivering constitutionally adequate care, we also 24 agree with Dr. Beard that [i]f you try to change the culture, you can t. You can t change the 25 culture until you reduce the population and can make the institution safe. Id. at 222:22-24. 26 Consequently, it is crowding and not culture that is the primary cause of the unconstitutional 27 system of health care delivery in California s prisons. 28 94 1 Third, we give less weight to the testimony of Dr. Thomas because he formed his 2 opinions and drafted his initial report before visiting even a single prison in California. Id. at 3 1220:20-22 (Thomas). Although he subsequently visited eight prisons and opined that those 4 visits supported his initial views, he took no notes during or after those tours; did not make 5 any audio or video recordings during the tours; reviewed fewer than ten medical records at 6 each prison and could not recall any details of any of the medical files he reviewed; and did 7 not recall how many staff members he talked to at each prison or whether he asked the staff 8 members at each prison any of the same questions. Id. at 1228:17-1229:3, 1229:21-1231:9, 9 1236:1-4, 1240:2-14 (Thomas). 10 Fourth, some of the testimony by Dr. Thomas was both internally inconsistent and 11 patently incredible. For instance, Dr. Thomas testified that he believed all eight prisons he 12 visited were richly staffed, yet he earlier testified that outcome measurements of work 13 study programs should be used to determine staffing ratios and he had not conducted or seen 14 any such studies of the California prison system. Id. at 1197:18-1198:6, 1251:2-17. He also 15 suggested that providing treatment in a men s restroom would be appropriate because one 16 has to be creative . . . in corrections, and that treatment could also be provided in closets, id. 17 at 1223:7-12, 1226:8-15, although he provided other, more plausible suggestions, such as 18 using space more frequently on weekends or adding modular buildings. 19 Finally, even if we were to credit Dr. Thomas s opinions in their entirety, we find 20 such opinions to be overwhelmingly outweighed by the testimony of the numerous other, 21 more qualified experts cited above. Defendants argue that the opinions of some of plaintiffs 22 experts must be discounted because of the role played by plaintiffs counsel in drafting the 23 expert reports. However, upon review of all of the relevant testimony, we are convinced that 24 the opinions contained in the expert reports are those of the experts themselves, and that 25 plaintiffs counsel did not impermissibly influence any of the experts opinions. See, e.g., id. 26 at 181:16-182:5 (Scott) (testifying that the opinions in his expert report were mine and only 27 mine and that, before signing his reports, he reviewed every word, [a]ll the way down to 28 the grammatical remarks, to ensure that they accurately reflected his opinions); Pls. Opp n 95 1 to Defs. Mot. in Limine No. 9 to Exclude Expert Reports at 2-5 (citing deposition testimony 2 by Director Scott, Dr. Shansky, Dr. Stewart, Secretary Lehman, and Secretary Woodford 3 concerning the preparation of their expert reports, including that counsel never asked the 4 experts to change any of their opinions); Marek v. Moore, 171 F.R.D. 298, 300-302 (D. Kan. 5 1997) (counsel s assistance in the preparation of expert reports is proper as long as the 6 reports reflect the testimony of the expert and are signed by the experts). 7 Defendants also suggest that the court should discredit the testimony of experts who 8 lack medical training and have never practiced correctional medicine. However, a medical 9 background is not required to opine on the cause of constitutional violations in the delivery 10 of medical care in a correctional environment, and plaintiffs experts wealth of experience in 11 managing prisons and prison systems, including experience in doing so under crowded 12 conditions, establishes their ability to form an expert opinion on that subject. In fact, the 13 CDCR has previously recognized the expertise of several of plaintiffs experts. In addition to 14 employing Secretary Woodford for twenty-seven years, culminating in her appointment as 15 acting Secretary, Nov. 9, 2007 Woodford Report ¶ 1, the CDCR named Dr. Beard and 16 Secretary Lehman to its Expert Panel on Adult Offender and Recidivism Reduction 17 Programming and employed Dr. Shansky as a medical consultant for five years. Ex. P2 at ii; 18 Rep. Tr. at 210:15-25 (Beard); Aug. 15, 2008 Lehman Report ¶ 4; Nov. 9, 2007 Shansky 19 Report ¶ 2. As noted earlier in this opinion and order, the CDCR employed Dr. Shansky as 20 its own medical expert during the Plata evidentiary hearings regarding whether a 21 receivership was necessary. Thus, we reject defendants suggestion that plaintiffs experts 22 are not qualified. To the contrary, we find their expertise far outweighs that of Dr. Thomas. 23 Defendants also offered mental health expert Dr. Ira Packer in support of their 24 position in Coleman. Dr. Packer who is board-certified in forensic psychology and has 25 worked for over twenty-eight years in correctional and forensic psychology, including as 26 Deputy Mental Health Program Director for the Massachusetts Department of Corrections 27 and as Assistant Commissioner for Forensic Mental Health in the Massachusetts Department 28 of Mental Health, Dec. 10, 2007 Packer Report at 4-6 testified that, with one exception, 96 1 crowding was not the primary cause of the constitutional violations with respect to mental 2 health care. Id. at 23-24. The exception was that, like all of plaintiffs experts, Dr. Packer 3 concluded that crowding is the primary cause of the particular difficulties in providing 4 services to the Coleman class at the reception centers, id. at 20; that issue is therefore 5 undisputed. As to mental health care delivery in other settings, Dr. Packer opined that 6 overcrowding in CDCR significantly contributes to the difficulties in providing adequate 7 mental health services, but is not the primary cause of the deficiencies. Id. at 8 (emphasis in 8 original). In Dr. Packer s opinion, the primary cause of the constitutionally inadequate 9 mental health care in California s prisons is that California now has many more acutely 10 mentally ill individuals and at a level of more severity than had been anticipated when the 11 prisons were built, and that the existing prison space was not designed to meet the needs 12 of a mentally ill population. Rep. Tr. at 1079:11-1080:4; see also Dec. 10, 2007 Packer 13 Report at 8-9.56 14 What Dr. Packer is actually saying is that lack of planning is the cause of the 15 overcrowding in California s prisons but that is not the question before us. Regardless of 16 the cause of the overcrowding, that condition is defined in terms of the capacity of the 17 prisons, and that capacity simply is not there. Dr. Packer s testimony principally supports 18 our conclusion that crowding is the primary cause of the constitutional violations in the 19 delivery of mental health care. For example, Dr. Packer testified that if crowding were 20 defined as not having enough mental health beds to serve the current population, then 21 crowding would be the primary cause of the ongoing mental health care violations in 22 California s prisons. Rep. Tr. at 1093:25-1094:6. Clear evidence establishes that, due to 23 crowding, there is insufficient room in California s prisons for necessary additional mental 24 health care beds and treatment space. Accordingly, Dr. Packer s opinion is congruent with 25 56 Dr. Packer also opined that difficulties with maintaining adequate medical records are a direct effect of overcrowding, as the number of charts in the institutions is proportional 26 to the population, Dec. 10, 2007 Packer Report at 19, and he testified that the medical record system is a paper system, and the prison is simply not able to keep up with the amount 27 of work and volume that s required in order to maintain an appropriate medical record system without going to an electronic process, which is not yet in place, Rep. Tr. at 28 1080:7-11. 97 1 our finding that crowding is the primary cause of the ongoing constitutional violations in 2 Coleman. 3 Additionally, while Dr. Packer s opinion on the unanticipated nature of the influx of 4 mentally ill prisoners into the correctional system might have had some merit at the time of 5 the Coleman trial in 1993, or even at the beginning of the Coleman remedial phase in 1996, it 6 is less persuasive at this late stage in the Coleman remedial process. The Coleman court has, 7 for almost a decade, directed defendants to make adequate projections of the size of the 8 mentally ill inmate population so that they can appropriately plan for that population s needs. 9 The fact that it has taken defendants years to comply with those orders does not render the 10 increasing size of the Coleman class unanticipated. As we explained when we denied 11 defendants motion for summary judgment: 12 Defendants inability or unwillingness to tackle the problem of the increasing prison population does not support the contention that overcrowding is not the primary cause of the unconstitutional delivery of medical or mental health care. It simply helps explain why overcrowding exists and has now become a problem that may be the primary cause of the constitutional violation. 13 14 15 16 Nov. 3, 2008 Order at 9-10. In fact, the efforts defendants have made since the Coleman 17 remedial process began, combined with the serious ongoing problems we have discussed in 18 this opinion, only bolster the inescapable conclusion that crowding is the primary cause of 19 defendants failure to deliver constitutionally adequate mental health care in their prison 20 system. 21 All of the steps defendants have taken under the Plata court s supervision, as well as 22 the steps taken under the Coleman court s supervision, have failed to remedy the 23 constitutional deficiencies. The crushing inmate population has strained already severely 24 limited space resources to the breaking point, and crowding is causing an increasing demand 25 for medical and mental health care services, a demand with which defendants are simply 26 unable to keep pace. It also, as the expert witnesses repeatedly told us, has created numerous 27 barriers to the delivery of constitutionally adequate medical and mental health care. 28 98 1 I. 2 On the basis of the clear and convincing, indeed overwhelming and overwhelmingly Findings and Conclusions 3 persuasive, evidence described above, we conclude that crowding is the primary cause of the 4 state s unconstitutional failure to provide adequate medical and mental health care to 5 California prisoners. Such is the opinion as well of some of the nation s foremost prison 6 administrators, who testified that they have never previously witnessed such appalling prison 7 conditions and that overcrowding is not only the primary cause of the constitutionally 8 inadequate medical and mental health care in California s prisons, but also that until the 9 problem of overcrowding is overcome it will be impossible to provide constitutionally 10 compliant care to California s prison population. No credible evidence to the contrary was 11 presented by defendants. 12 The evidence conclusively demonstrates the many ways in which crowding prevents 13 the state from providing constitutionally adequate medical and mental health care in its 14 prison system. Prison overcrowding has created a state of emergency in California s prisons, 15 as the Governor has proclaimed. It forces prison administrators to devote most of their 16 energy to addressing crises and has overwhelmed the prison system s management 17 infrastructure. Crowding of reception centers at levels approaching 300% design capacity 18 prevents the state from identifying the medical problems of entering inmates, and makes it 19 impossible to provide necessary medical and mental health care to incoming inmates, who 20 routinely remain in reception centers for more than sixty days and may serve their entire 21 sentence there. Crowding has also left the California prison system without the space, beds, 22 and medical, mental health, and custodial staff required to provide constitutionally adequate 23 medical and mental health care in all parts of the prison system, and has prevented proper 24 classification of inmates and appropriate housing according to their needs. Furthermore, 25 crowding has created conditions of confinement that contribute to the spread of disease, and 26 it requires the increased use of lockdowns as a method of prison control, further impeding the 27 prison authorities ability to provide needed medical and mental health care. In addition, 28 crowding has prevented the development of an adequate medical records system. The 99 1 consequences of crowding are often dangerous, and on many occasions fatal. Crowding 2 contributes to an alarming number of extreme departures from the standard of care and an 3 unacceptably high number of inmate deaths that are preventable or possibly preventable. 4 Likewise, crowding worsens many of the risk factors for suicide among California inmates 5 and increases the prevalence and acuity of mental illness throughout the prison system. 6 The history of the individual Plata and Coleman cases further demonstrates the role of 7 crowding in causing the constitutional violations at issue here. The extensive remedial 8 efforts in Plata over the last seven years, beginning with the stipulated relief and culminating 9 in the Receivership, have failed to bring the California prison system s medical care into 10 constitutional compliance. Likewise, fourteen years of remedial efforts in Coleman, directed 11 at every aspect of the mental health care problem, except crowding, have failed to ensure that 12 California prisoners have access to constitutionally adequate mental health care. In fact, by 13 2006, the progress that had been achieved during more than a decade of remedial work in 14 Coleman was being lost because of the inexorably expanding demand for services resulting 15 from the bulging population. Ex. D1108 at DEFS060303. 16 The only conclusion that can be drawn from the wealth of clear and convincing 17 evidence before this court is that the unconstitutional denial of adequate medical and mental 18 health care to California s inmates is caused, first and foremost, by the unprecedented 19 crowding in California s prisons. In reaching this conclusion, we need not, and do not, 20 conclude that crowding is the exclusive cause of those violations. We recognize that other 21 factors contribute to California s failure to provide its inmates with constitutionally adequate 22 medical and mental health care, and that reducing crowding in the prisons will not, without 23 more, completely cure the constitutional violations the Plata and Coleman courts have 24 sought to remedy. We need not find that crowding is the only cause, but simply that it is the 25 primary one. See supra Sections IV, IV.H. In the end, we agree with the former Executive 26 Director of the Texas Department of Corrections Doyle Wayne Scott, who testified that 27 [e]verything revolves around overcrowding, Rep. Tr. at 152:6. 28 100 1 In short, while other factors contribute to the unconstitutional state of the California 2 prisons medical and mental health care system, and while there are other steps the state must 3 take to meet its constitutional obligations, clear and convincing evidence establishes that 4 crowding is the primary cause of the constitutional violations, and that, therefore, this court 5 must consider, as we do below, what actions we may order be taken to remedy that condition. 6 7 V. NO OTHER RELIEF 8 The Governor has proclaimed that crowding in prisons constitutes an emergency that 9 poses a substantial risk to CDCR staff, inmates, and the general public, and that immediate 10 action is necessary to prevent death and harm caused by California s severe prison 11 overcrowding. Ex. P1 at 1-2, 6. Because crowding is the primary cause of the state s 12 inability to provide its inmates with constitutionally adequate medical and mental health care, 13 an order requiring a reduction in prison population is the most obvious and direct method by 14 which to bring the California prison system into constitutional compliance. 15 However, the PLRA makes such an order the remedy of last resort. H.R. Rep. No. 16 104-21, at 25 (1995). Before entering any prisoner release order, we must find that no other 17 relief could remedy the constitutional violations at issue here. 18 U.S.C. § 3626(a)(3)(E)(ii). 18 In context, it is clear that other relief refers to any form of relief other than a prisoner 19 release order. See id. § 3626(a)(3)(E) ( The three-judge court shall enter a prisoner release 20 order only if . . . no other relief will remedy the violation of the Federal right. ). In other 21 words, we must first determine whether the unconstitutional denial of adequate medical and 22 mental health care to California s prisoners can be remedied through an order that does not 23 have the purpose or effect of reducing or limiting the prison population and that does not 24 direct[] the release from or nonadmission of prisoners to a prison. Id. § 3626(g)(4). The 25 PLRA does not require that a prisoner release order, on its own, will necessarily resolve the 26 constitutional deficiencies found to exist in Plata and Coleman. All that the PLRA requires 27 is that a prisoner release order be a necessary part of any successful remedy. If all other 28 101 1 potential remedies will be futile in the absence of a prisoner release order, no other relief 2 will remedy the violation. Id. § 3626(a)(3)(E)(ii). 3 We conclude that the constitutional deficiencies in the California prison system s 4 medical and mental health system cannot be resolved in the absence of a prisoner release 5 order. Clear and convincing evidence establishes that none of the available alternatives to 6 such an order, including the continued efforts of the Plata Receiver and the Coleman Special 7 Master, can bring the California prison system into constitutional compliance within a 8 reasonable period of time. We agree with the numerous experts who testified that a prisoner 9 release order is a prerequisite to providing constitutionally adequate medical and mental 10 health care to California prisoners. Although the CDCR and the Receiver have implemented 11 a number of remedial programs as a result of the Plata and Coleman litigation, and 12 defendants have sought in various ways to improve the medical and mental health care 13 provided in California s prisons, these efforts cannot succeed in the absence of a prisoner 14 release order. 15 16 17 18 A. Alternatives to a Prisoner Release Order 1. Inadequacy of Construction as a Remedy a. Prison Construction In considering other alternatives to a prisoner release order, we first look to whether 19 the state has a feasible prison construction plan that would render a prisoner release order 20 unnecessary; if so, equity, if not law, would require that we refrain from entering that order. 21 In a case involving overcrowding, the construction of additional prisons always provides a 22 theoretical remedy because more prisons would necessarily reduce or eliminate 23 overcrowding. To construe the PLRA to preclude the entry of a prisoner release order based 24 on no more than such a theoretical remedy, however, would transform the conditions under 25 which the PLRA permits prisoner release orders into an absolute bar on such orders. In 26 short, it would eliminate overcrowding as a basis for a prisoner release order, and thus 27 prisoner release orders themselves, because the state could, in theory, always build more 28 prisons. Thus, what we must determine is not whether building prisons could solve the 102 1 problem, but whether prison construction offers an actual, feasible, sufficiently timely 2 remedy for the unconstitutional state of medical and mental health care in California s 3 prisons. Here, California has no plans to construct additional prisons in the near future and 4 has not suggested that it does. As a result, we need not consider further the construction of 5 additional prisons as an alternative remedy. 6 7 b. Construction of Re-entry Facilities The next question is whether building re-entry facilities could serve to reduce prison 8 overcrowding. The answer is that it could, if enough were constructed and if enough 9 prisoners were transferred to them. Thus, whether the state determined to build such 10 facilities voluntarily, or whether a court ordered or approved such construction, we would not 11 issue the type of order plaintiffs seek if the planned construction, like any prison-related 12 construction, offered an actual, feasible, and timely remedy that would render the relief 13 sought here unnecessary. Defendants point to only one existing proposal that might offer 14 such a partial remedy: construction of the additional re-entry facilities authorized by 15 Assembly Bill 900 ( AB 900 ). However, as we explain below, this construction plan does 16 not provide a feasible alternative to the order sought here. More than two years after AB 900 17 was signed into law, any reduction in the crowding of California s prisons resulting from the 18 construction of the AB 900 re-entry facilities remains years away and would in any event 19 likely not provide adequate relief. 20 In the first place, AB 900 construction has already been delayed for more than two 21 years due to the absence of funding. At the start of trial not a single facility had been 22 constructed under AB 900. E.g., Rep. Tr. at 1679:18-23 (Cate); id. at 2460:25-2465:7 23 (Spitzer); Ex. P750 (Sept. 17, 2008 CDCR press release following legislature s failure to 24 pass clean-up language to AB 900); Sept. 3, 2008 Hysen Dep. at 31:15-20 (state has not even 25 reached the preliminary-plan stage for any in-fill or re-entry construction under AB 900). 26 As far as we are aware, it remains the case today, eight months later, that there is no funding 27 for AB 900 and no ground has been broken on the AB 900-authorized re-entry facilities. 28 103 1 Second, even if funding were secured in the near future, other practical concerns 2 would lead to significant additional delays. Deborah Hysen, the CDCR s chief deputy 3 secretary for facility planning and construction management, Sept. 3, 2008 Hysen Dep. at 4 14:11-14, explained that environmental impact reviews, which have not yet been completed 5 for any of the proposed building sites, could potentially hang up projects for years, id. at 6 38:8-16, 56:1-2. Delays would also result from the need to obtain necessary construction 7 materials, id. at 38:17-25; permit public comment at each phase of construction, id. at 8 111:9-15; and provide for seismic retrofitting, id. at 112:16-21. Challenges in locating space 9 for re-entry facilities are also likely to significantly delay or prevent full implementation of 10 AB 900. Only one location, for 500 re-entering prisoners, has been secured, id. at 118:19-21, 11 but many obstacles to construction remain, and securing sites for other re-entry facilities is 12 likely to prove more difficult because of community opposition. E.g., Rep. Tr. at 221:11-16 13 (Beard); id. at 2750:16-2751:10 (Runner); id. at 2793:8-2794:3 (Meyer); Nov. 9, 2007 14 Woodford Report ¶ 36. Moreover, although some sites may offer up a renovation 15 alternative, most of the sites under consideration by the CDCR are raw land. Sept. 3, 16 2008 Hysen Dep. at 120:13-15. As to the latter sites, the CDCR itself estimates that we 17 could be looking at several years between the time that we make this recommendation to 18 acquire the land, and occupancy. Id. at 120:15-18. 19 Accordingly, it will be years before any re-entry facility construction pursuant to 20 AB 900 will be completed. It is thus clear that the proposed construction of re-entry facilities 21 cannot bring the sort of immediate action that the Governor has conceded is necessary to 22 resolve the present crisis caused by prison crowding. Ex. P1 at 6. As Dr. Haney explained, 23 prisoners in California s prisons with medical and mental health needs face emergency-like 24 conditions. Rep. Tr. at 945:25. 25 26 27 28 There are people, prisoners, suffering throughout the entire prison system, mentally ill and medically ill prisoners who are not able to get the level of care they need. . . . Those things are urgent problems, and only a solution which can be brought to fruition quickly can address the kind of immediate suffering which is taking place throughout the system which I saw and other experts saw as well. 104 1 Id. at 946:1-9. Any beneficial effects of defendants planned re-entry facility construction 2 are simply too distant to make such construction a meaningful remedy for the emergency-like 3 conditions in California s prisons. Moreover, it is unlikely that the number of re-entry 4 facilities that would be constructed would be sufficient to remedy the overcrowding 5 problems in any event. A prisoner release order would thus be necessary as well. 6 Given the serious inadequacy of the state s only existing facilities construction plan, it 7 is also clear that no other, yet-to-be-developed plan could remedy the constitutional 8 violations here within a reasonable period of time. The evidence before the court is thus 9 clear and convincing that the state has no feasible plan to remedy the constitutional violations 10 at issue in Plata and Coleman through either prison construction or re-entry facility 11 construction, and that such construction does not provide a meaningful alternative to the type 12 of order sought by plaintiffs in this case. 13 14 c. Medical Facilities and Prison Expansion Besides re-entry facility construction, defendants identify two additional proposals to 15 increase the capacity of the prison system: the Plata Receiver s medical facility construction 16 plan and prison expansion through the construction of space for in-fill beds, as authorized by 17 AB 900. Rep. Tr. at 1689:10-18 (Cate). For reasons similar to those discussed above, we 18 conclude that neither the Receiver s medical facility construction plans nor the proposed 19 AB 900 in-fill beds prison expansion can remedy the constitutional violations at issue in 20 Plata and Coleman. Like the AB 900 re-entry facilities, these proposed facilities will not be 21 realized at any point in the near future. Furthermore, their funding is threatened by the 22 present fiscal crisis, and the proposed construction would in any event likely fall far short of 23 remedying the problems created by the crowding of California s prisons. 24 As with the proposed re-entry facilities, any overcrowding relief resulting from the 25 construction of medical facilities or the addition of in-fill bed space as a result of prison 26 expansion is years away, at best. The Plata Receiver initially planned to start construction of 27 the first site in February 2009 and to complete construction of necessary additional facilities 28 by July 2013. Ex. D1100 at 64-65 (Plata Receiver s Sept. 15, 2008 Ninth Quarterly Report). 105 1 To date, however, no construction has started and no funding has been secured. Likewise, as 2 noted already, there is no available funding for AB 900, no ground has been broken on 3 AB 900 construction, and no new beds in-fill or re-entry have been made available. 4 The delays are compounded by the fiscal crisis now facing the state, which makes the 5 completion of any new construction even more distant and unlikely. The Receiver and the 6 CDCR were until a month or so ago negotiating a potential agreement concerning the 7 construction of health-care-focused prison facilities that would have provided funding for 8 necessary healthcare construction through the California Infrastructure and Economic 9 Development Bank (I-Bank). However, the state ultimately declined to sign the agreement. 10 Ex. 1 to Defs. July 1, 2009 Response to Court s June 18, 2009 Order, filed in Coleman, at 1. 11 Because the fiscal crisis has required severe and significant cuts to vital State programs, 12 the state refused to enter into any agreement that would require[] the State to seek I-Bank 13 funding, or any other additional funding not previously appropriated by the California 14 Legislature. Id. Although defendants did state that they would use a significant but 15 unspecified portion of the funds allocated by the legislature in AB 900 to build appropriate 16 beds for inmates with disabilities and/or other health needs, id. at 2, there is no indication as 17 to when such funds will be made available; when construction might begin; or what part, if 18 any, of the constitutional inadequacies in delivering medical and mental health care to 19 California inmates might be remedied by such construction. Because we have received no 20 evidence on any of these questions, we cannot conclude that the state has any actual, feasible, 21 timely plans for such construction, which in any event would be unlikely to render a prisoner 22 release order unnecessary. 23 As the state s failure to sign the agreement demonstrates, the present fiscal crisis 24 makes any remedy that requires significant additional spending by the state chimerical the 25 state has said that it will not procure any new funds for prison hospital construction. Even if 26 AB 900 funding were secured in the near future, however, the practical concerns described 27 above in relation to re-entry facilities environmental impact reviews, materials 28 procurement, public comment, and seismic retrofitting would lead to delays in the 106 1 construction of medical facilities and prison expansion. Accordingly, like the proposed re2 entry facilities, neither the Receiver s constructions plan nor AB 900 prison expansion will 3 provide inmates with relief from the emergency conditions in California s prisons in a timely 4 fashion. 5 Beyond any funding and timeliness issues, we have no reason to believe that 6 defendants proposed expansion of prison facilities would reduce crowding significantly or 7 lead to any improvements in the delivery of medical and mental health care to California 8 inmates. The Plata Receiver has found that the in-fill bed plan proposed by the CDCR 9 includes allocations of clinical space that are wildly disparate and, in many cases obviously 10 inadequate, and that the CDCR s plan ignor[es] the real life differences in clinical 11 requirement[s] based on the characteristics of the patient population, security level and escort 12 officers requirements, the need for clinical privacy, equipment requirements, and other 13 critical factors. Ex. D1092 at 37 (Plata Receiver s May 15, 2007 Report Re: 14 Overcrowding). On a more fundamental level, the AB 900 in-fill construction plan 15 essentially is a prison expansion measure which increases the number of prison cells 16 without addressing the fundamental structural issues that have caused the crisis and that have 17 created unconstitutional conditions within the prisons. Nov. 9, 2007 Woodford Report ¶ 31. 18 According to Secretary Woodford: 19 20 21 22 23 24 [t]he so-called in-fill beds will cause more problems than they will solve. Many of California s prisons are so big that they are effectively unmanageable. Wardens and other administrators spend much of their time responding to crises, rather than fulfilling their responsibilities to provide adequate medical and mental health care. Unless these in-fill beds stand alone with their own administrative and support facilities, adding thousands of additional prisoners to already overburdened facilities will only compound the burdens imposed on prison administrators and line staff. 25 Id. ¶ 39. Similarly, Director Scott explained that, because the in-fill bed numbers in 26 defendants construction plan are based on housing overcrowding capacity rather than 27 design build capacity, 28 107 1 [p]risoners in the new facilities . . . might not initially be living in gymnasiums or hallways, as they are now, but they will still be overcrowded. California will be in the same position with the new beds as with the old, replicating the same conditions that led to inadequate staffing and treatment space, inadequate out-of-cell time, and overworked and overstressed staff and violent, frustrated prisoners. 2 3 4 5 Aug. 13, 2008 Scott Supp. Report ¶ 17. Thus, while the construction of in-fill beds would 6 reduce the use of bad beds, the principal effects of the overcrowding in California s prisons 7 would remain unaddressed. 8 9 d. Construction as a Means of Compliance Given all of the above problems, we are convinced that neither prison expansion, nor 10 re-entry or medical facilities construction, nor any other construction effort offers a 11 meaningful and timely remedy for the constitutional deficiencies in the delivery of prison 12 medical and mental health care caused by crowding. Although it might be theoretically 13 possible for California to build its way out of its prison overcrowding problem, it is not 14 practical to anticipate that the state will do so in a timely manner, if ever, given the time that 15 it takes and . . . the huge costs that it takes to do things like this. Rep. Tr. at 254:25-255:12 16 (Beard). Dr. Beard concluded that although construction should be part of a plan, if you try 17 to rely on that alone, you are probably never going to get there, because they haven t been 18 able to get there over the last 20 years. Id. at 256:4-8; see also, e.g., id. at 219:11-25 19 (Beard). 20 Nonetheless, because our order requires defendants to reduce the prison population to 21 a specified percentage of the prison system s design capacity, any additional capacity 22 provided by completed construction could help the state meets its obligations and might 23 allow it to increase the number of prisoners who could constitutionally be housed in the 24 prison system. In such case an adjustment as to the specific terms of the population 25 reduction order, although not to the percentage cap itself, might conceivably be appropriate.57 26 We see little prospect for such an occurrence, however, in the reasonably near future, and 27 57 Likewise, should for some reason the design capacity of California s adult prison institutions decrease, the CDCR would be required to reduce the absolute population of its 28 adult prison institutions by a greater number. 108 1 thus no prospect of remedying the constitutional violations in a timely manner, other than in 2 accordance with the order we issue below. 3 4 2. Inadequacy of Additional Hiring Defendants do not suggest that the constitutional deficiencies in the CDCR s system 5 of medical and mental health care could be remedied by hiring additional medical, mental 6 health, and custodial staff. This is not surprising, given the serious and ongoing difficulty in 7 filling vacant positions encountered in both the Plata and Coleman remedial proceedings. 8 See supra Sections II.A.5, II.B.2.c. Furthermore, as noted already, crowding itself seriously 9 impedes the recruitment and retention of medical and mental health care staff. The working 10 conditions for such personnel in California s overcrowded prisons are uninviting, and many 11 potential staff members are unwilling to work under them. See, e.g., Nov. 9, 2007 Stewart 12 Report ¶ 41; Nov. 9, 2007 Shansky Report ¶ 23. Even if staff could be hired, they would 13 have almost nowhere to work because CDCR s facilities lack the physical space required to 14 provide medical and mental health care. See, e.g., Rep. Tr. at 272:1-13 (Lehman); id. at 15 501:3-7 (Shansky). Thus, the evidence is clear and convincing that hiring additional staff 16 could not bring the CDCR s medical and mental health care into constitutional compliance in 17 the absence of a reduction in prison crowding. 18 19 20 3. Insufficiency of the Plata Receivership and Coleman Special Mastership We next consider whether the existing remedial efforts of the Plata and Coleman 21 courts provide an alternative form of relief that could remedy the constitutional violations at 22 issue in Plata and Coleman. Defendants argue that the delivery of medical and mental health 23 care has improved and continues to improve under the direction of the Plata Receiver and the 24 oversight of the Coleman Special Master. However, the Plata and Coleman courts are barred 25 by the PLRA from ordering any remedy that involves a reduction in the prison population, 26 18 U.S.C. § 3626(a)(3)(B), and the Plata Receiver and Coleman Special Master therefore 27 lack the most direct and effective means of eliminating the fundamental problems that result 28 from overcrowding, see supra Section IV. While improvements have been and continue to 109 1 be made, and the Plata and Coleman courts have continued their efforts during this three2 judge court proceeding, it is clear that the Receiver and the Special Master cannot remedy the 3 constitutional violations in the absence of a prisoner release order. 4 The Plata Receiver has determined that adequate care cannot be provided for the 5 current number of inmates at existing prisons and that additional capacity is required to 6 remedy the medical care deficiencies that exist in California s prison system. See, e.g., 7 Ex. D1133 at 27-28 (Plata Receiver s June 6, 2008 Turnaround Plan of Action). Defendants 8 correctly note that the Plata Receiver has stated that [f]ailure is not an option and that 9 [o]ver time the CDCR s medical delivery system will be raised to constitutional levels. 10 Ex. D1092 at 41 (Plata Receiver s May 15, 2007 Report Re: Overcrowding). However, the 11 Receiver also noted that the time this process will take, and the cost and the scope of 12 intrusion by the Federal Court cannot help but increase, and increase in a very significant 13 manner, if the scope and characteristics of CDCR overcrowding continue. Id. According to 14 the Receiver, the creation of a system that could adequately deliver medical care to all of the 15 inmates moving through the reception center at the California Institution for Men under the 16 present level of overcrowding could all but bankrupt the State of California and create a 17 medical delivery problem in [surrounding] counties because there may not be enough 18 competent clinicians to provide medical care for an unlimited number of State prisoners and 19 for the public also. Id. Even assuming that the Receiver s comments are somewhat 20 overstated, relying on the authority that he possesses to resolve the medical care crisis in the 21 absence of a population reduction order does not offer a feasible alternative. There is no 22 question that in the absence of a population reduction order a fair number of new prisons and 23 medical facilities would be required. We have already explained that such construction 24 could not be completed in a timely manner, even if the legislature were willing to fund it. 25 The history of the Coleman case demonstrates even more starkly the impossibility of 26 establishing a constitutionally adequate mental health care delivery system at current levels 27 of crowding. For almost a decade the Coleman court has issued specific orders directing 28 defendants to develop sufficient beds for the delivery of mental health care at each level of 110 1 the mental health care delivery system. Despite all of those orders, defendants have far too 2 few mental health care beds to meet present demand. The CDCR s recent refusal to sign the 3 agreement it negotiated with the Receiver makes compliance even more unlikely, as the state 4 had previously offered the agreement as its primary method of developing the needed mental 5 health beds. Likewise, the Coleman court has issued numerous orders directing defendants 6 to decrease the time required to transfer seriously mentally ill inmates, including those who 7 are suicidal or otherwise in crisis, to appropriate levels of supervised care, but wait lists 8 remain at every level. These are but two examples of the Coleman court s ongoing inability, 9 despite tremendous effort, to bring the prison mental health care system into constitutional 10 compliance. In light of this history, the evidence is clear and convincing that defendants are 11 simply unable to meet the escalating demand for resources caused by the overcrowding in 12 California s prisons. 13 Defendants argue that a prisoner release order will not fix the constitutional violations 14 in the delivery of mental health care because they will need to develop appropriate treatment 15 space and hire sufficient staff even if the total inmate population is reduced. Defendants 16 point to the Coleman Special Master s findings that [e]ven the release of 100,000 inmates 17 would likely leave the defendants with a largely unmitigated need to provide intensive 18 mental health services to program populations that would remain undiminished by a 19 reduction of some 19,000 [CCCMS] inmates, and that the release of 50,000 inmates would 20 probably not raise staffing resources into equilibrium with the mental health caseload. 21 Ex. D1292 at 15. 22 We agree with the Special Master that the population reduction order sought by 23 plaintiffs is not by itself a panacea, and that defendants efforts to provide constitutionally 24 adequate mental health care must go beyond reducing prison overcrowding. Obviously, 25 simply creating additional space would not solve the problem; prison authorities would be 26 required to ensure that the space is used to provide prisoners with professionally sound 27 medical and mental health treatment, to administer necessary medications to prisoners, and to 28 remove the other barriers to constitutionally adequate medical and mental health care created 111 1 by overcrowding. However, the defendants cannot remedy the ongoing constitutional 2 violations without significant relief from the overcrowded conditions. We find the Special 3 Master s statement about 100,000 inmates somewhat hyperbolic. The comment about 50,000 4 inmates more nearly approximates the remedy we deem appropriate given our obligation to 5 adopt the least intrusive remedy. Nevertheless, as he and we have both noted, additional 6 steps will be required after the prison population is reduced. We believe that the Special 7 Master will be able to provide significant assistance to the state in that respect. 8 It is apparent from the extraordinary efforts undertaken by the Special Master and the 9 Receiver, as well as the fundamental constitutional inadequacies in medical and mental 10 health care, that a reduction in the present crowding of the California prisons is necessary if 11 the efforts of the Plata Receiver and the Coleman Special Master to bring the medical and 12 mental health care in California s prisons into constitutional compliance are ever to succeed. 13 In the absence of a prisoner release order, all other remedial efforts will inevitably fail. 14 15 4. Other Proposals As noted in our discussion of prison construction, equitable concerns would prevent 16 us from entering a prisoner release order if the state had plans in place that would reduce the 17 crowding of California s prisons sufficiently to allow the remedying of the constitutional 18 violations in the near future. However, the evidence at trial was clear and convincing that 19 none of the state s existing plans can reduce the prisoner population to the extent necessary 20 to permit the CDCR to bring its prison medical and mental health systems into constitutional 21 compliance. 22 The state and one of the defendant-intervenors have suggested two different means of 23 reducing the prison population. The first is already being implemented by the state through 24 its program to transfer California inmates to facilities in other states. E.g., Kernan Trial Aff. 25 ¶¶ 16-17.58 We do not comment on the merits of this program, although we have doubts 26 58 Defendants out-of-state transfer program, if ordered by the court, would fall within the PLRA s definition of a prisoner release order, because it directs the release [of inmates] 27 from . . . a prison, 18 U.S.C. § 3626(g)(4) (emphasis added). An order requiring the transfer of prisoners from California s adult prison institutions to out-of-state prisons would be 28 functionally identical to an order requiring the transfer of prisoners from a single prison or 112 1 about its efficacy as applied to the mentally ill and question its possible adverse effect on 2 prisoners moved to a location far removed from their families and friends. Still, as of 3 August 29, 2008, approximately 4852 California inmates had been housed in out-of-state 4 institutions, and the CDCR had plans to transfer up to a total of 3000 additional inmates to 5 such sites. Cate Trial Aff. ¶ 47. 6 Given the severely overcrowded conditions we have already described, this planned 7 additional reduction of 3,000 prisoners in the in-state inmate population is too small to 8 significantly affect the provision of medical and mental health care to California s inmates.59 9 Not surprisingly, defendants do not suggest that the transfer of even more additional inmates 10 to out-of-state facilities would provide a meaningful alternative to the population reduction 11 order proposed by plaintiffs. Furthermore, despite the small size of the existing transfer 12 program, the need to monitor out-of-state facilities to ensure that all California inmates are 13 receiving constitutionally adequate medical care has already hampered the in-state remedial 14 process. Ex. D1100 at 48-49 (Plata Receiver s Sept. 15, 2008 Ninth Quarterly Report) 15 (discussing on-site investigation and corresponding corrective action plan following the 16 death of a California inmate being housed at a private prison in Mississippi). As the 17 Receiver noted, out-of-state monitoring 18 19 20 21 has had a serious negative impact on the Office of the Receiver, drawing critical clinical personnel away from other important projects and delaying in-state remedial efforts. In essence, thousands of dollars of valuable clinical hours have been devoted to helping a private prison organization rework its medical delivery system (at the request of CDCR and State officials) in order to keep the out of state transfer process from collapsing. 22 jail to other institutions, and an order of that type has been held to be a prisoner release order, 23 notwithstanding that the state, county, or city could move the affected prisoners into other institutions rather than releasing them from incarceration. See Tyler, 135 F.3d at 595-98 24 (finding that injunction limiting the number of technical probation violators that could be housed in the city jail, thereby forcing the city to pay for their confinement elsewhere, was a 25 prisoner release order under the PLRA). 59 26 Defendants stated that CDCR will also seek authorization to transfer [additional] inmates out-of-state, if necessary, Cate Trial Aff. ¶ 47, but there is no evidence regarding 27 how difficult this authorization is to obtain; how long the authorization process or transfer process takes; what the cost of such a transfer would be; or how many additional inmates 28 could realistically be dispatched to out-of-state facilities. 113 1 Id. at 49. 2 Based on this clear and convincing evidence regarding the operation of the existing 3 out-of-state transfer program, we conclude that the transfer of inmates to out-of-state 4 facilities would not on its own begin to provide an adequate remedy for the constitutional 5 deficiencies in the medical and mental health care provided to California s inmates. 6 Moreover, given the need to ensure constitutionally adequate medical and mental health care 7 in states as distant as Mississippi, the program may be of questionable efficacy, given the 8 comparatively small number of prisoners who might be included. 9 A defendant-intervenor has suggested that the prison population might be reduced by 10 transferring inmates who do not have legal status in the United States to federal custody. 11 Runner Trial Decl. ¶ 19. However, the intervenor introduced no evidence suggesting that 12 this transfer program could soon be implemented, that the federal government would agree to 13 such an arrangement, or that any implementation of this program would result in a population 14 reduction sizable enough to allow the CDCR to remedy the constitutional violations in Plata 15 and Coleman.60 The program is thus too speculative to suggest that we should abstain from 16 entering the type of prisoner release order set forth below. 17 B. Expert Testimony 18 The testimony we received from the experts overwhelmingly rejected the claim that 19 alternatives such as construction of prisons or other facilities or the transfer of small numbers 20 of prisoners could render a prisoner release order unnecessary. Director Scott succinctly and 21 persuasively summarized the testimony of the experts in stating that unless the population is 22 [substantially] reduced, the state will remain in crisis verging on catastrophe and will remain 23 utterly unable to provide adequate medical and mental health care to the prisoners in its 24 custody. Nov. 9, 2007 Scott Report ¶ 6; see also id. ¶ 3 ( [W]ithout substantially reducing 25 its prisoner population, California will never be able to generate the custodial support 26 services necessary to provide prisoners with basic medical and mental health care. ). 27 Secretary Woodford, the former head of the CDCR who also served as warden at San 28 60 See infra note 82. 114 1 Quentin State Prison, Nov. 9, 2007 Woodford Report ¶ 1, similarly testified that, [u]ntil the 2 population is reduced substantially there is no realistic hope that the unconstitutional 3 conditions will be eliminated, id. ¶ 46; see also id. ¶ 6 ( [N]othing short of a reduction in 4 the prison population will effectively address these issues. ). Woodford explained that, in 5 her experience as a manager of both an individual institution and the entire department: 6 7 8 9 [W]e would come up with wonderful ideas and have great planning, but overcrowding interfered with our ability to implement any of those ideas, to bring resolution to any of the problems that we re facing in both [Plata and Coleman]. And the overcrowding was every day, more and more inmates coming into the system. 10 Rep. Tr. at 376:3-15. Thus, according to Woodford, without addressing the issue of 11 overcrowding, the Department of Corrections will never be able to provide appropriate 12 medical or mental healthcare and . . . sustain any kind of quality constitutionally-adequate 13 medical or mental healthcare. Id. at 385:6-10. 14 Other experts also agreed with Secretary Woodford s and Director Scott s opinions. 15 For example, Dr. Beard opined that, while he believes CDCR staff and leadership generally 16 want to do the right thing, he does not believe they are capable of providing 17 constitutionally adequate care under the current crowded conditions. Id. at 251:12-23, 18 259:5-12. Similarly, Secretary Lehman testified that you cannot provide adequate 19 healthcare and mental healthcare under the current situation of crowding within the State of 20 California, id. at 271:22-25, and that a reduction in the population is a necessary condition 21 for providing such care. Aug. 15, 2008 Lehman Report ¶ 11. And Dr. Shansky testified that: 22 23 24 25 The CDCR, in concert with the Receiver, cannot simultaneously develop a competent medical care delivery system in facilities that lack necessary space and staffing, and address the growing needs of an ever-increasing number of patients. Until the existing overcrowding situation is addressed, CDCR is locked into a crisis-response approach where it can focus only on putting out fires rather than system-building. 26 Nov. 9, 2007 Shansky Report ¶ 138. The limitations on the CDCR, including staffing, 27 administrative resources and especially treatment space, are so severe that the only avenue 28 for building a constitutional health care delivery system is to reduce the demand on the 115 1 system by lowering the number of patients it serves. Sept. 10, 2008 Shansky 2d Supp. 2 Report ¶ 8. One of defendant-intervenors experts agreed that the necessary constitutional 3 medical and mental health services can t be provided with today s overcrowding. Rep. Tr. 4 at 2202:4-6 (Bennett). 5 The mental health experts who testified also agreed that a reduction in crowding is a 6 prerequisite to providing constitutionally adequate care. Dr. Stewart testified that, due to 7 the extreme nature of the overcrowding, which negatively impacts all aspects of the mental 8 health and medical care system that is currently causing Coleman class members needless 9 suffering, as well as death, . . . the only remedy that would help the system move into 10 constitutional compliance is reducing the population. Id. at 2207:22-2208:2. Dr. Stewart 11 based his conclusion on the persistence of the [Eighth Amendment] violations [in Coleman] 12 after years of very close court monitoring, and on statements by the Coleman Special 13 Master in several places that the progress that was made early on in the Coleman matter has 14 been undermined by current population pressures that exist. Id. at 2208:12-19. Dr. Stewart 15 testified that defendants plans to remedy the persistent problems that pose barriers to 16 constitutional compliance are inadequate mainly because the plans will take years to 17 implement, if they are even able to be implemented at all, given the current degree of the 18 population pressures. Id. at 2208:21-2209:4. 19 Dr. Haney also concluded that the only remedy for the ongoing Eighth Amendment 20 violations in the delivery of mental health and medical care is a substantial reduction of the 21 CDCR inmate population, Aug. 15, 2008 Haney Report ¶¶ 364-378; Rep. Tr. at 945:14-19, 22 and provided several reasons for his conclusion. The first was the urgency of the problem 23 itself, and the unacceptably time-consuming nature of alternative solutions. Aug. 15, 2008 24 Haney Report ¶ 367. As Dr. Haney testified, mentally ill inmates suffering in the 25 emergency-like conditions of California s prisons cannot await relief for an additional four 26 or five years, the time projected by defendants best-case scenario for the construction of 27 additional mental health facilities. Rep. Tr. at 945:22-946:16; see also Aug. 15, 2008 Haney 28 Report ¶ 367. Dr. Haney also identified other problems with the proposed construction 116 1 plans: They are insufficient, by themselves, to address the range of mental health care 2 delivery problems caused by crowding; do not take into account the conditions in which 3 CCCMS inmates are housed; do not provide sufficient EOP space; and do not realistically 4 address the massive staffing increases that will be required. Rep. Tr. at 947:16-948:14. 5 Finally, Dr. Haney opined that, for the past twenty-eight years, the CDCR has taken the 6 same basic approach to overcrowding and its impacts on mental health and medical care and, 7 while conditions have occasionally improved over that period, the system has gotten worse 8 not better. Id. at 948:18-949:8. Delivery of services is now so stressed by the 9 overwhelming press of the numbers in the system that the CDCR s method of addressing it 10 has finally run its course, and it is time . . . to address the issue at its cause, and the cause of 11 it is overcrowding. Id. at 949:13-17 (Haney). 12 Defendants emphasize testimony that it is possible to provide constitutionally 13 adequate care in a crowded prison system. E.g., Rep. Tr. at 286:15-18 (Lehman) (testifying 14 that it is possible to provide adequate care at some level of overcrowding); id. at 1216:2115 1217:3 (Thomas) (testifying that such care can be provided in extremely overcrowded 16 conditions ); Nov. 9, 2007 Thomas Report ¶ 6 (same); Rep. Tr. at 1080:12-24 (Packer) 17 (testifying that, although overcrowding exacerbates the problems in providing appropriate 18 mental health care, such care can be provided if appropriate facilities and programs are 19 developed ). Although for the reasons previously stated we are skeptical of Dr. Thomas s 20 testimony, we credit the remaining testimony to the extent that it states that the inmate 21 population need not be reduced to 100% design capacity before constitutional levels of care 22 can be provided. We find, however, that California s prison system is now so overcrowded 23 that it is impossible to provide adequate care without a substantial reduction in crowding. As 24 Secretary Lehman persuasively explained, no state has experienced anything close to the 25 level of crowding in California a level that makes it impossible to provide constitutionally 26 adequate medical and mental health care. Id. at 286:19-287:1; see also id. at 297:1-17 27 (Haney) (testifying that California has been operating at 190% design capacity, which is an 28 unheard of amount of overcrowding ); Nov. 9, 2007 Scott Report ¶ 3 (overcrowding crisis in 117 1 California is unprecedented in scope ). Moreover, Dr. Packer s opinion that 2 constitutionally adequate mental health care can be provided in an overcrowded prison 3 setting is significantly qualified by his testimony that the provision of constitutionally 4 adequate care in such settings is contingent upon the development of appropriate facilities, 5 and that simply retrofitting prison space that was not originally designed for delivery of 6 mental health care is unlikely to lead to a program that is sufficient. See Rep. Tr. at 7 1080:18-1082:12. 8 Additionally, although defendants two experts testified that adequate care can be 9 provided in overcrowded settings and that they themselves have been able to do so, the 10 systems in which they worked had prison population controls in place. Dr. Packer testified 11 that he was able to provide appropriate mental health care in the overcrowded Massachusetts 12 jail facilities he supervised. Id. at 1086:6-12. However, he admitted that there was some 13 effort on the part of the courts to not send in some of the mentally ill inmates into the system. 14 And, frankly, in my opinion the most effective procedure we had was that we provided 15 mental health services at the courts, and we diverted mentally ill people away from the jail. 16 Id. at 1086:17-23. Thus, he explicitly opined that the diversion of mentally ill prisoners a 17 remedy falling within the PLRA s definition of a prisoner release order would be the most 18 effective interim remedy. Id. at 1086:25-1087:14; cf. id. at 1084:17-18 (testifying that new 19 construction is required to get the level of mental health care to the level that really needs to 20 be ). Similarly, during the time in which Dr. Thomas served as a physician with the Florida 21 Department of Corrections, the department operated under statutory population controls that 22 capped the population at prison hospitals and infirmaries at 100% design capacity and the 23 general prison population at 150% design capacity. Id. at 1250:1-1251:1. Consequently, in 24 light of the overwhelming expert testimony to the contrary, we do not find persuasive the 25 testimony by either Dr. Packer or Dr. Thomas that constitutional levels of medical and 26 mental health care can be established in California s prisons without first reducing the 27 California prisoner population to well below 190% design capacity. 28 118 1 C. 2 The evidence establishes that [r]educing the population in the system to a Findings and Conclusions 3 manageable level is the only way to create an environment in which other reform efforts, 4 including strengthening medical management, hiring additional medical and custody staffing, 5 and improving medical records and tracking systems, can take root in the foreseeable future. 6 Sept. 10, 2008 Shansky 2d Supp. Report ¶ 10. Other forms of relief are either unrealistic or 7 depend upon a reduction in prison overcrowding for their success. Accordingly, we find, by 8 clear and convincing evidence, that no relief other than a prisoner release order is capable of 9 remedying the constitutional deficiencies at the heart of these two cases. 10 11 VI. NARROWLY DRAWN, LEAST INTRUSIVE REMEDY THAT EXTENDS NO 12 FURTHER THAN NECESSARY 13 Plaintiffs have demonstrated that crowding is the primary cause of the 14 unconstitutional denial of medical and mental health care to California prisoners, and that no 15 relief other than a prisoner release order can remedy those constitutional violations. 16 Accordingly, plaintiffs have met the PLRA s requirements for the entry of a prisoner release 17 order. See 18 U.S.C. §§ 3626(a)(3)(E)(i), (ii). However, any relief this court orders must 18 also meet the PLRA s general standard for prospective relief. Specifically, the relief must be 19 narrowly drawn, extend[] no further than necessary to correct the violation of the Federal 20 right, and [be] the least intrusive means necessary to correct the violation of the Federal 21 right. Id. § 3626(a)(1)(A). Plaintiffs seek an order requiring the state to reduce the 22 population of its adult institutions to 130% of their combined design capacity. We find that 23 the scope and form of the relief proposed by plaintiffs comports with the PLRA. Although 24 we believe that plaintiffs request for a cap of 130% is reasonable and finds considerable 25 support in the record, there is some evidence that a reduction in the population to a level 26 somewhat higher than 130% of the system s design capacity but lower than 145% might 27 provide the relief from overcrowding necessary for the state to correct the constitutional 28 violations at issue. Notwithstanding the weight of the evidence, we cannot say with certainty 119 1 that a cap as low as 130% is necessary, although we are persuaded that the cap must not be 2 much higher. Because any relief we order must extend no further than necessary, and 3 because we are convinced that a cap of no higher than 137.5% is necessary, we order 4 defendants to reduce the prisoner population to 137.5% of the adult institutions total design 5 capacity.61 6 A. Scope of Relief 7 Our remedy must of course be limited to the inadequac[ies] that produced the 8 injur[ies] in fact that the plaintiff[s] ha[ve] established. Lewis, 518 U.S. at 357. In this 9 proceeding, those injuries involve the state s longstanding and knowing failure to provide its 10 prisoners with the minimal level of medical and mental health care required by the 11 Constitution. The Plata court found that the California prison medical system is broken 12 beyond repair ; that the future injury and death of California prisoners is virtually 13 guaranteed in the absence of drastic action ; and that the state had failed to address those 14 problems despite having every reasonable opportunity to do so. Oct. 3, 2005 FF&CL in 15 Plata, 2005 WL 2932253, at *1. Likewise, the Coleman court found that the state was 16 deliberately indifferent to the fact that seriously mentally ill inmates in the California Department of Corrections daily face an objectively intolerable risk of harm as a result of the gross systemic deficiencies that obtain throughout the Department. . . . [I]nmates have in fact suffered significant harm as a result of those deficiencies; seriously mentally ill inmates have languished for months, or even years, without access to necessary care. They suffer from severe hallucinations, they decompensate into catatonic states, and they suffer the other sequela to untreated mental disease. 17 18 19 20 21 22 Coleman, 912 F. Supp. at 1316, 1319. 23 With the identified constitutional violations in mind, we first consider the propriety of 24 plaintiffs request for a systemwide cap. The scope of injunctive relief is dictated by the 25 extent of the violation established. The key question . . . is whether the inadequacy 26 complained of is in fact widespread enough to justify system wide relief. Armstrong, 275 27 61 As noted already, our opinion and order is limited to the CDCR s thirty-three adult prison institutions and does not include camps, community correction centers, or Department 28 of Mental Health state hospitals. 120 1 F.3d at 870 (quoting Lewis, 518 U.S. at 359). In other words, a systemwide remedy like that 2 requested by plaintiffs is appropriate only if plaintiffs have established systemwide injury 3 and impact. See, e.g., Columbus Bd. of Educ. v. Penick, 443 U.S. 449, 463-65 (1979); 4 Armstrong, 275 F.3d at 871; Smith v. Ark. Dep t. of Corr., 103 F.3d 637, 645-46 (8th Cir. 5 1996). [I]solated violations affecting a narrow range of plaintiffs cannot support 6 systemwide relief. Armstrong, 275 F.3d at 870; see also Lewis, 518 U.S. at 359 (finding 7 systemwide relief inappropriate where plaintiffs had shown only two violations). 8 There can be no serious dispute that a systemwide remedy is appropriate in this case. 9 As we have already noted, the constitutional violations identified by the Plata and Coleman 10 courts exist throughout the California prison system and are the result of systemic failures in 11 the California prison system. See Nov. 3, 2008 Order at 7. Numerous reports issued by the 12 Plata Receiver and the Coleman Special Master document the systemic nature of those 13 problems. Not surprisingly, defendants have never contended that the problems at issue in 14 Plata and Coleman are institution-specific. Accordingly, a systemwide remedy is 15 appropriate. 16 Similarly, we conclude that a single systemwide cap rather than a series of institution- 17 specific caps or a combination of systemwide and institution-specific caps is appropriate. 18 Although institution-specific caps would be tailored to each institution s needs and 19 limitations, an institution-by-institution approach to population reduction would interfere 20 with the state s management of its prisons more than a single systemwide cap, which permits 21 the state to continue determining the proper population of individual institutions. Unless and 22 until it is demonstrated that a single systemwide cap provides inadequate relief, we will limit 23 the relief we order to that form of order. 24 To be certain, the relief sought by plaintiffs extends further than the identified 25 constitutional violations in one regard: Any population reduction plan developed by the state 26 is likely to affect inmates without medical conditions or serious mental illness. However, 27 there is no feasible prisoner release order that would reduce overcrowding without affecting 28 121 1 some inmates outside the Plata and Coleman classes. Thus, we have no doubt that the relief 2 we order contravenes no principle of law or equity in that regard. 3 Accordingly, the systemwide scope of plaintiffs requested relief is properly tailored 4 to the identified constitutional violations, at least at this first stage of the court s attempt to 5 bring the system into compliance with the Constitution s mandate. 6 B. 7 We next consider the form of relief proposed by plaintiffs. Plaintiffs seek an order Form of Relief 8 requiring the state to reduce its prison population to a specified percentage of the system s 9 design capacity within two years. Initially, the state would be required to develop a plan to 10 reduce the population to the designated percentage. After considering the proposed plan and 11 any objections from plaintiffs or intervenors, we would enter a final order incorporating the 12 state s proposal if it is feasible, with any appropriate modifications or amendments we may 13 deem necessary. We would then retain jurisdiction to ensure compliance with the order or 14 make further changes as necessary in order to allow the state to attain the actual reduction in 15 the prison population set forth in our order within the specified time. 16 The Supreme Court described the nearly identical procedure used in Bounds v. Smith, 17 430 U.S. 817 (1977), as an exemplar of what should be done in crafting systemwide 18 prospective relief. Lewis, 518 U.S. at 363. In Bounds, the district court found that the state s 19 failure to provide legal research facilities unconstitutionally denied its inmates access to the 20 courts. Bounds, 430 U.S. at 818. However, 21 22 23 24 [r]ather than attempting to dictate precisely what course the State should follow [to remedy the constitutional violation], the court charged the Department of Correction with the task of devising a Constitutionally sound program to assure inmates access to the courts. It left to the State the choice of what alternative would most easily and economically fulfill this duty. 25 Id. at 818-19 (internal quotations omitted). The State responded with a proposal, which the 26 District Court ultimately approved with minor changes, after considering objections raised by 27 the inmates. Lewis, 518 U.S. at 362-63 (citing Bounds, 430 U.S. at 819-20). 28 122 1 In both Bounds and Lewis, the Supreme Court praised the Bounds lower court s 2 remedial approach, finding that it scrupulously respected the limits on [the court s] role 3 and preserved the prison administrators wide discretion within the bounds of constitutional 4 requirements. Bounds, 430 U.S. at 832-33; Lewis, 518 U.S. at 363. The relief requested by 5 plaintiffs here demonstrates the same respect for this court s limited role and for the need to 6 preserve the state s wide discretion in managing its prisons. As in Bounds, plaintiffs 7 proposal would permit the state to develop the necessary population reduction plan in the 8 first instance. As we describe infra, the state would not be required to throw open the doors 9 of its prisons, but could instead choose among many different options or combinations of 10 options for reducing the prison population. The state s options include, inter alia, the 11 following: enhancing good time and program participation credits; diverting technical parole 12 violators and certain offenders with short sentences; reducing the length of parole 13 supervision; implementing evidence-based rehabilitative programming; or implementing 14 sentencing reforms, perhaps by means of a sentencing commission or by otherwise changing 15 outmoded or counterproductive sentencing practices. Many of these options have already 16 been proposed at various times by defendants themselves. See, e.g., Rep. Tr. at 1694:1917 1699:15 (Cate) (discussing Governor Schwarzenegger s proposed reforms, including the 18 elimination of parole supervision and enhanced good time and program participation credits); 19 see also Ex. P3 at 77 (noting that fifteen reports presented to the state between 1990 and 20 2007, some of which were prepared by state-established commissions or committees, 21 recommended sentencing reform and the establishment of a sentencing commission). 22 Plaintiffs proposed order would permit the state to choose among many available 23 means of achieving the prescribed population reduction, thereby maximizing the state s 24 flexibility and permitting the state to comply with the cap in a manner that best accords with 25 the state s penal priorities. For this reason, an order requiring a systemwide population 26 reduction to a specified percentage is preferable to an order or series of orders requiring 27 particular methods of population reduction, such as the reform of the parole system or the 28 overhaul of the state s sentencing policies. By asking the state to develop a remedial plan in 123 1 the first instance, the relief sought by plaintiffs exhibits the deference to state expertise 2 required by the PLRA and Lewis and limits this court s intrusion into the minutiae of 3 prison operations. Lewis, 518 U.S. at 362 (quoting Bell v. Wolfish, 441 U.S. 520, 562 4 (1979)). The population reduction order sought by plaintiffs is thus the least intrusive 5 means necessary to correct the [constitutional] violation[s] at issue in this proceeding. 6 18 U.S.C. § 3626(a)(1)(A).62 7 C. The Required Population Reduction 8 Finally, we consider plaintiffs specific request that we order defendants to reduce 9 California s prisoner population to 130% of the system s design capacity. At the outset, we 10 note that choosing the percentage of design capacity to which the prison population should 11 be reduced is not an exact science. Rep. Tr. at 976:3-4 (Haney). As plaintiffs expert 12 Dr. Craig Haney explained, there s nothing magical about any specific percentage, 13 including 100%, id. at 976:7-8, but the likelihood of bringing the system into constitutional 14 compliance increases as the prison population nears 100% design capacity, id. at 976:8-15. 15 Our task is further complicated by the fact that defendants have not presented any evidence 16 or arguments suggesting that we should adopt a percentage other than 130% design capacity. 17 Nonetheless, both the PLRA and general equitable principles require this court to ensure that 18 the population reduction sought by plaintiffs extends no further than necessary to rectify the 19 unconstitutional denial of medical and mental health care to California s prisoners. 20 Although plaintiffs seek a cap at 130% design capacity, the evidence at trial 21 demonstrated that even a prison system operating at or near only 100% design capacity faces 22 serious difficulties in providing inmates with constitutionally sufficient medical and mental 23 health care. First, California s prisons were not designed to provide medical and mental 24 health care for the numbers now housed therein. Instead, the physical space for health care 25 62 Of course, Bounds involved the initial remedial response to a recently identified constitutional violation, whereas Plata and Coleman have been in their remedial phases for a 26 number of years. We do not believe this distinction would justify a departure from the remedial model praised in Bounds and Lewis, in which the state is given the first opportunity 27 to develop a remedial plan. It may, however, along with the nature of the constitutional violations and of plaintiffs injuries, affect the length of time in which the state is required to 28 develop and implement the plan. 124 1 in California s prisons was devised on the assumption that the prisons populations would not 2 exceed 100% of their design capacity. Rep. Tr. at 271:8-10 (Lehman) ( The physical space 3 provided [in each institution] is based on the hundred percent population as opposed to 200 4 percent. ). As defendants witness Robin Dezember noted, the state s prisons were not 5 designed and made no provision for any expansion of medical care space beyond the initial 6 100% of capacity, and none of the 19 CDCR institutions planned and built in the boom of 7 the 80s and 90s gave any thought to the space that might be needed for mental health 8 purposes. Dezember Trial Aff. ¶ 72 (internal quotations omitted). Shockingly, this failure 9 to account for the effect of overcrowding on the ability of prisons to deliver medical and 10 mental health continued even after the state knew that they would be filled to 200% of their 11 design capacity. Ex. D1092 at 21-22 (Plata Receiver s May 15, 2007 Report Re: 12 Overcrowding) (noting that a new prison built in 2005 was designed to provide medical care 13 for a population equal to 100% design capacity notwithstanding the CDCR s existing plan to 14 house a population equal to 200% design capacity in the new prison). 15 The mismatch between the physical design of the prisons and their present 16 overcrowding accounts for many of the space-related obstacles to the provision of 17 constitutionally sufficient medical and mental health care. According to Secretary Lehman, 18 the former head of corrections in Washington, Maine, and Pennsylvania, this mismatch 19 leaves California s prisons without the physical space to provide medical and mental health 20 care to the number of prisoners now housed in those overcrowded institutions. In the 21 absence of sufficient space the prisons are simply not able to provide the [healthcare] 22 services that [are] required. Rep. Tr. at 271:10-11. 23 More generally, any prison operating at 100% design capacity stretches the limits of 24 its physical design. According to Dr. Haney, prisons were virtually always designed 25 sparsely . . . so that a prison that was reaching 100% of its capacity really was pushing 26 against the limits of the number of prisoners that it could safely and humanely hold. 27 Aug. 15, 2008 Haney Report ¶ 380. This is especially true in the context of space allocated 28 for purposes other than housing, including medical and mental health care. [P]rison design 125 1 traditionally maximized housing capacities and minimized space allocated to programming 2 needs, opportunities, and demands. Id. As a result, [w]hen a prison beg[ins] to operate at 3 or near its [design] capacity, there [is] typically little or no space available to pursue all but 4 the most basic programming options. Id. 5 Finally, numerous witnesses testified that a prison system must operate below 100% 6 design capacity to function properly. Secretary Woodford, former head of CDCR and 7 warden at San Quentin, stated that a five percent vacancy rate is necessary [t]o manage the 8 movement of prisoners appropriately. Nov. 9, 2007 Woodford Report ¶ 14. Without the 9 flexibility that this vacancy rate provides, it is very difficult to ensure that prisoners are 10 housed appropriately for their medical and mental health needs. Id. In addition, three 11 witnesses for the defendant-intervenors testified that jails require a vacancy rate of at least 12 five or ten percent to operate properly. According to San Mateo County Sheriff Gregory 13 Munks, jails operate properly only when at or below their functional capacity, which is 14 five to ten percent lower than their design capacity. See Rep. Tr. at 1776:15-20 (Munks); see 15 also id. at 1776:20-23 (functional capacity takes account [of] having the room for 16 classification, being able to move inmates around, [and] keep[ing] them separated based on 17 classification, based on needs, based on gang affiliation ). Lieutenant Stephen Smith of the 18 Los Angeles County Sheriff s Department testified that jails cannot operate safely or 19 properly if every bed is filled, and that he would expect the same result in prisons. Id. at 20 1837:5-1838:6 (Smith). According to Lieutenant Smith, A hundred percent of your 21 capacity is really a misnomer. . . . [Y]ou re at a hundred percent capacity when you are at 90 22 percent. You need a ten percent vacancy factor to just facilitate movement, and those type of 23 issues because of the margins. Id. at 1845:16-21. Likewise, Gary Graves, the acting 24 County Executive for Santa Clara County, testified that a fifteen percent vacancy rate is 25 generally necessary in Santa Clara County s jail system. Id. at 2275:3-6. 26 This testimony establishes that, when a prison or jail s population reaches 100% 27 design capacity, its administrators lose the flexibility required to classify inmates and to 28 move prisoners in accordance with their needs. We have already noted that overcrowding 126 1 prevents the state from providing constitutionally adequate medical and mental health care in 2 part by preventing the proper classification of inmate medical and mental health needs, 3 limiting the state s ability to bring inmates to required appointments and services, and 4 preventing the state from transferring inmates into necessary clinical placements. 5 Accordingly, the testimony suggests that the state s ability to provide constitutionally 6 adequate medical and mental health care is hampered at 100% design capacity. 7 Despite this evidence, plaintiffs do not seek an order capping the prison system s 8 population at 100% design capacity. Instead, they seek a cap at 130% design capacity, 9 acknowledging that constitutionally adequate medical and mental health care can be provided 10 in such circumstances. Plaintiffs proposed population limit is drawn from a 11 recommendation by the Governor s own prison reform personnel. To implement the prison 12 building and prison reform projects authorized by AB 900, the Governor established a series 13 of strike teams, and Deborah Hysen became head of the Facilities Strike Team in May 2007. 14 Sept. 3, 2008 Hysen Dep. at 12:8-10, 15-17. In that role, Ms. Hysen suggested that the 15 CDCR impose two limits on the state prison population. First, she suggested that new prison 16 beds built pursuant to AB 900 be allocated in a manner that would limit overcrowding to no 17 more than 145% design capacity. Ex. P128 at 1, 6 (Aug. 13, 2007 AB 900 Strike Team 18 memo). Hysen acknowledged that housing prisoners at 145% design capacity does not meet 19 federal guidelines nor national standards, but she nonetheless believed that a reduction in 20 overcrowding to 145% design capacity would begin to moderate and control the 21 department s overcrowding practices. Id. at 6. As a long-term goal, however, Hysen 22 suggested that the prison system s population should not exceed 130% design capacity, the 23 federal standard for prison overcrowding. Id.; Hysen Dep. at 94:13-24. Ms. Hysen also 24 suggested that the CDCR consider establishing planning capacity and oversight mechanisms 25 to prevent the occurrence of exceeding this [130%] threshold. Ex. P128 at 6. 26 Plaintiffs experts testified that the 130% cap recommended by Ms. Hysen would be 27 sufficient to remedy the constitutional violations here. Secretary Lehman testified that 28 housing California prisoners at 130% design capacity will give prison officials and staff the 127 1 ability to provide the necessary programs and services for California s prisoners. Aug. 15, 2 2008 Lehman Report ¶ 20. Doyle Wayne Scott, the former executive director of the Texas 3 Department of Criminal Justice, testified that Ms. Hysen s 130% recommendation was a 4 realistic and appropriate place for CDCR to be, to ensure that its prisons are safe and provide 5 legally required services, Aug. 13, 2008 Scott Supp. Report ¶ 18. Secretary Woodford also 6 agreed with Ms. Hysen s recommended 130% cap. Aug. 15, 2008 Woodford Supp. Report 7 ¶ 3. 8 Notably, however, both Director Scott and Secretary Woodford qualified their 9 endorsement of the 130% cap by stating that certain facilities could not provide 10 constitutionally sufficient medical and mental health care when filled to 130% design 11 capacity. Woodford noted that different (and particularly older) facilities might require 12 slightly lower population limitations, based on the quality of infrastructure and availability of 13 treatment space, for example. Id. According to Scott, [W]hile [130%] might be 14 appropriate for new construction, it should be used carefully in CDCR s old, decaying 15 facilities, with their failing infrastructure. Crowding prisoners at 130% is an appropriate goal 16 for CDCR, speaking broadly, but some facilities might only be able to support and provide 17 appropriate health care for smaller numbers. Aug. 13, 2008 Scott Supp. Report ¶ 18. 18 Although Director Scott and Secretary Woodford suggested that a 130% limit might 19 be too high in certain instances, other evidence suggested that a cap above 130% might be 20 sufficient. For example, Dr. Ronald Shansky testified that the Illinois prison medical system 21 was brought into constitutional compliance at 140% design capacity. Rep. Tr. at 479:2-16. 22 Similarly, the Corrections Independent Review Panel determined in 2004 that the California 23 prison system s operable capacity was 145% of its design capacity. Ex. P4 at 124. The 24 Panel s estimate was prepared by a group of experienced California prison wardens, who 25 suggested that a system operating at 145% design capacity could support full inmate 26 programming in a safe and secure environment. Id. 27 Numerous witnesses testified, however, that the Panel s operable capacity estimate 28 suffers from a potentially fatal flaw for purposes of measuring the constitutional 128 1 requirements relating to medical and mental health care. Operable capacity does not take 2 into account the ability to provide that care. Thus, the wardens did not consider prisoner 3 medical or mental health needs in reaching their estimate. See Ex. P4 at 161 n.3; Nov. 9, 4 2007 Scott Report ¶ 46 ( [The expert panel s] definitions [of design capacity, operational 5 capacity, and maximum safe and reasonable capacity], however, still fail to look at the 6 capability of a system or individual facility to adequately and legally care for the medical and 7 mental health needs of its population . . . . ). According to Dr. Stewart, The [maximum 8 operable capacity] incorporated educational, vocational, substance abuse, and other 9 rehabilitation programming, but did not account for programming associated with mental 10 health or medical treatment. . . . When mental health treatment needs are taken into account, 11 the maximum operable capacity will be lower. Aug. 15, 2008 Stewart Supp. Report 12 ¶¶ 126-27. Likewise, Dr. Haney reported that 13 14 15 16 17 18 the Panel s estimate of [maximum operable capacity] did not specifically contemplate, take into account, or attempt to calculate the additional space and staffing levels that would be required to provide constitutionally adequate mental health and medical care. . . . When these crucial mental health and medical treatment needs are taken into account as they must be in any calculation aimed at addressing the primary cause of these continuing constitutional violations then the appropriate percentage for maximum operable capacity would certainly be lower than the Panel s and wardens estimates of 145%. 19 Aug. 15, 2008 Haney Report ¶¶ 383, 385. 20 Plaintiffs experts convincingly demonstrated that, in light of the wardens failure to 21 consider the provision of medical and mental health care to California s inmates and in light 22 of their reliance on maximum operable capacity, which does not consider the ability to 23 provide such care, the Panel s 145% estimate clearly exceeds the maximum level at which 24 the state could provide constitutionally adequate medical and mental health care in its 25 prisons. Unfortunately, plaintiffs experts did not calculate the extent to which the operable 26 capacity of California s prisons exceeds the percentage necessary for the provision of 27 constitutionally adequate medical and mental health care. See Aug. 15, 2008 Stewart Supp. 28 Report ¶ 127 (stating only that the maximum operable capacity of California s prisons is 129 1 lower than 145%); Aug. 15, 2008 Haney Report ¶ 385 (same); see also Aug. 15, 2008 Haney 2 Report ¶ 385 (describing 145% as a very conservative estimate of [maximum operable 3 capacity] that is the outer limit or maximum capacity in a range that is intended to 4 eliminate the constitutional violations that are at issue here (last emphasis added)). Even 5 more unfortunately, as noted earlier, defendants introduced no evidence suggesting that the 6 population of California s prisons should be reduced to some level above 130%. 7 Although there is strong evidence that a prison system operating at even 100% design 8 capacity will have difficulty providing adequate medical and mental health care to its 9 inmates, the evidence before the court establishes that California s prisoner population must 10 be reduced to some level between 130% and 145% design capacity if the CDCR s medical 11 and mental health services are ever to attain constitutional compliance. The evidence in 12 support of a 130% limit is strong: Both national standards and the Governor s own strike 13 team, which adopted those standards, suggest 130% design capacity as a reasonable upper 14 limit on the prison system s population. However, we cannot determine from the evidence 15 whether the national standard selected by the Governor s strike team represents a judgment 16 regarding the mandates of the Constitution or whether it merely reflects a policy that ensures 17 desirable prison conditions. Other, far less persuasive evidence at trial suggested that 18 California might be able to remedy the constitutional violations at issue in Plata and 19 Coleman if the population of the CDCR s adult institutions were reduced to 140% or 20 somewhere else lower than 145% design capacity. Exercising the caution and restraint 21 required by the PLRA, we credit this evidence to the extent it suggests that the limit on 22 California s prison population should be somewhat higher than 130% but lower than 145%. 23 Rather than adopting the 130% limit requested by plaintiffs, we will out of caution require a 24 reduction in the population of California s adult prison institutions to only 137.5% of their 25 combined design capacity a population reduction halfway between the cap requested by 26 plaintiffs and the wardens estimate of the California prison system s maximum operable 27 capacity absent consideration of the need for medical and mental health care. At the adult 28 institutions present design capacity of 79,828, Ex. P135 (CDCR weekly population report as 130 1 of August 27, 2008), this equates to a population of just below 110,000.63 Should the state 2 prove unable to provide constitutionally adequate medical and mental health care after the 3 prison population is reduced to 137.5% design capacity, plaintiffs may ask this court to 4 impose a lower cap.64 Similarly, should it appear that the provisions set forth in the plan 5 adopted by the court will not achieve the expected population reduction, plaintiffs may seek 6 to have the plan amended. 7 8 VII. POTENTIAL POPULATION REDUCTION MEASURES AND THEIR 9 IMPACT ON PUBLIC SAFETY AND THE OPERATION OF THE CRIMINAL 10 JUSTICE SYSTEM 11 Before we enter a population reduction order, we must give substantial weight to any 12 adverse impact on public safety or the operation of a criminal justice system caused by the 13 relief. 18 U.S.C. § 3626(a)(1)(A). To aid us in meeting this requirement, the parties 14 devoted nearly ten days of trial to this issue and submitted hundreds of exhibits. The 15 impressive collection of evidence before the court included testimony from former and 16 current heads of corrections of five states; top academic researchers in the field of 17 incarceration and crime; CDCR officials; and county officials, district attorneys, probation 18 officers, and sheriffs from across California. We also had the benefit of many state19 commissioned reports that proposed various measures for safely reducing the overcrowding 20 in California s prison system. Indeed, four of plaintiffs experts Dr. Austin, Dr. Beard, Dr. 21 Krisberg, and Secretary Lehman had previously been appointed by the CDCR to serve as 22 members of the Expert Panel on Adult Offender Recidivism Reduction Programming. We 23 give substantial consideration to the report from this panel, which recommended a number of 24 63 Of course, our order is based on a percentage of design capacity. If the CDCR closes existing prisons or constructs new prisons or prison beds, the system s design capacity 25 will change, and our order will therefore require a prison population than just below 110,000. See supra Section V.A.1.d. 26 64 We recognize that certain institutions and programs in the system require a 27 population far below 137.5% design capacity. We trust that any population reduction plan developed by the state in response to our opinion and order will properly account for the 28 particular limitations and needs of individual institutions and programs. 131 1 measures that it believed would help to safely reduce overcrowding in California s prisons, 2 as a necessary first step to reducing recidivism; it included a list of ten related reforms that 3 have been repeatedly recommended to the state, Ex. P2 at 77, some of which we discuss 4 below. 5 We begin by emphasizing the nature of the order this court issues herein. The order 6 requires the state to reduce California s prison population to 137.5% design capacity within 7 two years and to submit a plan within 45 days to implement our order. As we discuss below, 8 there are a number of population reduction measures that will not have an adverse impact on 9 public safety and that in fact may improve public safety, all of which have been previously 10 recommended to the state, in various reports, by experts it retained to examine ways to 11 reduce California s high recidivism rate.65 Any or all of these measures may be included in 12 the state s plan. Whichever solutions it ultimately chooses, the evidence is clear that the state 13 can comply with our order in a manner that will not adversely affect public safety. Indeed, 14 the evidence is clear that the state s continued failure to address the severe crowding in 15 California s prisons would perpetuate a criminogenic prison system that itself threatens 16 public safety. 17 In addressing the potential impact on public safety of our population reduction order, 18 we do not ignore the serious fiscal crisis presently facing the state of California. We are 19 aware that California will not through its ordinary budget process increase its expenditures in 20 order to ameliorate or resolve the constitutional issues it confronts. However, as we explain 21 below, a reduction in California s prison population would produce significant savings, some 22 of which, even with a budget reduction, could be used to fund effective rehabilitative and re23 entry programming in the prisons and to help county and local governments meet any 24 additional costs resulting from their expansion of existing programs in order to meet the 25 65 The state recidivism rate is the ratio of the number of felons returned to prison 26 during a specific period to the number of felons paroled during the same period, times one hundred. Ex. DI-600 at 4. The CDCR s statistics on recidivism show return-to-prison rates 27 within three years, and they include returns for technical parole violations. Rep. Tr. at 1373:3-20 (Woodford). California s recidivism rate is one of the highest in the country. See 28 Aug. 15, 2008 Bennett Report ¶ 58; Aug. 15, 2008 Austin Report ¶¶ 9-11; Ex. P2 at 88. 132 1 needs of persons affected by a population reduction order who may require county or local 2 services. Even if the state were not to use any savings for such purposes, population 3 reduction could be accomplished without any significant adverse impact on public safety or 4 the operation of the criminal justice system. A number of the population reduction measures 5 that have been recommended by the various expert committees do not require any substantial 6 additional expenditures, and, in many instances, any additional burdens on county and local 7 governments resulting from the prison population reduction would fall within current 8 fluctuations in the demand for existing services. 9 In any event, we cannot now determine with finality whether the population reduction 10 plan the state will propose in response to our order would have an adverse impact upon 11 public safety or the operation of the criminal justice system. We do know, however, that the 12 state could comply with our population reduction order without a significant adverse impact 13 upon public safety or the criminal justice system s operation; the evidence before us clearly 14 establishes its ability to do so. We will consider the impact of the state s actual population 15 reduction plan before approving it or any modified or substitute plan. Whatever plan we do 16 adopt will be consistent with our obligation to accord substantial weight to any adverse 17 impact involved. 18 A. 19 As an initial matter, we conclude that the current combination of overcrowding and Criminogenic Nature of Overcrowded Prisons 20 inadequate rehabilitation or re-entry programming in California s prison system itself has a 21 substantial adverse impact on public safety and the operation of the criminal justice system. 22 A reduction in the crowding of California s prisons will have a significant positive effect on 23 public safety by reducing the criminogenic aspects of California s prisons. 24 Defendants do not credibly dispute the above conclusion, although they argue that 25 California s criminal justice system is no different from that of other jurisdictions. In a 26 certain sense they are correct. For example, California s incarceration rate for prisoners 27 sentenced to more than one year in state or federal prisons is about 475 per 100,000 28 residents, close to the national average. Cate Trial Aff. ¶ 22. California does not incarcerate 133 1 felons at an unusually high rate, id. ¶¶ 23-24, and the average prison sentence imposed and 2 served in California is lower than the national average, id. ¶ 25. However, as convincingly 3 explained by Professor Joan Petersilia, an expert on the California prison system and a 4 member of the CDCR s Rehabilitation Strike Team,66 the similarities end once an individual 5 has been sentenced to prison. California truly is different when it comes to the way inmates 6 are housed, the way they are treated while incarcerated, the way they are released, and the 7 way their parole is handled and revoked. Ex. P5 at 9 (May 2006 California Policy Research 8 Center Report, Understanding California Corrections ). As a consequence, although 9 California spends billions of dollars on its prison system, it has one of the highest return-to10 prison rates in the nation. Id. at ix. In 2005, 66% of offenders released from the California 11 prison system returned to prison within three years. Id. At least two experts reported that 12 California s recidivism rate is at 70 percent. Aug. 15, 2008 Bennett Report ¶ 58; Nov. 9, 13 2007 Austin Report ¶ 42. 14 The evidence clearly establishes that, because of overcrowding, the state is limited in 15 its capacity to classify inmates properly according to their security risk or programming 16 needs. See, e.g., Rep. Tr. at 2013:21-23 (Lehman); id. at 145:15-18 (Scott); id. at 225:2117 227:13 (Beard); Aug. 15, 2008 Lehman Report ¶ 8; Nov. 9, 2007 Woodford Report ¶ 13. In 18 addition, a December 2007 report from the CDCR s Rehabilitation Strike Team found that 19 fully 50% of all exiting California prisoners did not participate in any rehabilitation or work 20 program nor did they have a work assignment, during their entire prison term . . . . 21 Ex. P113 at 13 (December 2007 report, Meeting the Challenges of Rehabilitation in 22 California s Prison and Parole System: A Report from Governor Schwarzenegger s 23 Rehabilitation Strike Team ) (hereinafter Rehabilitation Strike Team Report ) (emphasis in 24 original); see also Sept. 22, 2008 Marquart Supp. Report ¶ 5 ( [O]f the 134,000 prisoners 25 who exited California s prisons in 2006, only 7% participated in substance abuse programs 26 and only 10% participated in vocational education while incarcerated. ). The CDCR s 27 66 The Rehabilitation Strike Team was established by Governor Schwarzenegger to 28 develop and implement prison and parole programs for the CDCR. Ex. P113 at 10. 134 1 Undersecretary of Programs Kathryn Jett believed that the same remained true as of August 2 2008. Rep. Tr. at 1731:4-8. 3 Witnesses for plaintiffs and defendant-intervenors with substantial experience 4 administering or studying correctional and law enforcement systems testified that, in such 5 conditions, high-risk inmates do not rehabilitate and low-risk inmates learn new criminal 6 behavior.67 E.g., Rep. Tr. at 1580:5-9 (Beard) ( They are probably getting worse with the 7 environment that they re in, associating with the higher risk people and with the 8 overcrowding, with the violence, those lower risk people are probably going to be more 9 likely to reoffend. ); id. at 1052:19-1053:9 (Powers) (Stanislaus County Chief Probation 10 Officer) (testifying that, as a probation officer, he would prefer to treat offenders in the 11 community because offenders come out of prison worse than when they went in); id. at 12 2777:2-19 (Meyer) (Yolo County Chief Probation Officer) ( When I toured the prisons with 13 the intervenors some time ago, I was actually shocked about how almost nothing positive is 14 going on, how crowded it was. It s an issue that that it seems like they produce additional 15 criminal behavior. ); see also id. at 385:23-25 (Woodford) ( I absolutely believe that we 16 make people worse, and that we are not meeting public safety by the way we treat people. ). 17 California s prisons, in other words, are serving as crime school[s]. Id. at 2014:1 18 (Lehman). 19 According to Secretary Lehman, the former head of corrections in Washington, 20 Maine, and Pennsylvania, there s only one term you can use to describe California s 21 overcrowded prisons: criminogenic. 68 Id. at 2013:18-2014:1. The criminogenic 22 environment in the prison system means that [e]ach year, California communities are 23 burdened with absorbing 123,000 offenders returning from prison, often more dangerous 24 than when they left. Ex. P3 at 17 (Jan. 2007 Little Hoover Commission Report, Solving 25 California s Corrections Crisis: Time is Running Out ). This situation presents a clear threat 26 27 67 Low-risk inmates are those with low risks of reoffending, whereas high-risk inmates are those with high risks of reoffending. See Rep. Tr. at 1170:25-1171:22 (Powers). 68 Something that is criminogenic contributes to the occurrence of crime. See Rep. 28 Tr. at 2013:19-20 (Lehman). 135 1 to public safety and the operation of the criminal justice system. See Rep. Tr. at 1580:17-19 2 (Beard); id. at 974:11-22 (Haney). 3 The state has recently attempted to reduce these criminogenic effects by implementing 4 a new case management system. However, the CDCR is still in the preliminary stages of 5 implementing this new programming. As of August 2008, there was no system in the prisons 6 to deliver the right inmate to the right program, Rep. Tr. at 1727:23-24 (quoting Jett 7 Dep.), and, at the time of trial, the case management system was in its infancy, id. at 8 1713:3-5 (Jett). 9 Defendants do not dispute the overwhelming evidence that overcrowding in prisons 10 itself threatens public safety, nor could they. In fact, in his 2006 Prison Overcrowding State 11 of Emergency Proclamation, Governor Schwarzenegger found that overcrowding causes 12 harm to people and property, leads to inmate unrest and misconduct, reduces or eliminates 13 programs, and increases recidivism as shown within this state and in others. Ex. P1 at 2. 14 The contention by defendants expert Dr. James Marquart that there is no clear evidence 15 that overcrowding by itself automatically leads to violence, Aug. 14, 2008 Marquart Prelim. 16 Report at 7-8, is unpersuasive, but in any event, it does not directly contradict the evidence 17 that crowding increases recidivism. Moreover, even if Dr. Marquart is correct that there is 18 no systematic empirical investigation confirming the pernicious effects of overcrowding, 19 Aug. 27, 2008 Marquart Rebuttal Report ¶ 3, we credit the testimony of correctional and law 20 enforcement experts who have seen and studied the conditions in California prisons and 21 convincingly opined that they do adversely affect public safety. 22 Accordingly, we find that California s overcrowded prisons are criminogenic and, as 23 the Governor declared in his State of Emergency Proclamation, Ex. P1 at 2, have an adverse 24 effect on public safety. Mitigating prison overcrowding could improve public safety by 25 rendering possible the proper classification of inmates and the expansion and targeting of 26 rehabilitation programming. See Ex. P2 at 9-10 (June 2007 CDCR Expert Panel on Adult 27 Offender Recidivism Reduction Programming Report to the California State Legislature, A 28 136 1 Roadmap for Effective Offender Programming in California ) (recommending population 2 reduction measures in order to reduce recidivism rates). 3 B. 4 There was overwhelming agreement among experts for plaintiffs, defendants, and Potential Population Reduction Measures 5 defendant-intervenors that it is absolutely possible to reduce the prison population in 6 California safely and effectively. Rep. Tr. at 2189:9-23 (Bennett) (Sonoma County 7 corrections expert); see, e.g., id. at 2101:24-2102:1 (Krisberg) (plaintiffs expert); id. at 8 1995:8-20 (Marquart) (defendants expert); id. at 2012:20-25 (Lehman) (plaintiffs expert); 9 id. at 1327:3-6 (Woodford) (plaintiffs expert). 10 Plaintiffs proposed several measures to reduce the prison population. The first, the 11 expansion of the good time credits system, would allow eligible low- to moderate-risk 12 inmates to be released a few months early in exchange for complying with prison rules and 13 participating in rehabilitative, education, or work programs. The second and third, the 14 diversion of technical parole violators and of low-risk offenders with short sentences, would 15 keep low-risk offenders in community correctional systems rather than incarcerating them in 16 prison for a few months. The fourth, the expansion of evidence-based rehabilitative 17 programming, would reduce the prison population by addressing offenders rehabilitative 18 needs, thus lowering their likelihood of reoffending.69 19 Many of the witnesses presented by defendant-intervenors objected to simply 20 throwing open the prison doors and releasing inmates early in a generic manner, erroneously 21 assuming that such a remedy might be contemplated or ordered by the court. See, e.g., Rep. 22 Tr. at 1087:16-22 (Packer) ( When they said prison release, I thought they were literally 23 releasing people from the prison. ); id. at 1052:8-12 (Powers) (stating that he prefers his 24 population reduction proposal to let s throw the door open, and in six months from now we 25 will be there ); Aug. 15, 2008 Bennett Report ¶ 13 ( The wholesale release of inmates would 26 69 Plaintiffs also propose shortening the length of parole supervision, which would 27 have a more immediate and direct impact on the distribution of parole resources than on the prison population. For that reason, we discuss this proposal separately. See infra 28 Section VII.C.3. 137 1 only shift the crowding problem to the counties and provide nothing more than temporary 2 relief to the state. ); Bay Stip. ¶ 7 (Director of San Mateo County Department of Housing) 3 ( I am assuming that the prisoner release order is a one-time event and not part of a pattern of 4 shorter sentences for a class of prisoners. ). However, many of the same witnesses, as well 5 as others presented by defendants and defendant-intervenors, testified that they supported 6 other measures for reducing the prison population, including measures substantially similar 7 to those proposed by plaintiffs. E.g., Rep. Tr. at 1086:20-1087:22 (Packer) (recommending 8 diversion of mentally ill inmates from the prisons); id. at 1041:12-1045:11 (Powers) (stating 9 that a prison population reduction could be achieved safely by investing in probation); Aug. 10 15, 2008 Bennett Report ¶¶ 68-71, 75-76 (recommending systemic changes, including 11 reducing return to incarceration as a sanction for technical parole violations and enhancing 12 community-based sanctions programs); see also Buddress Trial Decl. ¶ 3 (San Mateo County 13 Chief Probation Officer) (supporting population reduction measures proposed by plaintiffs 14 expert Dr. Krisberg); Dalton Am. Trial Decl. ¶¶ 17-26 (Los Angeles County Sheriffs 15 Department, Director of Bureau of Operations for Bureau of Offender Programs and 16 Services) (recommending diversion to community corrections, sentencing reform, diversion 17 of technical parole violators, and re-entry programming); Rep. Tr. at 2770:23-2771:10 18 (Meyer) (testifying that, if appropriate programs were funded, the population could be 19 reduced by about 30% while crime was also reduced); Dumanis Trial Decl. ¶¶ 16-20 (San 20 Diego District Attorney) (supporting re-entry programming and rehabilitative and diversion 21 programs); Boesch Trial Decl. at 13 (San Mateo County Assistant County Manager) 22 (supporting rehabilitation programs and graduated sanctions). 23 We do not suggest that plaintiffs proposed methods are the only ways to reduce the 24 prison population without adversely affecting public safety and the criminal justice system. 25 We have discussed some other methods earlier, supra Section V.A.4 (discussing the state s 26 proposals to transfer inmates out of state or into federal custody), and will discuss others 27 later, supra Section VII.B.5 (discussing, inter alia, sentencing reform and modifications of 28 criminal statutes). There are other proposals as well that have been recommended by various 138 1 state commissions or bodies that may be worthy of consideration. Our discussion here is not 2 necessarily exhaustive. 3 First, we consider plaintiffs four proposed population reduction measures. They are 4 substantially similar to those proposed by the Governor and many correctional experts. We 5 find credible the evidence that these measures, properly implemented, would not adversely 6 impact public safety or the operation of the criminal justice system. We also find that these 7 measures are feasible, and that they could achieve the population reduction required to 8 achieve constitutional levels of medical and mental health care delivery. 9 10 1. Early Release Through Expansion of Good Time Credits California, like the federal government and nearly every other state, has a system 11 through which inmates can earn credits to reduce their prison sentences by complying with 12 prison rules or by participating in rehabilitative, education, or work programs. See, e.g., Rep. 13 Tr. at 1398:6-15 (Austin); id. at 1549:23-1550:14 (Beard). California s inmates can earn 14 credits off their prison sentences through participation in work, educational, vocational, 15 therapeutic or other prison activities and for good behavior. Cal. Penal Code §§ 2931, 16 2933.70 CDCR Undersecretary Jett testified that the purpose of California s good time 17 credits system is to provide an incentive for inmates to participate in education and work 18 programs because those programs can reduce recidivism. Rep. Tr. at 1724:6-16. 19 Experts presented by plaintiffs, defendants, and defendant-intervenors all supported 20 the expansion of this good time credits system. Secretary Lehman and Secretary Woodford 21 both recommended the expansion of the credit system as a way to reduce the prison 22 population without adversely affecting public safety. See Aug. 15, 2008 Lehman Report 23 ¶ 13; Rep. Tr. 1326:21-1327:2, 1361:2-13 (Woodford). The public safety experts for 24 defendants and defendant-intervenors criticized generic early release programs but testified 25 that they were not opposed to the good time credits system. See Rep. Tr. at 1991:22-25 26 27 70 These credits are referred to, variously, as earned good time credits, good time 28 credits, earned credits, time credits, or earned time credits. 139 1 (Marquart) (stating that he is not opposed to granting earned credits for compliance with 2 prison rules);71 Aug. 15, 2008 Bennett Report ¶ 79; Rep. Tr. at 1015:21-1016:2 (Rodriguez). 3 Defendants themselves have proposed the expansion of earned good time credits, and 4 they would [o]f course not propose plans that would endanger public safety. Id. at 1685:35 15 (Cate); id. at 1921:14-1922:1 (Kernan). The Governor s 2008 and 2009 budget proposals 6 included an enhancement in the award of good time credits for up to four months for each 7 program successfully completed by an eligible inmate, reasoning that [i]ncentivizing 8 program participation and completion will reduce inmate violence within the CDCR and will 9 facilitate the inmate s reintegration into society. Ex. P780 at 18 (Governor s Budget, 10 Special Session 2008-09); Jan. 16, 2009 Sturges Decl., Ex. A at 28 (2009-10 Governor s 11 Budget).72 The Corrections Independent Review Panel chaired by former Governor 12 Deukmejian also recommended the expansion of the earned time credits system as one 13 component of comprehensive reform of the prison system. Ex. P4 at 122, 130. The CDCR 14 Expert Panel made the same recommendation. See Ex. P2 at ix, 92. 15 Expansion of the good time credits system would reduce the prison population by 16 allowing inmates to shorten their lengths of stay in prison by a few months. The evidence 17 indicates that such moderate reductions in prison sentences do not adversely affect either 18 recidivism rates or the deterrence value of imprisonment. According to Dr. Austin, a 19 correctional sociologist and plaintiffs expert, criminologists have known for many, many, 20 many years that generally there is no difference in recidivism rates by length of stay in 21 prison, so reducing the length of stay by a very moderate period of time four to six 22 months would have no effect on recidivism rates. Rep. Tr. at 1387:1-11. Other experts, 23 including an expert for defendants and an expert for defendant-intervenors, agreed with the 24 71 Dr. Marquart criticized the earned credit system implemented in Texas in his expert 25 report, see Aug. 14, 2008 Marquart Report at 19, but testified that he was not opposed to earned credits more generally. As we discuss below, Dr. Marquart s testimony on this issue 26 is contradictory and unreliable. 27 72 We do not consider here other proposed measures in budget messages submitted by the Governor post-trial. We are aware, however, that they contain additional proposals for 28 reducing the prison population safely. 140 1 proposition that there is no statistically significant relationship between an individual s 2 length of stay in prison and his recidivism rate. E.g., id. at 1325:11-16 (Woodford); id. at 3 1995:21-24 (Marquart); id. at 1154:18-24 (Powers); see also Ex. DI-204 at 1 (April 2007 4 National Council on Crime and Delinquency report, Effect of Early Release from Prison on 5 Public Safety: A Review of the Literature ). Dr. Austin s study of the CDCR data confirmed 6 that this is true of inmates in California s prisons. Aug. 27, 2008 Austin Supp. Report 7 ¶¶ 4-8.73 Similarly, a moderate reduction in an inmate s length of stay in prison would not 8 affect the deterrence value of imprisonment. According to two correctional experts, 9 including one presented by a defendant-intervenor, certainty of punishment and the 10 quickness with which penalties are brought to bear have a much bigger effect on deterrence 11 than a marginal difference in the level of sanctions. Rep. Tr. at 2106:2-7 (Krisberg) 12 ( There s a pretty large consensus that minor reductions [in sanctions] are not going to make 13 a big difference. ); id. at 2194:19-2195:18 (Bennett) (testifying that [i]t s not the severity of 14 the sanction that s important but the certainty and the immediacy of it ). 15 Defendants expert Dr. Marquart opined as a general proposition that shortening the 16 length of stay in prison by advancing good time credits could negatively impact recidivism 17 because it might reduce the opportunity for inmates to complete rehabilitation programming. 18 E.g., Aug. 14, 2008 Marquart Prelim. Report at 20-21.74 Dr. Marquart s opinion amounts, at 19 most, to a note about the factors that should be considered in designing an effective expanded 20 good time credits system. It is entitled to little, if any, weight as an observation about the 21 possible negative effect on public safety of such a system. First, as noted above, 22 approximately 50% of the 134,000 inmates released from California s prisons annually are 23 currently released without the benefit of any rehabilitation programming and, in fact, 24 73 In fact, some evidence suggests that properly targeted early release programs can 25 actually reduce recidivism rates. E.g., Sept. 8, 2008 Krisberg Report at 5. 74 26 Dr. Marquart stated that shortening the inmates lengths of stay might present a particular problem with respect to inmates in conservation camps who are trained to fight 27 fires. Aug. 14, 2008 Marquart Prelim. Report at 21. It appears, however, that his concern with respect to these inmates is not that their recidivism rates would increase but that their 28 release could severely impact the services these inmates render to the state. Id. 141 1 evidence shows that inmates with shorter sentences are especially unlikely to benefit from 2 such programming at this time.75 See Ex. P113 at 13 (Rehabilitation Strike Team Report); 3 Ex. P5 at 76. Thus, for at least 50% of the inmates released from California s prisons each 4 year, an expanded good time credits program would not, at present, reduce their 5 opportunities to complete rehabilitation programs. 6 More important, defendants are at the beginning stages of expanding rehabilitation 7 programming in the state s prison system. The CDCR has recently begun to implement an 8 evidence-based system of rehabilitation programming to reduce recidivism, and it has also 9 taken steps to increase utilization of existing educational, vocational, and substance abuse 10 programs. Rep. Tr. at 1710:20-1711:19, 1714:19-1715:1 (Jett). The evidence is clear that 11 expanded rehabilitation programming, and expanded inmate participation in such 12 programming, is a necessary component of California s goal of reducing its high recidivism 13 rate. One of the proposals advanced by Defendant Governor Schwarzenegger in his 2008-09 14 budget would authorize the CDCR to provide up to four months of earned credit for each 15 program successfully completed by an eligible inmate. Ex. P780 at 18. Thus, the Governor 16 contemplates completion of rehabilitation programs as one of the foundations of an expanded 17 earned good time credits system. 18 We also reject the testimony that inmates released early from prison would commit 19 additional new crimes. Even aside from the fact that many of these individuals would be less 20 likely to reoffend because they benefitted from completing rehabilitative programs, the 21 weight of the evidence showed that, because length of stay is unrelated to recidivism, all else 22 being equal the likelihood that a person who is released a few months before his original 23 release date will reoffend is the same as if he were released on his original release date. See, 24 e.g., Rep. Tr. at 1966:20 -1967:5 (Marquart); id. at 2653:2-15 (Yim) (Chief of Correctional 25 75 Inmates with short sentences who participate in San Diego County s community re26 entry program are receiving some rehabilitation programming in prison. San Diego, however, is the only county that has implemented such a program since the 2005 passage of 27 Senate Bill 618, which authorized them. At the time of the trial, San Diego s program had only 389 inmate-participants. Rep. Tr. at 987:9-16 (Rodriguez) (San Diego County Deputy 28 District Attorney). 142 1 Services Division of the Los Angeles County Sheriff s Department). Shortening the length 2 of stay in prison thus affects only the timing and circumstances of the crime, if any, 3 committed by a released inmate i.e., whether it happens a few months earlier or a few 4 months later. Id. at 1329:16-19 (Woodford); id. at 2319:1-23 (Dyer) (City of Fresno Police 5 Chief); id. 1569:11-20 (Beard); id. 2163:12-19 (Krisberg); see also id. at 1769:5-13 6 (Hoffman) (CDCR s Director of Adult Parole Operations) (testifying that returning technical 7 parole violators to prison only postpones victimization and crime). Although there might be 8 an increase in arrests in the initial months of an early release, see, e.g., Austin Aug. 15, 2008 9 Report ¶¶ 93-95; Austin Aug. 27, 2008 Report ¶ 9, this increase represents only a 10 concentration in the number of arrests that would have happened in any event and does not 11 affect the total number of arrests. While the victims of crimes may be different, and we do 12 not underestimate the significance of early release to those victims, our concern under the 13 PLRA is to consider the overall impact on public safety, which we find would be no 14 different. 15 Thus, the testimony by defendant-intervenors regarding the increased arrests that 16 followed early releases in two counties and one city does not undermine our conclusion that 17 early release through an expanded good time credits program could be implemented without 18 adversely affecting public safety. Defendant-intervenors witnesses offered the following 19 testimony. In Orange County in the 1990s inmates were released early due to court-ordered 20 population caps, and a number were rearrested for crimes committed during the time they 21 would otherwise have been in custody. Ex. DI-628 (July 1, 1997 Sheriff s Presentation on 22 Theo Lacy [Jail] Expansion to Orange County Board of Supervisors); see also Dostal Decl. 23 ¶ 11 (Executive Director of Administrative Services for Orange County Sheriff s 24 Department); Dostal Supp. Decl. ¶ 2. In Los Angeles County, 10% of those released from 25 jail pursuant to an early release program were rearrested during the period of early release, 26 including 16 for murder, over a five-year period. Rep. Tr. 1811:18-1812:4, 1824:14-15 27 (Smith) (Lieutenant in Los Angeles County Sheriff s Department Custody Support Services 28 Division). While this evidence may suggest an accelerated arrest pattern, it does not show an 143 1 increase in the overall crime rate. Chief Dyer of the Fresno Police Department testified that 2 when an increased number of parolees were released in 2005 as part of an earned credits 3 program, the city experienced an increase in crime, id. at 2329:20-2330:11, but his testimony 4 again did not reveal whether the crime represented only a temporary bulge, whether other 5 factors affecting crime remained unchanged, or whether a risk assessment tool which 6 measures the probability that an offender will recidivate, see id. at 2128:24-2129:1, 2132:67 2134:13 (Krisberg) was used to target low- to moderate-risk inmates for release. 8 In fact, empirical evidence from California s communities demonstrates that early 9 release programs as well as diversion, a population reduction measure we discuss in more 10 detail below do not increase crime. Dr. Krisberg reviewed data provided by California and 11 the FBI and concluded that such programs, which were instituted in twenty-one California 12 counties between 1996 to 2006, resulted in approximately 1.7 million inmates released by 13 court order but did not result in a higher crime rate. Sept. 8, 2008 Krisberg Report at 10. 14 This is persuasive evidence that the early release program proposed by plaintiffs poses no 15 threat to public safety or the operation of the criminal justice system. 16 Furthermore, if the good time credits system is expanded and programming enhanced, 17 it is likely that recidivism will decrease. Expansion of the good time credits system could 18 include an increase in the number of credits that prisoners can earn for participation in 19 programs, or being in compliance with a case management plan. Rep. Tr. at 1387:16-18 20 (Austin). Such an incentive contributes to a decline in recidivism because it gives [inmates] 21 what they need [in order] to keep them out of prison in the future, id. at 1549:21-22 (Beard), 22 as determined by an evidence-based assessment of the underlying factors, such as addiction 23 or lack of vocational skills, that may have driven the inmate s criminal behavior. Id. at 24 1550:18-1551:19 (Beard); see also id. at 1398:21-1399:1 (Austin) (reducing sentence length 25 due to inmate s completion of an education program is a win-win because it lowers the 26 length, and therefore the cost, of incarceration and lowers the likelihood of the inmate s 27 recidivism upon release). The evidence tendered thus confirms the conclusion of the CDCR 28 Expert Panel that the public safety benefits of [expanding good time credits] will be a vast 144 1 improvement over California s current practice of releasing offenders who have not 2 completed rehabilitation programming. Ex. P2 at 12. 3 Based on all of the above, we conclude that shortening an inmate s length of stay in 4 prison would not increase recidivism rates, and that shortening the length of stay through 5 earned credits would give inmates incentives to participate in programming designed to 6 lower recidivism. We credit the opinions of the numerous correctional experts that the 7 expansion of good time credits would not adversely affect but rather would benefit the public 8 safety and the operation of the criminal justice system. We also note that this is the view of 9 the Governor, who has recommended the adoption of an earned credit program as a means to 10 better facilitate the inmate[s ] reintegration into society. Ex. P780 at 18 (Governor s 11 Budget, Special Session 2008-09). 12 13 2. Diversion of Technical Parole Violators California has a very abnormal practice of sending a high number of technical 14 parole violators to prison for a short of amount of time.76 Rep. Tr. 1434:12-14 (Austin); 15 Aug. 15, 2008 Austin Report ¶ 13 (explaining that California s re-arrest rate for parolees is 16 similar to other states, but the high use of imprisonment for parole violations produces a high 17 return to prison rate). In California, more than 70,000 parolees are returned to prison each 18 year for technical parole violations, approximately 17,000 of whom are pure technical 19 violators who have not been arrested for a new crime but have only violated a term or 20 condition of their parole. Ex. P5 at 72-74; Rep. Tr. at 1739:18-19 (Hoffman). 21 Evidence including testimony from Thomas Hoffman, Director of the CDCR s 22 Division of Adult Parole Operations overwhelmingly showed that California s practice of 23 sending parole violators back into the state prison system for an average of four months and 24 incarcerating them during that time in crowded reception centers endangers public safety and 25 burdens the criminal justice system. See, e.g., Rep. Tr. at 1769:5-13 (Hoffman) (agreeing 26 76 Technical parole violators are those parolees who have violated their conditions of 27 supervised release but have not been convicted of new crimes. Aug. 15, 2008 Austin Report ¶ 12. The category includes those who have been arrested for new crimes but were not 28 prosecuted or convicted. Id. 145 1 that this churning pattern just postpones victimization and crime and testifying that we 2 know it s not working ); Ex. P113 at 78 (Rehabilitation Strike Team Report) ( This system 3 of catch and release makes little sense from either a deterrence, incapacitation, treatment, or 4 economic standpoint. ). According to research by Professor Petersilia, this high return-to5 prison rate for parole violators is creating a destructive situation by constantly cycling 6 offenders in and out of prison and their home communities in a way that blurs the distinction 7 between the two and combines the worst elements of each. Ex. P5 at 75. Professor 8 Petersilia found that, among other negative effects, this churning or catch-and-release 9 disrupts the inmate s ability to participate in community-based rehabilitative programs, 10 encourages the spread of prison-gang culture in communities, wastes parole processing 11 resources, and reduces the deterrent value of prison by transform[ing] a trip to prison into 12 . . . a trivial and short-lived intrusion on day-to-day criminality. Id. at 76. Secretary 13 Woodford, the former acting Secretary of the CDCR, agreed with this assessment, based on 14 her experience administering California prisons in various capacities, including as warden at 15 San Quentin. Rep. Tr. at 1316:23-1317:11; see also Ex. D1196 at DEFS021721 (Integrated 16 Strategy to Address Overcrowding in CDCR s Adult Institutions) ( [C]hurning is costly, 17 does little or nothing to promote public safety and frustrates real efforts at rehabilitation. ). 18 Dr. Gilligan, plaintiffs mental health expert, testified that this practice has a particularly 19 adverse impact on the mentally ill, who are not given adequate treatment or transition plans 20 because of the short length of their return to prison. Aug. 15, 2008 Gilligan Report ¶ 33. 21 This churning, and its adverse effects, could be stopped in several ways. One is to use 22 a parole revocation instrument to determine whether parole violators should be sent back to 23 prison. Rep. Tr. at 1385:11-21 (Austin); Aug. 15, 2008 Austin Report ¶ 52. Such an 24 instrument has been implemented in a number of states, including Pennsylvania, South 25 Carolina, New Jersey, Oregon, Georgia, Iowa, Kansas, South Dakota, and Texas.77 Id.; Rep. 26 Tr. at 1564:4-1565:8 (Beard). The CDCR has already started implementing reform of the 27 77 Washington State has a law that prevents technical parole violators from being 28 returned to prison. Aug. 15, 2008 Lehman Report ¶ 16. 146 1 parole system and has developed a Parole Violation Decision Making Instrument. Id. at 2 1678:15-25 (Cate); Hoffman Trial Aff. ¶¶ 8-13; Ex. D1198 (Sept. 30, 2008 Letter from 3 Jessica R. Devencenzi, Deputy Attorney General, to Michael Bien, Rosen Bien & Galvan, 4 LLP). Secretary Cate called the use of the parole revocation instrument one of the best 5 practices in the area of parole reform. Rep. Tr. at 1706:9-14 (Cate). Dr. Austin noted a 6 number of other ways to reduce the return of technical parole violators to prison, including 7 prohibiting parole violators from being readmitted to prison for technical violations, reducing 8 the period of parole supervision, and instituting an incentive program for parole agents. 9 Aug. 15, 2008 Austin Report ¶¶ 51-54. The Governor has proposed placing all non-serious, 10 non-violent, non-sex offenders on summary parole. See Ex. P780 at 18 (Governor s 11 Budget, Special Session 2008-09); Jan. 16, 2009 Sturges Decl. ¶ 2 & Ex. A at 28 (2009-10 12 Governor s Budget); Ex. P328 at 178 (Governor s Budget Summary 2008-09). All of these 13 options may be considered by the state and implemented in a manner that would be 14 consistent with its ultimate objectives. 15 The use of a best practices instrument, as well as other methods referred to above, to 16 reduce the number of parole violators returned to the state prison system, if properly 17 implemented, would not have an adverse impact on public safety or on the criminal justice 18 system. At the very least, slowing the flow of technical parole violators to prison would 19 mitigate the dangerous crowding at reception centers and ease the burden on the parole 20 processing system. It would free up space in the reception centers so that those centers could 21 be used for their original purpose: sorting inmates into the right correctional settings. It 22 would give parolees a better opportunity to participate in continued rehabilitative 23 programming in the community, and it would likely improve a system that currently 24 undercut[s] the deterrent effect of serving prison time. Ex. P5 at 76. We agree, for 25 example, with the assertion of the CDCR s Director of Adult Parole Operations that the use 26 of a parole revocation instrument in California would reduce future victimization, increase 27 public safety, and enhance the ability of offenders to become more productive members of 28 the community. Hoffman Trial Aff. ¶ 10. 147 1 Additionally, public safety would improve if technical parole violators who are not 2 returned to prison were diverted to alternative sanctions in the community, including drug 3 treatment, day reporting centers, electronic monitoring, and, if necessary, county jail.78 See 4 Rep. Tr. at 1318:21-1319:2 (Woodford). The CDCR already has alternative sanctions 5 programs and is working on expanding the scope and availability of such programs. 6 Hoffman Trial Aff. ¶¶ 19-25. Many of these programs address the offender s criminogenic 7 factors and can thus reduce recidivism. Id. The use of graduated sanctions would serve the 8 same deterrent purpose as imprisonment while effectively reducing recidivism. See 9 Woodford Aug. 15, 2008 Supp. Report ¶ 32 ( Sanctions other than incarceration are effective 10 in punishing many prisoners and at the same time reducing the risk of recidivism. ); Rep. Tr. 11 at 2194:19-2195:18 (Bennett) ( We need to have meaningful, immediate, certain sanctions. 12 And it doesn t have to be a return to prison. We can develop sanctions at the local level. . . . 13 We can have a more effective sanction without interrupting individuals lives and returning 14 them to prison. ). Former CDCR Secretary James Tilton stated that he believed that these 15 alternative community sanctions programs would improve public safety over time. Sept. 3, 16 2008 Tilton Dep. at 153:4-154:3. 17 CDCR officials and experts overwhelmingly supported the use of the parole 18 revocation instrument and the diversion of technical parole violators to alternative sanctions 19 in the community. Scott Kernan, the CDCR s Undersecretary of Operations, stated that 20 [t]hese efforts have proven to reduce prison population while maintaining public safety. 21 Kernan Trial Aff. ¶ 23. Experts for plaintiffs, defendants, and defendant-intervenors testified 22 in favor of the diversion of technical parole violators. E.g., Aug. 15, 2008 Lehman Report 23 ¶ 16; Rep. Tr. at 1993:6-8 (Marquart); id. at 2194:19-2195:18 (Bennett); Buddress Trial 24 Decl. ¶ 3. The Governor s Rehabilitation Strike Team urged the use of the parole violation 25 instrument and diversion, concluding that [s]imilar best practices proposals have worked 26 in other states to better prepare inmates for re-entry, reduce prison returns, protect public 27 safety, and reduce the costs of corrections. Ex. P113 at 17, 89-90. The three reports of 28 78 We discuss the likely impact of these measures on the counties infra Section VII.C. 148 1 independent commissions presented to the state the CDCR Expert Panel Report, the 2 Corrections Independent Review Panel Report, and the Little Hoover Commission Report 3 also recommended the diversion of technical parole violators. See Ex. P2 at 47-49; Ex. P3 at 4 31; Ex. P4 at 154, 158-59. In fact, according to the CDCR Expert Panel, fifteen reports 5 published since 1990 on California s prison crisis have recommended the diversion of 6 technical parole violators. Ex. P2 at 77. 7 We conclude that simply slowing the flow of technical parole violators to prison, 8 thereby substantially reducing the churning of parolees, would by itself improve both the 9 prison and parole systems, and public safety. Diversion of parole violators to community 10 alternative sanctions programs would serve to significantly reduce recidivism. We therefore 11 find that diverting parole violators to alternative community sanctions programs would 12 reduce the prison population while having a positive rather than a negative effect on public 13 safety and the operation of the criminal justice system. 14 15 3. Diversion of Low-Risk Offenders with Short Sentences Plaintiffs also propose reducing the prison population by diverting low-risk offenders 16 with short sentences for community sanctions. Rep. Tr. 1385:22-1386:21 (Austin); Aug. 15, 17 2008 Austin Supp. Report ¶¶ 58-61. 18 According to Dr. Austin, a substantial number of inmates enter the California prison 19 system with sentences of less than twenty-four months, the largest group of which are those 20 with a sixteen-month sentence, many of whom have already served up to seven months of 21 their sentence in a county jail. Rep. Tr. 1386:2-1386:12; Aug. 15, 2008 Austin Supp. Report 22 ¶ 60 & tbl. 5. Under current policies, these inmates can halve the remaining periods of their 23 sentences by earning work credits, with the result that these inmates serve only a few months 24 in state institutions an amount comparable to that served by technical parole violators. 25 Rep. Tr. 1386:2-1386:12. Like the technical parole violators, these inmates are unlikely to 26 participate in any meaningful programming during their short term of imprisonment. See 27 Aug. 15, 2008 Austin Report ¶ 60 ( A diversion program would eliminate a short period of 28 149 1 imprisonment within the CDCR (during which the prisoner is unlikely to become involved in 2 any meaningful programming). . . . ). 3 Instead of incarcerating all of these offenders, the CDCR could use risk assessment 4 instruments to identify low-risk offenders and divert these offenders to community 5 correctional programs to serve their sentences. See Rep. Tr. 1386:13-21 (Austin). The state 6 might also consider implementing incentive-based funding for community corrections, 7 similar to that adopted by California in the 1960s, when the state provided fiscal rewards to 8 counties that reduced the number of people being sent to prison. See Rep. Tr. at 1042:4-14 9 (Powers). This would require the diversion of only a portion of the funds that adoption of the 10 reforms discussed herein would save the state. 11 A number of correctional and law enforcement experts opined that the diversion of 12 low-risk offenders would not have an adverse impact on public safety or the operation of the 13 criminal justice system. Secretary Woodford stated, based on her prior experiences as the 14 chief probation officer of San Francisco, warden of San Quentin, and acting Secretary of the 15 CDCR, that California incarcerates many more prisoners than is necessary for the safety of 16 the public. Aug. 15, 2008 Woodford Supp. Report ¶ 32. She stated that there are 17 intermediate sanctions available, and that California would have safer communities if it used 18 those sanctions rather than incarceration in appropriate circumstances. Id. The use of such 19 intermediate sanctions would not significantly affect deterrence, as sanctions short of 20 imprisonment have deterrent value so long as they are meaningful, immediate, and certain. 21 See, e.g., Rep. Tr. at 2194:19-2195:18 (Bennett). 22 Law enforcement officials from the counties also testified that diversion could 23 improve public safety if implemented correctly. See, e.g., James Trial Decl. ¶¶ 6-7 (Orange 24 County Assistant Sheriff); Rep. Tr. at 2369:5-12 (Dyer); Buddress Trial Decl. ¶¶ 10-11. 25 According to these local law enforcement officials, offenders who have not been to prison 26 are easier to program [and] treat . . . before they have been exposed to (and potentially 27 trained by) more hardened and experienced criminals in the state prison system. James 28 Trial Decl. ¶ 20; see also Rep. Tr. at 1052:16-1053:10 (Powers) ( [S]o you put someone who 150 1 is a low risk, low level person into an environment[] with high risk individuals, they don t 2 naturally get better. They gravitate up. So when they come out, they are worse off. ). 3 According to the testimony of law enforcement and county officials, many counties now 4 successfully divert offenders from jail to substance abuse programs, correctional day 5 reporting centers, and electronic monitoring. See, e.g., id. at 2276:19-2277:1 (Graves); id. at 6 2798:3-24 (Hennessey) (City and County of San Francisco Sheriff); Dalton Am. Trial Decl. 7 ¶¶ 33-35. Thus, successful models for community corrections are already in place, and, 8 although the characteristics of the populations that they currently serve may be different from 9 the prison population, they can be expanded to serve an increase in diverted offenders with 10 proper funding and coordination between the state and the counties. See, e.g., Aug. 15, 2008 11 Garner Report at 6 (Director of Santa Clara County Department of Alcohol and Drug 12 Services) ( Local treatment systems exist in every county and with adequate state funding 13 they can be expanded to accommodate the proposed increase in clients resulting from early 14 release of prisoners. ); Meyer Am. Trial Decl. ¶ 69. An expert for the law enforcement 15 intervenors testified that if the state were to establish such programs on a statewide or 16 county-by-county level, the prison population could be reduced, by that reform alone, by 17 about 30%, as a conservative estimate, in two to five years. Rep. Tr. at 2771:4-10 (Meyer).79 18 The opinion of these California correctional and law enforcement experts was 19 confirmed by Dr. Beard, the Secretary of the Pennsylvania Department of Corrections, who 20 testified regarding Pennsylvania s success in implementing an intermediate punishment 21 program that diverts offenders from jails and prisons to substance abuse programs. Id. at 22 1554:20-1556:20. A study of that program found that inmates in the program had lower 23 recidivism rates than those sent to county jails or state prisons. Id. at 1555:2-5. Dr. Beard 24 testified that the research is really clear out there that community-based programming is 25 actually more effective than prison-based programming. Id. at 1555:21-23. Such 26 programming can contribute to rehabilitation without taking the offender away from the 27 79 The witnesses for defendant-intervenors expressed concern that neither the state nor the counties have the funds to expand the community correctional system. We address this 28 concern below. Infra Section VII.C. 151 1 community and creating the problems of re-entry upon release from prison. Id. at 1556:1-14. 2 According to Dr. Austin, other states, including Ohio and Michigan, have also successfully 3 adopted diversion programs without an adverse effect on crime. E.g., id. at 1399:2-15. 4 There was testimony that some individuals on electronic monitoring or in other 5 alternative programs have committed crimes, e.g., id. at 1179:23-1180:3 (Powers), and we 6 acknowledge that diversion programs cannot stop all crime. But, again, the individuals to be 7 diverted are those who would have been released from prison a few months later in any 8 event, after being exposed to more hardened and experienced criminals. James Trial Decl. 9 ¶ 20. Thus, the incidents that have occurred during participation in alternative programs do 10 not undermine the weight of the testimony that diversion programs have an overall positive 11 effect on public safety and the operation of the criminal justice system. 12 We therefore conclude that the diversion of offenders to community correctional 13 programs has significant beneficial effects on public safety and the operation of the criminal 14 justice system as compared to the current system, including preventing the exposure of 15 offenders to criminogenic conditions, providing effective rehabilitation, and avoiding a 16 disruption in the offender s life that creates re-entry problems upon release. 17 18 19 4. Expansion of Evidence-Based Rehabilitative Programming in Prisons or Communities Every witness, from the CDCR s Undersecretary of Programming to law enforcement 20 officers and former heads of correctional systems, testified that an increase in the availability 21 of evidence-based rehabilitative programming i.e., programs that research has proven to be 22 effective in reducing recidivism, Rep. Tr. at 1042:19-1043:14 (Powers) in the prisons or in 23 the communities would reduce the prison population and have a positive impact on public 24 safety. See, e.g., id. at 1721:16-22 (Jett); id. at 1159:14-19 (Powers); id. at 1962:15-23 25 (Marquart); id. at 2009:15-2010:1 (Lehman); id. at 2728:6-18 (Runner); id. at 2385:2-14 26 (Pacheco) (Riverside County District Attorney); Meyer Am. Trial Decl. ¶¶ 46-52. Research 27 from Washington State concluded that an expansion of evidence-based programming would 28 reduce the prison population, while leading to a net decrease in the crime rate. See 152 1 Ex. D1331 at 15 (Oct. 2006 Washington State Institute for Public Policy report, Evidence2 Based Public Policy Options to Reduce Future Prison Construction, Criminal Justice Costs, 3 and Crime Rates ). 4 Experience demonstrates the benefits of evidence-based programming. Missouri and 5 Washington have successfully and safely reduced prison populations through such 6 programming. See Rep. Tr. at 2767: 21-2768:11 (Meyer). Moreover, the evidence from the 7 law enforcement intervenors and county intervenors overwhelmingly showed that there are 8 already models for successful evidence-based programs all over California, from Yolo 9 County to San Diego County, that have reduced recidivism and thus improved public safety 10 in those communities. See, e.g., id. at 2784:25-2785:4 (Meyer); id. at 2803:19-2804:1 11 (Hennessey); Rodriguez Trial Decl. ¶¶ 20-21; Aug. 15, 2008 Bennett Report app. C ch.3 12 ( Chapter Three: Alternatives to Incarceration from July 2007 Sonoma County, California: 13 Corrections Master Plan ). As Chief Probation Officer Meyer stated, successful models are 14 on the shelf and ready to be implemented. Rep. Tr. 2784:25-2785:4. 15 As discussed above, the CDCR has also already begun to design and implement an 16 expansion of rehabilitation services for inmates and parolees. See Jett Trial Aff. ¶¶ 6-13; 17 Ex. P79 (July 15, 2008 California Rehabilitation Oversight Board Biannual Report). We 18 agree with Undersecretary Jett, who oversees this process, that its successful implementation 19 would lead to a reduction in recidivism and a reduction in the prison population. See Jett 20 Trial Aff. ¶ 13.80 Improvements in the implementation of the process will become 21 increasingly likely as the reduction in the prison population occurs. The two functions are, in 22 practicality, related. 23 Based on the overwhelming and uncontroverted evidence, we find that additional 24 rehabilitative programming would result in a significant population reduction while 25 improving public safety and reducing the burden on the criminal justice system. If 26 implemented in conjunction with any or all of the population reduction measures described 27 80 Successful implementation of such programming will, of course, require space that 28 is currently not available in California s prisons. 153 1 above, such programming would enhance the likelihood that recidivism will decline as the 2 prison population is decreased. Moreover, if implemented within the state prison institutions, 3 such programming would have a synergistic effect on the ability of inmates to reduce their 4 sentences by earning good time credits. 5 6 5. Sentencing Reform and Other Potential Population Reduction Measures The evidence at trial focused primarily on the potential effects of the population 7 reduction measures proposed by plaintiffs. However, there are other means as well by which 8 the state could reduce its prisoner population, and the state is in no way bound by plaintiffs 9 proposals. For example, Expert Panel co-chair Professor Joan Petersilia reported that fifteen 10 studies have been issued regarding California prisons since 1990, all containing essentially 11 the same ten recommendations. Ex. P2 at 77-79. Those recommendations include not only 12 the four population reduction measures proposed by plaintiffs, but also reformation of the 13 state s determinate sentencing regime to reward prisoners for participating in rehabilitation 14 programs and allow the system to retain prisoners who represent a continued public safety 15 risk, the creation of a sentencing reform commission authorized to design new sentencing 16 statues into a workable system that balances uniformity of sentencing with flexibility of 17 individualization, and the release or diversion of certain [s]ub-populations, such as women, 18 the elderly and the sick from prison to community-based facilities. Id. at 77. Also, as noted 19 above, the state has suggested that its prison population might be reduced through the 20 transfer of inmates out of state or into federal custody. The state is certainly free to include 21 any of these alternatives in its proposed population reduction plan should it be able to 22 establish the feasibility and the positive effects of such programs, especially their 23 compatibility with public safety. 24 Like plaintiffs proposed population reduction measures, the other measures discussed 25 by Professor Petersilia generally would have a positive effect on public safety. In particular, 26 the repeated recommendation that the state establish a sentencing commission and reform its 27 determinate sentencing regime reflects an urgent need for the state to reconsider its 28 counterproductive sentencing practices. As the Little Hoover Commission reported, 154 1 California s present sentencing regime is a chaotic labyrinth of [sentencing] laws with no 2 cohesive philosophy or strategy. Ex. P3 at 35. The state s sentencing laws promote 3 certainty in the length of sentences at the expense of public safety: Because release at a 4 particular date is certain, offenders have little incentive to improve themselves in prison or 5 while on parole,81 and offenders must be released even if they pose a serious threat to the 6 community. Id. at 34. In addition, sentencing judges and prison authorities have little ability 7 to ensure that sentences and conditions of incarceration reflect the circumstances of a 8 particular crime and offender. Similarly, characteristics suggesting that the offender presents 9 a low-risk of recidivism or would more effectively serve his sentence in a correctional setting 10 besides prison, including the fact that the offender is elderly or infirm, cannot be considered. 11 [California sentencing] law treats many crimes alike, even when the circumstances of an 12 individual case or the characteristics of the offender might warrant a different resolution that 13 would better benefit victims and the community. Id. at 36. Furthermore, the present system 14 leads to overreliance on the most expensive sanction state prison instead of local 15 correctional alternatives that could provide more effective and efficient punishment. Id. 16 Finally, the countless increases in the length of criminal sentences over the last few 17 decades do not reflect a coherent sentencing policy and also may not serve the state s 18 sentencing goals. Id. at 33, 35, 48. Public safety is not benefitted by blindly approving of 19 the continued incarceration of prisoners who pose little threat of committing further crimes. 20 Like a number of other official bodies, the Little Hoover Commission recommended that a 21 sentencing commission be established to develop sentencing guidelines, as well as post 22 release supervision and revocation guidelines that [would] become law unless rejected by a 23 majority vote of the Legislature, id. at 48. 24 The establishment of a sentencing commission is but one approach to addressing the 25 problems in the state s sentencing laws; there are undoubtedly others. Regardless of the 26 approach adopted by the state, however, it is clear that California s sentencing regime ill27 81 However, according to the Commission, incentives can be built into the existing 28 sentencing structure to improve public safety and offender outcomes. Id. at 37. 155 1 serves the state s interests, and that the overcrowding crisis in California s prisons provides 2 an opportunity for the state to reconsider its sentencing practices. Numerous reports have 3 recommended sentencing reform and established that such reform would, if implemented, 4 have a positive impact upon public safety. See id. at 38-42 (describing the positive public 5 safety effects in various states of their use of a sentencing commission). Given the fact that 6 legislative bodies tend to vote only to increase sentences and not to reduce them, however, 7 and given the questionable nature of California s initiative process, there appears to be little 8 or no hope of a serious review of sentencing laws or policies in the absence of some 9 extraordinary state action. 10 The state might also consider changing the criminal law itself. For example, the 11 Governor has proposed adjusting the threshold value at which certain property crimes 12 become felonies to reflect inflation since 1982. Ex. P780 at 18 (Governor s Budget: Special 13 Session 2008-09); Jan. 16, 2009 Sturges Decl. Ex. A at 28 (2009-10 Governor s Budget). 14 Such a change would reclassify crimes falling below the adjusted threshold as misdemeanors. 15 Likewise, the state might consider permitting low-risk offenders, such as the elderly or the 16 infirm, to serve the latter portions of their sentences in community corrections facilities or on 17 house arrest. Both of these proposals would reduce the prison population by diverting certain 18 offenders to alternative placements rather than prison. Both have also been endorsed by state 19 officials, a strong indication that the proposals would not have an adverse effect on public 20 safety. 21 The parties introduced no evidence as to the effect on public safety of the transfer of 22 inmates out-of-state or into federal custody, so we cannot consider those measures in detail at 23 this time. As we have already explained, however, the out-of-state transfer program 24 proposed by defendants is far too small, by itself, to make more than a dent in the problem of 25 overcrowding, and the additional resources required to monitor the medical and mental 26 health care provided to transferred inmates could eliminate any benefits that otherwise result 27 from such transfers. Furthermore, by moving inmates far away from their places of residence 28 and making contact with families and friends unavailable, such transfers may reduce the 156 1 inmates prospects for rehabilitation. The transfer of undocumented aliens to federal custody 2 might involve a larger number of inmates, but this suggestion was not sufficiently developed 3 to permit any extended analysis of its effect upon public safety and the operation of the 4 criminal justice system.82 C. 5 6 Impact of Proposed Measures on Communities Law enforcement and other witnesses from the communities testified that plaintiffs 7 proposed prisoner release order would result in an overwhelming increase in the number of 8 crimes, arrests, and jail inmates, thus adversely affecting their ability to investigate, 9 prosecute, and punish crime. We cannot accept their opinions, however, to the extent that 10 they are based on the assumption that a prisoner release order would involve such drastic 11 measures as a mass early release and/or a ban on the admission of new offenders to prison. 12 We credit the concern of some witnesses, however, that resources at the community level are 13 strained, particularly because of the current fiscal crisis. See, e.g., Cogbill Trial Decl. 14 ¶¶ 29-38 (Sonoma County Sheriff-Coroner); Boesch Trial Decl. at 8-9 (San Mateo County 15 Assistant County Manager); Aug. 15, 2008 Graves Report at 3-4. Nonetheless, as we discuss 16 below, the evidence demonstrates that the fears regarding increased crime, arrests, and jail 17 populations are largely unjustified, and that there are ways to achieve a reduction in 18 California s prison population without unduly burdening the already limited resources of 19 local communities. 1. 20 21 Investigation and Prosecution of Crime Defendant-intervenors presented credible evidence that California s local law 22 enforcement resources are currently overtaxed. There are not enough judges, prosecutors, 23 public defenders, police officers, or resources to support their necessary work, and the 24 25 26 27 28 82 The only evidence in the record on this proposal is testimony from one witness, California State Senator George Runner, that California prisons house roughly 30,000 illegal aliens, which is disproportionately high when compared to other states, and that federal reimbursement is insufficient to cover the costs of housing these inmates. Runner Trial Decl. ¶¶ 6, 19; Rep. Tr. at 2728:19-2729:12. We received no testimony on the feasibility of transferring all or even a portion of these inmates to federal custody, and no testimony regarding any potential impact on public safety or the operation of the criminal justice system of such a transfer, including whether the prisoners might be swiftly deported and just as swiftly re-cross the border into California shortly thereafter. 157 1 situation has worsened with the economic downturn. See, e.g., Rep. Tr. at 2197:5-2199:9 2 (Bennett); id. at 1856:13-21 (Word) (City of Vacaville Police Chief); Word Trial Decl. ¶ 25. 3 The courts are severely clogged with cases and are several years behind on trials. See Meyer 4 Am. Trial Decl. ¶ 43; Ryan Trial Decl. ¶ 28 (Amador County Sheriff-Coroner). Any 5 significant increase or concentration in crime would likely further hamper investigations and 6 prosecutions. See, e.g., Ryan Trial Decl. ¶¶ 27-28; Dumanis Trial Decl. ¶ 33; Dyer Am. 7 Report ¶ 28; Rep. Tr. at 1179:5-17 (Powers). 8 The population reduction measures described above, however, would not result in the 9 significant increase in crime that many witnesses opposed to the measures believe would 10 occur. As explained above, many witnesses wrongly assumed that this court would require a 11 sudden mass release of one-third of California s prisoners or a ban on accepting new or 12 returned prisoners. See, e.g., Rep. Tr. at 1052:8-12 (Powers); Aug. 15, 2008 Bennett Report 13 ¶¶ 13,18. That approach was not proposed by any party, nor would it be approved by the 14 court. 15 Many witnesses also testified that, at present, a large number of crimes are committed 16 by parolees, see, e.g., Rep. Tr. at 2331:1-8 (Dyer); parolees have a high rate of recidivism, 17 e.g., Meyer Am. Trial Decl. ¶¶ 39-40;83 and more crimes occur than are reported to the 18 police, e.g., Rep. Tr. at 1506:21-1507:20, 1508:11-19 (Austin). The parolees who would be 19 released early to communities under the proposed measures, however, are the ones who are 20 least likely to commit further offenses and who along with their fellow parolees would be 21 released in any event a few months later. Indeed, the evidence describing the criminogenic 22 nature of the California prisons suggests that the longer an inmate remains incarcerated, the 23 more likely he is to reoffend upon release. See, e.g., Rep. Tr. at 1580:5-9 (Beard); id. at 24 2013:14-2014:1 (Lehman); Ex. P3 at 17. The relevant question for us to examine is not the 25 absolute impact of the current population of parolees on local criminal justice systems, but 26 27 83 This testimony does not take into account that the recidivism rate for parolees is high in California in part because the state returns most technical parole violators to prison. See 28 Aug. 15, 2008 Austin Report ¶13. 158 1 the relative impact on the criminal justice system of the additional parolees in the community 2 because of the proposed population reduction order. 3 The evidence shows that any such impact would be small. The expanded award of 4 good time credits proposed by Dr. Austin, for example, would result in only a temporary 5 increase in the return of parolees to communities during the initial period of implementation. 6 Rep. Tr. at 1408:13-21 (Austin); Aug. 15, 2008 Austin Report ¶¶ 93-94. Although the 7 increase in parolees could result in a temporary increase in arrests during the initial period of 8 accelerated release, these arrests would represent an increase of only approximately 0.3% 9 during that period. Rep. Tr. at 1490:17-1491:25; see also Aug. 27, 2008 Austin Supp. Report 10 at 10; Rep. Tr. at 1479:13-1480:5. Similarly, the impact of the proposed diversion of 11 technical parole violators and low-risk offenders on the total number of arrests in each 12 county, and statewide, would be an increase of less than 1%. See Aug. 27, 2008 Austin 13 Supp. Report at 10. All of these individuals would in any event be released to the 14 community after a fairly short period of incarceration, following their going through the 15 churning process, in which they are subjected to criminogenic influences. Further, all of the 16 figures noted above are consistent with the testimony described earlier that plaintiffs 17 proposed population reduction measures do not threaten public safety or the operation of the 18 criminal justice system. 19 Any increase in the arrests of parolees resulting from the population reduction 20 measures would actually be smaller than that calculated by Dr. Austin and by many 21 defendant-intervenors. These witnesses assumed that prisoners released due to good time 22 credits or diverted to alternative sanctions would recidivate at a rate of 70% over a three-year 23 period, the average recidivism rate for all prisoners in California. See, e.g., Rep. Tr. at 24 2628:8-25 (Austin); Dyer Am. Report ¶ 18; Dostal Trial Decl. ¶ 14. However, if a risk 25 assessment instrument were used to implement such measures, the CDCR would be able to 26 identify low-risk inmates whose likelihood of recidivism would be considerably lower than 27 that of the average inmate. Rep. Tr. at 2628:8-25 (Austin); id. at 2133:8-11 (Krisberg) ( If 28 one is selecting low risk inmates, you would expect the recidivism rate would be lower 159 1 because that 70 percent rate consists of people with much higher risk and people with lower 2 risk. ). According to Director Hoffman, low-risk inmates have an average recidivism rate of 3 just 17%. Rep. Tr. at 1750:1-6. Furthermore, as we found above, it is likely that recidivism 4 rates would begin to drop as plaintiffs proposed measures were implemented. The proposed 5 population reduction measures would therefore not result in a significant additional burden 6 on the ability of law enforcement officers to investigate or prosecute crime. 7 8 2. Effect on Jail Population Defendant-intervenors also presented credible evidence that California s jails are, for 9 the most part, already overcrowded, resulting in adverse public safety and criminal justice 10 effects. Thirty-two of California s county jails are under some type of court-ordered 11 population cap, Rep. Tr. at 2198:3-9 (Bennett); Ex. DI-774,84 and many that are not have 12 inmate populations close to or above their design capacity. E.g., Rep. Tr. at 2684:22-23, 13 2686:15-22 (Ryan); Boesch Trial Decl. at 12. As expected, this overcrowding even at 14 levels much lower than in the state prison system has limited the counties capacity to 15 provide services in the jails or to maintain a safe correctional environment for the detainees, 16 the staff, and the community. See, e.g., Boesch Trial Decl. at 12; Munks Trial Decl. ¶¶ 7-9; 17 Rep. Tr. at 2702:5-17 (Ryan); Dostal Trial Decl. ¶¶ 15, 17. 18 As a result of this crowding problem, counties already routinely engage in the early 19 release of jail inmates. See, e.g., Rep. Tr. 1803:23-1804:9 (Smith) (stating that in 2007, Los 20 Angeles County released about 50,000 inmates early from its jails); Rep. Tr. at 2364:17-19 21 (Dyer); Rep. Tr. at 2378:13-18 (Pacheco); James Trial Decl. ¶ 19; Ingrassia Trial Decl. 22 ¶¶ 12-13 (Sheriff s Commander assigned to San Diego County Sheriff s Detention Services 23 Bureau). County law enforcement officials testified that any significant limit on the prison 24 population would force them to initiate the early release of jail inmates or to expand extant 25 early release programs to include higher-risk inmates. See, e.g., Rep. Tr. at 2388:8-2391:16 26 (Pacheco); id. at 2668:7-14 (Christianson) (Stanislaus County Sheriff-Coroner); Munks Trial 27 84 Many of the caps for the county jails are set at or near 100% design capacity. See 28 Ex. DI-774. 160 1 Decl. ¶¶ 11-12; Ingrassia Trial Decl. ¶ 12. According to these witnesses, such early releases 2 lower the deterrence value of incarceration, increase crime, reduce incentives for offenders to 3 participate in programming, and result in a high failure-to-appear rate for pre-trial defendants 4 who are not incarcerated. See, e.g., Aug. 15, 2008 Bennett Report ¶ 27; Rep. Tr. at 1179:185 1180:3 (Powers); id. at 1819:9-1821:19 (Smith). 6 We need not determine whether an acceleration of early release from jails would have 7 the pernicious effects anticipated by the law enforcement witnesses because evidence shows 8 that any increase in parolees and probationers resulting from plaintiffs proposed population 9 reduction measures would not have a significant effect on the population of the county jails. 10 These measures would adversely affect the jail population only if the additional parolees or 11 probationers in the community were incarcerated in jail for arrests for new crimes or as a 12 sanction for failing to complete community-based diversion programs. As Sheriff Munks of 13 San Mateo County noted, however, only a very, very small percentage of th[e] 14 overcrowding [in jails] is attributable to parolees who have been arrested and returned to 15 [the] jail. Id. at 1790:16-17 (Munks). Given the small adverse effect that the increase in 16 parolees and probationers would have on the total arrests in each county, this increase is not 17 likely to have a significant effect on the county jail population.85 See id. at 1409:2-23 18 (Austin); see also id. at 1830:21-1831:23 (Smith) (population reduction order of 52,000 19 inmates, even when calculated using the high 67.5% recidivism rate, would result in an 20 increase of only 20 admissions a day in the Los Angeles County jail system, which books 21 from 300 to 1,100 inmates every day). The diversion of technical parole violators could even 22 serve to reduce the jail population because those offenders would no longer have to be kept 23 in county jail pending their transfer to CDCR facilities. Aug. 15, 2008 Austin Report ¶ 88. 24 25 85 We reject some of the witnesses calculations of the impact of a population reduction 26 order on the county jails. Sheriff Munks, for example, agreed during trial that the method he used to calculate the impact of a population reduction order on the jail population was 27 inconsistent with the county s current experience with parolees. Rep. Tr. at 1794:19-22. Sheriff-Coroner Christianson admitted that he did not know how his staff calculated the 28 estimated impact on his jail population. Id. at 2680:4-7. 161 1 In any event, the implementation of plaintiffs proposed population reduction measures 2 would not significantly exacerbate overcrowding in the various county jails. 3 4 3. Effect on Parole Supervision Resources Plaintiffs proposed population reduction measures would result in an increase in the 5 population of parolees in the community at any given moment. Defendant-intervenors argue 6 that the parole departments would not be able to supervise the increased number of parolees, 7 and that inadequate supervision would lead to an increase in recidivism. They presented 8 evidence that, even at present, parole departments are overburdened and cannot adequately 9 supervise the parolees, leading to parolees failure to integrate into society. See, e.g., Dyer 10 Am. Report ¶¶ 6, 32; Rep. Tr. at 1856:13-21 (Word). 11 The evidence shows, however, that many of the current problems with parole 12 supervision are created by the poor allocation of resources. California s parole system is 13 significantly out of step with that of the other states. California is the only state that puts 14 every inmate leaving the prison system on parole, usually for one to three years. Rep. Tr. at 15 1756:16-22 (Hoffman); Ex. P113 at 75 (Rehabilitation Strike Team Report). The upshot is 16 that California s parole system is so overburdened that parolees who represent a serious 17 public safety risk are not watched closely enough, and those who wish to go straight cannot 18 get the help they need. Ex. P113 at 15. 19 The evidence conclusively showed that public safety would not be adversely affected 20 by releasing low-risk, nonserious, nonviolent offenders from the prison system without 21 placing them on parole supervision. Such individuals can be identified using a risk 22 assessment tool. See Rep. Tr. at 1406:6-1407:10 (Austin). Hoffman, the CDCR s Director 23 of Adult Parole Operations, testified that the science and evidence . . . do[] support a 24 conclusion that there is a percentage of the parole population that shouldn t be supervised or 25 supervised very little; that at the low end of the spectrum supervision is counter productive. 26 Id. at 1758:6-10. Secretary Woodford also opined that reducing the supervision of low-risk 27 offenders would reduce recidivism and crime, see id. at 1323:9-24 (Woodford), and the 28 Rehabilitation Strike Team s report reached the same conclusion, Ex. P113 at 15-17. Most 162 1 of the states in the country do not supervise low-risk offenders at all. Rep. Tr. at 2 1759:23-1760:7 (Hoffman). 3 Parole could also be shortened to one year for those who comply with their terms of 4 release and meet certain other criteria. This earned discharge strategy for parolees would 5 provide incentives for parolees to conform to their parole supervision requirements or to 6 participate in programming. Ex. P2 at 13 (CDCR Expert Panel Report); Ex. P113 at 82-84 7 (Rehabilitation Strike Team Report); Ex. P600 at CDCR015633 (CDCR Division of Adult 8 Parole Operations, White Paper: Earned Discharge ); see also Aug. 15, 2008 Austin Report 9 ¶ 53. At the same time, it would not adversely affect recidivism because there is no proven 10 relationship between time on parole and recidivism. Aug. 15, 2008 Austin Report ¶ 77. It 11 would also allow the CDCR to reallocate resources to moderate- and high-risk offenders 12 who require, and benefit from, improved supervision and evidence based programming. 13 Ex. P600 at CDCR015633. Such strategies have been successful across the nation in 14 lowering recidivism rates. Id. Both the Governor s Rehabilitation Strike Team and the 15 CDCR Expert Panel recommended implementing the earned discharge strategy for parolees 16 as a way to improve the parole system and reduce recidivism. Ex. P2 at 13; Ex. P113 at 17 16-17. 18 Based on this evidence, we find that shortening the length of parole or limiting the use 19 of parole for certain offenders would ease the present burden on the parole system. These 20 reform efforts would also improve the public safety impact of the parole system by 21 concentrating resources on high-risk offenders who need supervision and by offering 22 incentives to all offenders to participate in rehabilitative programming. 23 Both Dr. Austin and the CDCR Expert Panel included parole reform along the lines 24 described above in their packages of measures to reduce the prison population without 25 adversely affecting public safety or the operation of the criminal justice system. We find 26 their recommendations persuasive, and conclude that the implementation of parole reform 27 which is already in progress would allow local parole systems to safely absorb any increase 28 in the number of parolees resulting from the proposed population reduction measures. 163 1 4. Re-entry Programs 2 3 Impact on Community Corrections, Rehabilitative Services, and Defendant-intervenors also argued that the influx of parolees and probationers in 4 communities as a result of plaintiffs proposed population reduction measures would strain 5 the community corrections system, rehabilitative services, and re-entry programs. They 6 presented evidence that there are not enough community correctional resources to supervise 7 or provide services to offenders who are diverted from the prison system to the 8 communities.86 E.g., Rep. Tr. at 2384:3-14 (Pacheco); id. at 1030:3-21 (Powers); Cogbill 9 Trial Aff. ¶¶ 35-36. The caseload for probation officers in Los Angeles County, for example, 10 is upwards of 1000:1, while the recommended caseload is between 30:1 and 50:1. Dalton 11 Am. Trial Decl. ¶ 32; see also Meyer Am. Trial Decl. ¶¶ 18, 20. Many cases are largely 12 unsupervised, so that the officers can focus on cases that require more intense supervision or 13 on emergency situations. E.g., Meyer Am. Trial Decl. ¶¶ 20, 24; Rep. Tr. at 1030:7-21 14 (Powers). 15 Defendant-intervenors also presented evidence that both diverted offenders and 16 offenders coming out of California s prisons and reentering the communities have significant 17 needs in the areas of mental health, substance abuse treatment, other medical services, family 18 services, employment, and housing. See, e.g., Cogbill Trial Decl. ¶ 29; Dalton Am. Trial 19 Decl. ¶¶ 30-31; Johnson Trial Decl. ¶ 2 (Director of San Mateo County Human Services 20 Agency); Oct. 16, 2008 Bennett Supp. Report at 2-4; Ex. DI-218 at 1 (Report of the Re-Entry 21 Policy Council). Evidence shows that counties lack the resources to meet those needs even 22 now. See Rep. Tr. at 2073:15-2074:14 (Conklin) (San Diego County Sheriff s Department 23 Detentions Chief Mental Health Clinician); id. at 2456:7-14 (Pena) (Santa Clara County 24 Director of Mental Health); id. at 2492:13-22 (Garner); id. at 2511:25-2512:5 (Bataille) 25 (defendants expert); Aug. 15, 2008 Graves Report at 5-6; Cogbill Trial Decl. ¶ 7; Pena Trial 26 86 The increase in the population of probationers would not occur as a result of the 27 expansion of earned credits or the diversion of technical parole violators. It would occur only if the state decides to reduce the prison population by diverting low-risk offenders to 28 probation. 164 1 Decl. ¶¶ 11, 15; Aug. 15, 2008 Pena Report at 3-5; Word Trial Decl. ¶ 26; James Trial Decl. 2 ¶ 34.87 The gap between the needs and availability of services contributes to the high level of 3 recidivism among parolees. Cogbill Trial Decl. ¶ 7. 4 Because the community re-entry and rehabilitation services in most counties, if not all, 5 are inadequate to serve the current population, those released into the communities as a result 6 of the proposed population reduction measures would either not receive services in the 7 community promptly or would displace other people who are currently receiving services. 8 See Rep. Tr. at 2495:5-13 (Garner); id. at 2699:23-2700:3 (Ryan). Such a result could be 9 mitigated, however, through a population reduction plan that created only a gradual increase 10 in the number of parolees or probationers in each county. Moreover, the increased needs in 11 each county resulting from the population reduction measures proposed by plaintiffs are 12 likely to fall within normal fluctuations in the number of people served by the counties. See, 13 e.g., Rep. Tr. at 2442:2-8 (Pena) (stating that the Santa Clara mental health system serves a 14 dynamic population of between 17,000 and 19,000 clients each year); Pena Trial Decl. ¶ 18 15 (estimating that the proposed population reduction order would result in an additional 100 to 16 700 individuals in Santa Clara County needing mental health services). 17 Furthermore, overwhelming evidence establishes that diversion would be successful 18 and that the proposed population reduction measures would have no adverse effect and 19 would in fact improve public safety if the state were to divert some portion of the savings 20 generated by the population reduction to community corrections, rehabilitation, and re-entry 21 resources. See, e.g., Rep. Tr. at 1828:2-19 (Smith) (opining that his concerns would be 22 ameliorated if the state redirected funding to the counties); id. at 1573:1-1574:3 (Beard) 23 (testifying that funding community services could compensate for the 0.3% increase in 24 arrests of parolees). The programs are already in place, and better coordination between the 25 87 Although community public mental health programs are not intended to serve 26 parolees, Rep. Tr. at 2550:9-19 (Bataille), parolees still rely on county services at times. See, e.g., id. at 2432:16-22 (Pena) (testifying that in Santa Clara, approximately 60% of parolees 27 receiving state outpatient services also accessed county services); id. at 2550:24-25 (Bataille) (testifying that county systems still triage parolees in need of psychiatric emergency 28 services). 165 1 state and the counties, alongside additional funding, could make these services available to a 2 larger portion of the population. See, e.g., Aug. 15, 2008 Garner Report at 6; Meyer Am. 3 Trial Decl. ¶ 69. In any event, as noted already, the additional demand for community 4 resources created by a population reduction is likely to fall within existing fluctuations in 5 demand, and thus would not result in any significant changes at the county or local level. 6 We have no question that the entire criminal justice system and the state itself, as well 7 as the local communities, would be well-served if the state would help fund some of the 8 county programs that are designed to help parolees, probationers, and other persons 9 convicted of criminal offenses with problems such as drug and alcohol addiction, mental 10 illness, job training, and rehabilitation generally. Such programs would certainly help to 11 reduce the crime rate and make the local communities safer places in which to live. Whether 12 to do so, however, is a question as to how the state wishes to expend its resources that must 13 be answered by the state s elected officials and not by this court. We can only note that 14 should the officeholders of California and their constituents wish to raise the level of safety 15 of the state s communities by increasing the availability of programs that facilitate the 16 orderly re-entry into society by former prisoners, they are free to appropriate the necessary 17 funds to do so in a manner that will not divert such funds from other important societal 18 needs. There is no bar to the people s financing of projects they deem desirable through new 19 tax revenues or the issuance of additional state bonds. 20 21 5. Impact on Integrity of Criminal Justice System David Bennett, a criminal justice consultant and expert witness for Defendant- 22 Intervenor Sonoma County, opined that [t]he closing of the front door to the prisons and 23 resulting jail overcrowding, combined with a reduced capacity to locally sentence lower level 24 offenders (such as misdemeanants) will compromise the criminal justice system s ability to 25 hold offenders accountable. Aug. 15, 2008 Bennett Report ¶ 30 (emphasis in original). He 26 anticipated that this would result in a loss of system integrity because, among other negative 27 effects, offenders would not be held accountable for criminal behavior, district attorneys 28 166 1 might stop prosecuting certain crimes, and judges might modify sentences to accommodate 2 the overcrowding in jails. Id. ¶¶ 35-37. 3 Bennett s opinion was based on the assumption that a population reduction order 4 would involve closing the front door of the prisons. Id. ¶ 30. None of the measures proposed 5 by plaintiffs or considered here would require such an extreme result. Moreover, as 6 illustrated above, the measures would not result in a loss of deterrence or cause an increase in 7 jail overcrowding; they would simply affect where offenders serve their sentences and 8 whether they might be released a few months earlier, with no effect on the state s ability or 9 incentive to arrest, prosecute, or imprison new offenders. We thus find that a prison 10 population reduction could be achieved without the negative impact on the integrity of the 11 criminal justice system predicted by Bennett. 12 13 6. Weight To Be Given Public Safety As demonstrated above, we have given substantial weight to the question of the effect 14 of our order upon public safety and the operation of the criminal justice system. While we 15 conclude that there is no adverse effect, were we in error and were there in fact some adverse 16 effect, it would be small, given the number and types of individuals to be released early or 17 diverted to non-prison settings, and given the number of counties, and the size of the state 18 and its population. Even considering the possibility of a minor adverse effect, we would, in 19 view of the extremely serious injuries that continue to result from the long-standing 20 constitutional violations at issue, be required to grant (with the modification set forth in our 21 order) the relief that plaintiffs seek. 22 D. Feasibility Notwithstanding the Present Fiscal Crisis 23 In concluding that the plaintiffs proposed population reduction measures could safely 24 reduce the population of California s prisons, and that such a reduction would not have a 25 significant adverse effect in California s communities, we do not ignore the state s current 26 economic difficulties. The fiscal crisis does not, however, alter our conclusions. 27 There will be a substantial fiscal savings to the state as a result of the reduction in the 28 size of the prison population. According to Deputy Cabinet Secretary Robert Gore, the 167 1 approximate cost of housing a prisoner is $43,000 per year. Ex. P163 at DEFS036906 2 (Jan. 10, 2008 Mem. from Robert Gore re: Governor s CDCR Rehabilitation Strike Team 3 Final Report). Under the order establishing a population cap, the size of the prison 4 population will be reduced by approximately 46,000. The changes leading to that reduction 5 recommended by plaintiffs, such as an increase in good time credits followed by early 6 release, diverting technical parole violators and modifying parole requirements, and diverting 7 low-risk offenders with short sentences, involve no fiscal cost. Other changes recommended 8 by various state commissions and committees can also be adopted without any state funding. 9 There are other state actions that all agree would help reduce crime significantly on both a 10 short- and a long-term basis if taken along with the prisoner reduction measures. They 11 involve helping fund community re-entry programs, such as drug and alcohol treatment, job 12 training, mental health therapy, and half-way houses. Although California s prison 13 population could be reduced without adopting or strengthening such local programs, the 14 benefit to the state of investing in them would be considerable. Whether or not to make such 15 an investment, however, is, as we observed previously, a matter for state officials, not the 16 court, to decide. In any event, the present fiscal crisis would be alleviated rather than 17 worsened by a prisoner release order. 18 E. 19 The state has suggested that, should we issue a population reduction order, we should Inclusion of Mentally Ill Inmates in Any Population Reduction Order 20 nonetheless exempt seriously mentally ill inmates from release pursuant to our order. 21 However, there is no public safety reason to treat mentally ill inmates differently from other 22 inmates as a categorical matter. 23 Under the current system, mentally ill inmates are regularly released when their prison 24 sentences end. Although these inmates reportedly have higher recidivism rates than non25 mentally ill inmates, evidence shows that mentally ill inmates who are released do not, by 26 virtue of their mental illness, present any higher risk than other released inmates. Much of 27 the high recidivism is attributable to noncompliance with parole conditions related to the 28 disorganization produced by mental illness. Ex. P715 at 5 (July 2007 CDCR Division of 168 1 Adult Parole Operations report entitled Mentally Ill Parolee Population ). Dr. Gilligan, a 2 psychiatrist and an expert on mentally ill offenders, testified that, based on research 3 throughout the United States and also in California specifically, mentally ill parolees are not 4 more likely to commit violent crimes after discharge than are non-mentally ill parolees. 5 Aug. 15, 2008 Gilligan Report ¶¶ 34, 36-39; Rep. Tr. at 1608:12-25 (Gilligan). Rather, the 6 risk factors for violence, such as substance abuse, family dysfunction, and character 7 disorders, are comparable for the mentally ill and non-mentally ill. Aug. 15, 2008 Gilligan 8 Report ¶ 40. Defendants expert Dr. Packer agreed that the research literature does not 9 suggest that mentally ill offenders pose a higher risk of violence than their non-mentally ill 10 counterparts. Oct. 1, 2008 Packer Addendum at 1. Another expert for defendants Gale 11 Bataille, the former director of the of Behavioral Health and Recovery Services for San 12 Mateo County, testified that mental illness has a high rate of co-occurrence with substance 13 abuse, which is a predictor of violence, but agreed that mental illness by itself is not a 14 significant indicator of violence. Bataille Rebuttal Report at 2; Rep. Tr. at 2514:6-20; see 15 also Oct. 1, 2008 Packer Addendum at 2 (stating that mental illness is a risk factor for 16 violence, particularly if the individual also abuses substances and has acute psychotic 17 symptoms, but opining that [t]his does not mean that mentally ill inmates should, by virtue 18 of their mental illness, be considered higher risk than other inmates (emphasis in original)). 19 The testimony from the mental health care experts was unanimous that mentally ill 20 people who are receiving proper mental health treatment pose no greater risk to the 21 community than those who are not mentally ill. Rep. Tr. at 2209:25-2210:23 (Stewart); 22 Oct. 1, 2008 Packer Addendum at 1-2; Rep. Tr. at 1640:4-10 (Gilligan); Bataille Rebuttal 23 Report at 2; see also Ex. DI-219 at 6 (June 2006 UCLA Integrated Substance Abuse Program 24 Neuropsychiatric Institute report entitled Final Report on the Mental Health Services 25 Continuum Program of the California Department of Corrections and Rehabilitation Parole 26 Division ). Therefore, population reduction measures involving the successful diversion of 27 offenders and technical parole violators to community mental health programs instead of 28 prison would not have a negative impact on public safety. The diversion of mentally ill 169 1 technical parole violators might even improve public safety because the current churning of 2 mentally ill parole violators in and out of crowded prison reception centers is especially 3 disruptive to their treatment needs and re-entry success. Aug. 15, 2008 Gilligan Report 4 ¶¶ 32-33; Aug. 15, 2008 Stewart Supp. Report ¶ 136. 5 Numerous witnesses for defendants supported the diversion of mentally ill offenders. 6 Dr. Packer, defendants mental health expert, did not support the mass early release of the 7 mentally ill but recommended diversion of mentally ill offenders to community-based 8 programs as an effective population reduction measure. Rep. Tr. at 1086:15-1087:22. 9 Director Bataille also supported community diversion. See Aug. 15, 2008 Bataille Prelim. 10 Report at 19. Director Hoffman testified that the CDCR has, consistent with public safety, 11 already stopped returning parolees to custody for technical violations resulting from their 12 mental illness when programs are available. Rep. Tr. at 1766:15- 1767:19; Hoffman Trial 13 Aff. ¶ 29; Ex. D1195 (Jan. 12, 2007 Mem. from CDCR Secretary James E. Tilton to the 14 Division of Adult Parole Operations). He also stated that, like all other parolees, mentally ill 15 parole violators can be given intermediate sanctions using the Parole Violation Decision 16 Making Instrument. Hoffman Trial Aff. ¶ 30. 17 The disagreement among the experts centered not on whether diversion would be 18 harmful to public safety, but on whether California s communities had sufficient community 19 mental health programs to support the early release or diversion of mentally ill parolees. 20 Plaintiffs experts testified that the impact of the inclusion of some Coleman class members 21 in the population reduction measures would not be significant. Dr. Stewart calculated that, 22 assuming a reduction in the prison population by 50,000 inmates, there would be about 23 10,000 more Coleman class members in the community over a period of time. Out of that 24 group, about 8,500 people would be at the CCCMS level and would need minimal care in the 25 community. Rep. Tr. at 2211:3-15. About 650 additional people per year would need 26 enhanced outpatient care, but that would not pose a significant burden on the current system, 27 which serves 69,000 people. Id. at 2211:18-2212:7. Finally, only 100 additional people each 28 year would need DMH-level care, which would not be a significant additional burden on a 170 1 system that currently treats 43,000 people annually. Id. at 2212:8-21. Dr. Stewart also 2 testified that the number of people needing care may be lower because the class members 3 mental health conditions would improve once they left prison. id. at 2211:18-2212:21; see 4 also Aug. 27, 2008 Gilligan Rebuttal Report ¶¶ 10-11. 5 Defendants experts contested these numbers and their significance. Dr. Packer stated 6 that it is not necessarily true that mentally ill inmates will do better outside of prisons and 7 opined that it is more common for some mentally ill individuals to function at a higher level 8 while in prison. Oct. 1, 2008 Packer Addendum at 3. Dr. Packer also testified that elements 9 for successful release pre-release planning, coordination with community providers, access 10 to systems of care in the community, and availability of community programs are not 11 currently fully functioning within the CDCR, and that an accelerated release of mentally ill 12 prisoners would exacerbate those problems. Id. at 2. Director Bataille opined that most 13 California communities are not prepared for, or capable of, providing the community mental 14 health and treatment services necessary to support an accelerated release of mentally ill 15 inmates, and that the problem is not only funding but also a lack of trained professional staff. 16 See Aug. 15, 2008 Bataille Prelim. Report at 5-18. Other witnesses testified that counties are 17 unable to serve their mentally ill populations now. See, e.g., Rep. Tr. at 2456:7-17 (Pena); 18 Dalton Trial Decl. ¶ 31; Conklin Trial Decl. ¶ 41; Meyer Am. Trial Decl. ¶¶ 64-65. 19 We credit the testimony that community mental health programs are overburdened in 20 many, if not most, California communities. Still, the Coleman class may safely be included 21 in the state s population reduction measures in any number of ways. For example, as 22 Director Bataille suggested, a diversion or earned credits program could be structured so that 23 only those mentally ill individuals with the greatest level of psychiatric stability and the 24 greatest potential to voluntarily follow up on outpatient care would be eligible, at least 25 until appropriate community programming is in place. Aug. 15, 2008 Bataille Prelim. Report 26 at 4. 27 Moreover, credible evidence demonstrates that treating mentally ill offenders outside 28 prison is more effective and less costly than treating them in prison. See Gilligan Rebuttal 171 1 Report ¶ 11 (stating that mental health treatment in the community is more likely to be 2 successful and effective than similar treatment would be in the social environment of the 3 prison ); Rep. Tr. at 1747:9-16, 1753:24-1755:5 (Hoffman) (affirming that providing mental 4 health care for parolees is cheaper than providing it for inmates); id. at 2450:14-2451:7 5 (Pena) (acknowledging that it costs about $24,000 less per year to provide a therapeutic bed 6 in the community than to incarcerate a mentally ill person). There was also unrebutted 7 testimony that it is easier to recruit and hire qualified mental health professionals in civil 8 hospital and clinic settings than in prisons. Aug. 27, 2008 Gilligan Rebuttal Report ¶ 17. In 9 light of the abysmal qualify of the mental health care presently available to California s 10 inmates, it is unlikely that any mentally ill inmates released by the state will find their mental 11 health treatment seriously compromised by their release from prison. 12 We recognize that expanding community programming would require an increase in 13 professional staff at the community level; however, as with other types of programming, this 14 would require a shift in, rather than an infusion of, resources. The state has already begun to 15 expand parolee services, see Hoffman Trial Aff. ¶ 32, and also has a roadmap for further 16 expansion of programming in the CDCR Expert Panel Report. Defendants expert Director 17 Bataille agreed that a population reduction could be achieved and sustained by following the 18 recommendations contained in the CDCR Expert Panel Report, including its 19 recommendation for expanding the communities capacity to provide programming. 20 Aug. 15, 2008 Bataille Prelim. Report at 19. Collaboration between the mental health and 21 criminal justice systems could also begin to address the resource gap. See Rep. Tr. at 22 2534:7-2535:11 (Bataille). 23 On the basis of this evidence, we conclude that mentally ill inmates could, under 24 appropriate conditions, be included in the proposed population reduction measures without 25 any adverse effect on public safety or the operation of the criminal justice system. 26 F. 27 We acknowledge the concern of some law enforcement officials that incarceration Empirical Evidence on Incarceration and Crime Rates 28 serves the interest of incapacitation over the life of a repeat offender. See, e.g., id. at 172 1 1181:5-13 (Powers). To that extent, there is likely some correlation between incarceration 2 rates and crime rates. Indeed, according to plaintiffs experts, some studies have concluded 3 that every ten percent increase in the incarceration rate results in a two to four percent 4 decrease in the crime rate, id. at 1582:1-3 (Beard); id. at 2032:4-12 (Lehman), and that 5 massive incarceration rates have contributed to a 25% reduction in violent crime across the 6 United States, id. at 1447:18-1450:23 (Austin). 7 This testimony does not, however, persuade us that California s prison population 8 could not be reduced without adversely affecting public safety.88 First, even if we credit 9 these studies, population reduction measures could still have a net positive impact on the 10 crime rate. For example, defendants introduced Exhibit D1331, a report by the Washington 11 State Institute for Public Policy, for the proposition that incarceration rates and crime rates 12 correlate. See Rep. Tr. 2030:14- 2032:12 (questioning of Dr. Lehman by defendants counsel 13 and related colloquy with the court). That same report, however, concluded that the decrease 14 in recidivism resulting from an expansion of evidence-based programming would outweigh 15 any potential adverse impact on crime rates resulting from decreased incarceration rates. See 16 Ex. D1331 at 15. 17 Second, the evidence supported Dr. Austin s testimony that there is still disagreement 18 as to the validity of the research connecting incarceration rates to crime rates, Rep. Tr. at 19 1450:20-23, and that [r]esearch on crime and incarceration does not consistently indicate 20 that the massive use of incarceration has reduced crime rates, Aug. 15, 2008 Austin Report 21 ¶ 20. In fact, with regard to the relationship between incarceration and crime in California, 22 both defendants expert Dr. Marquart and Professor Petersilia concluded that the decline in 23 violent crime in California in the past decade is not likely to be a function of the state s 24 approach to corrections. Rep. Tr. 2001:9-2002:18 (Marquart) (agreeing that it would be a 25 88 We also note that the same studies referred to by plaintiffs experts found that increasing the incarceration rate becomes counterproductive once the incarceration rate 26 reaches a certain inflection point. E.g., Rep. Tr. at 1582:1-13 (Beard); see also id. at 1447:18-1450:23 (Austin). At its present incarceration rate of 470 per 100,000, California is 27 close to the inflection point at which further incarceration would not be productive. Id. at 28 1582:1-13 (Beard). 173 1 mistake to conclude that the decline in the California crime rate is a result of its incarceration 2 policies ); Ex. P5 at 2. As we have already noted, it is likely that [t]he overwhelming and 3 undisputed negative side effects of incarceration and crowding far outweigh the potential, 4 unproven benefits of incarceration in California. Aug. 15, 2008 Austin Report ¶ 23. 5 Moreover, Dr. Austin and Dr. Krisberg testified that the historical data and empirical 6 research regarding early release programs across the country show no significant relationship 7 between crime rates and early releases. Aug. 15, 2008 Austin Report ¶¶ 19, 27-42; Rep. Tr. 8 at 2159:20-2162:7 (Krisberg); see also Ex. DI-204 at 1. Their testimony, like that of 9 Dr. Beard and Secretary Lehman, who both implemented prison population reduction 10 measures in other prison systems, confirms that it is possible to lower the prison population 11 without an adverse impact on crime or public safety. For example, in Washington, the state 12 legislature prohibited sending technical parole violators to prison, instituted graduated 13 sanctions, and expanded good time credits. Rep. Tr. at 2004:24-2005:14, 2006:23-2007:18 14 (Lehman). Secretary Lehman, the former secretary of corrections in Washington, testified 15 that these measures did not have any deleterious effect on crime or public safety. Id. at 16 2008:18-2009:14. 17 Secretary Lehman further testified that, during his tenure as secretary of corrections in 18 Pennsylvania, sentencing reforms that made it more likely for an offender to be diverted into 19 the community did not have any adverse impact on public safety. Id. at 2007:19-2008:24. 20 Dr. Beard, the current secretary of corrections in Pennsylvania, testified that he had spent a 21 lot of time in the last seven years studying what other states have done and looking for ways 22 that we can better manage our population from a public safety perspective, from a population 23 control perspective, and from a cost perspective. Id. at 1552:19-24. He played a role in 24 passing legislation in Pennsylvania that allowed for, among other things, intermediate 25 punishment instead of incarceration, incentive credits for evidence-based programming, and 26 parole reform. Id. at 1549:10-1550:14, 1552:1-18. Rather than having an adverse impact, 27 these reform measures have served to improve public safety. Id. at 1552:19-1553:3. 28 174 1 Dr. Austin who has thirty years of experience in correctional planning and research 2 and has personally worked with correctional systems in eight states to reduce their prisoner 3 populations, Nov. 9, 2007 Austin Report ¶¶ 2, 5 similarly testified that a number of 4 population reduction measures have been adopted in various states without an adverse impact 5 on public safety: diversion of technical violators in Kansas and Washington, Rep. Tr. at 6 1392:21-1393:5, 1399:11-15; good time credits in Illinois,89 Nevada, Maryland, and Indiana, 7 id. at 1398:11-1399:1, 1399:11-15;90 and implementation of large community corrections 8 diversion programs in Ohio and Michigan, where the state basically is paying the counties 9 to hold people at the county level who otherwise would go to prison, id. at 1399:5-15. In 10 Nevada, the legislature expanded the award of good time credits to prisoners, probationers, 11 and parolees in 2007, which reduced the prison population without any known increase in 12 crime, arrests, or court filings as of July 2008. Aug. 15, 2008 Austin Report ¶ 36. In New 13 York, the prison population decreased due in part to the expansion of programs awarding 14 good time credits, and not only did the crime rate not increase, it declined substantially. 15 Id. ¶¶ 27-28. 16 Dr. Krisberg also reviewed empirical research analyzing early release programs over 17 the past twenty years in Canada, California, Washington, Wisconsin, Illinois, Texas, 18 Colorado, Montana, Michigan, and Florida, and found that such programs do not endanger 19 public safety. Sept. 8, 2008 Krisberg Report at 4-5. Dr. Krisberg reported that early release 20 produced lower recidivism rates for released inmates when the release targeted low-risk 21 offenders and made provisions for community-based supportive services. Id. 22 89 Dr. Austin s report regarding Illinois stated his opinion that early release should only be used as a short-term measure for prison overcrowding. Ex. DI-785 at 3614 (James Austin, 23 Using Early Release to Relieve Prison Crowding: A Dilemma for Public Policy, 32 Crime Delinquency 404 (1986)). Nonetheless, the article concluded that there was an overall cost 24 savings to the state as a result of early release, with relatively lower costs to local public criminal justice agencies stemming from arrests of the early releases. Id. at 3700. This is 25 not inconsistent with Dr. Austin s testimony in this case. 90 26 Although Dr. Austin stated that he did not endorse early release as a long-term remedy, Rep. Tr. 2610: 8-2611:1, it was not clear whether his testimony on that point related 27 to the expansion of good time credits or generic release. In any event, he testified that the prison population could be lowered safely through the expansion of good time credits and 28 other measures. Aug. 15, 2008 Austin Report ¶ 43. 175 1 District Attorney Pacheco of Riverside County opined that a generic early release 2 program from California prisons would increase crime, as it had in other jurisdictions like 3 Florida, Illinois, Philadelphia, and Los Angeles. Rep. Tr. at 2380:20-2381:9. His opinion, 4 however, appeared to be based largely on newspaper articles reporting specific crimes that 5 occurred during the early release period, and not on a broader analysis of crime rates. See 6 Pacheco Decl. ¶ 23 & Ex. C. Moreover, Mr. Pacheco discussed only a generic early release 7 and failed to consider whether some of the adverse impacts he fears would be mitigated by 8 basing early release decisions on an improved system of earned credits or by instituting a 9 diversion program or other measures proposed by plaintiffs. See Rep. Tr. at 2379:17-23 10 (discussing only generic early release). 11 Dr. Marquart, defendants sole witness on population reduction measures and public 12 safety, stated that he opposed any prisoner release order in part because the early release 13 measures implemented in Texas in the 1980s to meet a 95% population cap caused an 14 increase in crime.91 Id. at 1956:14-20, 23-24, 1957:12-18. However, he also testified that he 15 did not know how much of the increase in crime was attributable to the early release 16 program, as opposed to other factors. Id. at 1984:16-1985:9. Indeed, the basis for 17 Dr. Marquart s opposition to any reduction in the prison population appeared to be not the 18 Texas experience but, instead, his opinion that he didn t know what the consequences would 19 be, not that it would be a disaster, id. at 1990:22-24. According to Dr. Marquart, reducing 20 the prison population could have a negative impact on public safety, it could have no impact, 21 or it could have a positive impact. Compare id. at 1990:17-24; with id. at 1995:8-20. Such 22 equivocal testimony is not helpful to the court. In any event, Dr. Marquart stated that he was 23 not opposed to the expansion of good time credits, parole reform, or evidence-based 24 programming, and further stated that the prison population could be reduced in a safe manner 25 through proper programming. Id. at 1991:22-1993:18, 1994:17-25. The Texas prison 26 91 Neither the number of inmates who were released early nor the length of time by which their incarceration was shortened is apparent from Dr. Marquart s testimony. 27 Dr. Austin, however, stated that his recommended amount of good time credits is less than the amount awarded to Texas prisoners between 1980 and 1989, and the amount presently 28 awarded in that state. Aug. 27, 2008 Austin Supp. Report ¶ 20(e). 176 1 population, in fact, has recently been reduced safely by diverting technical parole violators 2 and increasing the state s parole grant rate using risk-based guidelines. Aug. 27, 2008 Austin 3 Supp. Report ¶ 20. 4 To the extent that District Attorney Pacheco, Dr. Marquart, or any other witness 5 opined that any population reduction measure applied to California prisons would result in an 6 adverse public safety impact, we reject that opinion. If anything, such testimony shows only 7 that the CDCR should implement population reduction measures mirroring those of the 8 jurisdictions that have successfully and safely reduced their inmate populations. We credit 9 the testimony from experts who, through careful study and experience in a number of 10 jurisdictions, arrived at the opinion that a population reduction, through a combination of 11 earned credits, parole reform, and diversion, could be accomplished in a manner that 12 preserves public safety and the operation of the criminal justice system. Moreover, 13 California s present system of churning inmates into and out of overcrowded and 14 criminogenic prisons itself poses a threat to public safety. Thus, any increase in the crime 15 rate associated with lowered incarceration rates could be substantially offset, and perhaps 16 entirely eliminated, by the public safety benefits of ridding the system of churning and 17 reducing the criminogenic effect of spending time in California prisons. 18 G. 19 We take seriously our duty to consider public safety, and we have done so. We do not Findings and Conclusions 20 construe this PLRA requirement, however, to preclude a population reduction order based on 21 a possibility that the order might have an adverse impact on public safety or the operation of 22 the criminal justice system, no matter how small. If that were enough to prevent the court 23 from ordering a population cap, no court would ever be able to impose such a remedy, thus 24 contravening the congressional intent that a population cap be ordered if it is truly necessary 25 to prevent an actual violation of a prisoner s federal rights. H.R. Rep. No. 104-21, at 25. 26 Based on our detailed findings examining the evidence from correctional and public 27 safety experts around the state and across the country, we are confident that a prison 28 population reduction to 137.5% design capacity can be achieved in California without a 177 1 meaningful adverse impact on public safety or the operation of the criminal justice system.92 2 The evidence and testimony from plaintiffs, defendants, and defendant-intervenors 3 overwhelmingly showed that there are ways for California to reduce its prison population 4 without such an adverse impact, and that a less crowded prison system would in fact benefit 5 public safety and the proper operation of the criminal justice system. 6 The population reduction measures that we specifically considered include the 7 expansion of earned credits, the diversion of technical parole violators, the diversion of low8 risk offenders to community corrections, and the expansion of evidence-based programming. 9 These measures were recommended not only by plaintiffs experts but also by experts for 10 defendants and defendant-intervenors, the Governor, CDCR officials, and the CDCR Expert 11 Panel. Because these measures either have no impact on or reduce the recidivism rate, they 12 would not adversely affect public safety. Furthermore, unlike measures such as 13 indiscriminately and suddenly releasing inmates or closing prison doors to further admission, 14 the measures we considered would not have a significant adverse impact on the operation of 15 the criminal justice system. Any adverse impact on community resources resulting from 16 these measures could readily be mitigated by parole reform and the reallocation of funding 17 and resources. It follows from the many reports we have discussed that other methods of 18 reducing the prison population such as sentencing reform and the release of members of 19 groups that are least likely to recidivate, such as the aged and the infirm as well as low-risk 20 prisoners nearing the end of their sentences, do not pose any threat to public safety. 21 Other jurisdictions have successfully reduced their prison populations through 22 measures similar to those proposed by plaintiffs and the other reforms discussed herein, and 23 we find that California could also do so. In fact, California could do so perhaps more easily 24 25 26 27 28 92 Duran, 760 F.2d 756, a pre-PLRA case, does not suggest a different outcome. In Duran, the court of appeals vacated the district court s order directing the release of pretrial detainees after finding that the order would adversely affect the public interest. Duran involved the release of detainees without the use of any risk-based instrument, and the uncontested evidence before that court showed that many of the released inmates would become fugitives or commit felonies while awaiting trial. See id. at 757-58. By contrast, the evidence before this court establishes that California could reduce its prison population without any adverse effect on public safety or the operation of the criminal justice system. Accordingly, the balance of interests in this case differs substantially from that in Duran. 178 1 than other jurisdictions because of its current, unproductive incarceration policies, such as 2 returning most technical parole violators to prison and denying judges the ability to tailor 3 sentences to the risks and needs of particular offenders. 4 One of the most persuasive pieces of evidence before us is the report of the Expert 5 Panel on Adult Offender Recidivism Reduction Programming, which was convened by the 6 CDCR in 2007 to suggest strategies for reducing California s high recidivism rate. Ex. P2 at 7 vii. The panel consisted of CDCR s Chief Deputy Secretary for Adult Programs, academic 8 researchers, consultants, and former and current secretaries of corrections in Pennsylvania, 9 Arizona, Washington, Ohio, and Maine. Id. at ii. The report recommended a comprehensive 10 set of measures that would reduce California s prison population while also reducing 11 recidivism. 12 The CDCR Expert Panel concluded that, if the CDCR were to follow its 13 recommendations to divert technical parole violators, implement parole reform, and expand 14 good time credits, these changes alone would serve to reduce the prison population by 15 between 38,500 and 43,500 inmates, and the parole population would be reduced by 6,500 to 16 11,500. Id. at 95. The panel expected an additional reduction in the prison population of 17 about 2,194 to 4,388 from evidence-based programming initiatives. Id. at 97.93 After 18 accounting for the costs of the additional programming recommended by the panel, full 19 implementation of its recommendations would still save the state between $561 and $684 20 million a year.94 Id. at 99. The proposed reduction resulting from the above measures alone 21 would fall within the range necessary to comply with a 137.5% population cap. Other means 22 suggested by the state and others, including the expert committees and the numerous other 23 official committees, could reduce the prison population even further. 24 93 As of August 27, 2008, the CDCR was housing 156,352 inmates in prison institutions designed to hold 79,828 inmates. Ex. P135 (CDCR weekly population report as 25 of August 27, 2008). 94 26 James Tilton, then the CDCR Secretary, endorsed the CDCR Expert Panel s recommendations, but with a reservation as to the estimated impact on the prison population. 27 Rep. Tr. at 2614:20-2615:2 (Austin); Ex. P49 (Sept. 25, 2007 Letter from Secretary James E. Tilton, California Department of Corrections and Rehabilitation, to the Hon. Denise 28 Ducheny). 179 1 Secretary Lehman, who was a member of the CDCR Expert Panel, testified that use of 2 the measures proposed in the Panel report could reduce California s prison population 3 without causing any adverse impact. Rep. Tr. at 2012:20-25. Secretary Woodford and 4 Dr. Austin testified that it is possible to reach 130% design capacity without adversely 5 impacting public safety. Id. at 1321:19-1322:5 (Woodford); id. at 1384:3-12 (Austin). 6 Dr. Austin called this a moderate reduction in the state s prison population, because 7 California has got this big bulge of unnecessary and unproductive incarceration, which is 8 an easier target for reduction. Id. at 1434:9-1435:4. Although Dr. Austin recommended 9 that, to achieve a reduction of 50,000 prisoners, California should change its sentencing laws 10 so that second strikers serve 65% to 70% of their sentences rather than 80% as required 11 currently, id. at 1436:18-20, 2568:2-3, he also stated that there are other ways to achieve that 12 reduction, id. at 2570:14-25, a reduction somewhat larger than that which we order. 13 Next, some law enforcement officials testified that the prison population could be 14 reduced safely by about 30% approximately the same size reduction we order here simply 15 by offering incentives for the communities to expand their local correctional systems. Id. at 16 2771:4-10 (Meyer); see also id. at 1042:4-14 (Powers). Their opinion was based on the 17 state s experience in the 1960s, when the state paid counties to reduce the number of people 18 being sent to prison, and the counties were able to achieve a 30% general reduction in the 19 state prison population through the expansion of community-level programming and 20 probation resources. See id. at 1042:4-14 (Powers). 21 We should note finally that, regardless of the conclusion of the overwhelming 22 majority of the experts that adoption of the population control measures described above 23 would not adversely affect public safety, they all strongly recommend that the state, in 24 addition to strengthening its own rehabilitative programs, should help establish or improve 25 local community programs designed to assist probationers, parolees, and released prisoners 26 (whether released as the result of the expiration of their terms or otherwise) to re-enter 27 society. Such programs, as noted earlier, should include drug and alcohol rehabilitation, 28 mental health treatment, and job training. 180 1 There is no doubt that the adoption of these programs would help increase public 2 safety above its current level, including after issuance of our population reduction order. 3 Clearly, a failure by the state to comply with the experts recommendations to take these 4 steps would be regrettable and would be contrary to the interests of public safety. Still, 5 unlike the population cap we order here, which our analysis shows is required by the United 6 States Constitution, the decision whether to adopt these rehabilitative measures is left to the 7 Governor and the Legislature. Whether a failure to adopt them would be acceptable, in view 8 of the effect on public safety, is a question that ultimately the people of California will be 9 required to answer. 10 In sum, the four recommendations in the CDCR Expert Panel report adopted as 11 proposals by plaintiffs provide a means for the state to safely reduce the prison population to 12 137.5% design capacity. The population could be reduced even further with the reform of 13 California s antiquated sentencing policies and other related changes to the laws. We are 14 therefore satisfied that the state has available methods by which it could readily reduce the 15 prison population to 137.5% design capacity or less without an adverse impact on public 16 safety or the operation of the criminal justice system. Accordingly, even after giving 17 substantial weight to any [potential] adverse impact on public safety or the operation of a 18 criminal justice system caused by our population reduction order, 18 U.S.C. 19 § 3626(a)(1)(A), we conclude that our order meets the requirements of the PLRA. 20 21 VIII. CONCLUSION 22 The massive 750% increase in the California prison population since the mid-1970s is 23 the result of political decisions made over three decades, including the shift to inflexible 24 determinate sentencing and the passage of harsh mandatory minimum and three-strikes laws, 25 as well as the state s counterproductive parole system. Unfortunately, as California s prison 26 population has grown, California s political decision-makers have failed to provide the 27 resources and facilities required to meet the additional need for space and for other 28 necessities of prison existence. Likewise, although state-appointed experts have repeatedly 181 1 provided numerous methods by which the state could safely reduce its prison population, 2 their recommendations have been ignored, underfunded, or postponed indefinitely. The 3 convergence of tough-on-crime policies and an unwillingness to expend the necessary funds 4 to support the population growth has brought California s prisons to the breaking point. The 5 state of emergency declared by Governor Schwarzenegger almost three years ago continues 6 to this day, California s prisons remain severely overcrowded, and inmates in the California 7 prison system continue to languish without constitutionally adequate medical and mental 8 health care. 9 Federal courts do not intervene in state affairs lightly. Principles of federalism, 10 comity, and separation of powers require federal courts to refrain from addressing matters of 11 state government in all but the most pressing of circumstances. Even then, federal courts 12 must proceed cautiously, giving the states every opportunity to meet their federal 13 constitutional and statutory obligations voluntarily. Unfortunately, during the 8 years of the 14 Plata litigation and the 19 years of the Coleman litigation, the political branches of 15 California government charged with addressing the crisis in the state s prisons have failed to 16 do so. Instead, the rights of California s prisoners have repeatedly been ignored. Where the 17 political process has utterly failed to protect the constitutional rights of a minority, the courts 18 can, and must, vindicate those rights. See John Hart Ely, Democracy and Distrust 103, 173 19 (1980). We do so here, recognizing the seriousness of our action and with the hope that 20 California s leadership will act constructively and cooperatively, and follow the mandate of 21 this court and the PLRA, so as to ultimately eliminate the need for further federal 22 intervention. 23 24 25 26 27 28 182 1 2 ORDER Within 45 days, defendants shall provide the court with a population reduction plan 3 that will in no more than two years reduce the population of the CDCR s adult institutions to 4 137.5% of their combined design capacity. Should any of defendants proposed population 5 reduction measures require the waiver of any provisions of state law, the state shall so advise 6 the court, and shall explain why the requested waiver is permissible under 18 U.S.C. 7 § 3626(a)(1)(B). In preparing their plan, defendants shall consult with plaintiffs, intervenors, 8 and other relevant stakeholders, including the Coleman Special Master and the Plata 9 Receiver. Should such consultation fail to resolve any objections to the proposed population 10 reduction plan, plaintiffs and intervenors shall file their objections no more than 20 days after 11 defendants file their proposed plan, and defendants shall file responses to such objections no 12 more than 10 days thereafter. Defendants shall set forth in their proposal the effective dates 13 of the various actions they propose to undertake and their estimate of the reduction in 14 population they expect to achieve after six, twelve, eighteen, and twenty-four months. The 15 court will consider all of the written submissions and make any necessary modifications or 16 changes to defendants proposed plan before issuing a population reduction plan as an order 17 of the court. The court may before doing so request clarification on any matters and conduct 18 any further hearings it deems necessary. However, given that this court issued a preliminary 19 ruling on this matter almost six months ago so as to give the parties notice of the likely 20 nature of [this] opinion, and [] allow them to plan accordingly, Feb. 9, 2009 Tentative 21 Ruling at 1, the court will look with disfavor upon any effort to postpone or delay an 22 expeditious resolution of the terms of the population reduction plan, including the submission 23 of a proposed plan by the state and the issuance of the order adopting the final plan. The 24 court will not grant any stay of the proceedings prior to the issuance of the final population 25 reduction plan, but will entertain motions to stay implementation of that plan pending the 26 resolution of any appeal to the Supreme Court. We will retain jurisdiction over this matter to 27 ensure compliance with the population reduction plan and to consider any subsequent 28 modifications made necessary by changed circumstances. 183

Some case metadata and case summaries were written with the help of AI, which can produce inaccuracies. You should read the full case before relying on it for legal research purposes.

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.