Stanford Criminal Justice Center Report on Parole for Lifers in Ca 2011
Download original document:
This text is machine-read, and may contain errors. Check the original document to verify accuracy.
Stanford Criminal Justice Center L IF E I N L IMBO : An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California Robert Weisberg, Debbie A. Mukamal and Jordan D. Segall September 2011 © 2011 Stanford Criminal Justice Center. Permission is granted for reproduction of this document, with attribution to the Stanford Criminal Justice Center. Stanford Criminal Justice Center Stanford Law School 559 Nathan Abbott Way Stanford, CA 94305 www.law.stanford.edu/program/centers/scjc L I F E I N LI M BO : An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California CONTENTS Acknowledgements ............................................................................... 2 Introduction ......................................................................................... 3 Who Are California’s “Lifers?” ................................................................. 6 Parole Process in a Nutshell ................................................................... 7 Determining Suitability for Parole Release ................................................ 9 Recent Disposition Rates ..................................................................... 11 Demographic Details of the Lifer Population ........................................... 15 Risk of Release ................................................................................... 17 The SCJC Lifer Transcript Analysis......................................................... 18 General Findings ................................................................................. 18 Specific Findings: Second-Pass Analysis ................................................ 20 Inmate Characteristics ......................................................................... 20 Life Crime................................................................................ 20 Prior Record............................................................................. 21 Age ......................................................................................... 21 Other Factors ........................................................................... 22 Other Factors Associated with Release ........................................ 22 Facility ............................................................................. 22 Behavior in Prison .............................................................. 23 Psychological Evaluations ................................................... 23 Drug Abuse ....................................................................... 24 Conclusion ......................................................................................... 24 Further Empirical Research on the Parole Release Process for Lifers .......... 25 Endnotes............................................................................................ 26 L I F E I N LI M BO : An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California ACKNOWLEDGEMENTS The Stanford Criminal Justice Center (SCJC) serves as Stanford Law School’s vehicle for promoting and coordinating the study of criminal law and the criminal justice system, including legal and interdisciplinary research, policy analysis, curriculum development, and preparation of law students for careers in criminal law. The center is headed by faculty co-directors Robert Weisberg and Joan Petersilia and executive director Debbie Mukamal. The authors gratefully acknowledge the Office of the Governor, the California Department of Corrections and Rehabilitation and the Board of Parole Hearings for providing data used and/or cited in this bulletin. We are also grateful to the many individuals who provided context for our understanding of the subject of this report. The authors would like to acknowledge Alexandra Lampert, SLS 2011, for the research and leadership she provided for the transcript analysis that underlies this report. Her analytical skills and logistical management were invaluable. Albert Gilbert, Diem Quynh (Cynthia) Ngoc Huynh, Nathan Pearl, Jeffrey Tai, and Jimmy Threatt, students at Stanford University, served as research assistants during the production of this report. We are grateful for their careful coding. L I F E I N LI M BO : An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California 2 INTRODUCTION In recent years, California’s prison system has been under federal judicial control because of severe overcrowding, which partly results from the recycling of revoked inmates under parole supervision. The federal litigation has cast a sharp focus on the mandatory parole system created by the 1976 Determinate Sentencing Law and viewed as the legal mechanism by which this recycling has developed. But far too little attention has been given to the prison population serving life sentences with the possibility of parole under older indeterminate sentencing principles, a population that as of 2010 represents a fifth of California state prisoners. More than 32,000 inmates comprise the “lifer” category, i.e., inmates who are eligible to be considered for release from prison after screening by the parole board to determine when and under what condition.1 (This group of prisoners is distinct from the much smaller population of 4,000 individuals serving life sentences without the possibility of parole (LWOP)). The goal of this project is to examine in empirical detail (a) the lifer population, covering key details of its demographics, and (b) the processes by which lifers are considered for release, including an examination of historical trends in grant and denial rates, the recidivism record of released inmates, and legal and policy analysis of the specific mechanisms of the parolee hearing process. Despite the importance of the lifer population in terms of its size and the major legal and policy changes that have occurred to the parole process for lifers in the last several years, little research has yet been devoted to this topic. We foresee the result to be a body of research that will generate both better public understanding and further academic examination of the lifer population and processes. In addition, we hope our study generates suggestions for legal and policy reform, including better ways of assessing the recidivism risks of lifers, the fairness of the hearing process, and possible budgetary savings from changes in the state’s legal rules governing lifers. This is the first in a series of reports the Stanford Criminal Justice Center (SCJC) will be issuing on this topic. It describes the scope of the population of prisoners serving life sentences with the possibility of parole, as well as the process by which they are considered for release. It also includes initial analysis from our research examining Board of Parole Hearings transcripts the factors that might correlate with grant and denial decisions. Finally, this report identifies important research questions we are now pursuing. L I F E I N LI M BO : An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California 3 Some highlights from our findings include: % The size of the lifer population has increased as a percentage of the overall California prison population from eight percent in 1990 to 20 percent in 2010. Most individuals serving life sentences with the possibility of parole are serving time for first- or second-degree murder. % In line with the increase in the size of the lifer population, the Board of Parole Hearings has steadily increased the number of lifer suitability hearings it has conducted in the last 30 years, representing a 745 percent increase from 1980 to 2010. The majority of the increase has occurred in the last decade. % More than twice as many hearings were scheduled than conducted in 2010, reflecting a trend that has appeared and grown since 2000. While efforts by the Board to address the backlog of hearings has increased the flow of hearings, the passage of Marsy’s Law and new regulations promulgated in 2008 have likely increased the number of hearings. % A lifer now stands an 18 percent chance of being granted parole by the Board of Parole Hearings. The grant rate has fluctuated over the last 30 years— nearing zero percent at times and never arising above 20 percent. The change in the rate could be attributed to changes in characteristics of the inmates appearing in a particular year, changes in the composition of the board, and court clarification of standards the Board should use in determining suitability or other factors. % The Governor’s rate in reversing decisions made by the Board has fluctuated over the last two decades, reflecting the individual policy orientation of the particular Governor in office. % As with the size of the lifer population and the number of hearings conducted by the Board, the number of parole decisions made by the Governor involving murder cases has increased by 1754 percent in the last 20 years, with the bulk of the increase occurring after 2000 (when the total number of suitability hearings conducted by the Board increased). % The likelihood of a lifer convicted of murder being granted parole by the Board and not having the decision reversed by the Governor is—and always has been—slim. In 2010, the probability was approximately six percent. % A major—perhaps the major—question in public debate about the current lifer population is their risk of recidivating. While data is limited, interim information suggests that the incidence of commission of serious crimes by recently released lifers has been minuscule, and as compared to the larger inmate population, recidivism risk—at least among those deemed suitable for release by both the Board and the Governor—is minimal. % In addition, while an inmate’s chance of being granted parole has increased in the last two years, the length of time he or she must wait for a subsequent hearing when denied parole has also increased (though there is a legal mechanism by which an inmate can petition the Board to advance his/her hearing by a showing of, among other things, changed circumstances). L I F E I N LI M BO : An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California 4 In particular, initial results from our research analyzing nearly 450 Board of Parole Hearings lifer suitability hearing transcripts from the time period 2007 through 2010 reveal the following significant findings: % Grant rates vary significantly year to year: the grant rate in 2010 was nearly triple what it was in 2007 and 2008. % Other factors like immigration status, whether an inmate has children, and marital status are not significantly associated with a release or denial. % Though commissioners become more lenient in one dimension—by increasing the grant rate in 2009 and 2010—they become more stringent on another dimension in those years, by setting lengthier periods of time until the subsequent parole hearing when denying parole. % More research is needed to determine grant rate variance across prison facility, and the reasons associated with it, including the security levels of and program availability at each facility. % When victims attend hearings, the grant rate is less than half the rate when victims do not attend. % There is no statistically significant difference in the grant rates of various types of offenses. One factor strongly associated with release is whether the life crime involved sexual violence. Other factors that do not relate in any statistically significant way include the use of a firearm in the life crime or the number of people the inmate victimized in the commission of the life crime. % Prior record does not appear to significantly affect release decisions, whether they are adult or juvenile records. % Most inmates committed their life crime between the ages of 20 and 25. Inmates who committed their life crimes between 20 and 30 were somewhat more likely to be paroled than inmates whose life crimes were committed in their forties. The average age of inmates at the time of the parole hearing is 50.8. The average age of inmates granted parole is 49.9 years, and the average age of inmates denied parole is 51. Surprisingly, age does not appear to be a significant factor in release decisions. L I F E I N LI M BO : % In-prison behavior can affect whether an inmate is granted or denied parole. CDC 115 infractions are strongly associated with the grant rate, though CDC 128 infractions are not significantly associated with the grant rate. Also, the seriousness of the disciplinary violation is dispositive: violent disciplinary infractions— regardless of when they occur—are significantly associated with parole denials. % Scores of psychological examinations administered to predict recidivism risk and inmate psychological stability are significantly correlated with the grant rate. Inmates who receive an average score or higher on these exams virtually never receive parole release. % History of drug or alcohol abuse is not correlated with the grant rate. However, whether an inmate is participating in a 12-step program and whether he or she can correctly answer questions about those steps does affect whether an inmate is granted or denied parole. An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California 5 WHO ARE CALIFORNIA’S “LIFERS?” As of 2010, 20 percent of the California prison population is serving a term-to-life prison sentence, more than twice the percentage 20 years ago, and the highest such percentage of any system in the country.2 Of the roughly 32,000 inmates serving life with the possibility of parole sentences, about 75 percent are serving so-called “term-to-life” sentences and 25 percent are serving threestrikes sentences. Chart 1 contextualizes the growth of these populations within the larger prison population. This bulletin concentrates on those inmates serving “term-to-life” or life sentences with the possibility of parole sentences (generally referred to as “lifers” by the California Department of Corrections and Rehabilitation (CDCR)). Note, however, that because the three-strikes law is less than two decades old, the percentage of the overall lifer population contributed by three-strikes will surely grow, regardless of any changes in the term-to-life population. It is presently unknown whether and how current policies and laws governing parole release for the term-to-life population will also presumably apply to the three-strike population, the first of whom will come before the Board of Parole Hearings for parole release in 2019.3 Although numerous crimes can lead to life sentences under the California Penal Code, the great majority of current lifers were convicted of first- or second-degree murder4 or attempted murder; the two other crimes with substantial numbers of lifers are rape and kidnapping. More details on the proportion of lifers representing the various crime categories, as well as the length of time typically served by category, appears in the “Detailed Demographics” section beginning on page 15. CHART 1 Sentencing Categories Comprising the CA Prison Population, 1990 – 2010 200,000 180,000 Number of CA Prisoners 160,000 140,000 Determinate Sentence 2nd Strike Offense 3rd Strike Offense Life with Parole Life without Parole Death Row 120,000 100,000 80,000 60,000 40,000 20,000 0 Year L I F E I N LI M BO : * Note: 2010 data as of 11/1/2010 An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California 6 PAROLE PROCESS IN A NUTSHELL The California Penal Code and Board of Parole Hearings regulations lay out the detailed rules that govern the parole decision-making process for individuals serving term-to-life sentences. The Board of Parole Hearings (“Board” or “BPH”, formerly called the “Board of Prison Terms”) is responsible for conducting suitability hearings to determine parole consideration for lifers. Its power vests from California Penal Code § 3040, et seq.: “The Board of Prison Terms shall have the power to allow prisoners imprisoned in the state prisons pursuant to subdivision (b) of Section 1168 to go upon parole outside the prison walls and enclosures.” As early as 1914, the court held that whether an inmate should be released on parole should “be left to the judgment and discretion of the [B]oard to be exercised as it might be satisfied that justice in the case of any particular prisoner required.”5 The Board is comprised of 12 full-time members, appointed by the Governor and confirmed by the Senate.6 Terms of service are three years, although Commissioners are eligible for reappointment. Membership is supposed to “reflect as nearly as possible a cross section of the racial, sexual, economic, and geographic features of the population of the state.”7 Some 70 Deputy Commissioners—civil servants— also participate in and make decisions at hearings to determine suitability for parole release, though they are not permitted to rule on objections at hearings.8 Commissioners and Deputy Commissioners participating in parole suitability hearings are required to receive 40 hours of annual training, including training in domestic violence and intimate partner battering.9 They are required to have a “broad background in criminal justice” and “…a varied interest in adult correction work, public safety, and shall have experience or education in the fields of corrections, sociology, law, law enforcement, medicine, mental health, or education.”10 The Board meets with and schedules initial parole suitability hearings with individuals serving life terms one year before their minimum parole eligibility dates (MEPD). Typically, one commissioner and one deputy commissioner preside over a hearing. Hearings are held L I F E I N LI M BO : MARSY’S LAW: AN EXPANSION OF VICTIMS’ RIGHTS In November 2008, California’s voters passed Proposition 9—also known as “Marsy’s Law”—a ballot initiative promoted as a “Victims’ Bill of Rights.” It was named for Marsy Nicholas, a 21-year-old college student who was murdered by her boyfriend in 1983 and whose perpetrator was released on bail without her family’s knowledge. The law amended the California Constitution by expanding victims’ rights in a number of important ways, including providing notice and granting participation in all proceedings. Specifically within the parole process for lifers, Marsy’s Law grants the victim, next of kin, members of the victim’s family, and two representatives designated by the victim the right to attend and make statements at suitability hearings which reasonably express their views concerning the prisoner, the effect of the crimes on the victim and the victim’s family, and the prisoner’s suitability for parole. It requires the Board to consider the entire and uninterrupted statements of victims, including victims of non-life crimes. It also forbids the prisoner or his/her attorney from asking the victim questions during the hearing. See: California Constitution Article I, Section 28 and California Penal Code §§ 3041.5 and 3043. As discussed within the text, another very important change made by Marsy’s Law was to lengthen the number of years by which individuals serving life sentences are granted subsequent hearings when denied parole by the Board. in person and at the institution in which the prisoner is currently housed. Before the hearing, the Board receives a case file consisting of the inmate’s central file, forensic evaluations (including the results of risk assessment instruments), behavior in prison, vocational and education certificates, letters of support and opposition, and statements from victims. An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California 7 The inmate is entitled to attend the hearing in person, ask questions, receive all non-confidential hearing documents at least 10 days in advance of the hearing, have his/her case individually considered, receive an explanation of the reasons for parole denial, and receive a transcript of the hearing proceedings.11 The inmate is also entitled to be represented by counsel at a suitability hearing.12 California pays appointed attorneys $50 per hour and a maximum of eight hours or $400 to represent inmates at parole hearings.13 Privately retained attorneys charge between $2000 and $5000 for parole board hearing representation.14 Some attorneys maintain that the amount of time necessary to review the inmate’s file, meet and prepare with the inmate, and provide representation far exceeds eight hours. Marsy’s Law in 2008, the victim, next of kin, members of the victim’s family, and two representatives have the right to receive notice 90 days prior to the hearing and to present uninterrupted testimony at the hearing either in person, by written statement, audio or video statement, or by video-conference appearance.17 The victim or his or her representative may speak about any of the crimes of which the inmate has been convicted, the effect of the crime, and the suitability of the inmate for parole. These individuals are also entitled to request and receive a stenographic record of all proceedings.18 The District Attorney from the county from which the inmate was committed has the right to participate in the hearing and be notified by the Board at least 30 days before the hearing date.15 The District Attorney is limited to asking clarifying questions of the inmate via the Board. As in nearly every jurisdiction in the United States, victims have the right to receive notice and participate in the parole hearing process in California.16 As expanded by In addition to the Board members, inmate, inmate’s attorney, the District Attorney, and victim(s), members of the press are permitted and sometimes attend hearings. In addition, at least 30 days before the hearing, the Board must send written notice to the judge of the court where the inmate was convicted; the attorney who represented the defendant at trial, the law enforcement agency that investigated the case, and, where the person was convicted of the murder of a peace officer, the agency which had employed that peace officer at the time of the murder.19 Any of these parties may submit written or recorded information to the Board.20 Appeal to Federal Court Governor Declines Review/ Inmate Paroled Governor Review (access severely limited by Swarthout v. Cooke) Governor Review (for non-murder cases) (for murder cases) Grant (requires review and approval by central BPH) GOVERNOR REVERSAL Appeal to State Court Full Board Review (en banc) Denial SPLIT DECISION Governor refers for full review & BPH schedules rescission hearing L I F E I N LI M BO : Suitability Hearing Conducted by Board of Parole Hearings BPH CAN RESCIND PAROLE DECISION An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California 8 DETERMINING SUITABILITY FOR PAROLE RELEASE Individuals serving life sentences with the possibility of parole—unlike those serving death or LWOP sentences—are presumed to receive a parole date unless the Board determines that the prisoner poses an “unreasonable risk of danger to society.”21 Regulations guide the Board in making these assessments. In particular, circumstances that weigh in favor of release include: (1) no juvenile record; (2) stable social history; (3) signs of remorse: (4) motivation for crime: (5) Battered Woman Syndrome; (6) lack of a significant violent criminal history; (7) age; (8) understanding and plans for the future; and (9) institutional activities that indicate an ability to function within the law upon release.22 Factors that weigh against release suitability for release include: (1) the commitment offense;23 (2) previous record of violence; (3) unstable social history; (4) sexual offense background; (5) severe mental problems; and 6) serious misconduct in prison.24 California law also lays out detailed due process rights for prisoners in regard to these hearings.25 In a series of key decisions (see “California Courts Clarify Standards for Determining Release” on this page), the California Supreme Court has shed light on the weight of the factors identified in the law and regulations. Notably, “although the Board exercises broad discretion in determining whether to rescind parole, such decisions are subject to a form of limited judicial review to ensure that they are supported by at least ‘some evidence.”26 By extension, the “some evidence” standard applies to Board decisions granting or denying parole. The nature of the prisoner’s offense, alone, can constitute a sufficient basis for denying parole. Although the parole authority is prohibited from adopting a blanket rule that automatically excludes parole for individuals who have been convicted of a particular type of offense, the authority properly may weight heavily the degree of violence use and the amount of viciousness shown by a defendant.27 For some time, the Board had relied heavily and primarily on the commitment offense itself in making L I F E I N LI M BO : CALIFORNIA COURTS CLARIFY STANDARDS FOR DETERMINING RELEASE While statute and regulation present the factors the Board—and by extension, the Governor—should consider in deciding whether to release individuals serving life sentences, case law over several decades has clarified the standards and the weight of the various criteria to be used by the Board and Governor in making their decisions. The Court most recently clarified that the relevant inquiry is whether there is “some evidence” showing that the prisoner is a current threat to public safety, and while the commitment offense is probative, in and of itself cannot serve as the sole reason to deny parole. Roberts v. Duffy (140 P.260 (Cal. 1914): Whether an inmate should be released on parole should “be left to the judgment and discretion of the [B]oard to be exercised as it might be satisifed that justice in the case of any particular prisoner required.” In re Minnis, 498 P.2d 997 (Cal. 1972): “Although a prisoner is not entitled to have his term fixed at less than maximum or to receive parole, he is entitlted to have his application for these benefits ‘duly considered;” based upon an individualized consideration of all relevant factors. In re Powell, 755 P.2d 881 (Cal. 1988): “[D]ue process requires only that there be some evidence to support a rescission of parole by the BPT.” In re Rosenkrantz, 59 P.3d 174 (Cal. 2002): ‘[U]nder California law the factual basis for a Board decision granting or denying parole is subject to a limited judicial review under the ‘some evidence’ standard of review.” Also: “The nature of the prisoner’s offense, alone, can constitute a sufficient basis for denying parole. Although the parole authority is prohibited from adopting a blanket rule that automatically excludes parole for individuals who have been convicted of a particular type of offense, the authority properly may weigh heavily the degree of violence used and the amount of viciousness shown by a defendant.” (continued next page) An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California 9 its decision, labeling nearly all offenses “heinous, atrocious, and cruel” and using that as the basis for denying inmates parole. But the Court has now clarified that the Board must grant parole unless it concludes that the inmate is still dangerous, and the Board cannot use the circumstances of the crime, standing alone, as a basis to deny parole.28 As a result, the trend has moved from reliance on the commitment offense to indicia that the inmate “lacks insight” (as shown by minimizing culpability or inconsistent statements of the crime itself) when determining unsuitability. In sum, the appropriate and governing standard of review of parole decisions for lifers is whether there exists “some evidence” that the inmate poses a current threat to public safety. In 1988, Proposition 89 amended the California Constitution and gave the Governor authority to review the parole board’s decisions in cases involving nonmurder cases and reverse the parole board’s decisions in cases involving murder convictions.29 For decisions involving non-murder cases, the Governor is limited to remanding the case back to the Board for full review if s/he disagrees with the decision made by the Board. California is one of only four states with gubernatorial review of parole board decision-making, though California is unique in limiting reversal power to decisions involving murder convictions.30 The Governor must apply the same legal standards as did the BPH itself when reviewing decisions. According to the California Supreme Court, the Governor’s decision should “reflect an individualized consideration of the specified criteria” that also must be considered by the Board in making parole decisions.31 Any judicial review of the Governor’s decision, in turn, “strictly is limited to whether some evidence supports the Governor’s assessment of the circumstances of petitioner’s crime—not whether the weight of the evidence conflicts with that assessment.”32 In re Dannenberg, 104 P.3d 783 (Cal. 2005): “[T]he Board, exercising its traditional broad discretion, may protect public safety in each discrete case by considering the dangerous implication of a life-maximum prisoner’s crime individually.” Also: [I]n order to prevent the parole authority’s casey-b-case suitability determinations from swallowing the rule that parole should be ‘normally” be granted, an offense must be ‘particularly egregious’ to justify the denial of parole.” In re Lawrence, 190 P.3d 535 (Cal. 2008): “[T]he relevant inquiry is whether the circumstances of the commitment offense, when considered in light of other facts in the record, are such that they continue to be predictive of current dangerousness many years after commission of the offense.” Also: “In some cases, such as those in which the inmate has failed to make efforts toward rehabilitation, has continued to engage in criminal conduct postincarceration, or has shown a lack of insight or remorse, the aggravated circumstances of the commitment offense may well continue to provide ‘some evidence’ of current dangerousness even decades after commission of the offense.” In re Shaputis, 190 P.3d 573 (Cal. 2008): “[T]he paramount consideration for both the Board and the Governor under the governing statutes is whether the inmate currently poses a threat to public safety…” depends chiefly on the original crime of conviction, according to rules set out in California Penal Code § 3000.1. If the original crime was murder and it was committed after 1982, the released person is presumptively on parole for his/her lifetime but can petition the Board to be discharged from parole after either five years (if second-degree) or seven years (if firstdegree). Most other lifers will serve between three and five years, but can petition for discharge earlier. Once a prisoner is released from custody onto parole supervision, the length of the parole period post-release L IF E I N L I M BO : An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California 10 RECENT DISPOSITION RATES As Chart 2 depicts, in the last 20 years the annual number of scheduled hearings to determine suitability for parole release for individuals serving life sentences has grown significantly though not at a consistent rate, with the annual number averaging about 1600 early in this period and over 6000 in the most recent years. But the annual number of hearings actually conducted has grown less significantly and has fluctuated much more, with the percentage of scheduled hearings actually ending up in conducted hearings dropping notably from about 75 percent to about 50 percent. In 2009, the Board of Parole Hearings scheduled 5,639 hearings to determine parole suitability and conducted 2,714 hearings.33 The reasons for this drop-off and increasing magnitude of the drop-off require further examination, including inquiry into whether resource constraints on BPH have played any role. But a key factor—at least in the last two years— appears to be a disincentive built into the system: If an inmate anticipates a high probability of denial of parole at a hearing, s/he often chooses to cancel the hearing as a formal denial by the Board could greatly delay his or her entitlement to a subsequent hearing. The mechanisms by which an inmate exercises this risk aversion is a stipulation to his/her own unsuitability for parole release; a waiver of the hearing; or a postponement. A stipulation is essentially the inmate’s concession that s/he is not suitable for parole release, while a waiver is a related but slightly different mechanism by which the inmate agrees to forego his/her entitlement to a hearing at which s/he could have argued suitability. The use of these procedural mechanisms has become much more significant since the passage of Marsy’s Law in 2008, which greatly increases the delay in entitlement to a new hearing after a denial and regulations promulgated in 2008 that give an inmate the right to waive his or her hearing without stipulating to unsuitability.34 The operation of these mechanisms and the inmate factors associated with them deserve special research emphasis, and the relationship between stipulations/waivers and the timing of later hearings and grant/release outcomes is an important question L IF E I N L I M BO : on which SCJC is now seeking to obtain and analyze empirical data. Meanwhile, we now have raw data on the frequency of stipulations/waivers. As Chart 3 illustrates, overall the grant rates by the Board of Parole Hearings have increased significantly in absolute numbers in recent years; these rates have fluctuated erratically as a percentage of conducted hearings, although in recent years that percentage has been higher than in previous ones. Currently the BPH grant rate is about 18 percent. In the last decade (2000-2010), the percentage of scheduled hearings resulting in denial has dropped from about 75 percent to about 40 percent, but the percentage resulting in grants has only increased a few percent. The explanation for the difference, as noted, has been a very large decrease in the percentage of scheduled hearings resulting in actually conducted hearings. More analysis is necessary to appreciate the difference in grant rates yearby-year. In particular, the more extreme differences in grant rates may be explained by differences in the profiles of appearing inmates, the composition of the board, or other factors. As Chart 4 illustrates, the average denial length (i.e. the numbers of years of delay before the inmate is entitled to a subsequent suitability hearing) has changed without pattern between 2000-08 but jumped dramatically after that. Proposition 9/Marsy’s Law mandates denial periods of three, five, seven, 10, and 15 years,35 the presumption starting with a 15-year denial period absent clear and convincing evidence that it should be shorter,36 Although litigation is pending on whether these deferral periods violate the ex post facto clause.37 An inmate may request that the Board advance a subsequent hearing once every three years. The Board has wide discretion to grant or deny these requests, the criteria including “the views and interests of the victim” and changed circumstances or “new information [that] establishes a reasonable likelihood that the additional period of incarceration is unnecessary.”38 According to statistics included by An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California 11 CHART 2 Number of Scheduled and Conducted Lifer Suitability Hearings, 1978 – 2010 8000 Number of Hearings 7000 Scheduled 6000 Conducted 5000 4000 3000 2000 1000 1978* 1979* 1980* 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 0 Year * There was only 1 lifer suitability hearing conducted in 1978 and in 1979, and 2 hearings in 1980. CHART 3 Board of Parole Hearings Grant Rate, 1978 – 2010 100% 90% 80% Grant Rate 70% 60% 50% 2008: Passage of Marsy's Law; Lawrence and Shaputis decisions 40% 30% 20% 0% 1978* 1979* 1980* 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 10% Year * There was only 1 lifer suitability hearing conducted in 1978 and in 1979, and 2 hearings in 1980. L IF E I N L I M BO : An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California 12 CHART 4 Duration of Parole Denials, 2000 – 2010 2500 Post-passage of Marsy's Law Number of Denials 2000 1500 1 to 2 years 3 to 5 years 1000 7 to 15 years 500 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Year plaintiffs in pending litigation, the Board denied 59 out of 61 or 97 percent of requests for advanced hearings submitted by prisoners between December 2008 and August 2010.39 Before the passage of Marsy’s Law in 2008, two-thirds of prisoners who were denied released received deferral dates of one or two years. Now most inmates denied release receive 3- and 5-year denials. A significant incidence of those long-term denials has occurred and will probably increase the number of inmates requesting waivers and making stipulations of unsuitability. As Chart 6 depicts, the Governor’s use of his power to reverse grants by the Board of Parole Hearings has changed dramatically with the identity of the Governor. Governor Pete Wilson (1991-1999), the first Governor to implement the new measure, rejected only 27 percent of grants, although he only considered a handful of cases. Governor Gray Davis (1999-2003)—who claimed he would not parole a single convicted murderer—reversed virtually all the grants during his term. Governor Arnold L IF E I N L I M BO : Schwarzenegger (2003-2011) reversed about 60 percent of grants, while remanding about 20 percent to the Board of Parole Hearings for further review (though Chart 6 illustrates the reversal rate within his term fluctuated). In his first few months in office, Governor Jerry Brown has reversed at the lowest rate of the three Governors. The Davis Administration is likely to remain a sharp anomaly—a virtual nullification of the law—since the Proposition 89 procedure was arguably designed as a kind of appellate review by the Governor. A lifer’s prospect of actually being granted parole by the Board and not having the decision reversed by the Governor is—and always has been—slim. Using the overall Board grant rate and the Governor’s non-reversal rate for murder cases, we have estimated the likelihood in Chart 7. An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California 13 CHART 5 Annual Number of Governor's Parole Decisions Involving Murder Cases, 1991 – 2010 450 Number of Decisions 400 350 300 250 200 150 100 50 0 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Wilson Davis Schwarzenegger Year CHART 6 Governor's Reversal Rate for Parole Decisions Involving Murder Cases, 1991 – 2010 Percent of All Cases that Were Reversed 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Wilson Davis Schwarzenegger Year L IF E I N L I M BO : An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California 14 CHART 7 Estimated* Likelihood of a Murder Case Being Granted Parole by BPH and Governor, 1991 – 2010 7.0% Percent 6.0% 5.0% 4.0% 3.0% 2.0% 1.0% 0.0% 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Wilson Davis Schwarzenegger Year *Estimated likelihood was calculated using the BPH's parole grant rate for all life-term sentences and the Governor's non-reversal rate for murder cases. DEMOGRAPHIC DETAILS OF THE LIFER POPULATION As discussed earlier and depicted in Chart 8, most lifers currently incarcerated were convicted of first- and second-degree murder.40 Of the people serving term to life sentences in California as of December 31, 2010, the largest categories by crime type are described in Chart 8. 20-year period from 1990-2010, the average number of years served was about 20 years. The average length served by the largest categories of crime type is depicted in Chart 9. CHART 8 CHART 9 Lifer Population by Type of Crime Lifer Population by Average Years Served TYPE OF CRIME NUMBER OF PRISONERS AS A PERCENTAGE OF LIFER POPULATION 19,360 81% 1st Degree 8,299 35% 2 Degree 8,654 36% Attempted 2,399 Rape & other sexual offenses Kidnapping Murder nd NUMBER OF PRISONERS MEAN (YEARS) PUNISHMENT PROSCRIBED BY CURRENT CA PENAL CODE41 2nd Degree Murder 701 19.87 15 years to life 10% 1st Degree Murder 375 20.14 25 years to life 1,467 6% Kidnapping for Robbery or Rape 120 17.13 7 years to life 1,057 4% Attempted Murder 107 13.85 7 years to life For the 1499 individuals who served term-to-life sentences who were released from custody between January 1, 1990 and December 31, 2010, the average amount of time served was 225 months or 18.75 years. Of approximately 1,000 lifers who had been sentenced for murder and were released from custody during the L IF E I N L I M BO : TYPE OF CRIME Obviously, because these individuals have committed more serious crimes, they are not typical of the larger California prison population, but the mix of similarities and dissimilarities in comparisons to the overall prison population is complex. An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California 15 The vast majority—96%—of lifers are male (as compared to 93% of the overall prison population). The percentages of lifers who are Black (31%) and Hispanic (38%) are very similar to the percentages for these groups in the overall inmate population.42 In terms of age, 85% of current lifers are 55 or under and 14% are 56 or older. In addition, note that the actual number of currently incarcerated lifers who are aged over 65 is 929. Unsurprisingly, this distribution is not similar to the age disproportion of the overall inmate population, since most lifers are serving lengthy prison sentences. In particular: 13% of California prisoners are 22-25, as compared to 5% of lifers. 33% of California prisoners are 26-35, as compared to 25% of lifers. 25% of California prisoners are 36-45, as compared to 30% of lifers. 18% of California prisoners are 46-55, as compared to 24% of lifers. 5% of California prisoners are 56-65, as compared to 10% of lifers. 1% of California prisoners are over 65, as compared to 4% of lifers. The distribution among lifers by mental health designations is closely proportionate to that in the general inmate population. The percentage of lifers “sentenced” by each county closely approximates the percentage of all prisoners coming from those counties and is also closely proportionate to the general population of those counties. In particular, Los Angeles (39%), San Diego (7%), and Orange (6%) and Riverside (6%) Counties comprise the biggest feeders of the state’s lifer population. Further analysis might factor in serious crime rates of those counties, as well as changes in the distribution over time. L IF E I N L I M BO : The distribution of lifers among across the state prisons is highly dispersed, ranging from one percent to eight percent in particular prisons, and is not a function of the differing sizes of the prisons: As a percentage of the prisoner population in particular prisons the lifer concentration differs drastically, with a huge concentration in California State Prison - Solano (63%), Calipatria State Prison (48%), Correctional Training Facility (38%) and California State Prison - Corcoran (36%). The reason for this variance may lie in noncontroversial decisions about logistics, resources, and classification status, but the issue merits further examination, including analysis of program availability at those institutions and whether place of imprisonment bears any distinct association with rates of hearings and grants/denials. Individuals serving life sentences with the possibility of parole are fairly evenly distributed among medium (30%) and high medium (29%) housing security levels, skewing them more toward the higher end than the general inmate population. On the other hand, 75% of lifers score as low risk and 90% as low or moderate risk by the Californa Static Risk Assessment instrument.43 These scores contrast sharply with the general inmate population (28% low, 28% moderate, 11% high property, seven% high drug, 22% high violent, and four percent none). These figures merit detailed further and secondary data gathering, including correlations to hearing/grant rates and consideration in light of recidivism analysis. An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California 16 RISK OF RELEASE Any indeterminate sentencing system—including California’s for individuals serving life sentences with the possibility of parole—purportedly has several important purposes. Among them is retribution which suggests that offenders should be punished in proportion to the harm they caused and their culpability in committing the crime. Thus, some portion of the time lifers serve is intended to satisfy the retributive purpose. The other portion meets other important purposes, including deterrence, rehabilitation, and incapacitation—all of which focus on using criminal penalties to minimize future criminal behavior by the individual offender and would-be offenders.44 In meeting these purposes, the Board is charged with assessing what the public safety risk is of each lifer’s release. Indeed, the criteria for release as articulated by governing statute and regulations and relevant case law reiterates that predicting and preventing recidivism is the primary concern. Few studies have been conducted documenting the recidivism rates for lifers specifically but the few that exist all suggest that the recidivism rate—as defined by recommitment for a new offense—is relatively low.45 In a cohort of convicted murderers released since 1995 in California, the actual recidivism rate is in fact minuscule. In particular, among the 860 murderers paroled by the Board since 1995, only five individuals have returned to jail or returned to the California Department of Corrections and Rehabilitations for new felonies since being released, and none of them recidivated for life-term crimes.46 This figure represents a lower than one percent recidivism rate, as compared to the state’s overall inmate population recommitment rate to state prison for new crimes of 48.7 percent.47 The variance between these two rates warrants additional analysis; in particular, a more nuanced examination of the 860 individuals granted parole release as compared to the overall lifer population might explain their low recidivism rates. L IF E I N L I M BO : Other sources of information shedding light on the recidivism risk of lifers are established studies of recidivism rates for non-lifers that focus on crime of conviction, criminal record, age at time of release, length of imprisonment and other factors. The factors examined in these studies can be used as proxies to help us gauge likely recidivism projections for lifers. A good example is the age factor. Some non-lifer studies demonstrate that as a general matter, people age out of crime. For most offenses—and in most societies—crime rates rise in the early teenage years, peak during the mid-to-late teens, and subsequently decline dramatically. Not only are most violent crimes committed by people under 30, but even the criminality that continues after that declines drastically after age 40 and even more so after age 50.48 More uncertain are the prospects for offenders between the ages of thirty and fifty. Determining when there is not an unreasonable risk to public safety to parole relatively young lifers will depend on the continuing improvement of risk-assessment instruments, as well as careful attention to the empirical evidence linking particular types of crimes to particular rates of re-offending. In California specifically, CDCR’s newest recidivism report (October 2010) documents that inmates designated as serious or violent offenders, older inmates and inmates who serve 15 years or more recidivated at a lower rate than those who were not.49 Two other sources of information are the risk levels classifications as assessed by both the California Static Risk Assessment instrument and the tools used by the Forensic Assessment Division (FAD) to predict current risk. Both indicate that lifers are relatively lower risk than other inmates, but more information is needed to understand the nature of instruments used and their ability to correlate recidivism rates with risk scores. An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California 17 THE SCJC LIFER TRANSCRIPT ANALYSIS In light of the rules governing and stakeholders participating in parole release for lifers, and the great variety of factors they bring into play in any hearing, the Stanford Criminal Justice Center decided to undertake the first empirical assessment of the actual conduct and circumstances of parole hearings to assess which factors play salient roles in predicting or determining outcomes. We received 754 hearing transcripts constituting a random sample of 10 percent of all parole suitability hearings conducted between October 1, 2007, and January 28, 2010 from the California Department of Corrections and Rehabilitation. Of the 754 hearings, 49 (6.5 percent) took place in 2007, 276 (36.7 percent) took place in 2008, 377 (50 percent) took place in 2009, and 52 (6.9 percent) took place in 2010. These transcripts ranged from less than 50 to more than 200 pages. To transform them into usable data, we used two procedures. First, we roughly summarized the data, gathering a basic set of information about all of the transcripts: hearing date, inmate name, result (grant or denial), persons present at the hearing, and so on. As a second, more comprehensive, process, we designed an extended codesheet to capture more than 180 variables of interest from the transcripts, ranging from inmate characteristics to details of the life offense to prison programming. We hired and trained Stanford University undergraduates to code the transcripts by carefully reading the text and making selections on a web-based form. To date, we have completed 448 transcripts in this second-pass process, or approximately 60 percent of the sample. The majority of the completed transcripts were from hearings conducted in 2009 (after the passage of Marsy’s Law and the court decisions in Lawrence and Shaputis), though we have coded some transcripts from 2007, 2008, and 2010 as well. GENERAL FINDINGS There are two types of parole suitability hearings: initial suitability hearings, in which the prospective parolee is appearing in front of the parole board for the first time, and subsequent suitability hearings, in which the prospective parolee has been denied parole at a past hearing. Almost 90 percent of the hearings were subsequent, rather than initial, parole hearings. Chart 10 summarizes the dispositions of the 754 hearings by whether the hearing was an initial or subsequent hearing. (Note that the table excludes one hearing in which the decision was postponed pending the receipt of a missing psychological evaluation, and a second in which the commissioners’ decision was not indicated in the transcript.) L IF E I N L I M BO : CHART 10 Disposition by Hearing Type, Full Sample INITIAL SUBSEQUENT TOTAL Denied 87 567 654 (87.0%) Granted 2 96 98 (13.0%) 89 664 752 Total In total, 87 percent of the hearings in our sample resulted in a denial of parole. Inmates in subsequent parole hearings fared much better than inmates appearing in front of the Board for the first time: nearly 15 percent of subsequent hearings resulted in a grant and 2.2 percent of initial hearings produced grants. An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California 18 Grant rates appear to vary significantly by year. Chart 11 reports the grant rate by year from the full sample of transcripts. (Our reporting on the grant rate here is not intended to expand upon or change our earlier analysis of the overall grant rate, but to contextualize our transcript analysis.) CHART 11 Grants by Year, Full Sample YEAR DENIED GRANTED GRANT RATE 2007 45 4 8% 2008 255 21 8% 2009 316 61 16% 2010 40 12 23% By the end of our sample, the grant rate was nearly triple what it was in 2007 and 2008. The result is highly statistically significant. Though commissioners became more lenient on one dimension, by increasing the grant rate in 2009 and 2010, they became more stringent on another dimension. Upon denying a parole applicant, parole commissioners must set a date until the next parole hearing but have discretion in determining the length of time. The commissioners most commonly set a date of one, three, or five years until the next parole hearing50, but in some cases in our dataset, the commissioners delayed the next parole hearing for as much as 15 years. Chart 12 summarizes the average number of years to next hearing, by the year the hearing was conducted. CHART 12 Years to Next Hearing, by Hearing Date 2007 2008 2009 2010 TOTAL 2.0 2.2 4.6 5.1 3.5 L IF E I N L I M BO : The result may reflect the impact of “Marsy’s Law,” an amendment to the California state constitution enacted by California voters via the ballot initiative process in November 2008. As discussed above, Marsy’s Law, also called Proposition 9, increased the maximum parole denial period to 15 years. After 2008, one- and two-year denial terms, which were common prior to the passage of Marsy’s Law, became prohibited. The result was a significant shift upward in denial periods: in the 2009 transcripts in our sample, 45 percent of denials were for periods of five years or more. Every hearing is led by a presiding commissioner, who is joined by a deputy. In total, there were 24 presiding commissioners in our dataset. The total number of hearings they presided over varied from a low of six hearings to a high of 89. Because of the relatively small amount of data we have about each commissioner, we cannot conclude that there is a statistically significant difference between the grant rates of the various commissioners. That said, the numerical differences are substantial: grant rates by commissioner varied greatly from a low of zero percent to a high of 31 percent. One commissioner, for instance, granted parole in twelve of the 61 hearings in our sample he presided over; Another commissioner, by contrast, granted parole in only one of the 43 hearings in our sample over which she presided. Additional study is necessary to understand possible reasons for these variances, including the classification status of inmates seen by each commissioner. The last piece of information we have collected about the complete sample is information about who attended the hearing. Specifically, we have data on whether a victim appeared at the hearing, with “victim” defined broadly as either the immediate victim of the crime or a friend, family member, or acquaintance of the victim of the crime. Chart 13 summarizes grant rates by the presence of victims. An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California 19 CHART 13 Grant Rate by Presence of Victim at Hearing Victim not present Victim present DENIED GRANTED GRANT RATE 586 94 13.8% 70 4 5% There is a statistically significant difference in the grant rate between hearings at which victims are present and hearings at which victims are not present. The effect is in the expected direction: when victims attending hearings, the grant rate is less than half the rate when victims do not attend. A more nuanced analysis of the relationship between victim participation and disposition rates might identify the reasons for this correlation. In particular, a better tracking of when victims most commonly participate in hearings—particularly whether they typically appear primarily at initial or first subsequent suitability hearings – could explain why their participation is associated with parole denials. SPECIFIC FINDINGS: SECOND-PASS ANALYSIS More detailed results can be obtained from our secondpass analysis. Because we have finished coding only around two-thirds of the transcripts, however, these analyses are necessarily preliminary. In what follows, we consider two general categories of results: first, the general characteristics of inmates serving life sentences and their relationship to release decisions; and second, other factors that are positively or negatively associated with parole release. INMATE CHARACTERISTICS CHART 14 Life Crime As Chart 14 indicates, the majority of parole-eligible life offenders are second- or first-degree murderers. There is no statistically significant difference in the grant rates of various types of offenders, although those serving sentences for attempted murder are the least successful inmates. Grant rates for first- and second-degree murderers are nearly identical. Chart 14 also includes the average time served by inmates in each offense category at the time of the hearing, in years. Offense Type by Decision AVERAGE YEARS SERVED DENIED INMATES GRANTED INMATES TOTAL Second Degree Murder 20.1 195 44 239 First Degree Murder 17.2 104 20 124 Attempted Murder 14.2 26 0 18 Kidnapping for Sex Crime/Robbery 21.7 27 2 29 Aggravated Mayhem 15.0 2 0 2 Kidnapping for Ransom 16.5 2 0 2 Conspiracy to Commit Murder 21.2 4 2 6 Rape 13.1 3 0 3 Drive-By Shooting 20.1 1 0 1 8.6 2 0 2 18.9 368 68 436 Torture Total L IF E I N L I M BO : An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California 20 Several factors related to the life crime are not related to release in a statistically significant way. First, the use of a firearm in the life crime does not appear to have a significant effect on the outcome of the parole hearing. In total, 182 prisoners did not use firearms in the commission of their life crime, and 214 prisoners did use a firearm. The release rates were 15 percent and 16 percent, respectively; the difference is not statistically significant. Commissioners’ decisions did not seem to vary according to the number of people the inmate victimized in the commission of the life crime. Thirteen of the 106 cases (12 percent) in which the inmate’s life crime involved multiple victims resulted in release; by contrast, 55 of the 333 (16.5 percent) single-victim cases resulted in release. The difference is not statistically significant. Prior Record Prior record does not appear to significantly affect release decisions, whether they are adult or juvenile records. Sixteen percent of inmates with juvenile records prior to the commission of their life crimes obtained parole release, compared to 15 percent of inmates without juvenile records. The difference is not statistically significant. The same holds true for the effect of prior adult criminal records. Almost 60 percent of inmates in our sample had prior adult convictions before committing their life crime, but the grant rate was 14 percent for inmates without adult criminal records and 16 percent for inmates with criminal records. L IF E I N L I M BO : Age Chart 15 shows the age at life crime, by whether the inmate was paroled. The figure shows that most inmates committed their life crime between the ages of 20 and 25. The pattern is similar for both paroled and nonparoled inmates, though inmates who committed their life crimes between 20 and 30 were somewhat more likely to be paroled than inmates whose life crimes were committed in their forties. Few of the latter type of inmates received parole grants. CHART 15 Age at Life Crime, by Parole Outcome .08 Density One factor that appears to be strongly associated with release is whether the life crime involved sexual violence. Only two of the 32 transcripts we have coded so far that involved sexual violence of any kind resulted in grants of parole; by contrast, 16 percent of parole cases not involving sexual violence (66 out of 404) resulted in release. .06 .04 .02 0 20 30 40 50 60 Age Grants Denials The average age of inmates at the time of the parole hearing is 50.8. The average age of inmates granted parole is 49.9 years, and the average age of inmates denied parole is 51. The difference is not statistically significant. Surprisingly, age does not appear to be a significant factor in release decisions: a simple logistic regression model using age at the hearing date to predict the probability of release shows a somewhat negative but statistically insignificant effect of age on the likelihood of parole release. An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California 21 Other Factors Chart 16 provides assorted demographic characteristics of the inmates in our sample. None of the characteristics presented in the table—immigration status, whether an inmate has children, and marital status—is significantly associated with a release or denial. Chart 17 presents the grant rate by facility. To avoid misleading findings, state prisons that are poorly represented in our sample—specifically, facilities with fewer than ten hearings in the sample— were omitted from this table, leaving a total of 13 facilities. CHART 16 CHART 17 Assorted Demographic Characteristics Grant Rate by Facility Immigration Status Citizen Illegal immigrant Legal resident Unknown* 51 DENIED GRANTED GRANT RATE % OF TOTAL 224 41 15.5% 63% Mule Creek 50 6 10.7% 13% 3 1 25.0% 1% California Institution for Women 78 19 19.6% 23% San Quentin 13 California Men’s Colony Children DENIALS GRANTS GRANT RATE 9 5 35.7% 10 5 33.3% 4 23.5% 26 6 18.8% 51 11 17.7% Has children 137 31 18.5% 41% Avenal Doesn’t have children 210 35 14.3% 59% Correctional Training Facility 45 9 16.7% Central California Women’s Facility 17 3 15.0% California Substance Abuse Treatment Facility 23 4 14.8% Solano 61 9 12.9% California Medical Facility 16 2 11.1% Chuckawalla Valley 20 2 9.1% Folsom 12 1 7.7% Pleasant Valley 10 0 0.0% Marital Status Divorced 84 10 10.6% 24% Married before prison 51 9 15.0% 15% Married, during prison Single Spouse deceased 32 15 31.9% 12% 156 23 12.9% 45% 13 4 23.5% 4% Though these characteristics are not significantly associated with the grant rate, some results are intrinsically interesting. First, 59 percent of the inmates in our sample have children. Of that population, only 35 percent are married, and only 22 percent were married before entering prison. Other Factors Associated with Release Facility Parole hearings are held on-site at most of California’s 33 state prisons. Grant rates might vary across facilities for a variety of reasons, such as systematic differences in the type of inmates held at various facilities, availability of rehabilitative programs at various facilities, or differences in the pool of commissioners who conduct hearings at various facilities. L IF E I N L I M BO : As the table indicates, grant rates differ dramatically by facility. Some prisons, like Chuckawalla, Folsom and Pleasant Valley, have grant rates below 10 percent, others, like Mule Creek and the California Institution for Women, grant more than a third of parole cases. As stated above, Solano houses the largest percentage of lifers as a percentage of its total prison population. A more robust analysis of grant rates by institution is warranted to better understand the reasons underlying variances. An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California 22 Behavior in Prison Inmate behavior during the prison term is a recurring theme in parole hearings. Parole commissioners typically scrutinize inmate’s disciplinary records, and often ask detailed questions about violations of prison rules. Psychological Evaluations Virtually all inmates who appear at parole hearings have undergone psychological evaluations. Parole commissioners always receive and often review the results of these evaluations carefully. In California prisons, disciplinary infractions are documented using two forms, the CDC 128 “Custodial Counseling Chrono” (or sometimes the CDC 128B “Informational Chrono”), and the CDC 115 “Rules Violation Report.” 128 infractions are typically minor conduct violations, including smoking, being in an unauthorized area, using foul language, or possessing non-serious contraband. 115 infractions, which trigger a notice-and-hearing process, can be either non-serious (“administrative”) or serious. Serious violations include violence toward inmates or prison personnel, possession of controlled substances or weapons, and other serious infractions. The two most common types of clinical opinions in our sample are the Axis V Global Assessment of Functioning Scale and the Clinician Generic Risk assessment.52 The Axis V GAF measures a patient’s overall level of psychological, social, and occupational functioning on a 100-point continuum, with higher scores indicating higher functioning. The Clinician Generic Risk, by contrast, assigns inmates a simple risk-of-recidivating score: low, low-moderate, moderate, moderate-high, and high. Both 115s and 128s are exceedingly common. Eightyone percent of inmates in our sample have at least one 115 in their record, and 89 percent of inmates have at least one 128. The 115 infractions are strongly associated with the grant rate; 25 percent of inmates with no 115 infractions received parole grants, while only 13 percent of inmates with at least one 115 infraction received a grant—a result significant at the .01 level. And the more 115s an inmate accumulates, the greater an effect the inmate’s disciplinary record has on the inmate’s chances for parole release. Just 16 of the 149 inmates with more than five 115s (11 percent) received parole release. On the other hand, 128 infractions are not significantly associated with grant rate. One inmate received a grant of parole despite accumulating sixty 128 infractions. Preliminary evidence also suggests that the seriousness of the disciplinary violation has a substantial effect on commissioners’ decisions. For example, violent disciplinary infractions, regardless of when they occur, are significantly associated with parole denials. Only 11 of the 128 (8.5 percent) inmates with violent disciplinary records in prison were released, compared to 20 percent of inmates with no violent disciplinary infractions. L IF E I N L I M BO : CHART 18 Grant Rate by psychological Evaluation Instrument DENIED GRANTED TOTAL GRANT RATE Clinician Generic Risk Low 107 42 149 28% Low-Moderate 50 8 58 14% Moderate 47 2 49 4% Moderate-High 12 0 12 0% High 14 0 14 0% Axis V-GAF 0-74 37 0 37 0% 75-84 78 18 96 19% 85-100 66 12 78 15% Both the Clinician Generic Risk and the Axis V-GAF are significantly correlated with grant rate. This is especially true of the Clinician Generic Risk assessment, which is statistically significant at the .001 level. As Chart 18 indicates, inmates who receive an average score or higher virtually never receive parole release. Similarly, none of the inmates in our sample who received below 75 on the Axis V-GAF enjoyed favorable release outcomes. An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California 23 These results suggest that the psychological evaluation tools used to assess risk potential and inmate psychological stability play an influential role in the parole process. Drug Abuse During parole hearings, commissioners often discuss inmates’ records of drug and alcohol abuse at considerable length. A history of drug or alcohol abuse is not correlated with grant rate. What is highly associated with grant rate, however, is whether an inmate is participating in a “twelve-steps” program (that is, Alcoholics Anonymous, Narcotics Anonymous, or some similar program), and whether he or she can correctly answer questions about those steps, which commissioners often ask to test inmates’ commitment to drug and alcohol treatment Conclusion The foregoing analyses are necessarily preliminary, but they shed important light on how the parole hearing process functions in California. Some results, like the importance of in-prison conduct and psychological evaluations, confirm standard presuppositions about what matters to parole commissioners. Other results, like the irrelevance of age and offense type, are counterintuitive. As the study proceeds, we will continue to analyze factors that contribute to parole release decisions, with the goal of developing a comprehensive model of parole decisionmaking in California. In total, 159 inmates were asked whether they could identify one or more of the 12 steps. Of the 56 inmates who failed to correctly answer the commissioners’ question, only one was paroled. By contrast, 37 of the 141 who correctly responded to commissioners’ queries received parole—a grant rate double that of inmates who were not asked about their treatment program. It therefore appears that commissioners mostly do not discriminate between inmates who have or have not abused drugs or alcohol. For those inmates with substance-abuse problems, however, the ability to demonstrate a commitment to a recovery program is a key component of obtaining parole release. L IF E I N L I M BO : An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California 24 FURTHER EMPIRICAL RESEARCH ON THE PAROLE RELEASE PROCESS FOR LIFERS The Stanford Criminal Justice Center is working on the following other research projects related to lifers and will be issuing subsequent bulletins on a quarterly basis: THE ROLE OF THE DISTRICT ATTORNEY: One key factor in the course and ultimate outcome of lifers seeking release is the role of the District Attorney. SCJC is currently undertaking an innovative survey consisting of interviews with district attorneys in a broad sample of California counties. The goal of the survey is to determine particular offices’ approach to these hearings, including what resources and staff they devote, whom they assign to the hearing, what role the designated District Attorney representatives are expected to play, how they prepare for the hearings, what factors they consider important in opposing release, their role in judicial review, and other information. THE ROLE OF VICTIM(S): We are currently reviewing the role victims play in the hearing process, including how their rights have expanded since the passage of Marsy’s Law, how frequently and in what manner victims participate and whether victim participation has any bearing on Board decision-making. In addition, our research will identify model practices for victim participation used in other jurisdictions. FORENSIC EXAMINATIONS: The governing standard for granting parole is whether the inmate presents a current risk to public safety. The Forensic Assessment Division (FAD) is charged with conducting forensic examinations on lifer inmates prior to their meeting with the Board. We are currently researching the tools and procedures used by the FAD to determine the role the examinations play and the weight they get—and should get—in assessing current and future risk. JUDICIAL REVIEW OF PAROLE DECISIONS: Given that the majority of decisions made by the Board result in denial and the relatively high reversal rate among Governors, the court serves as an effective and default vehicle for lifers seeking parole release through habeas appeals. Since the 2011 Swarthout v. Cooke decision, which virtually precludes federal habeas corpus review, state judicial review offers inmates an opportunity to challenge the decisions of BPH and the Governor.53 Tracking the number of cases brought before the court and the results of these habeas petitions will help us gain important understanding into the flow of parole release for lifers. THE ROLE OF COMMISSIONERS: Given the enormous role commissioners and deputy commissioners play in the parole suitability hearing process, we are investigating the nature of training received by commissioners who preside over suitability hearings; how commissioners prepare for and approach suitability hearings; and the roles assumed by commissioners versus deputy commissioners. L IF E I N L I M BO : An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California 25 ENDNOTES 1 Under California’s Determinate Sentencing Law, most felonies carry a “determinate” prison sentence consisting of a specific number of months or years the offender must serve in prison before s/he can be released. See California Penal Code § 1170. The death sentence can only be imposed for first-degree murder when certain special and aggravating circumstances are charged and proved. For a few very egregious crimes, the sentence may be life without the possibility of parole (LWOP). Individuals serving LWOP sentences can only be released from prison by Governor pardon or commutation. See California Penal Code §§ 4801-4802; 15 California Code of Regulations § 2816. The “lifers” who are the subject of this study are prisoners who have been sentenced to a “life sentence with the possibility of parole.” These sentences are also sometimes called “indeterminate” because, by definition, the trial judge cannot pre-determine the exact time the prisoner will be released; that time is subject to the parole process. Any sentence of life with the possibility of parole has a minimum sentence that must be served before the Board can even consider release. The default rules for the minimum term are established by California Penal Code § 3046: (a) No prisoner imprisoned under a life sentence may be paroled until he or she has served the greater of the following: (1) A term of at least seven calendar years or (2) A term as established pursuant to any other provision of law that establishes a minimum term or minimum period of confinement under a life sentence before eligibility for parole. For many specific crimes that authorize life sentences, the specific criminal statute expressly includes a minimum prison term that constitutes “any other provision of law” under § 3046 (a) (2). Thus, the punishment for second degree murder is ordinarily “a term of 15 years to life,” while first degree murder generally carries “a term of 25 years to life.” (California Penal Code § 190 (a)). Other statutes specifying indeterminate sentences do not mention a minimum term, describing the sentence simply as “imprisonment in the state prison for life with the possibility of parole” or “imprisonment in the state prison for life.” In this category are sentencing provisions for attempted premeditated murder (California Penal Code §§ 664(a), 187, 189) as well as aggravated mayhem (California Penal Code § 205), torture (§ 206.1), kidnap for ransom without bodily harm (§ 209, subd. (a)), kidnap for robbery or sexual assault (§ 209, subd. (b)), kidnap during carjacking (§ 209.5, subd. (a)), nonfatal train wrecking (§ 219), attempted murder of peace officer or firefighter (§§ 664, subd. (e), 187), exploding a destructive device with intent to kill (§ 12308), and exploding a destructive device that causes mayhem or great bodily injury (§ 12310, subd. (b)). These statutes would then incorporate the default minimum term of seven years under California Penal Code § 3046(a)(1). Finally, note that if a person is convicted of a crime carrying an indeterminate term that does not specify a minimum term but is also convicted of a separate crime that does carry a fixed term, that latter term can establish the minimum number of years that must be served before parole eligibility. Thus, the operative minimum term can depend on any of the numerous complex determinate sentencing laws and enhancements. California Penal Code § 1168(b), cross-referenced in § 3040, states: “For any person not sentenced under [a determinate term], but who is sentenced to be imprisoned in the state prison … the court imposing the sentence shall not fix the term or duration of the period of imprisonment.” L IF E I N L I M BO : 2 For instance, as comparison to other large systems, lifers (i.e. people serving life sentences with the possibility of parole) comprise nearly three percent of the federal prison population, four percent of the Florida prison population, nearly five percent of the Texas prison populations, 10 percent of the Ohio prison population, and nearly 15 percent of the New York prison population. 3 Because the Three-Strikes law was passed in 1994, the first inmates sentenced under that law will come before the Board of Parole Hearings in 2019 after they have served 25 years of their sentences. See California Penal Code § 667(e)(2): “If a defendant has two or more prior felony convictions …the term for the current felony conviction shall be an indeterminate term of life imprisonment with a minimum term of the indeterminate sentence calculated as the greater of…imprisonment in the state prison for 25 years.” 4 Under California Penal Code §§187-189, a person commits first-degree murder when s/he kills with deliberation and premeditation, or otherwise causes death in the course of commuting or attempting to commit one of several enumerated felonies, including arson, rape, sexual assault against a minor, carjacking, robbery, burglary. A person commits second-degree murder if s/he kills intentionally, although without premeditation, or if s/he causes death with “an abandoned and malignant heart,” which means that s/he has acted with a conscious disregard for— or indifference to—human life. 5 Roberts v. Duffy, 140 P.260 (Cal. 1914) at 264. 6 California Penal Code § 5075. 7 California Penal Code § 5075(b). The list of current Commissioners and their biographies is available on the California Department of Corrections and Rehabilitation website at: http://www.cdcr.ca.gov/BOPH/commissioners.html. 8 The minimum qualifications for a Deputy Commissioner include either: (1) two years of experience in the California state service with equivalent responsibility to a Parole Administrator I; (2) three years of experience within the last five in the California Department of Corrections and Rehabilitation or Board of Parole Hearings in an equivalent class to Parole Agent III; (3) three years of experience in the field of administrative or criminal law plus equivalent to graduation from college; or (4) three years of experience in the administrative plus equivalent to graduation from college. Unlike the Commissioners, the list of Deputy Commissioners is not made public. 9 California Penal Code §§ 5075.5, 5075.6(b)(2). 10 California Penal Code § 5075.6(b)(1). 11 California Penal Law Code §§ 3041.5, 3041.7. 12 California Penal Law Code § 3041.7; 15 California Code of Regulations § 2256. 13 “Board of Parole Hearings “Lifer Attorney Packet’ Application for Attorney Appointment Roster Life Parole Consideration Hearings” at page 5. Available at: http://www.cdcr.ca.gov/BOPH/attorney_ employment.html 14 California Lifers Newsletter “The Parole Board Hires ‘Your’ Attorney” Volume 5, Number 6 at 10 (December 2009). 15 California Penal Law Code § 3041.7; 15 California Code of Regulations § 2030. An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California 26 16 California Penal Law Code § 3043. 17 California Penal Law Code § 3043, as expanded by Proposition 9 or “Marsy’s Law” (2008). Also 18 California Penal Law Code § 3041.5(a)(4). 19 15 California Penal Law Code § 3042(a). 20 15 California Penal Law Code § 3042 (f). 21 California Penal Code § 3041(b); 15 California Code of Regulations § 2402(a). 22 15 California Code of Regulations § 2281(d). 23 In particular, the regulations spell out the following factors that should be considered in determining whether the prisoner committed the offense in an “especially heinous, atrocious or cruel manner”: “(A) multiple victims were attacked, injured or killed in the same or similar incidents. (B) The offense was carried out in a dispassionate and calculated manner, such as executionstyle murder. (C) The victim was abused, defiled or mutilated during or after the offense. (D) The offense was carried out in a manner which demonstrates an exceptionally callous disregard for human suffering. (E) The motive for the crime is inexplicable or very trivial in relationship to the offense.” 15 California Code of Regulations § 2281(c)(1). 24 15 California Code of Regulations § 2281(c). 25 California Penal Code §§ 3041.5, 3041, 5011. 26 In re Powell, 45 Cal.3d 894 (Cal. 1988) at 904. 27 In re Rosenkrantz, 59 P.3d 174 (Cal. 2002) at 222. 28 In re Lawrence (44 Cal. 4th 1181 (2008)) and In Re Shaputis (44 Cal. 4th 1241 (2008), See W. David Ball, Heinous, Atrocious, and Cruel: Apprendi, Indeterminate Sentencing, and the Meaning of Punishment, 109 Colum. L. Rev, 893, 900 (2009). 29 Cal. Const. Article V, Section 8(b). See also California Penal Code § 3041.1. 30 The other states are Louisiana, Maryland and Oklahoma. In 2009, Louisiana allowed offenders sentenced to life on certain heroin offenses to be eligible for parole. All other life sentences are imposed without the possibility of parole. The Governor must approve all parole decisions. Interestingly, the Texas Constitution was amended in 1984 to remove Governor review of parole decisions. See www.tdcj.state.tx.us/bpp/publications/PG%20 AR%202010.pdf 31 In re Rosenkrantz, supra, 29 Cal.4th at p. 677-79. The factors to be considered in determining parole suitability as set forth in Title 15 of the California Code of Regulations, Section 2402, include “the absence of serious misconduct in prison and participation in institutional activities that indicate an enhanced ability to function within the law upon release are factors that must be considered on an individual basis by the Governor in determining parole suitability. The Governor also must consider any evidence indicating that the prisoner has expressed remorse for his crimes, as well as any evidence demonstrating that “[t]he prisoner has made realistic plans for release or has developed marketable skills that can be put to use upon release.” (§ 2402, subd. (d)(8).) . 32 In re Rosenkrantz, supra, 29 Cal.4th at page 679. 33 As noted below, the differences between the number of hearings scheduled and conducted in a given year are primarily due to stipulations by the inmate to unsuitability for parole, waivers to the right of a hearing, cancellations by the Board, and postponements by either the inmate or the Board. 34 Prior to 2008, the governing regulation required the inmate to stipulate to unsuitability and waive his/her right to a hearing simultaneously. See 2 California Code of Regulations § 2253(b). 35 California Penal Law Code § 3041.5(b)(3) as amended by Marsy’s Law (Proposition 9, 2008). 36 California Penal Law Code § 3041.5(b)(3). 37 The California Court of Appeals recently held that the application of the mandated denial periods enacted pursuant to Marsy’s Law to inmates convicted prior to the effective date of Marsy’s Law violates ex post facto principles and therefore cannot be applied. In re Michael Vicks, No. D056998, slip op. (Calif Ct. App., May 1, 11 2011). The decision is likely to be appealed. 38 California Penal Law Code § 3041.5(b)(4). 39 See Notice and Motion for Preliminary Injunction to Entire Class, Gilman v. Schwarzenegger, December 20, 2010 40 A person commits first-degree felony murder if s/he causes a death in the perpetration or attempt to perpetrate robbery, rape, burglary, kidnapping, mayhem, or sexual assault on a minor. A person commits second-degree felony murder if s/he causes death in the course of perpetrating or attempting to perpetrate certain other inherently dangerous felonies, such as providing heroin to a minor, distributing methamphetamine, or discharging a weapon in an inhabited building. 41 Because the individuals whose sentences comprise the mean may be serving terms under varying historical iterations of the California Penal Code that carry different punishments, there may be discrepancies between the punishment proscribed by current California Penal Code and the mean years served. 42 Blacks represent a much higher than their share of the resident population at 6.2 percent, whereas Hispanics comprise 37.6 percent of California’s resident population. See U.S. Census Bureau: State and County QuickFacts at: http://quickfacts.census. gov/qfd/states/06000.html 43 This instrument computes the risk to re-offend by using static risk indicators: gender, age, and offense history. See: http:// www.cdcr.ca.gov/Regulations/Adult_Operations/docs/ NCDR/2010NCR/10-02/CSRA%2012-09.pdf 44 Frase, R. “Punishment purposes” Stanford Law Review 58: 67-83. 45 A study of females in Canada found that only 6.3 percent of paroled lifers recidivated. See Bonta, J., B. Pang, and S. WallaceCapretta 1995 “Predictors of recidivism among incarcerated female offenders” The Prison Journal 75: 277-294. Also a study that tracked a group of “Furman inmates” who had their death sentences commuted to life in 1972 found very low recidivism rates for the subset that were eventually paroled (though the sample size – 47 individuals – was very small. See Marquant, J. and J. Sorensen 1988 “Institutional and postrelease behavior of furman-commuted inmates in Texas” Criminology 26(4): 677-693. 46 The data does not reflect any new misdemeanors committed or any crimes committed in other states by this cohort. 47 For releasees in FY 2005-6. L IF E I N L I M BO : An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California 27 48 Sex crimes are somewhat anomalous, with a bimodal distribution: a peak in the teen years, then a drop, and then another rise, but that later rise is in the offender’s late 20s. 49 See page 26 of 2010 Adult Institutions Outcome Evaluation Report (October 11, 2010) at: http://www.cdcr.ca.gov/Adult_Research_ Branch/Research_Documents/ARB_FY0506_Outcome_ Evaluation_Report.pdf 50 Our sample includes transcripts of suitability hearings conducted before the implementation of Marsy’s Law when commissioners could delay hearings for one or two years. 51 Inmates whose immigration statuses are “unknown” are likely U.S. citizens. In the vast majority of parole hearings involving noncitizens, citizenship status is explicitly discussed by the commissioners. In many hearings involving citizens, however, citizenship status is not discussed in the course of the hearing. 52 As of 2008, BPH stopped relying on the Axis V GAF. Riskassessment tools now used include the PCL-R, HCR-20, LS-CMI, and STATIC-99-R. 53 562 U.S., _, 131 S. Ct. 859 (2011) L IF E I N L I M BO : An Examination of Parole Release for Prisoners Serving Life Sentences with the Possibility of Parole in California 28