In the Matter of Jorge L. Linares, Appellant,v.Andrea W. Evans,, Respondent.BriefN.Y.September 8, 2015 To be argued by: Alfred O’Connor (15 minutes) Albany County Index No. 5033-12 State of New York Court of Appeals In the Matter of Jorge Linares, Appellant, -against- Andrea W. Evans, Chair of the New York State Board of Parole, Respondent. Brief for Petitioner-Appellant APL 2014-00076 Alfred O’Connor New York State Defenders Association 194 Washington Avenue, Suite 500 Albany, New York 12210 (518) 465-3524 Counsel of Record Alan Rosenthal Center for Community Alternatives 115 East Jefferson Street, Suite 300 Syracuse, NY 13202 (315) 422-5638 Dated: June 2, 2014 TABLE OF CONTENTS Table of Authorities ........................................................................................................................ i-iv Preliminary Statement ........................................................................................................................ 1 Summary of the Case and Argument .................................................................................................... 1 Statement of Facts .............................................................................................................................. 3 The 1978 guidelines for parole release decision-making .................................................... 4 The advent of evidence-based practice in criminal justice................................................. 6 New York State Commission on Sentencing Reform recommendations ....................... 7 2009 amendment to Executive Law § 259-c (4) .................................................................. 8 2011 amendment to Executive Law § 259-c (4) .................................................................. 8 Linares’ November, 2011 parole release hearing ................................................................ 9 COMPAS risk and needs assessment instrument ........................................................... 10 Parole denied ......................................................................................................................... 11 Administrative appeal .......................................................................................................... 12 Article 78 Petition ................................................................................................................. 12 The Evans memo .................................................................................................................. 13 The Board implements COMPAS but fails to establish written procedures governing its use ............................................................................................... 16 Supreme Court’s decision .................................................................................................... 17 Matter of Garfield v. Evans, 108 A.D.3d 830 (3d Dept. 2013) ...................................... 17 Appellate Division proceedings .......................................................................................... 18 Appellate Division order ..................................................................................................... 19 Motion for leave to appeal to the Court of Appeals ....................................................... 19 Proposed Rulemaking........................................................................................................... 19 Commentators criticize the proposed rule as non-compliant with the enabling legislation ................................................................................................................ 21 ARGUMENT POINT ONE Petitioner is aggrieved because the Appellate Division failed to accord him all the relief he is entitled to: a direction that the Board comply with Executive Law § 259-c (4) by establishing legislatively required written procedures incorporating risk and needs principles governing use of the COMPAS instrument on remand. ............................... 23 POINT TWO By failing to specify a governing framework for consideration of the COMPAS instrument at parole hearings, and failing to require an explanation for determinations that run counter to COMPAS’ validated results, the Board of Parole has failed to comply with Executive Law § 259-c (4). Neither the “Evans memo” nor the recently approved regulation “memorializing” it establish written procedures that “incorporate risk and needs principles to measure the rehabilitation of persons appearing before the board, the likelihood of success of such persons upon release, and assist members of the state board of parole in determining which inmates may be released to parole supervision.” In view of the Board’s delay and default, this Court should now establish minimum requirements for use of the COMPAS instrument at parole hearings. ........................................................................................................................................... 24 a. The Evans memo did not establish written procedures governing use of COMPAS. ............................................................................................................................ 26 b. The recently approved regulation fails to comply with Executive Law § 259-c (4) ............................................................................................................................. 27 c. Risk and Needs Principles. .................................................................................................. 29 d. Evidence-based practice requires that overrides based on professional judgment be explained. ....................................................................................................... 32 e. The Board’s opaque “procedure” is inconsistent with the legislative history and intent of Executive Law § 259-c (4). ............................................................. 34 1. Early 1970s - pre-guidelines parole release decision-making. ................................... 35 2. Origin of the 1978 Guidelines. ..................................................................................... 36 3. Parole Reform Act of 1977. .......................................................................................... 37 4. The Board’s new rule would turn the clock back forty years to the age of unregulated discretion in parole-release decision-making. ................................... 39 CONCLUSION The order of the Appellate Division should be modified by directing the Board of Parole to provide detailed reasons for determinations that run counter to COMPAS’ empirically-based results. ............................................................................................. 41 i TABLE OF AUTHORITIES Cases Matter of Boyea v. Board of Parole, Sup. Ct. Onondaga County, May 23, 2013 (Index No. 2013-1751) ........................................................................................... 31 Matter of Davis v. Travis, 292 A.D.2d 742 (3d Dept. 2002) ..................................................... 39 Matter of Garfield v. Evans, 108 A.D.3d 830 (3d Dept. 2013) .................................... 17, 27, 28 Matter of King v. New York State Board of Parole, 83 N.Y.2d 788 (1994) ........................... 22 Matter of Montane v. Evans, 981 N.Y.S.2d 866 (3d Dept. 2014) ............................................. 27 Matter of Morris v. NYS Dept. of Corrections and Community Supervision, 40 Misc.3d 226 (Sup. Ct. Columbia County 2013) ...................................................................... 20 Matter of Patterson v. Evans, 106 A.D.3d 1456 (4th Dept. 2013) ............................................. 29 Matter of Shark v. NYS Div. of Parole, 110 A.D.3d 1134 (3d Dept. 2013) ............................ 29 Matter of Siao-Pao v. Dennison, 51 A.D.3d 105 (1st Dept. 2008) ............................................ 29 Matter of Thomches v. Evans, 108 A.D.3d 124 (2d Dept. 2013) ............................................. 29 Parochial Bus Systems, Inc. v. Board of Education, 60 N.Y.2d 539 (1983) ............................ 23 People v. Johnson, 11 N.Y.3d 416 (2008) .................................................................................... 34 Statutes & Chapter Laws Correction Law § 71-a ...................................................................................................................... 14 Correction Law § 213 (former) ..................................................................................................... 35 Correction Law §213 (4) (former) ................................................................................................. 35 Executive Law § 259-b (7) ........................................................................................................ 13, 26 ii Executive Law § 259-c (4) ...................................................................................................... passim Executive Law § 259-i (2)(c) ............................................................................................... 4, 15, 28 Executive Law § 259-i (2)(c)(A) ............................................................................................... 13, 26 L.1977, chap. 904, § 1 ...................................................................................................................... 37 L.2009, ch. 56, Part N, § 2 ................................................................................................................. 8 L.2011, ch. 62, subpart A, § 38-b ...................................................................................................... 4 Regulations and Directives 9 NYCRR Part 8001 ........................................................................................................................... 1 9 NYCRR § 8001.3 ........................................................................................................................... 13 9 NYCRR§ 8001.3 (b) ....................................................................................................................... 5 9 NYCRR § 8001.3(b)(3) .................................................................................................................. 4 9 NYCRR § 8001.3(3)(c) ............................................................................................................ 6, 38 9 NYCRR § 8001.3(3)(d) ................................................................................................................ 12 9 NYCRR § 8006.4(2)(c) ................................................................................................................ 12 DOCCS Directive no. 8500 ...................................................................................................... 11, 33 Other Authorities Austin, The Proper and Improper use of Risk Assessment in Corrections, Federal Sentencing Reporter, Vol. 16, no. 3 (Feb., 2004) .......................................................... 33 Caher, Dismantling Parole - Parole Release Rates Plunge Under Pataki’s Tough Policy, New York Law Journal, Jan. 31., 2006 ................................................... 39 Caher, Parole Board’s Plan to Enact Risk Analysis Criticized, New York Law Journal, Feb. 3, 2014 ................................................................................................................ 21 iii A Framework for Evidence-Based Decision Making in Local Criminal Justice Systems, U.S. Dept. of Justice, National Institute of Corrections (2010) ................................................................................................................................................ 32 The Future of Sentencing in New York State: Recommendations for Reform ..................................................................................................................................... 7, 30 Genty, Changes to Parole Laws Signal Potentially Sweeping Policy Shift, New York Law Journal, Sept. 1, 2011 ..................................................................................... 39-40 Kamins, Restoring Fairness to the Parole System, Forty-Fourth Street Notes (April, 2007) .......................................................................................................................... 39 Karger, Powers of the New York Court of Appeals § 11: 66 (c) (3d ed.) ................................ 23 Implementing Evidence Based Policy and Practice in Community Corrections, U.S. Dept. of Justice, Natl. Inst. of Corrections, 2d ed. (Oct., 2009) ................................. 27-28 New York’s Broken Parole System, New York Times, Feb. 16, 2014 ...................................... 39 New York State COMPAS-Probation Risk Needs Assessment Study: Examining the Recidivism Scale’s Effectiveness and Predictive Accuracy (Division of Criminal Justice Services, 2012) ......................................................................... 10-11 New York State Special Commission on Attica. Attica: The Official Report of the New York State Special Committee on Attica (New York: Bantam Books 1972) ....................................................................................................................... 36 New York State Parole Board Decision-Making Guidelines: Their Flexibility and Their Value as a Policy-Setting Tool (New York State Div. of Parole - April, 1985) ...................................................................................................... 6, 38 Practitioner’s Guide to COMPAS (Northpointe) .......................................................... 10, 11, 34 Recommendations for Parole Reform in New York State ......................................................... 39 A Report on Board Initiatives to Strengthen Decisionmaking Policy and Practice, The Center for Effective Public Policy, Feb., 2008 ............................................. 32 iv A Report on Parole Reform by the Staff of the Codes Committee of the New York State Assembly (Nov., 1976) ...................................................................................... 36 Rosenthal & Warth, New York Still in Need of Parole Reform, Atticus, Spring 2014 ........................................................................................................................................ 40 Webster’s Third New International Dictionary ............................................................................ 29 1 Preliminary Statement This is an appeal from an order of the Appellate Division, Third Department, dated December 5, 2013, reversing a judgment of the Supreme Court, Albany County (McDonough, J.), which dismissed petitioner’s Article 78 challenge to the denial of his application for parole release. The Appellate Division granted petitioner partial relief – a remand for a new parole release hearing at which the Board must consider a risk and needs assessment instrument called COMPAS. But the Appellate Division did not rule on whether the Board had complied with a 2011 amendment to Executive Law § 259-c (4), requiring the Board to establish written procedures governing the use of COMPAS at parole hearings. Consequently, petitioner is aggrieved by the Appellate Division’s order because it granted him only part of the relief he is entitled to on appeal. This Court granted petitioner’s motion for leave to appeal on April 3, 2014. Summary of the Case and Argument This case is about the Board of Parole’s intransigence in response to a 2011 amendment to Executive Law § 259-c (4) that directed it to modernize its parole release decision-making process. The legislature directed the Board to replace the 1978 “guidelines for parole release decision-making” (9 NYCRR Part 8001) with a new evidence-based paradigm for determining which inmates are suitable for release on parole. By October 1, 2011, the Board was required to establish new written procedures incorporating “risk and needs principles” to “measure” an inmate’s “rehabilitation” and “likelihood of success . . . upon release.” 2 The Board began using the COMPAS risk and needs assessment instrument in early 2012. But board members apparently could not agree on procedures governing appropriate use of the instrument, and so the Board never issued the required written procedures. Instead, in response to mounting legal challenges in 2012 and 2013, the Board claimed that an October 5, 2011 inter-office memo represented its procedures. But that memo was written before the Board had even decided to use COMPAS, and it included no procedures governing use of the instrument. Petitioner, Jorge Linares, challenged his November, 2011 parole release denial on the ground that the Board had failed to consider a COMPAS instrument and had failed to establish the required written procedures. The Supreme Court dismissed his petition but, on December 5, 2013, the Appellate Division reversed and remanded for a new parole release hearing at which the COMPAS instrument would be considered. But the Appellate Division did not rule on whether the Board had established the required written procedures for use of COMPAS. A few weeks after the Appellate Division’s decision, the Board filed a proposed rule purporting to comply with the 2011 law. Following the notice-and-comment period, the Board approved the rule at its April, 2014 business meeting. It will become effective in June, 2014. But the proposed rule is structurally inconsistent with the enabling legislation. It treats COMPAS risk and needs assessment results as merely additional, undifferentiated information for board members’ consideration, rather than as a new framework for assessing rehabilitation. Under the proposed rule, 3 Board members are not required to give COMPAS results any weight, and are under no obligation to explain their rejection of empirically-based assessments of an inmate’s likelihood of reoffending. Risk assessment instruments like COMPAS are designed to work in tandem with professional judgment and overrides of their empirically-based results are normal and expected. But it is a first principle of evidence-based decision-making that a validated risk assessment instrument should be the starting point for all risk evaluations. Such instruments provide a structured, consistently applied framework for all decision-making and a baseline from which overrides grounded in professional judgment can be applied and documented. By failing to establish any meaningful procedures governing use of COMPAS at parole hearings, the Board has failed to implement Executive Law § 259-c (4). The legislative history of the statute reveals it is designed to promote transparency, consistency and fair treatment of similarly- situated parole candidates. The Board’s opaque new rule completely undermines these objectives. In view of the Board’s delay and default in meeting its statutory obligations, this Court should now establish minimum requirements for use of COMPAS at parole hearings. Board members should be required to give detailed reasons for determinations that run counter to COMPAS’ empirically-based results. Statement of Facts Petitioner, Jorge Linares, became eligible for parole release consideration for the first time in November, 2011, having served the minimum of a 16 2/3 – 40 year 4 aggregate sentence imposed on 1996 convictions in the Suffolk County Court. Approximately one-month earlier, Linares met with a facility parole officer at Mohawk Correctional Facility in anticipation of the parole board hearing. Linares asked whether his application for parole would be evaluated under recent amendments to the Executive Law (17). Linares was referring to a 2011 amendment to Executive Law § 259-c (4) that required the Board of Parole to replace the “guidelines for parole release decision- making,” which had been in effect since 1978, with a new framework for evaluating candidates for discretionary parole release. Enacted as part of the 2011-2012 budget (L.2011, ch. 62, subpart A, § 38-b), the amendment directed the Board to replace the 1978 guidelines with “written procedures” governing evaluation of the statutory criteria for parole release decision-making: i.e., whether there is a “reasonable probability that, if such inmate is released, he will live and remain at liberty without violating the law, and that his release is not incompatible with the welfare of society and will not so deprecate the seriousness of his crime as to undermine respect for law.” Executive Law § 259-i (2)(c). The 1978 guidelines for parole release decision-making The 1978 parole release guidelines set forth recommended prison time ranges (expressed in months) before an inmate’s first release to parole supervision. The guidelines [9 NYCRR § 8001.3 (b)(3)] applied to two functions exercised by the Board: 1) the Board’s authority to decide whether inmates should be released on parole, and 2) the Board's long-since moribund authority to set a minimum period of imprisonment (MPI) for certain inmates whose crimes were committed prior to September 1, 1980.1 The guidelines were expressed in a grid with a horizontal axis for an inmate's prior criminal history score (0- 1low, 2- 5 moderate, 6- 11 serious), and a vertical axis for an offense severity score (1least severe through 8-9 most severe). At the intersection of these two scores, the grid established an inmate's guidelines time range, which represented the "policy of the Board concerning the customary total time to be served before release." 9 NYCRR§ 8001.3(b). 2 1 In 1980, the legislature required courts to set minimum terms for all indeterminate sentences. 2 As last revised in 1985, the guidelines time matrix provided as follows: GUIDELINES FOR PAROLE BOARD DECISIONMAKING Prior criminal history score Offense severity score 0-1 (Low) 2-5 (Moderate) 6-11 (Serious) 8-9 most severe Specific ranges are not given due to the limited number of cases and the extreme variation possible within the category. 7 30-60. 40-72 60-96 months months months 6 22-40 30-50 46-72 months months months 4-5 16-30 20-38 30-54 months months months 2-3 14-24. 18-30 20-36 months months months 1 12-18 14-24 16-28 least severe months months months 5 6 Board members were not required to adhere to the guidelines; regulations acknowledged that “[m]itigating or aggravating factors may result in decisions above or below the guidelines.” But commissioners were required to explain any decision rendered outside the guidelines. The Board’s regulations specified, “In any case where the decision rendered is outside the guidelines, the detailed reason for such decision, including the fact or factors relied on, shall be provided to the inmate in writing.” See 9 NYCRR § 8001.3(3)(c). In 1985, the last time the guidelines were revised, the Board reported that 70% of its first-time release decisions were consistent with the revised guidelines scheme. Overall, the Board reported, similarly-situated inmates were being treated similarly, a major goal of the 1977 enabling legislation. See New York State Parole Board Decision-Making Guidelines: Their Flexibility and Their Value as a Policy-Setting Tool (April, 1985) (369). The advent of evidence-based practice in criminal justice The guidelines time ranges were based on two unchangeable factors, the seriousness of the crime and inmate’s prior criminal record. Because the prevailing view in the 1970s and 1980s was that an inmate’s risk of reoffending could not be reliably measured, rehabilitation was not factored into the guidelines scheme. But views began to change in the 1990s with the advent of “evidence-based” practices in the criminal justice system. Risk and needs assessment instruments were developed to measure an offender’s statistical likelihood of re-offending, measurements that could help criminal justice professionals make better and more informed decisions about 7 such matters as pre-trial release, sentencing, probation and parole supervision, and parole release. Risk and needs assessment instruments identify static (unchangeable) and dynamic (changeable) factors empirically shown through longitudinal studies to be positively or negatively correlated with reoffending. By grouping these factors in meaningful ways and subjecting them to weighted scoring systems based on how strongly each factor correlates with recidivism, risk and needs assessment instruments can achieve “validated” results. That is, these instruments can be shown to accurately measure what they purport to measure. Risk and needs assessment instruments can accurately and consistently differentiate between high, medium and low risk offenders in a way that is superior to judgments based on unstructured, intuitive assessments about the risk of re-offense. New York State Commission on Sentencing Reform recommendations In 2009, the New York State Commission on Sentencing Reform recommended adoption of evidence-based practices throughout the criminal justice system, including the parole release decision-making process. See The Future of Sentencing in New York State: Recommendations for Reform.3 The Commission stressed that the “cornerstone of evidence-based practices is the use of a validated risk and needs assessment instrument,” which can “significantly improve upon the predictive accuracy” of judgments about who is or is not likely to reoffend (205). The 3 http://www.criminaljustice.ny.gov/pio/csr_report2-2009.pdf (last visited May 27, 2014) 8 Commission recommended the “Division of Parole . . . use a risk and needs instrument to help determine which offenders are appropriate for release into the community and which offenders continue to pose a significant threat to public safety” (209). However, the Sentencing Commission observed, “this [recommendation] may require a change in the statute regarding how the Board of Parole is authorized to make release decisions.” Id. 2009 amendment to Executive Law § 259-c (4) A few months later the legislature responded to the Sentencing Commission by granting the Board of Parole discretion to revise the 1978 parole release guidelines to incorporate the use of a risk and needs assessment instrument. (L.2009, ch. 56, Part N, § 2) The legislature amended Executive Law § 259-c (4) to provide that “such written guidelines may consider the use of a risk and needs assessment instrument to assist members of the state board of parole in determining which inmates may be released to supervision.” But the Board failed to act on the legislature’s invitation and did not revise its guidelines. 2011 amendment to Executive Law § 259-c (4) The legislature took up the Sentencing Commission’s recommendation again in a 2011 budget bill, but now directed the Board to replace the 1978 guidelines. The legislature instructed the Board to establish “written procedures” for a new evidence- based paradigm for evaluating whether an inmate was suitable for parole release: i.e., whether there was a reasonable probability he or she would “live and remain at liberty 9 without violating the law.” The written procedures, the 2011 amendment specified, “shall incorporate risk and needs principles to measure the rehabilitation of persons appearing before the board, the likelihood of success of such persons upon release, and assist members of the state board of parole in determining which inmates may be released to parole supervision.” Executive Law § 259-c (4). The new scheme was to be in place by October 1, 2011. The legislature gave the Board six months to revamp its parole release regulations for the first time in more than thirty years. Linares’ November, 2011 parole release hearing When Jorge Linares inquired about this new parole release decision-making scheme in October, 2011, the facility parole officer told him the Board would not be ready to apply the new law at his upcoming parole hearing. He explained the Board would not be ready to do so until “possibly January 2012” (17). Instead, Linares’ application would apparently be governed by the 33 year-old – and now legislatively countermanded – guidelines for parole release decision-making. Consequently, the parole officer did not generate a risk and needs assessment instrument called COMPAS (Correctional Offender Management Profiling for Alternative Sanctions), which was then being developed for possible use in the parole release decision-making process. Instead, he prepared an inmate status report that referenced the old guidelines (17, 63). 10 COMPAS risk and needs assessment instrument COMPAS is a risk and needs assessment instrument developed by the Northpointe Institute for Public Management. It is in widespread use in New York’s criminal justice system in different formats keyed to the offender populations under review. It is employed by probation departments and the Department of Corrections and Community Supervision to determine supervision levels for probationers and persons released on parole and post-release supervision. COMPAS offers actuarially- based estimates, expressed in decile scores of 1 (lowest) through 10 (highest), of an offender’s 1.) risk of felony violence; 2.) risk of re-arrest; and 3.) risk of absconding from supervision. Within an adult offender inmate population, COMPAS also provides a ten-point scale of an inmate’s criminal involvement, history of violence and prison misbehavior. COMPAS also assesses criminogenic needs, evaluating an inmate’s risk of encountering circumstances conducive to criminality, such as substance abuse, unemployment, low family support, or other negative social conditions or attitudes that might interfere with his or her successful re-entry. COMPAS has been validated in studies involving probation and parole offender populations in New York State, its results falling in the “moderate to strong range of predictive accuracy.” See Practitioner’s Guide to COMPAS, p. 15;4 see also New York State COMPAS-Probation Risk Needs Assessment Study: Examining the Recidivism 4 http://www.northpointeinc.com/files/technical_documents/FieldGuide2_012813.pdf (last visited May 27, 2014) 11 Scale’s Effectiveness and Predictive Accuracy (Division of Criminal Justice Services, 2012).5 Importantly, COMPAS is not designed to replace the exercise of professional judgment. The instrument is based on statistical probabilities that may be overridden by professional evaluators based on articulated reasons. The developers of COMPAS report that “due to either aggravating or mitigating circumstances not detected by COMPAS one may expect override rates of from 8% to 15%.” See Practitioner’s Guide to COMPAS, p. 27. The Office of Probation and Correctional Alternatives and the Department of Corrections and Community Supervision have established procedures governing proper use of COMPAS. Evaluators may override the COMPAS risk assessment when they provide a “rationale” for doing so. See DOCCS Directive no. 8500 (COMPAS Assessments/Case Plan) (375). Parole denied As noted, no COMPAS risk and needs assessment instrument was prepared for Linares’ November, 2011 hearing. The panel denied his application for parole release and directed that he be held two years before reconsideration (84). Linares’ inmate status report referred to the old guidelines time range of 16 – 30 months applicable to his felony crime of conviction and first-time offender status (63). By 2011, Linares, then 49, had already served 196 months in prison. But the Board did not give a 5 http://www.criminaljustice.ny.gov/crimnet/ojsa/opca/compas_probation_report_2012.pdf (last visited May 27, 2014) 12 reason for holding him beyond the applicable guidelines time range, as required by its still un-repealed regulations. See 9 NYCRR § 8001.3 (3)(d). This wasn’t an oversight. The Board had not updated the guidelines in more than twenty-five years. The time ranges were considered outdated and board members had long ago stopped using or referring to them. Administrative appeal Linares perfected an administrative appeal raising several claims of error, including that a new hearing was required because the Board failed to comply with Executive Law § 259-c (4) (98-100). The Board of Parole’s appeals unit failed to respond within four months, thereby triggering an administrative exhaustion rule and Linares’ option to seek judicial review. See 9 NYCRR § 8006.4(2)(c). Article 78 Petition In a petition filed on September 7, 2012, Linares sought Article 78 relief in Supreme Court, Albany County, raising the Executive Law § 259-c (4) claim along with several others (5). The Board opposed in an answer dated March 29, 2013. Conceding a COMPAS risk and needs assessment instrument was not considered at Linares’ parole hearing, the Board argued no error had occurred because the hearing nevertheless complied with its “written procedures” promulgated in accordance with Executive Law § 259-c (4). The Board insisted that an inter-office memo written by then Board chair Andrea Evans, dated October 5, 2011 (the “Evans memo”), represented the required “written procedures” (41-44). However, the Board did not 13 explain how an inter-office memo qualified as action by the Board of Parole, which requires a “majority of the board [members] . . . to constitute a quorum for the transaction of all business of the board” [Executive Law § 259-b (7)]. Nor did the Board claim the Evans memo had ever been “adopted” by the Board, as required by Executive Law § 259-i (2)(c)(A). The Board similarly failed to explain how the Evans memo superseded the now legislatively countermanded “guidelines for parole release decision-making,” which remained un-repealed in the Official Compilation of Codes, Rules and Regulations of the State of New York. See 9 NYCRR § 8001.3 The Evans memo Despite the Board’s dubious claims about its legal status, the Evans memo was simply a memorandum describing the new law and possible interim responses parole commissioners might take until the Board was ready to formally adopt new written procedures to replace the 1978 guidelines. In fact, in testimony before the Assembly Committee on Correction one month after writing the memo, Andrea Evans restated the substance of the memo and described it as “an interim measure” pending formal adoption of written procedures, which had not yet occurred (310). The two-page memo advised Board members that Executive Law § 259-c (4) had been amended to require written procedures that “incorporate risk and needs principles” to measure an inmate’s rehabilitation, likelihood of success, and to assist Board members to determine which inmates may be released (281-282). Referring to a case management document called a Transition Accountability Plan (TAP) then in 14 development by the Department of Corrections and Community Supervision, Chairwoman Evans wrote that the TAP instrument “incorporates risk and needs principles” and “will provide a meaningful measurement of an inmate’s rehabilitation” (281).6 Noting that Board members had been “afforded training in July, 2011 in the use of a TAP instrument where it exists,” she advised, “as we proceed, when staff have prepared a TAP instrument for a parole-eligible inmate, you are to use that document when making your parole release decisions.” Id. The memo made only brief mention of COMPAS, reminding board members they had been “afforded training” on it “to understand the interplay between that instrument and the TAP instrument, as well as understanding what each of the risk levels mean.” Id. The memo was notably silent on any anticipated future use of the COMPAS instrument itself. When a TAP instrument was not available, the memo suggested, board members could simply proceed without one (282). The chairwoman pointed out that the legislatively-prescribed factors that board members must review at parole hearings 6 Development of the TAP case management plan was part of the 2011 budget bill. Correction Law § 71-a was added to provide: “Transitional Accountability Plan. Upon admission of an inmate committed to the custody of the department . . . the department shall develop a transitional accountability plan. Such plan shall be a comprehensive, dynamic and individualized case management plan based on the programming and treatment needs of the inmate. The purpose of such a plan shall be to promote the rehabilitation of the inmate and their successful and productive reentry and reintegration into society upon release. To that end, such plan shall be used to prioritize programming and treatment services for the inmate during incarceration and any period of community supervision. The commissioner may consult with the office of mental health, the office of alcoholism and substance abuse services, the board of parole, the department of health, and other appropriate agencies in the development of transitional case management plans.” 15 had not changed (e.g., institutional record, release plans, statements of crime victims, seriousness of the offense, pre-sentence report, inmate’s prior criminal record).7 Observing that “there is a strong rehabilitative component” in these factors, she suggested Board members could comply with the 2011 law by “ascertain[ing] what steps an inmate has taken toward their [sic] rehabilitation and the likelihood of their success once released to parole supervision.” Board members could do so, she explained, by asking about an inmate’s rehabilitative efforts and release plans at parole hearings as they had always done (282). 7 In full, these factors are: (i) the institutional record including program goals and accomplishments, academic achievements, vocational education, training or work assignments, therapy and interactions with staff and inmates; (ii) performance, if any, as a participant in a temporary release program; (iii) release plans including community resources, employment, education and training and support services available to the inmate; (iv) any deportation order issued by the federal government against the inmate while in the custody of the department and any recommendation regarding deportation made by the commissioner of the department pursuant to section one hundred forty-seven of the correction law; (v) any statement made to the board by the crime victim or the victim’s representative, where the crime victim is deceased or is mentally or physically incapacitated; (vi) the length of the determinate sentence to which the inmate would be subject had he or she received a sentence pursuant to section 70.70 or section 70.71 of the penal law for a felony defined in article two hundred twenty or article two hundred twenty-one of the penal law; (vii) the seriousness of the offense with due consideration to the type of sentence, length of sentence and recommendations of the sentencing court, the district attorney, the attorney for the inmate, the presentence probation report as well as consideration of any mitigating and aggravating factors, and activities following arrest prior to confinement; and (viii) prior criminal record, including the nature and pattern of offenses, adjustment to any previous probation or parole supervision and institutional confinement. N.Y. Exec. Law § 259-i (2)(c). 16 The tentative character of the memo was explained by Chairwoman Evans on November 10, 2011. Testifying before the Assembly Correction Committee, she explained the Board was not yet ready to use TAP or COMPAS at parole hearings: Currently, the use of these instruments is being piloted in three . . . correctional facilities for the purpose of establishing appropriate conditions of supervision. When the pilot phase of these instruments is concluded, the Board looks forward to their use when assessing the appropriateness of inmate’s release to community supervision. (310). The Board implements COMPAS but fails to establish written procedures governing its use. The Board began widespread deployment of COMPAS in early 2012. But more than one year later, it had still not adopted the statutorily-required written procedures. When the Board answered Linares’ Article 78 petition in March, 2013, it insisted the Evans memo – informal and tentative as it was – “now serves as the written procedures of the Board pursuant to the amended version of Executive Law § 259-c (4), and is in full compliance with all statutory mandates” (45). Addressing the merits of his petition, the Board argued that the amendment to Executive Law § 259-c (4) “did not specify the means by which” risk and needs principles were to be incorporated into parole release decision-making, and that the Board had complied with the new law even though no risk and needs assessment instrument had been available at Linares’ hearing (41-45). 17 Supreme Court’s decision The Supreme Court (McDonough, J.) agreed, ruling that the Board’s “promulgation of the October 5, 2011 Memorandum from its Chairwoman, as well as the record as a whole” satisfied the requirements of Executive Law 259-c (4). Rejecting Linares’ other claims for relief, the court dismissed his pro se petition on June 17, 2013 (216-219). Linares appealed. Matter of Garfield v. Evans, 108 A.D.3d 830 (3d Dept. 2013) A few weeks later on July 3, 2013, the Appellate Division, Third Department, held that an inmate-petitioner was entitled to a new parole release hearing because the Board had failed to use COMPAS at an October, 2011 parole release hearing (i.e., one month prior to Linares’ hearing). The ruling was based on the Board’s concession that it had been using COMPAS at parole hearings since early 2012: Significantly, Executive Law § 259-c (4) requires that the Board "establish written procedures for its use in making parole decisions as required by law," and the Board acknowledges that the statute requires it to incorporate risk and needs principles into its decision-making process. According to the record, the Board was trained in the use of the COMPAS instrument prior to petitioner’s hearing. Moreover, the Board acknowledges that it has used the COMPAS instrument since February 2012 and will use it for petitioner's next appearance. Under these circumstances, we find no justification for the Board's failure to use the COMPAS instrument at petitioner's October 2011 hearing. Accordingly, we agree with petitioner that he is entitled to a new hearing. Matter of Garfield v. Evans, 108 A.D.3d 830 (3d Dept. 2013) 18 Appellate Division proceedings In his August, 2013 pro se brief to the Third Department, Linares claimed Garfield entitled him to a new hearing (243-244). He additionally argued the Garfield error was compounded by the Board’s continued failure to “promulgate or file the written procedures” required by Executive Law § 259-c (4). Linares cited lower court decisions that held the Evans memo did not comply with Executive Law § 259-c (4) in form or in substance (244-245). Acknowledging a split in the lower courts, Linares argued “it is imperative that this Court resolve the controversy” (251). In light of Garfield, the Board advised the Appellate Division on October 18, 2013 that it would not submit a brief or defend the lower court’s judgment. The Board noted that Linares “is scheduled to reappear before the Board on November 5, 2013, at which time a COMPAS assessment will be available for the Board’s use.” Hence, the Board concluded, Linares’ “appeal will not only become moot at that time, but the error identified by the Court in Garfield will be corrected” (248). Linares advised the Appellate Division that he had already requested a six- month postponement of his regularly scheduled parole hearing8, and implored the court to decide the merits of his appeal. Even though the Board conceded it must now use COMPAS, Linares argued, there remained the “unanswered question” of whether the Board is “still violating the legislative intent in Executive Law § 259-c (4) 8 The parole release hearing was, in fact, postponed to July, 2014 (253). Petitioner plans to seek further adjournments pending resolution of this appeal. 19 by failing to ‘promulgate and file’ new procedures for making discretionary release determinations” (251). In other words, Linares argued, the Garfield error that occurred at his 2011 parole hearing wouldn’t be cured by later use of COMPAS unless the Board first issued the mandated written procedures directing how the instrument was to be used. Appellate Division order Citing Garfield, the Appellate Division reversed on December 5, 2013, and ordered the Board to give Linares a new hearing. Matter of Linares v. Evans, 112 A.D.3d 1056 (3d Dept. 2013) (3-4). But the court did not decide the related aspect of Linares’ Executive Law § 259-c (4) claim: that the Board had failed to adopt written procedures governing use of COMPAS on remand. Motion for leave to appeal to the Court of Appeals Linares moved for leave to appeal to this Court, seeking review of - among other claims - the Appellate Division’s failure to provide meaningful relief by directing the Board to establish written procedures in accordance with Executive Law § 259-c (4) (262-264). The Board opposed, maintaining that Linares was not aggrieved by the Appellate Division’s order (300). On April 3, 2014, this Court granted Linares’ motion for leave to appeal (2). Linares then acquired pro bono counsel. Proposed rulemaking In possible reaction to lower court holdings that the written procedures required by Executive Law § 259-c (4) were subject to the rule-making provisions of 20 the State Administrative Procedure Act [see, e.g., Matter of Morris v. NYS Dept. of Corrections and Community Supervision, 40 Misc.3d 226 (Sup. Ct. Columbia County 2013)], the Board filed a proposed rule with the Secretary of State. It was published in the December 18, 2013 New York State Register (353).9 While the Board described the proposed rule as “memorialization” of the Evans memo, in substance it effectively rewrites the 2011 legislation. The proposed regulation recasts the “risk and needs principles” requirement as simply additional information to be reviewed by the Board under Executive Law § 259-i (2)(c)(A), rather than as the legislatively-designed framework for decision-making under Executive Law § 259-c (4), in replacement of the 1978 guidelines. There was nothing subtle about this recasting effort. The proposed regulation repeals the old guidelines and simply adds two non-statutory items to the list of eight legislatively-prescribed factors the Board must consider. In abbreviated form, the statutory list includes: 1. institutional record; 2. performance in temporary release programs; 3. release plans; 4. any deportation order; 5. crime victim’s statement; 6. the length of alternative available sentences in certain drug cases; 7. the seriousness of the crime; recommendations of the sentencing court, district attorney and defense counsel; pre-sentence report; and 8. inmate’s prior criminal record. 9 http://docs.dos.ny.gov/info/register/2013/dec18/pdf/rulemaking.pdf (last visited May 27, 2014) 21 Subdividing the statutory list into ten factors, the proposed rule tacks on two non-statutory “factors:” 11. the most current risk and needs assessment that may have been prepared by the Department of Corrections and Community Supervision; and 12. the most current case plan that may have been prepared by the Department of Corrections and Community Supervision pursuant to section seventy-one-a of the Correction Law. (359) (emphasis added). Commentators criticize the proposed rule as non-compliant with the enabling legislation Legislators10, bar associations, re-entry organizations, and even a former member of the Board of Parole sharply criticized the proposed rule during the notice- and-comment period. See Caher, Parole Board’s Plan to Enact Risk Analysis Criticized, New York Law Journal, Feb. 3, 2014. Critics stressed that the proposed rule is structurally inconsistent with the enabling legislation because it treats the risk and needs assessment instrument as undifferentiated material for review, rather than as a uniform, consistent framework for evaluating whether inmates were likely to “live and remain at liberty without violating the law.” The proposed rule, critics pointed out, fails to provide any guidance about how COMPAS should be used. Rather, it leaves that question to the unfettered discretion of individual board members. While board members must consider the legislatively- 10 Assemblyman O’Donnell, Chair of the NYS Assembly Corrections Committee, and Assemblyman Zebrowski, Chair of the NYS Assembly Administrative Regulations Review Commission 22 prescribed sources of information listed in Executive Law § 259-i (2)(c), they may give these items any weight they choose, and need not address each item in written decisions denying parole. Matter of King v. New York State Board of Parole, 83 N.Y.2d 788, 791 (1994) (Board “need not expressly discuss each of these [items] in its determination”). Thus, under the proposed regulation, board members would be free to discount or even ignore an evidence-based risk assessment and continue to make wholly intuitive judgments about an inmate’s risk of reoffending, without revealing they are doing so or explaining why. In this way, commentators argued, the proposed rule fails to “incorporate risk and needs principles” and establish a rational and consistent framework for assessing rehabilitation in the parole release decision-making process.11 Without discussion, the Board approved this much-criticized rule at its April, 2014 business meeting and will publish formal notice of its adoption sometime in June, 2014. 11 The comments are collected here: http://nylawyer.nylj.com/adgifs/decisions14/020314comments.pdf and here http://www.correctionalassociation.org/resource/public-comments-in-support-of-parole- reform (last visited May 27, 2014) 23 ARGUMENT POINT ONE Petitioner is aggrieved because the Appellate Division failed to accord him all the relief he is entitled to: a direction that the Board comply with Executive Law § 259-c (4) by establishing legislatively required written procedures incorporating risk and needs principles governing use of the COMPAS instrument on remand. The Appellate Division granted petitioner partial relief. It ordered the Board to give him a new parole release hearing at which a COMPAS risk and needs assessment instrument must be considered. But the court did not rule on the related part of Linares’ Executive Law § 259-c (4) claim – that the Board has failed to establish required written procedures governing use of COMPAS on remand. Since neither the October 5, 2011 “Evans memo” nor the Board’s recently approved regulation complies with Executive Law § 259-c (4) (see Point Two, infra), the Appellate Division’s remedial order is inadequate to assure that Linares will now be provided a proper parole release hearing under the 2011 amended statute. Accordingly, Linares is aggrieved by the Appellate Division’s order. See Karger, Powers of the New York Court of Appeals § 11: 66 (c), p. 403-404 (3d ed.) (“[A] party is generally considered to be aggrieved, for the purpose of appeal, if the order of the Appellate Division grants him only part of the relief he requested.”); Parochial Bus Systems, Inc. v. Board of Education, 60 N.Y.2d 539, 544-545 (1983). 24 POINT TWO By failing to specify a governing framework for consideration of the COMPAS instrument at parole hearings, and failing to require an explanation for determinations that run counter to COMPAS’ validated results, the Board of Parole has failed to comply with Executive Law § 259-c (4). Neither the “Evans memo” nor the recently approved regulation “memorializing” it establish written procedures that “incorporate risk and needs principles to measure the rehabilitation of persons appearing before the board, the likelihood of success of such persons upon release, and assist members of the state board of parole in determining which inmates may be released to parole supervision.” In view of the Board’s delay and default, this Court should now establish minimum requirements for use of the COMPAS instrument at parole hearings. In 2011 the legislature directed the Board of Parole to revise its practice and procedure for deciding which inmates should be released to parole supervision. In place of the “guidelines for parole release decision-making” - regulations in effect since 1978 governing the Board’s parole release function - the legislature directed the Board to “establish written procedures for its use in making parole decisions.” “Such written procedures,” the legislature directed, “shall incorporate risk and needs principles to measure the rehabilitation of persons appearing before the board, the likelihood of success of such persons upon release, and assist members of the state board of parole in determining which inmates may be released to parole supervision” (emphasis added). The effective date for this part of the law was October 1, 2011. After initial uncertainty about how it would proceed, the Board chose to use the 25 COMPAS risk and needs assessment instrument, and by early 2012 began providing these actuarial instruments to board members at parole release hearings. However, the Board apparently could not reach consensus on procedures governing its use of COMPAS. More than two years elapsed before the Board filed a proposed rule to finally repeal the old guidelines for parole release decision-making. The proposed rule is entirely non-committal, specifying only that board members consider a COMPAS instrument that “may” have been prepared. Board members need not give COMPAS results any weight, and need not explain when professional judgment causes them to discount or override its risk estimates, an otherwise routine protocol for proper use of evidence-based risk assessment instruments. Risk assessment instruments are designed to work in tandem with professional judgment and overrides are normal and expected. But it is a first principle of evidence-based decision-making that a validated risk assessment instrument should be the starting point for all risk evaluations. The instrument provides a structured, consistently-applied framework for all decision-making and a baseline from which overrides grounded in professional judgment can be applied and documented. By failing to establish meaningful procedures governing use of COMPAS at parole release hearings, the Board has failed to implement Executive Law § 259-c (4). Moreover, conferring unbridled discretion on individual board members to use COMPAS in any manner they see fit is inconsistent with the legislative history and intent of Executive Law § 259-c (4). First enacted in 1977 to usher in the guidelines 26 for parole-release decision-making, the section was intended to promote transparency, consistency, and fair treatment of similarly-situated inmates at parole release hearings, values completely missing in the Board’s feckless responses to the 2011 parole reform law thus far. a. The Evans memo did not establish written procedures governing use of COMPAS. The Board has approved its new rule and it will go into effect in June, 2014 (i.e., after this brief has been filed). Once this occurs, the Evans memo will no longer matter, except insofar as the Board’s dubious characterization of it as the required “written procedures” betrays the agency’s long, failed effort to reach consensus on how to structure its use of COMPAS. The Evans memo is perfectly understandable when considered in light of its original limited purpose. The memo was written to advise Board members how to proceed after the effective date of the new law but before the agency was fully prepared to implement COMPAS. But once the Board implemented COMPAS in early 2012 it disingenuously restyled the Evans memo as its “written procedures” under Executive Law § 259-c (4). The two-page document was simply a memo written by the chairwoman; it clearly did not reflect the required “adoption” of written procedures by a quorum of the Board [Executive Law §§ 259-b (7); 259-i (2)(c)(A)]. Chairwoman Evans’ contemporaneous public statements made clear the memo was an “interim measure” 27 pending future adoption of the required written procedures.12 The memo had to be an interim measure because until the Board amended its regulations parole hearings would continue to be governed by the 1978 guidelines, which the legislature had unquestionably directed the Board to repeal. As for its content, the memo did not even mention use of COMPAS; it focused exclusively on the Board’s anticipated use of the TAP instrument. The memo simply advised Board members to use the TAP instrument when it was available, nothing more. And the memo’s expedient suggestion that parole release hearings could go forward after October 1, 2011without a risk and needs assessment instrument was later rejected in Matter of Garfield v. Evans, 108 A.D.3d 830 (3d Dept. 2013), a holding the Board apparently now accepts.13 b. The recently approved regulation fails to comply with Executive Law § 259-c (4). The proposed regulation likewise fails to qualify as required written procedures. “Implementing evidence based policy and practice is not a simple task; it requires a fundamental change in the way [an agency] does business,” and a shift away from “that’s always how we’ve done it” attitudes on the part of decision-makers. See Implementing Evidence Based Policy and Practice in Community Corrections, U.S. 12 Testifying before the Assembly Standing Committee on Correction on November 10, 2011, Chairwoman Evans discussed the COMPAS and TAP instruments and stated “the Board’s written procedures will call for the use and careful consideration of these documents” (emphasis added) (310). 13 The Third Department held the Evans memo complied with Executive Law §259-c (4) in Matter of Montane v. Evans, 981 N.Y.S.2d 866 (3d Dept. 2014), leave granted, May 13, 2014. Montane’s holding is wrong for the reasons indicated herein. 28 Dept. of Justice, Natl. Inst. of Corrections, 2d ed. (Oct., 2009), p. xv. Regrettably, this statement’s truth is borne out in the Board’s proposed regulation - filed more than two years late. The Board was evidently unable to reach consensus about how to meaningfully structure its use of COMPAS. Effectively giving up, the Board has now proposed a rule that authorizes board members to use COMPAS any way they wish. First, the rule’s text permits the possibility that parole hearings may be conducted without use of COMPAS or TAP. It provides that Board members shall consider the “most current risk and needs assessment” (COMPAS) and “case plan” (TAP) that “may have been prepared.” This is a puzzling word choice because the Board apparently accepts Garfield’s holding that COMPAS must now be used at all parole release hearings. To the extent the Board’s new rule suggests otherwise, it is deficient. More troubling, the rule is structurally at-odds with the enabling legislation. The regulation incongruously lists the risk and needs assessment and case plan as “factors” board members must consider along with the eight statutorily-prescribed factors listed in Executive Law § 259-i (2)(c). But the legislature didn’t instruct the Board to consider additional factors; it directed the Board to adopt written procedures incorporating risk and needs principles to “measure” an inmate’s “rehabilitation” and “likelihood of success . . . upon release.” The written procedures were intended to replace the long-standing organizational framework for the Board’s parole release decision-making authority – the 1978 guidelines. 29 By mischaracterizing COMPAS as a “factor” to be reviewed along with many others, the proposed rule nullifies this risk assessment instrument as the agency-wide organizing scheme for assessing rehabilitation the legislature intended it to be. Under the rule, Board members are free to give COMPAS as much or as little weight as they wish, with no obligation to explain their rejection of evidence-based risk assessments, even ones on the lowest end of COMPAS’ ten-point risk scale.14 Board members could continue to conclude that an inmate is “likely” to reoffend based on intuition alone, a decision-making process that is substantially inferior to ones that incorporate evidence-based practices. No matter what board members did with COMPAS, they would never be in violation of agency procedure. This is not a procedure: “a series of steps followed in a regular orderly definite way.”15 It is carte blanche for board members to do whatever they wish. c. Risk and Needs Principles While the legislature delegated authority to the Board to write the procedures, it provided some guidance about their required content. Executive Law § 259-c (4) directs that the procedures “shall incorporate risk and needs principles.” Therefore, the written procedures must conform to basic principles governing use of risk assessment instruments in evidence-based decision-making. The Board’s selection of 14 The Board is “not required to articulate every statutory factor it considered nor to give each factor equal weight.” Matter of Shark v. NYS Div. of Parole, 110 A.D.3d 1134 (3d Dept. 2013); Matter of Thomches v. Evans, 108 A.D.3d 124 (2d Dept. 2013); Matter of Patterson v. Evans, 106 A.D.3d 1456 (4th Dept. 2013); Matter of Siao-Pao v. Dennison, 51 A.D.3d 105 (1st Dept. 2008). 15 Webster’s Third New International Dictionary 30 COMPAS was an important first step because “[t]he cornerstone of evidence-based practice is the use of a validated risk and needs assessment instrument.” The Future of Sentencing in New York State: Recommendations for Reform, p. 135 (205). But choosing a risk assessment instrument alone is not enough. The agency’s procedures must guide its use by decision-makers. As the Sentencing Commission pointed out, “While practitioners intuitively understand that [some factors] . . . will affect the offender’s likelihood of recidivism, risk instruments significantly improve upon the predictive accuracy of that assessment” (206-207). But no significant improvements are possible when decision-makers don’t use available risk assessment instruments, or use them in undisclosed, widely divergent, or idiosyncratic ways. Evidence-based decision-making is not designed to replace professional judgment; it is intended to improve it by helping to eliminate the personal biases of decision-makers and by providing baseline assessments of all parole-eligible inmates under the same empirically-driven criteria. It is a basic principle of evidence-based practice that aggravating and mitigating circumstances will sometimes call for a decision to be made outside the framework of the risk assessment instrument. But when a decision is made outside the framework, decision-makers must give a reason for the override. Otherwise, there can be no assurance the risk assessment instrument 31 is actually being considered, and no meaningful way to review decisions for rationality on administrative appeal or in an Article 78 proceeding.16 There are many ways to incorporate risk assessment instruments and professional judgment into a coherent and transparent decision-making procedure. The Office of Probation and Correctional Alternatives developed a COMPAS-based matrix to structure decisions about the supervision levels of probationers. The matrix includes a vertical axis for the COMPAS risk of re-offense score, and a horizontal axis for the risk of violence score. At the intersection of the two scales, the matrix provides a recommended level of supervision (low, medium, high) with certain zones indicating possible consideration of an override to a higher level based on the evaluator’s professional judgment.17 Pennsylvania has developed a highly- structured system for integrating risk assessment instruments and professional 16 See e.g., Matter of Boyea v. Board of Parole, Sup. Ct. Onondaga County, May 23, 2013 (Index No. 2013-1751) (Discussing the impossibility of engaging in rationality review where Board provided no explanation for its rejection of a COMPAS-based estimate that the inmate-petitioner posed a low risk of re-offending) (286). 17 32 judgment into its parole release decision-making process.18 The number of ways to do this is limited only by the ingenuity of those who develop the procedures. During the two-plus years it has been delinquent in meeting its statutory obligation, the Board of Parole presumably considered various approaches to integrating COMPAS into its decision-making process. But in the end it came up short. Obviously, this Court cannot write the required procedures for the Board. But this Court can recognize minimum requirements for the Board’s use of COMPAS within the meaning of Executive Law § 259-c (4). At a minimum, the Board should be required to adhere to basic risk and needs principles by providing a rationale for determinations that run counter to COMPAS’ empirically based results. d. Evidence-based practice requires that overrides based on professional judgment be explained. The first principle of evidence-based practice is that “the professional judgment of criminal justice system decision makers is enhanced when [it is] informed by evidence-based knowledge.” See A Framework for Evidence-Based Decision Making in Local Criminal Justice Systems, U.S. Dept. of Justice, National Institute of Corrections (2010), p. 25.19 Thus, “[p]rofessional judgment should take into account both evidence-based knowledge and individual circumstances; and where decisions are made that counter empirical evidence, the rationale for those exceptions should be 18 See Pennsylvania Board of Probation and Parole – A Report on Board Initiatives to Strengthen Decisionmaking Policy and Practice, The Center for Effective Public Policy, Feb., 2008. 19 http://www.cepp.com/documents/EBDM%20Framework.pdf (last visited May 29, 2014) 33 explained.” Id. By openly addressing risk assessment results in decisions, board members can demonstrate that they are using COMPAS, a precondition for any rational procedure involving integration of professional judgment and evidence-based knowledge. As one commentator has written about this integration process in the context of parole-release decision-making: The instrument and the risk assessment process need to be credible with all of the parties who are directly impacted. Staff assigned to the risk assessment process must believe that the instrument actually works and will help inform the decision process for sentencing, release, and supervision decisions. The decision makers (judges, parole boards, and correctional administrators) must also have confidence in the risk assessment process and demonstrate through their decisions that they are using it . . . Finally, the offender must believe that the process is credible and will be used by decision makers. Austin, The Proper and Improper use of Risk Assessment in Corrections, Federal Sentencing Reporter, Vol. 16, no. 3 (Feb., 2004), p. 5. Unsurprisingly, other departments and agencies that utilize evidence-based practices in New York’s criminal justice system require a statement of reasons whenever professional judgment overrides an empirically-based risk assessment. The Department of Corrections and Community Supervision’s rules for establishing supervision levels require a stated “rationale” for COMPAS overrides.20 The Office of 20 See DOCCS Directive no. 8500 (Aug. 21, 2013) (“Override: The override process is intended to assist with placement of releasees in supervision levels and allow for adjustments to the assessment instrument’s recommended Supervision Status. The PO must seek approval from the Senior Parole Officer (SPO) and the final decision will be made by the Bureau Chief (BC). Any changes in the Supervision Status must be accompanied by a 34 Probation and Correctional Alternatives likewise requires documentation of overrides on its COMPAS-based supervision level matrix. The Board of Examiners of Sex Offenders acknowledges that departures from presumptive risk levels are appropriate when “there exists an aggravating or mitigating factor of a kind, or to a degree, that is otherwise not adequately taken into account by the guidelines.” People v. Johnson, 11 N.Y.3d 416, 421 (2008). The Northpointe Practitioner’s Guide to COMPAS points out that “disagree[ment] with an actuarial risk assessment (e.g., COMPAS) [is expected] in about 10% of the cases due to mitigating or aggravating circumstances.” In those situations, the manual instructs evaluators to “use their professional judgment and override the computed risk as appropriate – documenting it in COMPAS with the Override Reason.” Practitioner’s Guide to COMPAS, p. 30. Only the Board of Parole has devised a “procedure” wherein decision-makers can ignore, severely discount, or legitimately override empirically-based risk assessments without providing any rationale for doing so. e. The Board’s opaque “procedure” is inconsistent with the legislative history and intent of Executive Law § 259-c (4). The Board’s opaque “procedure” is not only unprecedented; it is at-odds with its own practice under the 1978 guidelines. For the past 36 years, the Board’s rules required a “detailed reason” whenever an inmate was not released in accordance with the applicable guidelines time range. The Board has now proposed to repeal this rule rationale and a recommended level. Overrides can increase or decrease the level of supervision” – emphasis added) (395). 35 along with the 1978 guidelines, and replace it with one that requires no explanation for decisions that run counter to COMPAS’ empirically-based results. The legislative history of Executive Law § 259-c (4) reveals a clear legislative intent to promote transparency, consistency and fair treatment in parole release decision-making, values that the Board’s recently approved rule irresponsibly casts aside. 1. Early 1970s - pre-guidelines parole release decision-making Before 1977, the Board of Parole’s decision-making function was largely unstructured. Then as now, board members were required to review certain factors (e.g., an inmate’s crime and criminal history, prison discipline and employment records, “a report of the warden” 21), and the statutory standard for release was substantially similar to the contemporary one (“a reasonable probability that, if such prisoner is released, he will live and remain at liberty without violating the law, and that his release is not incompatible with the welfare of society”22). But the decision to release or hold an inmate was left to the unguided and unfettered discretion of parole commissioners. Disparate treatment of similarly-situated offenders was an intractable problem under this unregulated scheme. The perceived unfairness of parole release decision making became a persistent and bitter source of controversy and prison turmoil. The McKay Commission, which conducted the official investigation into the 1971 Attica prison uprising, reported that 21 Former Cor. Law §213 (4) 22 Former Cor. Law § 213 36 “disparities in sentences imposed for identical offenses leave those who are convicted with a deep sense of disgust and betrayal.” The Commission concluded that “parole procedures are unfair, and appear to inmates to be even more inequitable and irrational than they are.” It recommended that “the grant or denial of parole must be measured by clear and comprehensible standards, disseminated to inmates in advance.”23 2. Origin of the 1978 Guidelines By 1976, the Assembly Codes Committee observed that “no concrete improvement can ever be made in the administration of criminal justice without some action to restore public confidence, and we cannot restore that confidence until . . . we have developed a more rational system of parole than presently exists.” A Report on Parole Reform by the Staff of the Codes Committee of the New York State Assembly (Nov. 1976), p. 18. It recommended adoption of parole release guidelines modeled on ones recently adopted by the United States Board of Parole. A guidelines-based system, the Committee argued, could “provide a scientific and objective means of structuring and institutionalizing discretion in parole release decisionmaking . . . [thus] minimize[ing] the effects of sentencing disparity.” Id. at 34. 23 New York State Special Commission on Attica. Attica: The Official Report of the New York State Special Committee on Attica (New York: Bantam Books 1972), at xviii – available at http://www.nysl.nysed.gov/mssc/attica/atticareport.pdf (last visited May 27, 2014) 37 The Committee’s recommendation was enacted into law the following year as part of a major reorganization of corrections and parole. 3. Parole Reform Act of 1977 The Parole Reform Act of 1977 re-established the Department of Correctional Services and the Division of Parole as separate executive agencies. And for the first time the legislature directed that parole release decisions be structured to promote transparency, consistency and fair treatment of inmates. The language enacting this important change was deceptively simple. Executive Law § 259-c (4) directed the Board to “establish written guidelines for its use in making parole decisions as required by law, including the fixing of minimum periods of imprisonment or ranges thereof for different categories of offenders.” The Act’s preamble spelled out the clear purpose of the soon-to-be-written guidelines: It is the legislative intent that these . . . changes will create an appropriate framework within which the parole system can arrive at individual determinations that are just and proper to the particular individual while at the same time consistent with the treatment of others similarly situated (emphasis added). L.1977, chap. 904, § 1. The guidelines and accompanying regulations were “guided by three major objectives”: 1.) “equitable treatment of similarly-situated offenders, 2.) the importance of articulating an explicit and open policy for inmates and criminal justice practitioners, and 3.) preserving the Board’s discretionary power within a structure 38 that also safeguards against abuse.” See New York State Parole Board Decision- Making Guidelines: Their Flexibility and their Value as a Policy-Setting Tool (NYS Division of Parole, April 1985), p. 2 (364). To that purpose, the Board’s rules provided that when an inmate was not released in accordance with the guidelines time range, a detailed reason was required: Decisions outside the guidelines. The time ranges indicated above are merely guidelines. Mitigating and aggravating factors may result in decisions above or below the guidelines. In any case where the decision is outside the guidelines, the detailed reason for such decision, including the fact or factors relied on, shall be provided to the inmate in writing. 9 NYCRR § 8001.3 (3)(c). The Board complied with this rule for many years. But after the last revision of the guidelines in 1985, the Board slowly drifted away from using the guidelines. The 1990s witnessed major changes in New York’s sentencing laws, but the Board never revised the guidelines and the time ranges grew outdated. The Board abandoned the guidelines fifteen or twenty years ago, doing so quietly and without legislative permission.24 24 After the Board effectively stopped using the guidelines, it defended the practice in court by claiming the “detailed reason” requirement applied to MPI determinations only, not parole release decisions, an assertion belied by the regulation’s plain language (“The guidelines will be considered in each MPI and release decision. . . In any case where the decision is rendered outside the guidelines, the detailed reason for such decision . . . shall be provided to the inmate in writing”), and years of the Board’s prior practice and pronouncements on the subject. See e.g., New York State Parole Board Decision-Making Guidelines: Their Flexibility and their Value as a Policy-Setting Tool (NYS Division of Parole, April, 1985, p. 8) (370) (Extolling the value of the “detailed reason” requirement in 39 4. The Board’s new rule would turn the clock back forty years to the age of unregulated discretion in parole-release decision-making. In recent years as parole release rates have plummeted, there has been renewed focus on the unfairness of the system and persistent calls for reform. See New York’s Broken Parole System, New York Times, Feb. 16, 2014; Kamins, Restoring Fairness to the Parole System, Forty-Fourth Street Notes (April, 2007);25 Caher, Dismantling Parole - Parole Release Rates Plunge Under Pataki’s Tough Policy, New York Law Journal, Jan. 31. 2006; see also, Recommendations for Parole Reform in New York State (various signatories) (“Board of Parole release guidelines should be updated and modified to require the Board to give appropriate weight to the extent of an individual’s rehabilitation and the lack of risk to public safety if the individual is released”).26 To many observers, the 2011 amendment to Executive Law §259-c (4) offered the promise of real change and a chance to modernize the harsh and rudderless machinery of New York’s parole release decision-making process. See e.g., Genty, Changes to Parole Laws Signal Potentially Sweeping Policy Shift, New York the context of parole release decision-making.) Unfortunately, the Board raised this argument in the context of pro se inmate litigation and the Appellate Division wrongly ruled in the Board’s favor. See Matter of Davis v. Travis, 292 A.D.2d 742 (3d Dept. 2002). 25 http://www2.nycbar.org/email/44StreetNotes/04_07.pdf (last visited may 27, 2014) 26 http://www.jjay.cuny.edu/centersinstitutes/pri/pdfs/Recommendations%20NYS%20Parol e%20Reform%20With%20Endorsements.pdf (last visited May 27, 2014) 40 Law Journal, Sept. 1, 2011; Rosenthal & Warth, New York Still in Need of Parole Reform, Atticus, Spring 2014.27 But now, three years later, the Board’s dilatory response to the 2011 amendment threatens to turn back the clock forty years. Thus far, the Board has failed to replace the guidelines with written procedures governing use of evidence- based risk assessments to measure rehabilitation. Board members are once again cast adrift without clear guidance on how to properly exercise their discretionary release power – just as they were prior to 1977. This appalling lack of transparency and accountability was certainly not what the legislature intended when it revised the parole release decision-making process for the first time in more than three decades. In view of the Board’s delay and default, this Court should now declare minimum requirements for consideration of a risk and needs assessment instrument at parole release hearings. Board members should be required to provide detailed reasons for determinations that run counter to COMPAS’ empirically-based results. And this Court should direct the Board of Parole to finally comply with its obligations under the law. 27 http://www.nysacdl.org/wp/wp-content/uploads/2010/09/atticus-spring-2012-web.pdf (last visited May 27, 2014). 41 CONCLUSION The order of the Appellate Division should be modified by directing the Board of Parole to provide detailed reasons for determinations that run counter to COMPAS’ empirically-based results. Respectfully submitted, __________________ Alfred O’Connor New York State Defenders Association 194 Washington Ave., Suite 500 Albany, New York 12210 (518) 465-3524 Counsel of Record Alan Rosenthal Center for Community Alternatives 115 E. Jefferson Street Syracuse, New York 13202 (315) 422-5638 June 2, 2014