The Missouri Project: A Study of the Missouri Defender System and Attorney Workload Standards January 2014 Prepared by: On Behalf of the American Bar Association’s Standing Committee on Legal Aid and Indigent Defendants This publication has been prepared by RubinBrown on behalf of the American Bar Association’s Standing Committee on Legal Aid and Indigent Defendants. The views expressed herein, unless otherwise noted, have not been approved by the House of Delegates or the Board of Governors of the American Bar Association and, accordingly, should not be construed as representing the policy of the American Bar Association. RubinBrown was engaged by the American Bar Association (“ABA”) on behalf of its Standing Committee on Legal Aid and Indigent Defendants (“SCLAID”) to provide consulting services and assist in the development of a process to calculate workload standards for the Missouri State Public Defender System. This report presents the results of the engagement. Our services were performed in accordance with the Statements on Standards for Consulting Services as prescribed by the American Institute of Certified Public Accountants. RubinBrown LLP Preface This report is the result of a collaborative effort between RubinBrown LLP (“RubinBrown”)1, the American Bar Association (“ABA”) 2, and the Missouri State Public Defender System (“MSPD”). The ABA has denominated the effort, “The Missouri Project.” The effort was led by Michael T. Lewis (PartnerIn-Charge of the Business Advisory Services group at RubinBrown LLP). The research team consisted of Jason Mannello (RubinBrown LLP), Josh Leesmann (RubinBrown LLP), and Kent Bausman, Ph.D. (Maryville University). We would like to thank Stephen F. Hanlon 3, Chair of the Indigent Defense Advisory Group of the ABA’s Standing Committee on Legal Aid and Indigent Defendants (“SCLAID”) 4 for his assistance in this effort. We would also like to thank Cathy Kelly and Peter Sterling (Director and General Counsel, respectively, Missouri State Public Defender System) for providing guidance and insight into the MSPD’s practices and technical systems. We also appreciate the input and editorial comments of Norman Lefstein.5 We would like to thank each member of the MSPD for their cooperation and input throughout the project. We would also like to thank the group of private bar attorneys and public defender experts asked to participate in numerous surveys and meetings. The project would not be possible without their input and efforts. 1 Founded in 1952, RubinBrown (www.RubinBrown.com) is one of the nation’s leading accounting and professional consulting firms. RubinBrown helps its clients build and protect value, while at all times honoring the responsibility to serve the public interest. RubinBrown’s vision statement is: One firm, highly respected and nationally prominent with a solid foundation of core values, inspired team members and totally satisfied clients. 2 The ABA (www.AmericanBar.org) is the largest voluntary professional membership organization and the leading organization of legal professionals in the United States. Its nearly 400,000 members come from all 50 states and other jurisdictions. They include attorneys in private law firms, corporations, nonprofit organizations, government agencies, and prosecutorial and public defender offices, as well as judges, legislators, law professors and law students. Since its founding, the ABA has actively worked in the fields of legal ethics and indigent defense. In 1908, the ABA adopted its first Canons of Professional Ethics (now the Model Rules of Professional Conduct) (“ABA Model Rules”). In 1913, the ABA created the entity now known as the ABA Standing Committee on Ethics and Professional Responsibility (“ABA Ethics Committee”). The ABA Ethics Committee publishes formal ethics opinions on professional and judicial conduct, provides informal responses to ethics inquiries, and, upon request, assists courts in their development, modification, and interpretation of ethical standards such as the ABA Model Rules and the ABA Model Code of Judicial Conduct. 3 For the last 20 years, much of Mr. Hanlon’s work has involved systemic challenges to indigent defense systems. Mr. Hanlon currently limits his practice to advising and representing public defenders with excessive caseloads. He currently teaches indigent defense at St. Louis University School of Law. 4 The ABA created SCLAID in 1920 and charged the entity with examining the delivery of legal services to assist the poor. 5 Mr. Lefstein is the Dean Emeritus of the Indiana University Robert H. McKinney School of Law, former Chairman of the ABA Section of Criminal Justice, Reporter for the Second Edition of ABA Criminal Justice Standards Relating to The Prosecution Function and The Defense Function, Providing Defenses Services, and Pleas of Guilty, and former Chairman of the ABA Committee on Criminal Justice Standards. Table of Contents Executive Summary ....................................................................................................................................................5 Introduction ................................................................................................................................................................7 Literature Review .......................................................................................................................................................9 Methodology & Analysis.......................................................................................................................................... 11 A. System Analysis ........................................................................................................................................... 12 B. Case Type / Case Task Summary ................................................................................................................. 13 C. Time Study ................................................................................................................................................... 15 D. Time Sufficiency Survey ............................................................................................................................... 16 E. Delphi Process ............................................................................................................................................. 17 Attorney Workload Standard Conclusion ................................................................................................................ 21 The Missouri Project Executive Summary This report lays out the methodology, analysis, and results of the joint efforts of the MSPD, the ABA, and RubinBrown to develop data-supported workload standards. Although this effort is not the end of the process, it is a critical first step in establishing supportable, data-driven workload standards that can assist the MSPD in assessing staffing requirements and provide empirical support to determine maximum workloads. The Department of Justice’s National Advisory Commission on Criminal Justice Standards and Goals established criteria for assessing maximum case loads in 1973 (the “NAC Standards”). However, MSPD’s recent application of the NAC Standards has been criticized by the Missouri State Auditor and the National Center for State Courts (“NCSC”). Both the Missouri State Auditor and the NCSC concluded: • • • MSPD’s protocol “suffers significantly from its failure to depart from NAC caseload standards,” The protocol suffers “from its apparent inability to make fuller use of a 2006 time study,” and “Our review of the calculations and available data supporting the caseload protocol noted the MSPD lacks sufficient support for the data and methodology used for protocol calculations.” 6 These critiques were at the forefront of the analysis to establish new workload standards for the MSPD. This study does not rely upon the 1973 NAC Standards and utilizes MSPD’s current time data, combined with a data-driven survey process, to calculate new workload standards. Daily time entry became a mandatory function for all MSPD practitioners as of March 1, 2013. 7 This study utilized MSPD time data for a 25-week period beginning in March 2013 and ending August 2013 as the foundation for workload standards. Excessive workloads result in insufficient time available to provide reasonably effective assistance of counsel to all clients. As in prior studies of this type, this study required a means to identify areas where MSPD attorneys reported they often did not have sufficient time to complete certain tasks with reasonable effectiveness. 6 7 NCSC, “Brief Comment on the Missouri State Public Defender Commission’s Caseload Standard Protocol: Technical Assistance Paper”, (2010). At the time this report was issued, MSPD personnel did not record sick leave, vacation, or holidays in the time log system. However, this information was tracked in a separate form. Further, the Appellate division did not begin tracking time until April 1, 2013. 5 The Missouri Project The study surveyed MSPD practitioners directly, and was used to identify which case-related tasks they reported they often had either sufficient or insufficient time to perform based on current practices and staffing levels. The study then utilized a Delphi method to estimate the amount of time that should be allotted for those tasks that MSPD line defenders identified as often not having sufficient time to complete with reasonable effectiveness. 8 The Delphi method used in this study was an iterative process that included both experienced private practice criminal defense attorneys as well as MSPD line defenders. 9 The resulting attorney workload standards (shown below) reflect estimates of the average amount of time 10 an attorney can expect to spend on a category of Case Tasks for a particular type of case to provide reasonably effective assistance of counsel. 11 Case Type Murder/Homicide A/B Felony C/D Felony Sex Felony Misdemeanor Juvenile Appellate/PCR Probation Violation Controllable Case Task Hours per Case 106.6 47.6 25.0 63.8 11.7 19.5 96.5 9.8 The conclusion shown in the above chart reflects the consensus time expectations (under prevailing professional norms and standards) of a group of both private practice and public defender experts from across the state of Missouri. However, the above workload standards are not the end of the process. Stakeholders throughout the criminal justice system recognize that the study and calculation of workload and related workload standards is a continuous process. 8 MSPD line defenders were asked to identify tasks that they often did not have sufficient time to complete with reasonable effectiveness in the current environment (i.e., current practices and staffing levels). 9 The study excluded MSPD personnel solely tasked with the administration and management of the system, focusing only on MSPD practitioners that carry a caseload. 10 Of course, some cases will take less time and some cases will take more time, but in each case, as the recent decision of the United States District Court for the Western District of Washington makes clear, reasonably effective representation “presumes a certain basic representational relationship” and the system of public defense must “[allow]…counsel to give each case the time and effort necessary” to ensure reasonably effective representation (Wilbur v. City of Mount Vernon, Case No. C11-1100RSL, December 4, 2013), so that the prosecutor’s case can be subjected to “meaningful adversarial testing” (U.S. v. Cronic, 466 U.S. 648, 654, 659 (1984). 11 The workload standards include only case-related tasks over which an attorney has some control (i.e., exclude in-court, travel, training, and administrative time). Further, the workload standards assume adequate support staff and attorney resources are available. Private practice defense counsel reported utilizing 2 support staff resources per attorney, on average. By contrast, the MSPD currently has approximately 2 attorneys for every 1 support staff resource (0.55 support th staff per attorney, or approximately 1/4 of the support staff available to private practice defense counsel). 6 The Missouri Project These standards should be revisited periodically to account for potential changes in technology and technology usage, indigent defendant demographics and crime patterns, Missouri criminal code, and the staffing and organizational structure within the MSPD and the larger criminal justice system. In addition, as the MSPD’s time entry system matures and the amount of data contained within it increases, the ability to utilize that data to examine organization practices, study different types of complexity factors, and quantify time sufficiency will increase. Introduction The guarantee of assistance of counsel in one’s defense against criminal prosecution is a right afforded under the Sixth Amendment to the U.S. Constitution and Article 1, Section 18(a) of the Missouri Constitution. In 1963, the Supreme Court’s Gideon decision extended the right to counsel to felony cases in state criminal courts; subsequently the Supreme Court extended the right to counsel in misdemeanor cases ending with the defendant being imprisoned. 12 However, a majority of states recognize the right to a lawyer if the defendant is charged with a misdemeanor. In Missouri, the right to counsel is extended to defendants in misdemeanor cases if incarceration is probable. Despite the promise provided by the Gideon decision, many academics and legal scholars have concluded that much of that promise has remained unfulfilled. 13 Simply assigning an attorney to a defendant does not ensure a fair outcome.14 Further, pursuant to the Rules of Professional Conduct, proper defense representation requires that lawyers: 1) be competent to represent the client, 2) offer prompt and diligent representation of the client’s interests, and 3) productively engage with the client while exercising independent judgment.15 12 Gideon v. Wainwright, 372 U.S. 335 (1963) and Argersinger v. Hamlin, 407 U.S. 25 (1972). Bruce R. Jacob, “50 Years Later: Memories of Gideon V. Wainwright”, The Florida Bar Journal (March 2013), Cara H. Drinan, “Getting Real About Gideon: The Next Fifty Years of Enforcing the Right to Counsel” (2013), Roger A. Fairfax, Jr., “Searching for Solutions to the Indigent Defense Crisis in the Broader Criminal Justice Reform Agenda”, The Yale Law Journal (2013), “Justice Denied: America’s Continuing Neglect of Our Constitutional Right to Counsel”, Report to National Right to Counsel Committee (2009), and “Gideon’s Broken Promise: America’s Continuing Quest for Equal Justice”, ABA Standing Committee on Legal Aid and Indigent Defendants (2004). 14 Wilbur v. City of Mount Vernon, Case No. C11-1100RSL, December 4, 2013. 15 Phyllis E. Mann, “Ethical Obligations of Indigent Defense Attorneys to Their Clients”, Missouri Law Review, Vol. 75 (2010). 13 7 The Missouri Project To guide policy makers and criminal defense practitioners in ensuring a fair and proper process, the American Bar Association has developed Criminal Justice Standards (“ABA Standards”)16 and the state of Missouri has developed the Missouri State Public Defender Guidelines for Representation (“Missouri Guidelines”). The fulfillment of these obligations is dependent upon having sufficient resources available. In the years since Gideon, the consensus is that indigent defense in the United States has been inadequately supported, creating a crisis in the state of indigent defense. 17 According to Drinan (2013), “From the start, states have failed to fund the indigent defense function adequately, and as the volume of criminal cases has grown over the years, too few lawyers have faced ever-increasing workloads. The result has been what many have called “assembly-line justice” – in other words, egregious and persistent violations of the right to counsel.” 18 To address the potential violations of the right to counsel, several states and counties throughout the United States have begun to search for data-driven workload standards to assess and manage attorney resources in an attempt to provide adequate criminal defense for the poor. From our discussions with the MSPD, it was also clear that the system would benefit from a consistent methodology that would allow public defenders to quantify workload expectations necessary to provide reasonably effective assistance of counsel to indigent defendants. 16 The ABA Standards are the result of a lengthy process that has been in continual development since 1964. Specifically, the ABA Standards “are the result of the considered judgment of prosecutors, defense lawyers, judges, and academics who have been deeply involved in the process, either individually or as representatives of their respective associations, and only after the [ABA] Standards have been drafted and repeatedly revised on more than a dozen occasions, over three or more years.” Marcus, Martin, The Making of the ABA Criminal Justice Standards, Criminal Justice Volume 23, Number 4 (Winter 2009). 17 “Justice Denied: America’s Continuing Neglect of Our Constitutional Right to Counsel”, Report to National Right to Counsel Committee, (2009). 18 Cara H. Drinan, “Getting Real About Gideon: The Next Fifty Years of Enforcing the Right to Counsel”, (2013), p 1311. 8 The Missouri Project Literature Review The Delphi method was introduced in 1962 by researchers at the Rand Corporation. The method was described as a “new” research technique utilized by the Air Force in the 1950s to gather expert opinion and generate a reliable consensus. 19 As a methodological strategy, the Delphi method proposed that a succession of surveys be given to a group of experts, with structured feedback presented to the experts at each interval stage. 20 The surveying practices applied by the Delphi method could be interviews and/or questionnaires that focus on some fundamental question of significance to the group of experts convened for feedback. The features of this method include “anonymity, iteration, controlled feedback, and the statistical aggregation of group response.” 21 At the onset of the process, participants in a Delphi group are largely anonymous from one another. The purpose of anonymity is to ensure that solicited experts are not influenced by the responses of other participants and that the ideas presented are judged on their own merit. This technique is believed to be conducive to the exercise of independent thought on the part of participating experts, and to aid experts in forming well-thought-out opinions. The reliance on expert opinion as data is built on the premise that an expert is “able to select the needed items of background information, determine the character and extent of their relevance, and apply these insights to the formulation of the required personal probability judgments.” 22 Experts typically complete a questionnaire over multiple iterations with the goal of allowing participants to change their opinions and judgments when presented with controlled feedback regarding the opinions and judgments of their fellow participants. This controlled feedback is normally presented as a statistical summation of the group’s responses, e.g. a mean or median. The structured feedback at each successive iteration consists of “available data previously requested by…the experts…, or of factors and considerations suggested as potentially relevant by one or another respondent.” 23 19 Norman Dalkey and Olaf Helmer, “An Experimental Application of the Delphi Method to the Use of Experts”, Memorandum RM-727/1-Abridged, (1962). 20 Norman Dalkey and Olaf Helmer, “An Experimental Application of the Delphi Method to the Use of Experts”, Memorandum RM-727/1-Abridged, (1962). 21 Gene Rowe and George Wright, “The Delphi technique as a forecasting tool: issues and analysis”, International Journal of Forecasting, (1999), p 354. 22 Olaf Helmer and Nicholas Rescher, “On the Epistemology of the Inexact Sciences”, P-1513, (1958), p 42. 23 Norman Dalkey and Olaf Helmer, “An Experimental Application of the Delphi Method to the Use of Experts”, Memorandum RM-727/1-Abridged, (1962). p 2. 9 The Missouri Project 10 The goal of the feedback at each stage is to assist in limiting mistaken beliefs an expert may have on the question at hand or to increase their awareness of other information they may not have previously considered. 24 At the conclusion of the final iteration, the final iteration’s mean or median response is used as the measure of the group’s opinion. 25 In theory, the number of iterations required of the Delphi method can be unlimited until consensus among participants is achieved, however it has been found that three to four iterations is usually all that is required to reach consensus. 26 Rowe and Wright (1999) systematically reviewed studies that explored the effectiveness of the Delphi method. Their focus was on how well the Delphi method worked in producing a consensus of opinions and judgments and to assess how accurate those opinions and judgments were. Overall, they found that the majority of these evaluative studies showed support for the Delphi method in reducing variances in opinion and judgment, thus indicating that greater consensus had been achieved. As for the concern over the accuracy of those opinions and judgments, Rowe and Wright (1999) again found that the majority of studies provide compelling evidence in support of the Delphi method. Compared to other methodological techniques utilized for similar purposes, the Delphi method was found to “lead to improved judgments over staticized groups and unstructured interacting groups.” 27 Since its introduction, the Delphi method has been employed across a diverse array of industries, such as health care, education, information systems, transportation, and engineering. 28 The purpose of its use beyond forecasting has ranged from “program planning, needs assessment, policy determination, and resource utilization.” 29 Within the legal system, early examples of use of the Delphi method can be traced back a couple of decades. 24 Norman Dalkey and Olaf Helmer, “An Experimental Application of the Delphi Method to the Use of Experts”, Memorandum RM-727/1-Abridged, (1962). 25 Gene Rowe and George Wright, “The Delphi technique as a forecasting tool: issues and analysis”, International Journal of Forecasting, (1999). 26 Chia-Chien Hsu and Brian A. Sandford, “The Delphi Technique: Making Sense of Consensus”, Practical Assessment Research & Evaluation, (2007). 27 Gene Rowe and George Wright, “The Delphi technique as a forecasting tool: issues and analysis”, International Journal of Forecasting, (1999), p 366. 28 Harold A. Linstone and Murray Turoff, “The Delphi Method: Techniques and Applications” (2002) and Gene Rowe and George Wright, “The Delphi technique as a forecasting tool: issues and analysis”, International Journal of Forecasting, (1999). 29 Chia-Chien Hsu and Brian A. Sandford, “The Delphi Technique: Making Sense of Consensus”, Practical Assessment Research & Evaluation, (2007), p. 1. For detailed examples of the application of the Delphi method see Harold A. Linstone and Murray Turoff, “The Delphi Method: Techniques and Applications” (2002). The Missouri Project 11 Examples of these attempts were sponsored by both the National Association of Court Management (“NACM”) and the National Center for State Courts (“NCSC”). These efforts were principally charged with assessing judicial and court support staff needs. 30 In the 2000’s, the NCSC started using Delphi techniques in addressing the caseload/workload crisis of indigent defense in the United States. In a recent book, Lefstein (2011) comments on the use of the Delphi method noting: The technique is recommended when a problem does not lend itself to precise measurement and can benefit from collective judgments. This would seem to be precisely the situation when a defense program seeks to determine how much additional time, on average, its lawyers need to spend on a whole range of activities involving different kinds of cases.31 The Delphi method has been recommended as a necessary complement to time-based studies that seek to determine appropriate caseloads for defense lawyers. 32 What the Delphi method is believed to offer is a method to adjust preliminary case weights based on time studies while avoiding the institutionalization of potentially sub-optimal current practices. Methodology & Analysis Past workload studies33 were reviewed and assessed in developing the methodology advanced in this study, which sought to quantify the amount of time a public defender should expect to spend on a particular task in a particular case type through an application of the Delphi method. As in prior studies, the Delphi methodology was used to provide an estimate of what workload standards should be in order for a public defender to provide reasonably effective assistance of counsel. However, among other things, this study expands upon prior work in this field in that it focuses on both the amount of time that should be spent on a task, as well as how often a task should be completed. Further, this study expands on prior work in that it utilizes the input of private practice defense counsel. 30 For example, see Victor E. Flango, Ph.D. and Brian J. Ostrom, Ph.D, “Assessing the Need for Judges and Court Support Staff”, National Center for State Courts, (1996). 31 Norman Lefstein, “Securing Reasonable Caseloads: Ethics and Law in Public Defense”, American Bar Associations (2011), p 146. 32 Norman Lefstein, “Securing Reasonable Caseloads: Ethics and Law in Public Defense”, American Bar Associations (2011). 33 Such as National Center for State Courts’ (“NCSC”) reports: Virginia Indigent Defense Commission Attorney and Support Staff Workload Assessment – Final Report (2010); A Workload Assessment Study for the New Mexico Trial Court Judiciary, New Mexico District Attorneys’ Offices, and the New Mexico Public Defender Department – Final Report (2007); and Maryland Attorney and Staff Workload Assessment (2005). See also Elizabeth Neeley, PhD. Lancaster County Public Defender Workload Assessment, University of Nebraska Public Policy Center (2008). The Missouri Project 12 To determine workload standards, a multi-step process was used that first analyzed the current, “actual” state of affairs, as a starting point. After an introduction of the definitions and key terms utilized throughout this study, the methodology can best be explained as the performance of the following steps: A. B. C. D. E. System Analysis Case Type / Case Task Summary Time Study Time Sufficiency Survey Delphi Process Standards, Definitions and Key Information Sufficient time to complete the specific task: the amount of time, on average, reasonably required to complete the task with reasonable effectiveness. Reasonable effectiveness: effective representation under prevailing professional norms. Typical case: average, or typical, case considering that each case may have significant variability in the level of complexity (i.e., language, mental health, and other issues). ABA Criminal Justice Section Standard 4-6.1(b): "Defense counsel may engage in plea discussions with the prosecutor. Under no circumstances should defense counsel recommend to a defendant acceptance of a plea unless appropriate investigation and study of the case has been completed, including an analysis of controlling law and the evidence likely to be introduced at trial." Missouri v. Frye (132 S.Ct. 1399, 2012): “…ninety-four percent of state convictions are the result of guilty pleas.” A. System Analysis The MSPD provides “direct representation to over 98% of the indigent defendants accused of state crimes.” 34 The system is comprised of approximately 585 employees, of which 376 (64%) are attorneys and 209 (36%) are support staff. In other words, there is approximately one support staff resource for every two attorneys. The line defender attorney group consists of 312 public defenders in the trial division, 36.5 public defenders in the Appellate/PCR division, and 17 public defenders in the capital division. The trial and appellate/PCR division support staff group consists of 43 legal assistants, 56.5 investigators, 6.5 paralegals, 3 mitigation specialists, and 67.5 secretaries. 34 State of Missouri Public Defender Commission Fiscal Year 2012 Annual Report. 13 The Missouri Project The MSPD maintains a case management system that tracks basic case information such as open date, close date, charge type, disposition type, jurisdiction, and assigned attorney. This study utilized the current 2–year extract of the case management database consisting of over 120,000 cases and over 300 different charge types. Since March of 2013, the MSPD has required that all of its public defenders enter their time in a time log system. This system captures the amount of time across over 50 types of tasks for all MSPD practitioners on each case. This study linked the case management system with the time log system as a basis for the workload analysis. Based upon MSPD’s current systems, cases were grouped by type and attorney time spent on each case was grouped by task. B. Case Type / Case Task Summary Working with the MSPD, 11 different Case Types were initially identified to use in the development of new workload standards. The 11 Case Types are: 1. 2. 3. 4. 5. 6. Murder / Homicide Sex Felony AB Felony CD Felony Misdemeanor Juvenile 7. Probation Violation 8. Sexual Predator Proceeding 9. Appeals/PCR 10. Release Petitions 11. Special Writs From an assessment of current cases, it was determined that Special Writs, Release Petitions, and Sexual Predator Proceedings were highly specialized and infrequent. As a result, these Case Types were excluded from the study. The remaining 8 Case Types35 are as follows: 1. 2. 3. 4. 5. 6. 7. 8. 35 Murder / Homicide Sex Felony AB Felony CD Felony Misdemeanor Juvenile Probation Violation Appeals/PCR The remaining Case Types capture the vast majority of case specific time in the MSPD time log system. For this study, the “Murder / Homicide” Case Type did not include capital murder cases. Juvenile cases primarily encompass juvenile delinquency cases; however, there are a small percentage of juvenile status offenses. The MSPD does not handle abuse or neglect cases involving juveniles. The Missouri Project 14 Case Type is currently the primary way complexity is addressed in this workload study, though it is recognized that case complexity can be impacted by a variety of factors. Language barriers, mental health history, and family issues are just a few examples of factors that can impact case complexity. All else equal, however, different Case Types generally have different degrees of average complexity. For example, without any prior case specific knowledge, it is reasonable to assume that a Class A felony will be more complex than a misdemeanor. Since both the case management database and the time log system consistently report Case Type, this data was utilized to assess current actual time and resource utilization, which provides the foundation to draw conclusions about time and resource allocation by Case Type. The 8 Case Types form the foundation for the workload standards and are used to identify how MSPD practitioners are actually spending their time under current conditions and practice. This study was particularly interested in the tasks that are best performed by attorneys (versus support staff), and in those attorney tasks that may be affected by excessive workloads. In other words, this study sought to identify which attorney Case Tasks are impacted by excessive caseloads and time constraints. An attorney typically has more control over the time spent on certain Case Tasks, such as trial preparation, research, interviews, etc. (“Controllable Tasks”) than the time spent on other Case Tasks, such as travel, court, etc. (“Non-Controllable Case Tasks”). Therefore, Case Tasks were segregated into two different categories for purposes of this study: Controllable Case Tasks • Client Communication: 1. 2. 3. 4. In Person Over the Phone Written Family/Other Communications • Discovery/Investigation: 5. 6. 7. 8. State's Discovery Disclosure Records and Transcripts Depositions and Witness Interviews Experts and Technical Research • Case Preparation: 9. Legal Research 10. Drafting and Writing 11. Plea Negotiation 12. Court Preparation 13. Case Management 14. Alternative sentencing research The Missouri Project 15 Non-Controllable Case Tasks 15. In Court - Pretrial 16. In Court – Trial 17. In Court - Appellate Argument 18. Travel 19. Miscellaneous Case Administration Once identified, the Case Type and Case Task classifications were utilized to measure how MSPD attorneys are currently spending their time on case-related work. C. Time Study The Time Study combines MSPD’s time entry database with its case management database to present a picture of how much time MSPD practitioners are spending on case-related work. Time data was extracted from the time entry database for a 25-week period beginning March 2013 and ending August 2013. This data showed how much time, in total, MSPD practitioners spent on case-related tasks. This data was combined with case count information from the MSPD case management database to calculate average time spent per Case Type, shown below. 36 Case Type Murder/Homicide A/B Felony C/D Felony Sex Felony Misdemeanor Juvenile Appellate/PCR Probation Violation Average Reported Controllable Case Task Hours per Case 84.5 8.7 4.4 25.6 2.3 4.6 30.3 1.4 The calculation first annualized the total case-related time incurred over the 25-week period (by dividing the total time by 25 weeks, and then multiplying the result by the 52 weeks in a year). We then estimated the average time per Case Type by using MSPD’s case management database to estimate the average number of cases for which that time is incurred. 36 The chart summarizes current average reported time on case-related tasks by Case Type. Further, the reported average excludes travel, in-court, and administrative time. This Delphi study has focused on the Controllable Case Tasks for each of the 8 referenced case types, excluding all Non-Controllable Case Tasks (which account for a significant portion of an attorney’s time), because the time required for the Non-Controllable Case Tasks is predominantly dictated by the court’s schedule and the geography of the district. The Missouri Project 16 Specifically, we determined the average number of open cases between March 31, 2013 and August 31, 2013 (to reflect a workload that is concurrent with the time data), and then annualized that figure by dividing the open workload by the average length of case (based on fiscal year 2012 and fiscal year 2013 case management database). The resulting figure is an estimate of the number of cases per year. The annualized total time incurred divided by the estimated number of cases per year provides the average reported time per case, as shown below. Case Type Annual Hours Murder/Homicide 22,677 A/B Felony 53,855 C/D Felony 113,002 Sex Felony 26,916 Misdemeanor 30,127 Juvenile 7,085 Appellate/PCR 44,719 Probation Violation 24,405 ÷ ÷ ÷ ÷ ÷ ÷ ÷ ÷ Annual Case Count 269 6,196 25,910 1,051 13,322 1,554 1,477 16,977 = = = = = = = = Average Reported Controllable Case Tasks Hours per Case 84.5 8.7 4.4 25.6 2.3 4.6 30.3 1.4 The Time Study quantified how MSPD attorneys are actually spending their time. However, it does not indicate if this actual time is sufficient to provide reasonably effective assistance of counsel. D. Time Sufficiency Survey A “Time Sufficiency Survey” was conducted on MSPD line defenders 37 where MSPD practitioners were asked what percentage of the time for specific Case Type / Case Task combinations they had sufficient time to complete the task with reasonable effectiveness. The Time Sufficiency Survey results were utilized to exclude certain Case Type and Case Task combinations from the Delphi process.38 That is to say, if MSPD practitioners indicated that they often had sufficient time to adequately perform the specified Case Task for the Case Type based on current practices and staffing levels, the particular task was excluded from the Delphi process. The Time Sufficiency Survey was performed by creating a questionnaire that was distributed via email to all MSPD line defenders by a third-party, professional survey provider. The survey asked respondents in what percentage of cases do they have sufficient time to complete the Case Task with reasonable effectiveness. 37 The survey excluded operations personnel and senior management, focusing on public defenders currently carrying a case load. 38 See the attached Appendix 2 for the list of excluded Case Type and Case Task combinations. The Missouri Project 17 The questions were asked for each Controllable Case Task within each Case Type and the respondents were able to indicate their response by choosing either 0-20%; 21-40%; 41-60%; 61-80%; or 81-100% of the time. An example of the survey instrument is shown in the illustration below. The Time Study, combined with the Time Sufficiency Survey, described the “current state of affairs.” In other words, these two components of the analysis allowed for 1) the quantification of how MSPD practitioners currently spend their time on cases, broken down by Case Type / Case Task combinations and, 2) the identification of those Case Type / Case Task combinations where MSPD practitioners report that there is not enough time to perform those functions with reasonable effectiveness. To move from the “current state of affairs” to a “sufficient state” required a methodology to gauge how much time should be allowed for performance of certain Case Type / Case Task combinations. The “Delphi Process” was utilized to obtain this data. E. Delphi Process The Time Sufficiency Survey, in combination with the Time Study, provides critical information about current practice. However, the Time Sufficiency Survey results indicated that the MSPD defenders may be operating under excessive workloads. Thus, current practice may provide very little useable information about how much time attorneys should be spending and how often attorneys should be performing particular tasks in order to provide reasonably effective assistance of counsel. The Missouri Project 18 The Delphi process used in this study leverages the expertise of both private practice and public defenders to provide a consensus estimate of the amount of time defense counsel should expect to spend on a particular case in order to provide reasonably effective assistance of counsel. Further, in providing estimates of the amount of time an attorney should expect to spend on a particular case, the Delphi panel was asked to consider prevailing professional norms and standards of practice. That is to say, the standards resulting from this process should reflect the prevailing professional norms and standards, such as the Missouri Guidelines and the ABA Standards. As a first step in this process, the time an attorney spends on a particular case can be broken out into two components, time and frequency, as follows: 1. time incurred on the performance of specific Case Tasks (“Task Time”); and 2. the actual performance (or non-performance) of certain Case Tasks (“Task Frequency”). 39 Then, criminal defense experts (private, as well as public defense practitioners) from across the state of Missouri were identified and asked to participate in an iterative study of the time associated with the Case Tasks and Case Types. The expert panel was then asked to provide an estimate of the amount of time that should be spent on each Case Task for each Case Type, assuming that the task must be performed. An example of the survey instrument for this step is shown in the illustration below. 39 For example, Task Frequency component seeks to answer: In how many cases does an attorney speak to a client over the phone? This is contrasted with the number of phone calls an attorney may make within a particular case, which was not in the scope of this study. The Missouri Project The response estimates of Task Time were then summarized across the entire group of experts. The group was then provided summary statistics on the point estimate and range of time from this first survey. An example of the summary information from the first survey round is shown in the table below. CD Felony (Estimated Minutes) Client Communication - In person Lower 30 Average 309 Upper 606 Using the same survey instrument as the first round, the expert panel was then asked to again provide an estimate of the Task Time, this time taking into account the summary of responses from the first round of the process. Similar to the first iteration, the responses were summarized across all participants for this round as well. This round also presented the summarized responses of the private practice and MSPD attorney groups. An example of the summary from this second round is shown below for the same CD felony Case Type and in person client communication Case Task. Average Estimate from the Delphi Panel 246 124 Low MINUTES 4.1 hours High 369 one standard deviation range Average Estimate from MSPD Attorneys 254 119 Low Average Estimate from Private Practice Attorneys MINUTES High 233 388 o n e s t a n d a r d d e vi a t i o n r a n g e 131 Low MINUTES High 336 o n e s t a n d a r d d e vi a t i o n r a n g e 19 The Missouri Project 20 As a third and final iteration, the expert panel was invited to participate in a live meeting to discuss a summary of the second round of the survey process and to reach a group consensus for each Task Time estimate. The information in the above illustration was presented to the group during the inperson discussion. The group was reminded to keep in mind that the time estimate should: 1. assume adequate support staff (and that attorneys would only perform tasks not appropriate for support staff), 2. assume that the task must be performed, 3. apply prevailing professional norms, and 4. provide an estimate of the amount of time defenders should expect to perform the Case Task. Given the above information, the group was asked to either confirm the time estimate from the second survey round, or provide a new estimate to replace the second round average. Although this initial survey process provided an estimate of the typical amount of time required for a particular task, it did not provide insight into the frequency with which these tasks should be performed. To collect this data, the Delphi panel was asked to repeat the same process (making the same assumptions as above and applying professional norms) as with the Task Time estimates, only now providing an estimate of the frequency that defenders should expect to perform the Case Tasks. Again, the three stage process resulted in a group consensus of the Task Frequency for each Case Type. As a final step in the Delphi process, the Task Time was combined with the Task Frequency to arrive at an expected time for each Case Task. That is to say, the Task Time was multiplied by the Task Frequency as follows: 𝐸𝑥𝑝𝑒𝑐𝑡𝑒𝑑 𝑇𝑖𝑚𝑒 𝑝𝑒𝑟 𝑇𝑎𝑠𝑘 = 𝑇𝑎𝑠𝑘 𝑇𝑖𝑚𝑒 × 𝑇𝑎𝑠𝑘 𝐹𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 The expected time per task is interpreted as the amount of time that a public defender should expect to spend on any one Case Task and Case Type combination for the typical case. The Delphi panel was then convened for a final meeting for a presentation of the results of the Delphi process to confirm the time expectation resulting from the study. The expected time was then summarized for each Case Type in arriving at the final workload standards. The Delphi panel’s frequency weight was applied to the time estimate to arrive at the estimated amount of time that an attorney can reasonably expect to spend on a particular task for a given Case Type. The resulting workload standards for each Case Type are shown below. 40 40 The reported workload standards include only time for Controllable Case Tasks (i.e. exclude in court, travel, and th administrative time). The workload standard total shown in the table is rounded to the nearest 10 of an hour. The Missouri Project Case Type Murder/Homicide A/B Felony C/D Felony Sex Felony Misdemeanor Juvenile Appellate/PCR Probation Violation Average Reported Hours 84.5 8.7 4.4 25.6 2.3 4.6 30.3 1.4 + + + + + + + + Delphi Panel Adjustment 22.2 38.9 20.7 38.2 9.5 15.0 66.2 8.3 = = = = = = = = Workload Standard for Controllable Case Tasks 106.6 47.6 25.0 63.8 11.7 19.5 96.5 9.8 Attorney Workload Standard Conclusion The following table shows the Delphi-estimated time required for controllable Case Tasks for an attorney to provide reasonably effective defense by Case Type. 41 Case Type Murder/Homicide A/B Felony C/D Felony Sex Felony Misdemeanor Juvenile Appellate/PCR Probation Violation Controllable Case Task Hours per Case 106.6 47.6 25.0 63.8 11.7 19.5 96.5 9.8 This study sought to quantify the amount of time a public defender should expect to spend on a particular type of case through an application of the Delphi method. In other words, this study sought to quantify what workload standards should be in order for a defender to provide reasonably effective assistance of counsel. 41 The workload standards include only case-related tasks over which an attorney has some control (i.e., exclude in-court, travel, training, and administrative time). Further, the workload standards assume adequate support staff and attorney resources are available. Private practice defense counsel reported utilizing 2 support staff resources per attorney, on average. By contrast, the MSPD system currently has approximately 2 attorneys for every 1 support staff resource (0.55 th support staff per attorney, or approximately 1/4 of the support staff available to private practice defense counsel). 21 The Missouri Project 22 Exhibit 1 (attached to this report) shows the estimated time by both Case Type and Case Task group. The conclusion shown in the above chart reflects the consensus time expectations (under prevailing norms and standards) of a group of both private practice and public defender experts from across the state of Missouri. Exhibit 1: Concluded Workload Standards by Case Type and Case Task group Murder/Homicide AB Felony CD Felony Sex Felony Misdemeanor Juvenile Appellate/PCR Probation Violation The Missouri Project Client Communication1 Discovery/Investigation2 Case Preparation3 34.6 33.5 38.5 13.1 18.3 16.2 6.3 8.4 10.3 22.5 17.8 23.6 3.5 4.1 4.1 5.4 6.8 7.3 20.3 31.5 44.7 2.9 2.6 4.2 Total 106.6 47.6 25.0 63.8 11.7 19.5 96.5 9.8 1. The client communication Case Task group includes: in person conversations, phone calls, written communication, and communication with family. 2. The discovery/investigation Case Task group includes: State's discovery disclosures, records and transcripts, depositions and witness interviews, and expert and technical research. 3. The case preparation Case Task group includes: legal research, drafting and writing, plea negotiations, alternative sentencing research, court preparation, and case management. 23 The Missouri Project Exhibit 2: Current Average Reported Case-Related Hours by Case Type and Case Task group Murder/Homicide AB Felony CD Felony Sex Felony Misdemeanor Juvenile Appellate/PCR Probation Violation Client Communication1 Discovery/Investigation2 Case Preparation3 14.8 33.5 36.2 3.0 2.1 3.6 1.8 0.8 1.7 6.0 7.3 12.4 0.9 0.4 0.9 1.4 1.0 2.1 3.1 7.5 19.6 0.7 0.2 0.5 Total 84.5 8.7 4.4 25.6 2.3 4.6 30.3 1.4 1. The client communication Case Task group includes: in person conversations, phone calls, written communication, and communication with family. 2. The discovery/investigation Case Task group includes: State's discovery disclosures, records and transcripts, depositions and witness interviews, and expert and technical research. 3. The case preparation Case Task group includes: legal research, drafting and writing, plea negotiations, alternative sentencing research, court preparation, and case management. 24 Appendix 1: Case Task Descriptions The Missouri Project Case Task Client Communication - In person Task Description Attorney's time for privileged client interviews and consultations conducted face-to-face. Client Communication - Phone Client Communication - Written Attorney's time for privileged client interviews and consultations conducted via phone. Attorney's time for privileged client interviews and consultations conducted by written correspondence. Includes drafting and reviewing correspondence prepared by others. Non-privileged communications with client's family and friends, not including potential witnesses. Attorney's time receiving, organizing and reviewing the state's disclosure to all discovery requests including special discovery by motion. Attorney's time in requesting, acquiring and reviewing records which were not part of the state's disclosure, e.g., client's medical records. Attorney's time preparing for and conducting depositions or witness interviews where the attorney is investigating the case. Identifying, contracting, and consulting experts including testimony prep and also attorney's time doing self research on a technical (not legal) subject. Case related legal research for arguments, motions or briefs. Attorney's time actually drafting, typing or reviewing legal documents including motions and briefs. Plea negotiation with the state's attorney or representative whether verbal or written. Attorney's time reflecting, analyzing, brainstorming and outlining court case presentation. Also includes subpoenas, writs ad testificandum, and pre-charge representation. Client Communication - Family/other communications Discovery/Investigation - State's discovery disclosure Discovery/Investigation - Records and transcripts not included in state's discovery Discovery/Investigation - Depositions and witness interviews Discovery/Investigation - Experts and technical research Case Preparation - Legal research Case Preparation - Drafting and writing Case Preparation - Plea negotiation Case Preparation - Court preparation Case Preparation - Case management Case Preparation - Alternative sentencing research Attorney's time for case related office administrative tasks, e.g., time keeping, billing, and docket management tasks. Attorney's time identifying, locating, and engaging alternative sentencing resources, e.g., treatment programs. 25 The Missouri Project Appendix 2: List of Tasks Identified as Often Having Sufficient Time to Perform1 Murder/ Homicide Client Communication In person Phone Written Family/other communications Discovery/Investigation State's discovery disclosure Records and transcripts Depositions & interviews Experts and technical research Case Preparation Legal research Drafting and writing Plea Negotiation Court Preparation Case management Alternative sentencing research AB Felony X CD Felony Sex Felony X X X X X X X X X X X X Misdemeanor Juvenile Appellate/ PCR Probation Violation X X X X X X X X X 1. The table shows the 22 Case Type and Case Task combinations that MSPD practitioners identified as often having sufficient time to adequately perform based on current practices and staffing levels. If the Case Type and Case Task combination was identified by MSPD practitioners and the practitioner estimated sufficient time was consistent with actual time spent on the particular task (from the time log system), the combination was excluded from the Delphi process. It is important to note that it is anticipated that this study will be the first of many performed by the MSPD. An evaluation should be performed in each subsequent study to determine which Case Type and Case Task combinations should be included, or excluded, based upon the results of the Sufficiency Survey. 26 The Missouri Project Appendix 3: Example Questions from Delphi Time Survey Felony CD Cases Below, you will be asked to provide your estimate of the amount of time that is reasonably required to perform the respective task with reasonable effectiveness. Please enter your response in minutes. ------------------------------------------------------------------------------------------------------------------------------------------CLIENT COMMUNICATION - IN PERSON: Time for privileged client interviews and consultations conducted face-toface. How much time, on average, is reasonably required to perform this task with reasonable effectiveness? Minutes: (Optional) Please provide an explanation of your time estimate. ------------------------------------------------------------------------------------------------------------------------------------------CLIENT COMMUNICATION - PHONE: Same as above only by phone. How much time, on average, is reasonably required to perform this task with reasonable effectiveness? Minutes: (Optional) Please provide an explanation of your time estimate. 27 The Missouri Project Appendix 4: Example Questions from Delphi Frequency Survey Felony CD Cases Below, you will be asked to provide your estimate of the percentage of cases, on average, that attorneys should perform the specific task to provide reasonably effective representation. Please enter your response as a percentage (for example, please enter 25 for 25%) ------------------------------------------------------------------------------------------------------------------------------------------CLIENT COMMUNICATION - IN PERSON: Time for privileged client interviews and consultations conducted face-toface. Please enter an estimate of the percentage of cases, on average, that attorneys should perform the above task to provide reasonably effective representation. Percent of cases: (Optional) Please provide an explanation of your percentage estimate. ------------------------------------------------------------------------------------------------------------------------------------------CLIENT COMMUNICATION - PHONE: Same as above only by phone. Please enter an estimate of the percentage of cases, on average, that attorneys should perform the above task to provide reasonably effective representation. Percent of cases: (Optional) Please provide an explanation of your percentage estimate. 28