The Pennsylvania State University. The Graduate School. College of the Liberal Arts PLEA BARGAINING IN PENNSYLVANIA: EFFECTS OF CASE, COURT, AND

Transcription

1 The Pennsylvania State University The Graduate School College of the Liberal Arts PLEA BARGAINING IN PENNSYLVANIA: EFFECTS OF CASE, COURT, AND COUNTY FACTORS A Thesis in Sociology by Leigh A. Tinik 2010 Leigh A. Tinik Submitted in Partial Fulfillment of the Requirements for the Degree of Master of Arts May 2010

2 The thesis of Leigh A. Tinik was reviewed and approved* by the following: R. Barry Ruback Professor of Crime, Law and Justice and Sociology Thesis Advisor John Kramer Professor of Crime, Law and Justice and Sociology Lori Burrington Assistant Professor of Crime, Law and Justice Cynthia Kempinen Deputy Director of The Pennsylvania Commission on Sentencing John McCarthy Professor of Sociology Head of the Department of Sociology and Crime, Law and Justice *Signatures are on file in the Graduate School

3 ABSTRACT iii This study is a quantitative examination of plea bargaining in approximately 8,000 burglary and 5,000 robbery cases in Pennsylvania during the years 2006 and Plea bargaining was measured as the reduction in the severity and the number of charges from initial charging to conviction. This analysis was possible because of initial charge information from the Administrative Office of Pennsylvania Courts and conviction information from the Pennsylvania Commission on Sentencing. The reduction in the severity and the number of charges were analyzed simultaneously using multinomial logistic regression in HLM. A multilevel analysis was appropriate because of the nested nature of the data (cases nested within courts nested within counties). Moreover it permitted an analysis of the effects of social, economic, and political contextual factors, created from the 2000 Census, on plea bargaining. In addition to examining county contextual factors on plea bargaining, this research examined the effect of county- and court-level indicators of procedural and distributive justice on plea bargaining. Procedural justice, fairness of procedures, was measured using the severity of the court (including the average number of charges, the average number of convictions, the average severity of convictions, and the average prior criminal history), and the day-to-day operations of the court (including the percentage of cases withdrawn, the percentage of guilty pleas, the percentage of jury trials, and the percentage of bench trials). Distributive justice, fairness of the outcome, was measured using departures above the guidelines, departures below the guidelines, and the going rate, or average sentence of a court. Results indicated significant differences with respect to the reduction in the severity and the number of charges for burglary and robbery cases. A majority of burglary and robbery cases received some type of charge reduction. For burglary cases, 63% of the sample received a charge reduction: 57% received a reduction in the number of charges and 6% received a reduction in the

4 severity of charges. For robbery cases, 61% of the sample received a charge reduction: 42% iv received a reduction in the number of charges and 19% received a reduction in the severity of charges. Consistent with most sentencing research, the strongest predictors of charge reductions were two legal variables: offense severity and prior criminal history. But also consistent with much sentencing research, extra-legal variables including gender, race, and age had some effect on the likelihood of a reduction in charges. For burglary only, males were less likely than females to receive a reduction in the severity and the number of charges, and Blacks were less likely than Whites to receive a reduction in the severity of charges. Age had a significant effect for robbery, but not for burglary. Younger offenders were less likely to receive a charge reduction than older offenders. Although there were significant county-level contextual variables, these variables did not explain a significant portion of the variance across counties in the reduction of charges for either burglary or robbery cases. However, county- and court-level measures of procedural justice did explain a significant proportion of the variance across counties in the reduction of charges for burglary. Specifically, measures of procedural justice for burglary were better at explaining this variance across counties than were measures of distributive justice. These results suggest that efforts to decrease disparity in distributive justice by limiting judicial discretion through sentencing guidelines has been relatively successful, but there is still disparity in procedural justice.

5 Table of Contents v LIST OF FIGURES... viii LIST OF TABLES... ix ACKNOWLEDGEMENTS... xi Chapter Introduction... 1 History of Plea Bargaining... 3 Criticism and Support of Plea Bargaining... 6 Supreme Court Cases and the Constitutionality of Pleading Guilty... 7 Actors in the Plea Bargaining Process Chapter Literature Review on Plea Bargaining Offender characteristics Case characteristics Contextual characteristics Chapter Literature Review on Justice Distributive Justice Procedural Justice Descriptive Model Chapter Description of the Current Study Chapter Hypotheses Individual-level Hypotheses Court-level Hypotheses County-level Hypotheses Chapter 6 Methods... 51

6 Overview of the Data vi Chapter 7 Analyses and Results Overview of the Analyses Descriptive Analyses Bivariate Analyses Multivariate Analyses Two-level HLM Analyses for Burglary Two-level HLM Analyses for Robbery Comparison of Two-level Multivariate Analyses for Burglary and Robbery Three-level HLM Analyses Three-level HLM Analyses Three-level HLM Analyses for Burglary Three-level HLM Analyses for Robbery Multivariate Analyses of the Reduction in Sentence Chapter 8 Discussion Appendix A Sentencing Guideline Form Appendix B Sentencing Guideline Matrix Appendix C County-level Contextual Descriptives [N=67] Appendix D County-level Justice Descriptives for Burglary [N=67] Appendix E County-level Justice Descriptives for Robbery [N=67] Appendix F Two-level Hierarchical Multinomial Logistic Models Predicting the Reduction in the Severity and Number of Charges for Burglary: Comparing Model Results for All Counties and Counties Excluding Philadelphia and Allegheny Counties Appendix G Two-level Hierarchical Multinomial Logistic Models Predicting the Reduction in the Severity and Number of Charges for Robbery: Comparing Model Results for All Counties and Counties Excluding Philadelphia and Allegheny Counties Appendix H Two-level Hierarchical Multinomial Logistic Models Predicting the Reduction in the Severity and Number of Charges for Burglary: Comparing Model Results using MeanMinimum Sentence and Median Minimum Sentence Appendix I Two-level Hierarchical Multinomial Logistic Models Predicting the Reduction in the Severity and Number of Charges for Robbery: Comparing Model Results using MeanMinimum Sentence and Median Minimum Sentence Appendix J Court-level Justice Descriptives for Burglary [N=136]

7 vii Appendix K Court-level Justice Descriptives for Robbery [N=69] Appendix L Two-level Hierarchical Regression Models Predicting the Reduction in Sentence for Burglary: Comparing Model Results for All Counties and Counties Excluding Philadelphia and Allegheny Counties Appendix M Two-level Hierarchical Regression Models Predicting the Reduction in Sentence for Robbery: Comparing Model Results for All Counties and Counties Excluding Philadelphia and Allegheny Counties

8 LIST OF FIGURES viii Figure 1: Descriptive Model Predicting the Effects of County Contextual Factors and Court Justice Measures on Plea Bargaining Figure 2: Results of the Hypothesis Test

9 LIST OF TABLES ix Table 1: Unmatched Descriptives for Burglary [N=3,345] and Robbery [N=2,244] Table 2: Offender and Case Descriptives for Burglary [N=8055] and Robbery [N=4861] Table 3: Bivariate Results for Burglary [N=8055] Table 4: Bivariate Results for Robbery [N=4861] Table 5: County-level Contextual Descriptives [N=67] Table 6: County-level Justice Descriptives [N=67] Table 7: Correlations for Burglary Among County-Level Contextual Factors and County-level Justice Factors Table 8: Correlations for Robbery Among County-level Contectual Factors and Countylevel Justice Factors Table 9: Estimation of Level 1 Random Effects for Burglary Table 10: Two-level Hierarchical Multinomial Logistic Models Predicting a Reduction in Number and a Reduction in Severity of Charges for Burglary Table 11: Estimation of Level 1 Random Effects for Robbery Table 12: Two-level Hierarchical Multinomial Logistic Models Predicting a Reduction in Number and a Reduction in Severity of Charges for Robbery Table 13: Descriptive Information About Judges in Burglary Cases [N=136] and Robbery Cases [N=69] Table 14: Court-level Justice Descriptives for Burglary [N=136] and Robbery [N=69] Table 15: Correlations for Burglary Among Court-Level Characteristics and Court-level Justice Factors Table 16: Correlations for Robbery Among Court-Level Characteristics and Court-level Justice Factors Table 17: Three-level Hierarchical Multinomial Logistic Models Predicting a Reduction in Number and a Reduction in Severity of Charges for Burglary Table 18: Three-level Hierarchical Multinomial Logistic Models Predicting a Reduction in Number and a Reduction in Severity of Charges for Robbery

10 Table 19: Two-level Hierarchical Regression Models Predicting the Reduction in Sentence Between the Most Serious Charge and the Most Serious Conviction for Burglary Table 20: Two-level Hierarchical Regression Models Predicting the Reduction in Sentence Between the Most Serious Charge and the Most Serious Conviction for Robbery Table 21: Differences in Offender and Case Characteristics by Type of Disposition for Burglary and Robbery Cases x

11 ACKNOWLEDGEMENTS xi First, I would like to thank my advisor, Barry Ruback, for giving so generously of his time and his unending patience. Second, I would like to thank the other members of my committee: John Kramer, for his wealth of knowledge and invaluable comments; Lori Burrington, for her helpful tips on HLM, and, especially, Cynthia Kempinen, for her unending encouragement and support while pursuing my graduate studies.

12 1 Chapter 1 Introduction Plea bargaining is the process by which defendants waive their right to go to trial in exchange for a reduction in charge or sentence (Acker & Brody, 2004; Alschuler, 1979; 1979; O'Hear, 2007). Guilty pleas account for over 90% of all criminal convictions (Acker & Brody, 2004). Property (96%) and drug offenses (96%) have higher rates of guilty pleas than violent offenses (91%). One reason plea bargaining is so prevalent is that all parties benefit from striking a deal. The prosecutor is guaranteed a conviction, the defendant receives a reduction in sentence, and the court is spared resources because no trial is needed. According to Chief Justice Burger, The disposition of criminal charges by agreement between the prosecutor and the accused, sometimes loosely called plea bargaining, is an essential component of the administration of justice. Properly administered, it is to be encouraged. If every criminal charge were subjected to a full-scale trial, the States and the Federal Government would need to multiply by many times the number of judges and court facilities (Santobello v. New York, 1971). Most research on plea bargaining was conducted in the late 1970 s and early 1980 s and is limited in several ways. First, most research on plea bargaining is qualitative rather than quantitative, and the few quantitative studies use relatively small samples. Second, most research on plea bargaining has focused on the effects of particular offender and offense characteristics rather than the effects of contextual factors on plea bargaining. Third, rarely do studies differentiate between the different types of plea bargaining, and no studies have examined the

13 different types of plea bargaining simultaneously. Finally, only rarely has plea bargaining been 2 put into a theoretical framework. The purpose of this study was to further the understanding of plea bargaining. I used hierarchical modeling to examine how contextual factors influence plea bargaining. In addition to examining county contextual factors on plea bargaining, this research examined county-and court-level indicators of procedural justice on plea bargaining practices, an important issue because research has shown that procedures have as much effect as outcome on defendant satisfaction.

14 History of Plea Bargaining 3 Most people believe that plea bargaining has always been a significant component of the criminal justice system. In fact, however, plea bargaining attitudes and practices have significantly changed over time because of such factors as urbanization, the rise of professional police and prosecutors, and the expansion of criminal law. These factors led to an increase in court caseloads, the need to dispose of court cases quickly and efficiently, and ultimately the practice of plea bargaining. Before 1800, guilty pleas were essentially unknown. Guilty pleas were allowed, but were discouraged by the courts, and often courts would persuade defendants to withdraw their guilty plea. The earliest recorded court case regarding a guilty plea conviction was Commonwealth v. Battis in The court initially refused to accept the defendant s plea of guilty and accepted the guilty plea only after allowing the defendant a reasonable amount of time to reconsider, and after the defendant was examined for signs of insanity and coercion. The courts were reluctant to accept guilty pleas for three reasons. First, they feared coercion of the innocent either through fear of punishment or promise of leniency. Second, because the punishment for felony offenders was severe, including, for many felonies, death, judges were likely to refuse guilty pleas. Third, most defendants lacked defense counsel. Although the Bill of Rights was ratified in 1791, the Supreme Court did not affirm the right to counsel in felony cases until 172 years later. 2 For all three reasons, it was common for the judge to advise the defendant to go to trial. By the early 1800 s, guilty pleas were still relatively uncommon. In 1839, only 25% of felony defendants in New York pled guilty, and in Manhattan and Brooklyn this percentage was U.S. 355 (1963) 1 Commonwealth v. Battis, 1 Mass. 95 (1804) 2 The Supreme Court affirmed the right to counsel in felony cases in Gideon v. Wainwright, 372

15 4 as low as 15% (Moley, 1928). Prior to the 19 th century there was no administrative need for plea bargaining because trials were so efficient. The trial process originated in the criminal court in London (the Old Bailey), which tried over a dozen cases a day (Langbein, 1979). There were three reasons for this efficiency. First, trials were not adversarial in nature. Trials were conducted by judges, and lawyers were infrequently involved in the prosecution or defense. Second, there were no formal rules regarding jury selection. Juries were selected quickly without challenges, and the same jury would hear multiple unrelated cases before being dismissed. Third, there were no formal laws of evidence, such as cross examination of witnesses, that would slow down the trial process. The number of guilty pleas increased by the mid 1800 s. In New York, the percentage of guilty pleas steadily increased by 1850 to 50% (Moley, 1928). This jump in plea bargaining rates resulted from an increase in criminal court cases, brought about by higher crime rates following urbanization, the rise of professional police and prosecutors, and the rapidly expanding number and severity of crimes. Beginning in 1830 urban areas grew much more rapidly than rural areas. With increased urbanization came administrative services such as the creation of police departments. The first police department was established in 1845 in New York and then shortly thereafter in Boston and Philadelphia. With police patrolling the streets, there was an increase in the number of arrests, and soon there was a need to hire full-time prosecutors. The professionalization of the police also led to a rise in plea bargaining rates. Prior to professionalization, police lacked formal education and police training. With technological and scientific advancements in policing, trials became less necessary, due to the mounting physical evidence that essentially proved guilt (Friedman, 1979). Beginning in the twentieth century, a third factor multiplying the number of court cases, was the rapid expansion of criminal law (Alschuler, 1979). Aside from a surge in number, these

16 5 laws were different in nature than previous laws, in that they attempted to control social behavior (Mather, 1979a). For example, the increase in the rate of plea bargaining coincided with the implementation of liquor laws during Prohibition in the 1920 s. Prosecutors pled these cases because juries would frequently refuse to convict defendants guilty of liquor violations. A final factor that led to the rise in plea bargaining was an increase in the severity of sanctions for violating the criminal law (Haller, 1979). Laws requiring incarceration for all felony convictions, led to increasing prison populations. Because of the resulting prison overcrowding, it was often not possible to imprison all felony offenders. Thus, there was some discretion regarding who would go to prison, discretion that resulted in prosecutorial plea bargaining. Compounding the effects of the increased number of court cases was the decreased efficiency of the trial. The rise of adversarial justice and due process severely impaired the courts ability to function. No longer was it possible to try multiple cases in a day, let alone a single case in a day. Defendants were provided counsel, who urged them not to testify because of self-incrimination. Jury selection processes became more complex, and rules of evidence were established to protect defendants from abuse. Together, these changes in the law produced a backlog of court cases. One way to reduce the backlog was to plea bargain cases, rather than taking them to trial. By 1926, guilty pleas accounted for 90% of all convictions (Moley, 1928), a rate that has remained relatively constant since. Today, across jurisdictions, the percentage of guilty pleas is estimated to be 95% (Statistics, 2004). For most states the percentage of guilty pleas varies by less than 10 percentage points ranging from 88-98% (Jones, 1979). States with the highest rates of guilty pleas (above 97%) include Vermont, South Dakota, and Connecticut.

17 Criticism and Support of Plea Bargaining 6 Although plea bargaining in the United States has existed for more than 200 years, critics make two arguments against it. First, critics of plea bargaining fear innocent defendants plead guilty because of coercion or offers of leniency. They claim that if defendants refuse to plead guilty, they will be punished more severely for going to trial. Second, critics suggest that plea bargaining reduces the legitimacy of the criminal justice system because it violates procedural and distributive justice (Brunk, 1979). Procedural justice is the fairness of the process by which decisions are made, and distributive justice is the fairness of the outcome. Critics argue that plea bargaining threatens procedural justice by forcing defendants to forfeit their Fifth and Sixth Amendment rights, including the right against self-incrimination, the right to a jury trial, the right to counsel, and the right to cross-examine witnesses. They also claim that plea bargaining violates distributive justice because the outcome is unjust. Justice is achieved when people receive the exact punishment they deserve, no more or less (Kipnis, 1979). By this definition of justice, plea bargaining is inherently unjust, in that defendants who plead guilty in return for a reduction in sentence receive less punishment than they deserve. Some critics have called for the abolition of plea bargaining because they believe it violates constitutional rights. The National Advisory Commission on Criminal Justice Standards and Goals (NAC), created to increase efficiency of the criminal justice system, called for the total abolition of plea bargaining. In 1975, following this proposal, Alaska s Attorney General instituted a ban on plea bargaining. However, this and other attempts to ban plea bargaining have failed. Evaluations of Alaska s ban found that rather than prosecuting cases as the law required, prosecutors were rejecting cases at initial screening (McDonald, 1979). Contrary to critics of plea bargaining, supporters argue that the fear of convicting an innocent person is not real (Casper, 1979). According to their studies, defendants in 90% of

18 7 cases are factually guilty (i.e. they committed the crime), and of these a sizable percentage are legally guilty, meaning that there is enough evidence to secure a conviction at trial (Heumann, 1977). Moreover, for four reasons supporters believe plea bargaining enhances legitimacy in the criminal justice system. First, plea bargaining increases procedural justice by guaranteeing a timely disposition. Without plea bargaining the amount of time spent in pre-trial detention would drastically increase, thus denying defendants their Sixth Amendment right to a speedy trial. Second, plea bargaining increases distributive justice by mitigating excessively harsh penalties and tailoring the sentence to the needs of the offender (Casper, 1979). Third, the lenient sentences from plea bargained outcomes act as an effective deterrent, in that in these cases conviction is assured (Church, 1979). Finally, supporters believe that any problems with plea bargaining can be overcome by establishing institutional safeguards, such as a more visible bargaining process. Supreme Court Cases and the Constitutionality of Pleading Guilty There have been five major Supreme Court cases upholding the constitutionality of plea bargaining and outlining what constitutes a valid guilty plea. Taken together, these decisions make plea bargaining an appropriate and legitimate means of handling criminal cases. The first Supreme Court case pertaining to guilty pleas, Hallinger v. Davis (1892) 3, considered the constitutionality of guilty pleas based on a claim that plea bargaining violated the Fourteenth Amendment, which asserts that no state shall deprive any person of life, liberty or property without due process of law (p. 314). Hallinger, the defendant, pleaded guilty to murder, and was sentenced to death. Prior to accepting the defendant s guilty plea, the trial court twice adjourned for several days and assigned the defendant counsel. The Supreme Court ruled 3 Hallinger v. Davis, 146 U.S. 314, 324, 1892

19 8 that the defendant was deprived of no right or privilege within the protection of the Fourteenth Amendment (p. 324). Given the constitutionality of plea bargaining, the Supreme Court, in a series of four cases, outlined the terms necessary for a valid guilty plea. In Boykin v. Alabama (1969) 4, the Court stated that a guilty plea is more than a mere confession, it is a conviction. And because of the severe consequences of a conviction, the Court held that a guilty plea must be entered voluntarily and intelligently. If a guilty plea is not entered voluntarily and intelligently, it is in violation of due process and therefore void. After Boykin, two cases expanded on the meaning of the term voluntary. In Brady v. United States (1970) 5, the Court held that a defendant can voluntarily enter a guilty plea when faced with a possible death sentence, and in North Carolina v. Alford (1970) 6, the Court held that a defendant can voluntarily enter a guilty plea without an admission of guilt. In the final major case on plea bargaining, Santobello v. New York (1971) 7, the Supreme Court held prosecutors (and defendants) accountable for their promises made while plea bargaining. In Santobello, the prosecutor promised to make no sentence recommendation to the judge in return for a plea of guilty from the defendant. However, at sentencing a different prosecutor recommended the maximum sentence to the judge. The Supreme Court held that a guilty plea entered voluntarily and intelligently is like a contract, and if either prosecutors or defendants do not fulfill their obligations, the plea is null and void. In Santobello, the Court confirmed not only that the practice of plea bargaining is appropriate and legitimate, but also desirable: Disposition of charges after plea discussions is not only an essential part of the process but a highly desirable part for many reasons (p. 261). The 4 Boykin v. Alabama, 395 U.S. 238, 89 S. Ct. 1709, 23 L. Ed. 2d 274 (1969) 5 Brady v. United States, 397 U.S. 742 (1970) 6 North Carolina v. Alford, 400 U.S. 25, 91 S. Ct. 160, 27 L. Ed. 2d. 162 (1970) 7 Santobello v. New York, 404 U.S. 257, 92 S. Ct. 495, 30 L. Ed. 2d 427 (1971)

20 9 Court outlined four benefits of plea bargaining. First, plea bargaining is efficient and leads to the swift conviction of many cases. Second, plea bargaining avoids lengthy periods of pre-trial confinement for those defendants denied bail. Third, plea bargaining protects the public from dangerous defendants granted bail. Fourth, plea bargaining quickens the rehabilitation of the defendant. In addition to the Supreme Court ruling, in federal courts the practice of plea bargaining is governed by Rule 11 of the Federal Rules of Criminal Procedure. Rule 11 outlines plea agreement procedures that include advising and questioning defendants to determine whether they understand their waived rights and the full consequence of the plea, determining the factual basis for the plea, and ensuring that the plea is voluntary (Figueira-McDonough, 1985; Hagan & Bernstein, 1979). States courts also have rules of criminal procedure regulating the practice of plea bargaining. In Pennsylvania, Rule 590, outlines plea agreement procedures similar to those in Federal Rule 11. There are six questions that a judge in Pennsylvania must ask before accepting a plea: (1) Does the defendant understand the nature of the charges to which he or she is pleading guilty or nolo contendere? (2) Is there factual basis for the plea? (3) Does the defendant understand that he or she has the right to a trial by jury? (4) Does the defendant understand that he or she is presumed innocent until found guilty? (5) Is the defendant aware of the permissible range of sentences and/or fines for the offenses charged? (6) Is the defendant aware that the judge is not bound by the terms of any plea agreement tendered unless the judge accepts such agreement? Only after these six questions are asked and answered can the judge determine the validity of the defendant s plea.

21 Actors in the Plea Bargaining Process 10 There are four actors associated with plea bargaining; the prosecutor, the judge, the defense attorney, and the defendant. These actors have specific roles and different motives for entering into the plea bargaining process. Prosecutors The prosecutor is the most important and powerful actor in the plea bargaining process. Even more, Supreme Court Justice Robert Jackson claimed, The prosecutor has more control over life, liberty and reputation than any other person in America. His discretion is tremendous (Jackson, 1940). Prosecutors decide what charges to file and whether or not to dismiss or reduce those charges, and they make these decisions with relatively unfettered discretion. In the past 20 years, there has been an increase in prosecutorial discretion because of the restrictions on judicial discretion resulting from sentencing guidelines and mandatory minimum sentencing statutes. This shift in discretion from judges to prosecutors, known as the hydraulic displacement of discretion, maintains flexibility in sentencing (Miethe, 1987). Some critics have argued that prosecutors have too much power in the plea bargaining process, because the required punishments are so severe. Thus, when prosecutorial lenience is the only reliable means to avoid a draconian sentence, the prosecutor can effectively dictate the terms of the deal (O'Hear, 2007). Historically, in Pennsylvania the police, not prosecutors, have been responsible for filing the initial charges. Because of their lack of legal knowledge, the police often overcharged defendants. This overcharging gave the false appearance of increased plea bargaining, in that the dismissal or reduction of charges by the prosecutor reflected how the case would have initially been handled had the prosecutor controlled the entrance into the system (Katz, 1979).

22 In the early 1980 s the Pennsylvania Rules of Criminal Procedure were changed Today, prosecutors have the opportunity to review cases before they enter the system, but the extent of this review varies across jurisdiction. In some urban jurisdictions the decision to file charges is generally made quickly based on limited information contained on the outside of the case file, including the arrest offense and the defendant s prior criminal history (Hephner, 2002). After prosecutors determine what charges to file, they then consider whether or not to dismiss or reduce those charges through plea bargaining. There are five reasons prosecutors bargain with defendants. First, prosecutors plea bargain to dispose of cases. Criminal courts are reactive, responding as efficiently as possible to a large number of cases brought to them by police (Nagel & Bernstein, 1978). Because the average prosecutor can handle only about 200 trials a year, prosecutors must assess their time and resources and decide which cases are important enough to take to trial (Jones, 1979). Second, prosecutors plea bargain in weak cases (Albonetti, 1987). Prosecutors are more likely to bargain in cases lacking physical evidence or eyewitness testimony placing the defendant at the scene of the crime, because without such evidence it is difficult to prove guilt beyond a reasonable doubt. Rather than risk an acquittal at trial, prosecutors are willing to plea bargain with the defendant to ensure the defendant is convicted and receives some type of punishment. Third, prosecutors plea bargain less serious cases committed by less serious offenders (Albonetti, 1987; Ryan & Alfini, 1979). Prosecutors are more likely to plea bargain misdemeanor rather than felony cases due to the decreased seriousness of the crime. Prosecutors are also more likely to offer plea bargains to first time offenders with a less serious prior criminal record, because of their perception that such offenders are more likely to be successfully rehabilitated Pa.R.Crim.P. 507.

23 12 Fourth, prosecutors offer to dismiss or reduce charges in return for information from the defendant regarding more serious offenders. This process is commonly referred to as turning state s evidence, but is formally known as providing substantial assistance to the prosecutor (McDonald, 1979). For example, prosecutors will use information provided by a drug user to make a case against a drug dealer. This type of plea bargaining occurs less frequently at the state level than the federal level. Lastly, prosecutors plea bargain to secure a conviction. Prosecutors are elected officials, and the high conviction rate that follows from plea bargaining gives the appearance that they are tough on crime. Judges In most jurisdictions, judges play a minimal role in the plea bargaining process. The rationale for restricting judicial participation is that judges are supposed to be unbiased and their participation in the plea bargaining process would threaten the voluntariness of the plea. The American Bar Association Standards minimize judicial participation in plea bargaining, although there are no specific guidelines as to what constitutes participation (Houlden, 1980; Lee, 2005; Ryan & Alfini, 1979). 9 The ABA Standards are not legally binding, but they have been adopted by 48 of the 50 states and have a substantial influence on judges and prosecutors behavior. Depending on the jurisdiction, the judicial role in plea bargaining can significantly vary from direct involvement in the plea bargaining process to no involvement. Direct involvement, which is the minority position, involves a judge verbally or in writing committing to a sentence prior to sentencing if the defendant pleads guilty. Some states, like Florida, encourage judicial 9 ABA Standards for Criminal Justice (c) (3rd Ed. 1999)

24 13 participation, while other states, including Pennsylvania, strictly prohibit judicial participation. 10 In most jurisdictions, the judge is informally involved in the plea bargaining process through unannounced, but known, sentencing discounts for guilty pleas. Out of respect for judges and their ongoing relationships with them, prosecutors and defense attorneys will generally not recommend a plea bargain outside of the judges going rate. Nationally, two-thirds of trial judges do not attend plea discussions and restrict their role to accepting or rejecting plea bargains (Ryan & Alfini, 1979). Ultimately, the judge has the final decision about whether to accept or reject the plea bargain. Rejection of a plea bargain is rare (less than 2% of cases) and usually involves a question about the voluntariness of the plea rather than the sentence recommendation (Ryan & Alfini, 1979; Turner, 2006). Judges accept plea bargains for two reasons. First, plea bargaining is an efficient way to clear their overcrowded dockets. Judges understand the need for a steady rate of guilty pleas to ensure rapid justice for defendants. Without plea bargaining, the court would become backlogged and defendants would spend even more time in pre-trial detention awaiting trial. Second, judges believe trials should be reserved for cases in which the outcome is uncertain. If a case is clearcut, it should be disposed of by a guilty plea, rather than by trial (Rossett & Cressey, 1976). Defense Attorneys Defense attorneys have little discretion compared to prosecutors, who essentially dictate the terms of the plea bargain. Defense attorneys merely advise their defendants whether to reject 10 PA Commonwealth v Evans, 252 A.2d 689, 691 (Pa. 1969) states, we feel compelled to forbid any participation by the trial judge in the plea bargaining prior to the offering of a guilty plea. Davis v. State, 308 So. 2d 27 (Fla. 1975) states, we refrain from condemning the practice per se since we are confident that the trial judges of this state will take all necessary precautions to assure that defendants rights are protected by appropriate safeguards. This sentiment was confirmed in Warner v. State, 721 So. 2d 767 (Fla. 1998).

25 14 or accept the prosecutor s plea bargain based on their knowledge and experience working with prosecutors and judges in the courtroom. Defense attorneys enter into the plea bargaining process for the same reasons prosecutors do. First, like prosecutors, defense attorneys plea bargain to dispose of cases. They too, have limited time and resources and must decide which cases are important enough to take to trial. Plea bargaining requires minimal time and resources compared to taking a case to trial. Second, defense attorneys plea bargain when it is in the best interest of their clients. A majority of defendants admit their guilt, but are unwilling to accept plea bargains because they believe them to be unfair. This is especially relevant today with the proliferation of sentencing guidelines and mandatory minimum statutes, under which a repeat offender can receive twentyfive years to life for any subsequent felony conviction. Defense attorneys must advise their client that going to trial poses a greater risk if convicted than accepting the plea bargain. Based on their knowledge and experience, defense attorneys know whether or not the plea bargain is a good deal. However, when defense attorneys believe the plea bargain is not a good deal, they can advise their clients to go to trial. If prosecutors are unwilling to renegotiate the plea bargain, defense attorneys have the ability to overload the court by taking all their cases to trial (Ditzen, 1997). Third, defense attorneys plea bargain for economic reasons. Most private defense attorneys are paid per defendant. The more defendants who plead guilty, the more defendants the defense attorney can represent, and the more money the defense attorney can make (Mather, 1979b). It is important to distinguish between public and private attorneys, not only with respect to economic incentives for pleading guilty, but also with respect to their relationships with the other actors in the courtroom. Public defense attorneys establish strong personal and professional relationships with prosecutors and judges because of daily interactions in the courtroom. In some jurisdictions, there are private defense attorneys who are regulars in the courtroom. However,

26 15 in most jurisdictions private defense attorneys do not have personal and professional relationships with the other actors in the plea bargaining process (Mather, 1979b). Defendants There are five reasons defendants choose to plead guilty. First, and most important, defendants plead guilty to avoid a more severe sentence. Although defendants ultimately decide whether to plead guilty or go to trial, this is not much of a choice (O'Hear, 2007). Most defendants are unwilling to go to trial because they fear they are likely to receive a longer sentence. The trial penalty is a well-known incentive to plea bargain (LaFree, 1985; Ulmer & Bradley, 2006), and the practice of punishing defendants who go to trial more severely than those who plea bargain is constitutional (Santobello v. New York, 1971). Second, defendants plead guilty to have control over the outcome and avoid the uncertain outcome of a trial (Mather, 1979b). Trials are unpredictable, and many defendants are unwilling to give up their control over the outcome to a jury. Third, defendants plead guilty to avoid the embarrassment and emotional stress of a lengthy trial, particularly defendants with strong ties to the community. Defendants who live and work in the community will quickly plead guilty to resolve the case with as little attention as possible. Fourth, defendants plead guilty for economic reasons. Plea bargaining saves the defendant expensive attorney s fees that would arise from going to trial. Moreover, if defendants cannot afford to post bail, plea bargaining can shorten the length of time they spend in pre-trial confinement. And lastly, defendants plead guilty to a lesser offense in order to avoid a more serious charge on their record. For example, a misdemeanor carries drastically different social costs than a felony conviction. A felony conviction can prohibit defendants from holding certain occupations, and in some states, from voting. Also pleading guilty to a reduced number of

27 charges can make a difference in the defendant s prior criminal history. This reduction in 16 severity or number of charges is significant if the defendant is ever convicted of another offense, because sentencing guidelines are based in part on prior criminal history. Courtroom Communities Although prosecutors, judges, and defense attorneys act independently, their decisions are also influenced by one another. The most common explanation for plea bargaining is attributed to interpersonal relationships between the prosecutor, defense attorney, and judge, commonly referred to as the courtroom workgroup. These actors work together on a daily basis and eventually develop a shared acceptance of what is just when dealing with specific kinds of offenses and offenders (Rossett & Cressey, 1976). Through years of working together in the same courtroom, prosecutors, judges, and defense attorneys have established going rates or agreed upon sentencing ranges for certain crimes. The plea bargaining practice of lower courts has been compared to pricing in supermarkets (Feeley, 1979). The prices in supermarkets are set in advance and people rarely bargain over the price of an item. Occasionally, there is a change in the price of an item due to a trial, which challenges the going rate established by courtroom communities. Courtroom communities place an emphasis on conformity, or not rocking the boat, and violating established norms set by the courtroom community can have both professional and personal consequences, such as being denied promotions and losing golf partners (Nardulli et al., 1988). The cohesiveness of a courtroom community varies by county. Contextual factors, such as county size, and political and demographic characteristics, influence the cohesiveness of the courtroom community (Nardulli et al., 1988). Courtroom communities are based on consensus rather than conflict. Most view the courtroom community as a legitimate and effective method to handle cases, but others view this

28 17 relationship as a charade, in that the plea negotiation is made to appear too good to pass up, when in reality it is the established going rate of the courtroom community (Feeley, 1979).

29 18 Chapter 2 Literature Review on Plea Bargaining In order to understand how plea bargaining practices affect procedural justice in Pennsylvania, it is important to understand the factors that affect plea bargaining in general. This review focuses on how offender, case, and contextual factors affect plea bargaining practices. Research on plea bargaining closely resembles sentencing research, not surprisingly, because more than 90% of all sentences in the United States are the result of guilty pleas. Thus, the legally relevant variables predicting sentencing outcomes (i.e., offense severity, prior criminal history), are also used to predict the prosecutor s decision to dismiss or reduce charges. Moreover, the same extra-legal variables (i.e., gender, race, age) used to investigate unwarranted judicial disparity in sentencing are included in plea bargaining research to determine unwarranted prosecutorial disparity in plea bargaining. Offender characteristics Most sentencing studies assume that, for the same conviction offense, a finding of no difference in sentence severity across gender, race, and age is proof of nondiscrimination in the criminal justice system. However, since the plea bargaining process determines the conviction offense, the conviction offense is an incomplete method of determining discrimination (Figueira- McDonough, 1985). Discrimination can easily occur at earlier stages in the criminal justice system: at arrest, the filing of charges, the initial court appearance where charges are dismissed or

30 19 nolle prossed, or during plea negotiations. Therefore, it is imperative that these earlier stages in the process also be examined for discrimination. Gender There has been much research on the effects of gender at sentencing, but less research on the impact of gender at decision points prior to sentencing, such as plea bargaining. Sentencing research typically finds that judges treat female offenders more leniently than male offenders (Steffensmeier et al., 1998). However, contrary to sentencing research, plea bargaining research finds no significant gender difference in the likelihood of the prosecutor s decision to reduce the severity (Albonetti, 1992) or the number (Ball, 2006) of charges. But, it is important to note that these findings cannot lead to the conclusion that prosecutorial decision making is free of unwarranted gender disparity, because it is possible that prosecutorial disparity exists prior to plea bargaining at arrest or the filing of charges. Although plea bargaining research finds no significant gender difference in the likelihood of the prosecutor s decision to reduce the severity and the number of charges, there is a significant gender difference in the defendant s decision to plead guilty. Not only are females less likely to go to trial (either a bench trial or jury trial) than males (Johnson, 2003), females are also less likely to bargain and more likely to plead guilty to the original charges than males (Figueira-McDonough, 1985). It is important to note that this finding is mediated by offense severity, in that females are less likely than males to commit serious crimes, and serious crimes are more likely to be bargained (Figueira-McDonough, 1985).

31 Race 20 Sentencing research typically finds that Black offenders receive harsher sentences than White offenders (Steffensmeier et al., 1998). A study using downward and upward departures from sentencing guideline ranges found Blacks and Hispanics were less likely to receive a downward departure across all modes of conviction, and they were more likely to receive an upward departure for negotiated pleas (Johnson, 2003). But, again contrary to sentencing research, plea bargaining research finds no significant racial difference in the likelihood of the prosecutor s decision to reduce the severity (Albonetti, 1992) or the number (Ball, 2006) of charges. However, studies have found racial disparity when it comes to the magnitude of the sentence reduction. Being Black decreases the magnitude of the sentence reduction (Albonetti, 1997). Past research regarding racial differences and plea bargaining has found a significant difference in the likelihood of pleading guilty. Black defendants are less likely than White defendants to plead guilty, even after controlling for offender, offense, and evidence variables (Albonetti, 1990, 1997; Johnson, 2003). The most likely explanation for this difference is that Black defendants do not trust the criminal justice system (Albonetti, 1990). Plea bargains are often implicit, and are always based on trust. If defendants lack trust in the system, they are more likely to opt for a trial, where there are procedural safeguards such as due process and an appeal process. Age The age of the defendant is positively related to the likelihood of a charge reduction; the younger the defendant, the less likely a charge reduction (Albonetti, 1992; Bernstein et al., 1977).

32 21 Youthful offenders are seen as more dangerous than their older counterparts and more likely to reoffend. Although age is a significant predictor of charge reductions, the impact of age is significantly greater for trials than pleas (Johnson, 2003). Case characteristics In sentencing research, the strongest predictors of sentencing outcomes are two legally relevant case characteristics: offense severity and prior criminal history. They are also significant predictors of the prosecutor s decision to reduce charges and the defendant s decision to plead guilty. A third case characteristic that influences the prosecutor s decision to reduce charges and the defendant s decision to plead guilty are evidence variables measuring the strength of the case. Offense severity Studies of sentencing and plea bargaining find offense severity to be the single most influential factor predicting both sentencing and plea bargained outcomes (Meyer & Gray, 1997). For plea bargaining, higher offense severity increases the likelihood of a charge reduction (Bernstein et al., 1977). One explanation for this finding that felony offenders are significantly more likely to get charge reductions is that felony offenses have a greater potential for a charge reduction than misdemeanor offenses (Bernstein et al., 1977). Because of the excessively harsh punishment of repeat felony and violent offenders, offenses with higher severity and potentially lengthy sentences give prosecutors greater flexibility to plea bargain. However, although felony offenders are more likely to get charge reductions, they are less likely than misdemeanor offenders to plead guilty because they face longer incarceration sentences (Albonetti, 1990; Figueira-McDonough, 1985; Jones, 1979).

33 Prior Criminal History 22 Prior criminal history influences prosecutors willingness to negotiate, because it affects their judgments of both the defendant s culpability and the defendant s ability to bargain. Past research supports the hypothesis of a non-linear relationship between prior criminal history and the likelihood of a charge reduction. Defendants with no prior criminal history (Bernstein et al., 1977) and habitual offenders (more than one prior felony conviction) are less likely to receive a charge reduction (Albonetti, 1992). Most likely to receive a charge reduction are defendants with prior arrests but no prior convictions (Bernstein et al., 1977), or defendants with only one prior conviction (Albonetti, 1992). This non-linear relationship suggests that experienced-but not too experienced-defendants fare better in the plea negotiation process because of their prior knowledge of the criminal justice system (Bishop & Frazier, 1984). With respect to prior criminal history and the defendant s decision to plead guilty, studies consistently find felony convictions decrease the probability of pleading guilty (Albonetti, 1990). The severe sentences for habitual offenders make it logical for defendants to take their chances at trial rather than plead guilty. Offense Type Numerous studies investigate the likelihood of charge reductions among different types of offenses. Although type of offense is highly associated with offense severity, type of offense remains a significant predictor of charge reductions net of offense severity, and, in fact, some research finds it to be the most significant predictor of determining a reduction in number of charges (Ball, 2006).